- Overview
- Components
- Build & Installation
- How to use
- Performance
- OPLIST
- Memory Allocation
- Emplace construction
- Errors & compilers
- External Reference
- API Documentation
- Generic methods
- List
- Array
- Deque
- Dictionary
- Tuple
- Variant
- Red/Black Tree
- B+ Tree
- Generic Tree
- Priority queue
- Fixed buffer queue
- Atomic Shared Register
- Shared pointers
- Intrusive Shared Pointers
- Intrusive list
- Concurrent adapter
- Bitset
- String
- Core preprocessing
- Thread
- Worker threads
- Atomic
- Generic algorithms
- Function objects
- Exception handling
- Memory pool
- JSON Serialization
- Binary Serialization
- Generic interface
- Byte String
- Global User Customization
- License
M*LIB (M star lib) is a C library enabling to define and to use generic and
type safe container in C, aka handling generic
containers in pure C language.
The encapsulated objects can have their own constructor, destructor, operators
or can be basic C type like the C type 'int': both are fully supported.
This makes it possible to construct fully
recursive container objects (container-of[...]-container-of-type-T
)
while keeping compile time type checking.
This is an equivalent of the C++ Standard Library, providing vector, deque, forward_list, set, map, multiset, multimap, unordered_set, unordered_map, stack, queue, shared_ptr, string, variant, option to standard ISO C99 / C11. There is not a strict mapping as both the STL and M*LIB have their exclusive containers:
See here for details. M*LIB provides also additional concurrent containers to design properly multi-threaded programs: shared register, communication queue, ...
M*LIB is portable to any systems that support ISO C99. Some optional features need at least ISO C11.
M*LIB is only composed of a set of headers. There is no C file, and as such, the installation is quite simple: you just have to put the header in the search path of your compiler, and it will work. There is no dependency (except some other headers of M*LIB and the LIBC).
One of M*LIB design key is to ensure safety. This is done by multiple means:
- in debug mode, defensive programming is extensively used: the contracts of the function are checked, ensuring that the data are not corrupted. For example, strict Buffer overflow are checked in this mode through bound checking or the intrinsic properties of a Red-Black tree (for example) are verified. Buffer overflow checks can still be kept in release mode if needed.
- as few cast as possible are used within the library (casts are the evil of safety). Still the library can be used with the greatest level of warnings by a C compiler without any aliasing warning.
- the genericity is not done directly by macro (which usually prevent type safety), but indirectly by making them define inline functions with the proper prototypes: this enables the user calls to have proper error and warning checks.
- extensive testing: the library is tested on the main targets using Continuous Integration with a coverage of the test suite of more than 99%. The test suite itself is run through the multiple sanitizers defined by GCC/CLANG (Address, undefined, leak, thread). The test suite also includes a comparison of equivalent behaviors of M*LIB with the C++ STL using random testing or fuzzer testing.
- static analysis: multiple static analyzer (like scan-build or GCC fanalyzer or CodeQL) are run on the generated code, and the results analyzed.
Other key designs are:
- do not rewrite the C library and just wrap around it (for example don't rewrite sort but stable sort),
- do not make users pay the cost of what they don't need.
Due to the unfortunate weak nature of the C language for pointers, type safe means that at least a warning is generated by the compiler in case of wrong type passed as container arguments to the functions.
M*LIB is still quite-efficient: there is no overhead in using this library rather than using direct C low-level access as the compiler is able to fully optimize the library usage and the library is carefully designed. In fact, M*LIB is one of the fastest generic C/C++ library you can find.
M*LIB uses internally the malloc
, realloc
and free
functions to handle
the memory pool. This behavior can be overridden at different level.
Its default policy is to abort the program if there is a memory error.
However, this behavior can also be customized globally.
M*LIB supports also the exception error model by providing its own implementation of the try / catch mechanism.
This mechanism is compatible with RAII programming:
when an exception is thrown, the destructors of the constructed objects are called (See m-try for more details).
M*LIB may use a lot of assertions in its implementation to ensure safety:
it is highly recommended to properly define NDEBUG
for released programs.
M*LIB provides automatically several serialization methods for each containers. You can read or write your full and complex data structure into JSON format in a few lines.
M*LIB is distributed under BSD-2 simplified license.
It is strongly advised not to read the source to know how to use the library as the code is quite complex and uses a lot of tricks but rather read the examples.
In this documentation,
- shall will be used to indicate a user constraint that is mandatory to follow under penalty of undefined behavior.
- should will be used to indicate a recommendation to the user.
All pointers expected by the functions of the library shall expect non-null argument except if indicated.
The following headers define containers that don't require the user structure to be modified:
- m-array.h: header for creating dynamic array of generic type,
- m-list.h: header for creating singly-linked list of generic type,
- m-deque.h: header for creating dynamic double-ended queue of generic type,
- m-dict.h: header for creating unordered associative array (through hashmap) or unordered set of generic type,
- m-rbtree.h: header for creating ordered set (through Red/Black binary sorted tree) of generic type,
- m-bptree.h: header for creating ordered map/set/multimap/multiset (through sorted B+TREE) of generic type,
- m-tree.h: header for creating arbitrary tree of generic type,
- m-tuple.h: header for creating arbitrary tuple of generic types,
- m-variant.h: header for creating arbitrary variant of generic type,
- m-prioqueue.h: header for creating dynamic priority queue of generic type.
The available containers of M*LIB for thread synchronization are in the following headers:
- m-buffer.h: header for creating fixed-size queue (or stack) of generic type (multiple producer / multiple consumer),
- m-snapshot: header for creating 'atomic buffer' (through triple buffer) for sharing synchronously big data (thread safe),
- m-shared-ptr.h: header for creating shared pointer of generic type,
- m-c-mempool.h: WIP header for creating fast concurrent memory allocation.
The following containers are intrusive (You need to modify your structure to add fields needed by the container) and are defined in:
- m-i-list.h: header for creating doubly-linked intrusive list of generic type,
Other headers offering other functionality are:
- m-string.h: header for creating dynamic string of characters (UTF-8 support),
- m-bstring.h: header for creating dynamic string of BYTE,
- m-bitset.h: header for creating dynamic bitset (or "packed array of bool"),
- m-algo.h: header for providing various generic algorithms to the previous containers,
- m-funcobj.h: header for creating function object (used by algorithm generation),
- m-try.h: header for handling errors by throwing exceptions,
- m-mempool.h: header for creating specialized & fast memory allocator,
- m-worker.h: header for providing an easy pool of workers on separated threads to handle work orders (used for parallel tasks),
- m-serial-json.h: header for importing / exporting the containers in JSON format,
- m-serial-bin.h: header for importing / exporting the containers in an adhoc fast binary format,
- m-generic.h: header for using a common interface for all registered types,
- m-genint.h: internal header for generating unique integers in a concurrent context,
- m-core.h: header for meta-programming with the C preprocessor (used by all other headers).
Finally, headers for compatibility with non C11 compilers:
- m-atomic.h: header for ensuring compatibility between C's
stdatomic.h
and C++'s atomic header (provide also its own implementation if nothing is available), - m-thread.h: header for providing a very thin layer across multiple implementation of mutex/threads (C11/PTHREAD/WIN32).
The following headers are obsolete:
- m-shared.h: header for creating shared pointer of generic type,
- m-concurrent.h: header for transforming a container into a concurrent container (thread safe),
- m-i-shared.h: header for creating intrusive shared pointer of generic type (Thread Safe).
Each containers define their iterators (if it is meaningful).
All containers try to expose the same common interface: if the method name is the same, then it does the same thing and is used in the same way. In some rare case, the method is adapted to the container needs.
Each header can be used separately from others: dependency between headers have been kept to the minimum.
M*LIB is only composed of a set of headers, as such there is no build for the library. The library doesn't depend on any other library than the LIBC.
To run the test suite, run:
make check
You can also override the compiler CC or its flags CFLAGS if needed:
make check CC="gcc" CFLAGS="-O3"
To generate the documentation, run:
make doc
To install the headers, run:
make install PREFIX=/my/directory/where/to/install [DESTDIR=...]
Other targets exist. Mainly for development purpose.
To use these data structures, you first include the desired header,
instantiate the definition of the structure and its associated methods
by using a macro _DEF
for the needed container.
Then you use the defined types and functions. Let's see a first simple example
that creates a list of unsigned int:
#include <stdio.h>
#include "m-list.h"
LIST_DEF(list_uint, unsigned int) /* Define struct list_uint_t and its methods */
int main(void) {
list_uint_t list ; /* list_uint_t has been define above */
list_uint_init(list); /* All type needs to be initialized */
list_uint_push_back(list, 42); /* Push 42 in the list */
list_uint_push_back(list, 17); /* Push 17 in the list */
list_uint_it_t it; /* Define an iterator to scan each one */
for(list_uint_it(it, list) /* Start iterator on first element */
; !list_uint_end_p(it) /* Until the end is not reached */
; list_uint_next(it)) { /* Set the iterator to the next element*/
printf("%d\n", /* Get a reference to the underlying */
*list_uint_cref(it)); /* data and print it */
}
list_uint_clear(list); /* Clear all the list (destroying the object list)*/
}
[!NOTE] Do not forget to add -std=c99
(or c11) to your compile command to request a C99 compatible build
This looks like a typical C program except the line with LIST_DEF
that doesn't have any semi-colon at the end. And in fact, except
this line, everything is typical C program and even macro free!
The only macro is in fact LIST_DEF
: this macro expands to the
good type for the list of the defined type and to all the necessary
functions needed to handle such type. It is heavily context dependent
and can generate different code depending on it.
You can use it as many times as needed to defined as many lists as you want.
The first argument of the macro is the name to use, e.g. the prefix that
is added to all generated functions and types.
The second argument of the macro is the type to embed within the container.
It can be any C type.
The third argument of the macro is optional and is the oplist to use.
See below for more information.
You could replace LIST_DEF
by ARRAY_DEF
to change
the kind of container (an array instead of a linked list)
without changing the code below: the generated interface
of a list or of an array is very similar.
Yet the performance remains the same as hand-written code
for both the list variant and the array variant.
This is equivalent to this C++ program using the STL:
#include <iostream>
#include <list>
typedef std::list<unsigned int> list_uint_t;
typedef std::list<unsigned int>::iterator list_uint_it_t;
int main(void) {
list_uint_t list ; /* list_uint_t has been define above */
list.push_back(42); /* Push 42 in the list */
list.push_back(17); /* Push 17 in the list */
for(list_uint_it_t it = list.begin() /* Iterator is first element*/
; it != list.end() /* Until the end is not reached */
; ++it) { /* Set the iterator to the next element*/
std::cout << *it << '\n'; /* Print the underlying data */
}
}
As you can see, this is rather equivalent with the following remarks:
- M*LIB requires an explicit definition of the instance of the list,
- M*LIB code is more verbose in the method name,
- M*LIB needs explicit construction and destruction (as plain old C requests),
- M*LIB doesn't return a value to the underlying data but a pointer to this value:
- this was done for performance (it avoids copying all the data within the stack)
- and for generality reasons (some structure may not allow copying data).
Note: M*LIB defines its own container as an array of a structure of size 1. This has the following advantages:
- you effectively reserve the data whenever you declare a variable,
- you pass automatically the variable per reference for a function call,
- you can not copy the variable by an affectation (you have to use the API instead).
M*LIB offers also the possibility to condense further your code, so that it is more high level:
by using the M_EACH
& M_LET
macros (if you are not afraid of using syntactic macros):
#include <stdio.h>
#include "m-list.h"
LIST_DEF(list_uint, unsigned int) /* Define struct list_uint_t and its methods */
int main(void) {
M_LET(list, LIST_OPLIST(uint)) { /* Define & init list as list_uint_t */
list_uint_push_back(list, 42); /* Push 42 in the list */
list_uint_push_back(list, 17); /* Push 17 in the list */
for M_EACH(item, list, LIST_OPLIST(uint)) {
printf("%d\n", *item); /* Print the item */
}
} /* Clear of list will be done now */
}
Here is another example with a complete type (with proper initialization & clear function) by using the GMP library:
#include <stdio.h>
#include <gmp.h>
#include "m-array.h"
ARRAY_DEF(array_mpz, mpz_t, (INIT(mpz_init), INIT_SET(mpz_init_set), SET(mpz_set), CLEAR(mpz_clear)) )
int main(void) {
array_mpz_t array ; /* array_mpz_t has been define above */
array_mpz_init(array); /* All type needs to be initialized */
mpz_t z; /* Define a mpz_t type */
mpz_init(z); /* Initialize the z variable */
mpz_set_ui (z, 42);
array_mpz_push_back(array, z); /* Push 42 in the array */
mpz_set_ui (z, 17);
array_mpz_push_back(array, z); /* Push 17 in the array */
array_it_mpz_t it; /* Define an iterator to scan each one */
for(array_mpz_it(it, array) /* Start iterator on first element */
; !array_mpz_end_p(it) /* Until the end is not reached */
; array_mpz_next(it)) { /* Set the iterator to the next element*/
gmp_printf("%Zd\n", /* Get a reference to the underlying */
*array_mpz_cref(it)); /* data and print it */
}
mpz_clear(z); /* Clear the z variable */
array_mpz_clear(array); /* Clear all the array */
}
As the mpz_t
type needs proper initialization, copy and destroy functions
we need to tell to the container how to handle such a type.
This is done by giving it the oplist associated to the type.
An oplist is an associative array where the operators are associated to methods.
In the example, we tell to the container to use
the mpz_init
function for the INIT
operator of the type (aka constructor),
the mpz_clear
function for the CLEAR
operator of the type (aka destructor),
the mpz_set
function for the SET
operator of the type (aka copy),
the mpz_init_set
function for the INIT_SET
operator of the type (aka copy constructor).
See OPLIST chapter for more detailed information.
We can also write the same example shorter:
#include <stdio.h>
#include <gmp.h>
#include "m-array.h"
// Register the oplist of a mpz_t.
#define M_OPL_mpz_t() (INIT(mpz_init), INIT_SET(mpz_init_set), \
SET(mpz_set), CLEAR(mpz_clear))
// Define an instance of an array of mpz_t (both type and function)
ARRAY_DEF(array_mpz, mpz_t)
// Register the oplist of the created instance of array of mpz_t
#define M_OPL_array_mpz_t() ARRAY_OPLIST(array_mpz, M_OPL_mpz_t())
int main(void) {
// Let's define `array` as an 'array_mpz_t' & initialize it.
M_LET(array, array_mpz_t)
// Let's define 'z1' and 'z2' to be 'mpz_t' & initialize it
M_LET (z1, z2, mpz_t) {
mpz_set_ui (z1, 42);
array_mpz_push_back(array, z1); /* Push 42 in the array */
mpz_set_ui (z2, 17);
array_mpz_push_back(array, z2); /* Push 17 in the array */
// Let's iterate over all items of the container
for M_EACH(item, array, array_mpz_t) {
gmp_printf("%Zd\n", *item);
}
} // All variables are cleared with the proper method beyond this point.
return 0;
}
Or even shorter when you're comfortable enough with the library:
#include <stdio.h>
#include <gmp.h>
#include "m-array.h"
// Register the oplist of a mpz_t. It is a classic oplist.
#define M_OPL_mpz_t() M_OPEXTEND(M_CLASSIC_OPLIST(mpz), \
INIT_WITH(mpz_init_set_ui), EMPLACE_TYPE(unsigned int))
// Define an instance of an array of mpz_t (both type and function)
ARRAY_DEF(array_mpz, mpz_t)
// Register the oplist of the created instance of array of mpz_t
#define M_OPL_array_mpz_t() ARRAY_OPLIST(array_mpz, M_OPL_mpz_t())
int main(void) {
// Let's define `array` as an 'array_mpz_t' with mpz_t(17) and mpz_t(42)
M_LET((array,(17),(42)), array_mpz_t) {
// Let's iterate over all items of the container
for M_EACH(item, array, array_mpz_t) {
gmp_printf("%Zd\n", *item);
}
} // All variables are cleared with the proper method beyond this point.
return 0;
}
There are two ways a container can known what is the oplist of a type:
- either the oplist is passed explicitly for each definition of container and for the
M_LET
andM_EACH
macros, - or the oplist is registered globally by defining a new macro starting with the prefix
M_OPL_
and finishing with the name of type (don't forget the parenthesis and the suffix _t if needed). The macros performing the definition of container and theM_LET
andM_EACH
will test if such macro is defined. If it is defined, it will be used. Otherwise, default methods are used.
Here we can see that we register the mpz_t
type into the container with
the minimum information of its interface needed, and another one to initialize a mpz_t
from an unsigned integer.
We can also see in this example so the container ARRAY provides also a macro to define the oplist of the array itself. This is true for all containers and this enables to define proper recursive container like in this example which reads from a text file a definition of sections:
#include <stdio.h>
#include "m-array.h"
#include "m-tuple.h"
#include "m-dict.h"
#include "m-string.h"
TUPLE_DEF2(symbol, (offset, long), (value, long))
#define M_OPL_symbol_t() TUPLE_OPLIST(symbol, M_BASIC_OPLIST, M_BASIC_OPLIST)
ARRAY_DEF(array_symbol, symbol_t)
#define M_OPL_array_symbol_t() ARRAY_OPLIST(array_symbol, M_OPL_symbol_t())
DICT_DEF2(sections, string_t, array_symbol_t)
#define M_OPL_sections_t() DICT_OPLIST(sections, STRING_OPLIST, M_OPL_array_symbol_t())
int main(int argc, const char *argv[])
{
if (argc < 2) abort();
FILE *f = fopen(argv[1], "rt");
if (!f) abort();
M_LET(sc, sections_t) {
sections_in_str(sc, f);
array_symbol_t *a = sections_get(sc, STRING_CTE(".text"));
if (a == NULL) {
printf("There is no .text section.");
} else {
printf("Section .text is :");
array_symbol_out_str(stdout, *a);
printf("\n");
}
}
return 0;
}
This example reads the data from a file and outputs the .text section if it finds it on the terminal.
Other examples are available in the example folder.
Internal fields of the structure are subject to change even for small revision of the library.
The final goal of the library is to be able to write code like this in pure C while keeping type safety and compile time name resolution:
M_LET(list, list_uint_t) {
push(list, 42);
push(list, 17);
for each (item, list) {
M_PRINT(*item, "\n");
}
}
See the example and M-GENERIC header for details.
M*LIB performance is compared to the one of GNU C++ STL (v10.2) in the following graphs. Each graph is presented first in linear scale and then in logarithmic scale to better realize the differences. M*LIB is on par with the STL or even faster.
All used bench codes are available in this repository The results for several different libraries are also available in a separate page.
An OPLIST
is a fundamental notion of M*LIB that hasn't be seen in any other library.
In order to know how to operate on a type, M*LIB needs additional information
as the compiler doesn't know how to handle properly any type (contrary to C++).
This is done by giving an operator list (or oplist in short) to any macro that
needs to handle the type. As such, an oplist has only meaning within a macro.
Fundamentally, this is the exposed interface of a type:
that is to say the association of the operators defined by the library to the
effective methods provided by the type, including their call interface.
This association is done only with the C preprocessor.
Therefore, an oplist is an associative array of operators to methods in the following format:
(OPERATOR1(method1), OPERATOR2(method2), ...)
It starts with a parenthesis and ends with a parenthesis. Each association is separated by a comma. Each association is composed of a predefined operator (as defined below) a method (in parentheses), and an optional API interface (see below).
In the given example,
the function method1
is used to handle OPERATOR1
.
The function method2
is used to handle OPERATOR2
, etc.
In case the same operator appears multiple times in the list, the first apparition of the operator has priority, and its associated method will be used. This enables overloading of operators in oplist in case you want to inherit oplists.
The associated method in the oplist is a preprocessor expression that shall not contain a comma as first level.
An oplist has no real form from the C language point of view. It is just an abstraction that disappears after the macro expansion step of the preprocessing. If an oplist remains unprocessed after the C preprocessing, a compiler error will be generated.
When you define an instance of a new container for a given type, you give the type OPLIST
alongside the type for building the container.
Some functions of the container may not be available in function of the provided interface of the OPLIST
(for optional operators).
Of if some mandatory operators are missing, a compiler error is generated.
The generated container will also provide its own oplist, which will depends on all the oplists used to generate it. This oplist can also be used to generate new containers.
You can therefore use the oplist of the container to chain this new interface with another container, creating container-of-container.
Note
An operator shall fail only on abnormal error and if it is marked as potentially raising asynchronous errors. In this case it shall throw exceptions only if exceptions are configured. Otherwise, the program shall be terminated.
In this case, the objects remain initialized and valid but in an unspecified state. In case of in constructors, the object is not constructed and the destructor of the object has not to be called.
Note
If an operator is not marked as raising asynchronous errors, it shall not fail or throw any exceptions in any cases.
Not all operators are needed for an oplist to handle a container. If some operator is missing, the associated default method of the operator is used if it exists.
The following classic operators are usually expected for an object:
INIT(obj)
: initialize the objectobj
into a valid state (constructor). It may raise asynchronous error.INIT_SET(obj, org)
: initialize the objectobj
into the same state as the objectorg
(copy constructor). It may raise asynchronous error.SET(obj, org)
: set the initialized objectobj
into the same state as the initialized object org (copy operator). It may raise asynchronous error.CLEAR(obj)
: destroy the initialized objectobj
, releasing any attached memory (destructor).
Other documented operators are:
NAME()
-->prefix
: Return the base nameprefix
used to construct the container.FIELD()
-->field name
: Return the name of the field used when constructing the container.TYPE()
-->type
: Return the base type associated to this oplist.SUBTYPE()
-->type
: Return the type of the element stored in the container (used to iterate over the container).GENTYPE()
-->type
: Return the type representing TYPE suitable for a _Generic statement.OPLIST()
-->oplist
: Return the oplist of the type of the elements stored in the container.KEY_TYPE()
-->key_t
: Return the key type for associative containers.VALUE_TYPE()
-->value_t
: Return the value type for associative containers.KEY_OPLIST()
-->oplist
: Return the oplist of the key type for associative containers.VALUE_OPLIST()
-->oplist
: Return the oplist of the value type for associative containers.NEW(type)
-->type pointer
: allocate a new object (with suitable alignment and size) and return a pointer to it. The returned object is not initialized (a constructor operator shall be called afterward). The default method isM_MEMORY_ALLOC
(that allocates from the heap). It returns NULL in case of failure.DEL(&obj)
: free the allocated uninitialized objectobj
. The destructor of the pointed object shall be called before freeing the memory by calling this method. The object shall have been allocated by the associated NEW method. The default method isM_MEMORY_DEL
(that frees to the heap).obj
shall not be NULL and shall be of the proper type.REALLOC(type, type pointer, number)
-->type pointer
: reallocate the given array referenced by type pointer (either a NULL pointer or a pointer returned by the associatedREALLOC
method itself) to an array of the number of objects of this type and return a pointer to this new array. Previously objects pointed by the pointer are kept up to the minimum of the new size and old one but may have moved from their original positions (if the array is reallocated otherwhere). New objects are not initialized (a constructor operator shall be called afterward). Freed objects are not cleared (A destructor operator shall be called before). The default isM_MEMORY_REALLOC
(that allocates from the heap). It returns NULL in case of failure in which case the original array is not modified.FREE(&obj)
: free the allocated uninitialized array objectobj
. The destructor of the pointed objects shall be called before freeing the memory by calling this method. The objects shall have been allocated by the associated REALLOC method. The default isM_MEMORY_FREE
(that frees to the heap).INC_ALLOC(size_t s)
-->size_t
: Define the growing policy of an array (or equivalent structure). It returns a new allocation size based on the old allocation size (s
). Default policy is to get the maximum between2*s
and 16.
Note
It doesn't check for overflow: if the returned value is lower than the old one, the user shall raise an overflow error (memory error).
INIT_MOVE(objd, objc)
: Initializeobjd
to the same state thanobjc
by stealing as many resources as possible fromobjc
, and then clearobjc
(constructor ofobjd
+ destructor ofobjc
). It is semantically equivalent to callingINIT_SET(objd,objc)
thenCLEAR(objc)
but is usually way faster. Contrary to the C++ choice of using "conservative move" semantic (you still need to call the destructor of a moved object in C++) M*LIB implements a "destructive move" semantic (this enables better optimization). By default, all objects are assumed to be trivially movable (i.e. using memcpy to move an object is safe). Most C objects (even complex structure) are trivially movable and it is a very nice property to have (enabling better optimization). A notable exception are intrusive objects. If an object is not trivially movable, it shall provide anINIT_MOVE
method or disable theINIT_MOVE
method entirely
Note
Some containers may assume that the objects are always trivially movable (like array). Moved objects shall use the same memory allocator.
MOVE(objd, objc)
: Setobjd
to the same state thanobjc
by stealing as resources as possible fromobjc
and then clearobjc
(destructor ofobjc
). It is equivalent to callingSET(objd,objc)
thenCLEAR(objc)
orCLEAR(objd)
and thenINIT_MOVE(objd, objc)
. SeeINIT_MOVE
for details and constraints. TBC if this operator is really needed as callingCLEAR
thenINIT_MOVE
is what do all known implementation, and is efficient.INIT_WITH(obj, ...)
: Initialize the objectobj
with the given variable set of arguments (constructor). The arguments are variable and can be of different types. It is up to the method of the object to decide how to initialize the object based on this initialization array. This operator is used by theM_LET
macro to initialize objects with their given values and this operator defines what theM_LET
macro supports. It may raise asynchronous error.
Note
In C11, you can use API_1(M_INIT_WITH_THROUGH_EMPLACE_TYPE)
as method to automatically use the different emplace functions defined in EMPLACE_TYPE
through a _Generic switch case. The EMPLACE_TYPE
shall use the LIST format. See emplace chapter.
SWAP(objd, objc)
: Swap the states of the objectobjc
and the objectobjd
.
Note
The swapped objects shall use the same allocator.
RESET(obj)
: Reset the object to its initialized state (Emptying the object if it is a container object).EMPTY_P(obj)
-->bool
: Test if the container object is empty (true) or not.FULL_P(obj)
-->bool
: Test if the container object is full (true) or not. Default if not defined is to be not full.GET_SIZE (container)
-->size_t
: Return the number of elements in the container object.HASH (obj)
-->size_t
: return a hash of the object (not a secure hash but one that is usable for a hash table). Default is performing a hash of the memory representation of the object. This default implementation is invalid if the object holds pointer to other objects or has spare fields.EQUAL(obj1, obj2)
-->bool
: Compare the two objects for equality. Return true if both objects are equal, false otherwise. Default is using the C comparison operator. 'obj1' may be an OOR object (Out of Representation) for the Open Addressing dictionary (seeOOR_*
operators): in such cases, it shall return false.CMP(obj1, obj2)
-->int
: Provide a complete order the objects. return a negative integer ifobj1 < obj2
, 0 ifobj1 = obj2
, a positive integer otherwise. Default is C comparison operator.
Note
The equivalence between EQUAL(a, b)
and CMP(a, b) == 0
is not required, but is usually welcome.
ADD(obj1, obj2, obj3)
: Set obj1 to the sum of obj2 and obj3. Default is+
C operator. It may raise asynchronous error.SUB(obj1, obj2, obj3)
: Set obj1 to the difference of obj2 and obj3. Default is-
C operator. It may raise asynchronous error.MUL(obj1, obj2, obj3)
: Set obj1 to the product of obj2 and obj3. Default is*
C operator. It may raise asynchronous error.DIV(obj1, obj2, obj3)
: Set obj1 to the division of obj2 and obj3. Default is/
C operator. It may raise asynchronous error.GET_KEY (container, key)
-->&value
: Return a pointer to the value object within the container associated to the keykey
if an object is associated to this key. Otherwise it may return NULL or terminate the program with a logic error (depending on the container). The pointer to the value object remains valid until any modification of the container.SET_KEY (container, key, value)
: Associate the key objectkey
to the value objectvalue
in the given container. It may raise asynchronous error.SAFE_GET_KEY (container, key)
-->&value
: return a pointer to the value object within the container associated to the keykey
if it exists, or create a new entry in the container and associate it to the keykey
with the default initialization before returning its pointer. The pointer to the object remains valid until any modification of the container. The returned pointer is therefore never NULL. It may raise asynchronous error.ERASE_KEY (container, key)
-->bool
: Erase the object associated to the keykey
within the container. Return true if successful, false if the key is not found (nothing is done).PUSH(container, obj)
: Pushobj
(of typeSUBTYPE()
) into the containercontainer
. How and where it is pushed is container dependent. It may raise asynchronous error.POP(&obj, container)
: Pop an object from the containercontainer
and set it in the initialized object*obj
(of typeSUBTYPE()
) if the pointerobj
is not NULL. Which object is popped is container dependent. It may raise asynchronous error.PUSH_MOVE(container, &obj)
: Push and move the object*obj
(of typeSUBTYPE()
) into the containercontainer
(*obj
destructor). How it is pushed is container dependent.*obj
is cleared afterward and shall not be used anymore. SeeINIT_MOVE
for more details and constraints. It may raise asynchronous error.POP_MOVE(&obj, container)
: Pop an object from the containercontainer
and init & move it in the uninitialized object*obj
(aka constructor). Which object is popped is container dependent.*obj
shall be uninitialized. SeeINIT_MOVE
for more details and constraints.
Note
When using a POP
operator (or any derived operator) on a container, this container shall have at least one object.
The iterator operators are:
IT_TYPE()
-->type
: Return the type of the iterator object of this container.IT_FIRST(it_obj, obj)
: Set the iterator it_obj to the first sub-element of the containerobj
. What is the first element is container dependent (it may be front or back, or something else). However, iterating from FIRST to LAST (included) or END (excluded) throughIT_NEXT
ensures going through all elements of the container. If there is no sub-element in the container, it references an end of the container.IT_LAST(it_obj, obj)
: Set the iterator it_obj to the last sub-element of the containerobj
. What is the last element is container dependent (it may be front or back, or something else). However, iterating from LAST to FIRST (included) or END (excluded) throughIT_PREVIOUS
ensures going through all elements of the container. If there is no sub-element in the container, it references an end of the container.IT_END(it_obj, obj)
: Set the iterator it_obj to an end of the containerobj
. Once an iterator has reached an end, it can't use PREVIOUS or NEXT operators. If an iterator has reached an END, it means that there is no object referenced by the iterator (kind of NULL pointer). There can be multiple representation of the end of a container, but all of then share the same properties.IT_SET(it_obj, it_obj2)
: Set the iterator it_obj to reference the same sub-element as it_obj2.IT_END_P(it_obj)
-->bool
: Return true if the iterator it_obj references an end of the container, false otherwise.IT_LAST_P(it_obj)
-->bool
: Return true if the iterator it_obj references the last element of the container (just before reaching an end) or has reached an end of the container, false otherwise.IT_EQUAL_P(it_obj, it_obj2)
-->bool
: Return true if both iterators reference the same element, false otherwise.IT_NEXT(it_obj)
: Move the iterator to the next sub-element or an end of the container if there is no more sub-element. The direction ofIT_NEXT
is container dependent. it_obj shall not be an end of the container.IT_PREVIOUS(it_obj)
: Move the iterator to the previous sub-element or an end of the container if there is no more sub-element. The direction of PREVIOUS is container dependent, but it is the reverse of theIT_NEXT
operator. it_obj shall not be an end of the container.IT_CREF(it_obj)
-->&subobj
: Return a constant pointer to the object referenced by the iterator (of typeconst SUBTYPE()
). This pointer remains valid until any modifying operation on the container, or until another reference is taken from this container through an iterator (some containers may reduce theses constraints, for example a list). The iterator shall not be an end of the container.IT_REF(it_obj)
-->&subobj
: Same asIT_CREF
, but return a modifiable pointer to the object referenced by the iterator.IT_INSERT(obj, it_obj, subobj)
: Insertsubobj
after 'it_obj' in the containerobj
and update it_obj to point to the inserted object (as perIT_NEXT
semantics). All other iterators of the same container become invalidated. If 'it_obj' is an end of the container, it inserts the object as the first one.IT_REMOVE(obj, it_obj)
: Remove it_obj from the containerobj
(clearing the associated object) and update it_obj to point to the next object (as perIT_NEXT
semantics). As it modifies the container, all other iterators of the same container become invalidated. it_obj shall not be an end of the container.SPLICE_BACK(objd, objs, it_obj)
: Move the object of the containerobjs
referenced by the iterator 'it_obj' to the containerobjd
. Where it is moved is container dependent (it is recommended however to be like thePUSH
method). Afterward 'it_obj' references the next item in 'containerSrc' (as perIT_NEXT
semantics). 'it_obj' shall not be an end of the container. Both objects shall use the same allocator. It may raise asynchronous error.SPLICE_AT(objd, id_objd, objs, it_objs)
: Move the object referenced by the iteratorit_objs
from the containerobjs
just after the object referenced by the iteratorit_objd
in the containerobjd
. Ifit_objd
references an end of the container, it is inserted as the first item of the container (See operatorIT_INSERT
). Afterwardit_objs
references the next item in the containerobjs
, andit_objd
references the moved item in the containerobjd
.it_objs
shall not be an end of the container. Both objects shall use the same allocator. It may raise asynchronous error.
Note
An iterator doesn't have a constructor nor destructor methods. Therefore, it cannot not allocate any memory.
Note
A reference to an object through the pointer get from the iterator is only valid until another reference is taken from the same container (potentially through another iterator), or the iterator is modified or the container itself is modified. This reference is therefore extremely local and should not be stored anywhere else. Some containers may lessen this constraint (for example list or RB-Tree).
Note
If the container is modified, all iterators of this container become invalid and shall not be used anymore except if the modifying operator provided itself an updated iterator. Some containers may lessen this constraint.
The I/O operators are:
OUT_STR(FILE* f, obj)
: Outputobj
as a custom formatted string into theFILE*
streamf
. Format is container dependent, but is human readable.IN_STR(obj, FILE* f)
-->bool
: Setobj
to the value associated to the formatted string representation of the object in theFILE*
streamf
. Return true in case of success (in that case the streamf
has been advanced to the end of the parsing of the object), false otherwise (in that case, the streamf
is in an undetermined position but is likely where the parsing fails). It ensures that an object which is output in a FILE throughOUT_STR
, and an object which is read from this FILE throughIN_STR
are considered as equal. It may raise asynchronous error.GET_STR(string_t str, obj, bool append)
: Setstr
to a formatted string representation of the objectobj
. Append to the string ifappend
is true, set it otherwise. This operator requires the module m-string. It may raise asynchronous error.PARSE_STR(obj, const char *str, const char **endp)
-->bool
: Setobj
to the value associated to the formatted string representation of the object in the char streamstr
. Return true in case of success (in that case if endp is not NULL, it points to the end of the parsing of the object), false otherwise (in that case, if endp is not NULL, it points to an undetermined position but likely to be where the parsing fails). It ensures that an object which is written in a string through GET_STR, and an object which is read from this string throughPARSE_STR
are considered as equal. It may raise asynchronous error.OUT_SERIAL(m_serial_write_t *serial, obj)
-->m_serial_return_code_t
: Outputobj
into the configurable serialization streamserial
(See m-serial-json.h for details and example) as per the serial object semantics. ReturnM_SERIAL_OK_DONE
in case of success, orM_SERIAL_FAIL
otherwise. It may raise asynchronous error.IN_SERIAL(obj, m_serial_read_t *serial)
-->m_serial_return_code_t
: Setobj
to its representation from the configurable serialization streamserial
(See m-serial-json.h for details and example) as per the serial object semantics.M_SERIAL_OK_DONE
in case of success (in that case the streamserial
has been advanced up to the complete parsing of the object), orM_SERIAL_FAIL
otherwise (in that case, the streamserial
is in an undetermined position but usually around the next characters after the first failure). It may raise asynchronous error.
The final operators are:
OOR_SET(obj, int_value)
: Some containers want to store some information within the uninitialized objects (for example Open Addressing Hash Table). This method stores the integer value 'int_value' into an uninitialized objectobj
. It shall be able to differentiate between uninitialized object and initialized object (How is type dependent). The way to store this information is fully object dependent. In general, you use out-of-range value for detecting such values. The object remains uninitialized but sets to of out-of-range value (OOR). int_value can be 0 or 1.OOR_EQUAL(obj, int_value)
: This method compares the objectobj
(initialized or uninitialized) to the out-of-range value (OOR) representation associated to 'int_value' and returns true if both objects are equal, false otherwise. SeeOOR_SET
.REVERSE(obj)
: Reverse the order of the items in the containerobj
. It may raise asynchronous error.SEPARATOR()
-->character
: Return the character used to separate items for I/O methods (default is ',') (for internal use only).EXT_ALGO(name, container oplist, object oplist)
: Define additional algorithms functions specialized for the containers (for internal use only).PROPERTIES()
-->( properties)
: Return internal properties of a container in a recursive oplist format. Use M_GET_PROPERTY to get the property.EMPLACE_TYPE( ... )
: Specify the types usable for "emplacing" the object (initializing the object in-place, constructor). See chapter Emplace construction. THe referenced initializing functions may raise asynchronous error.
Note
The operator names listed above shall not be defined as macro.
More operators are expected.
Properties can be stored in a sub-oplist format in the PROPERTIES
operator.
The following properties are defined:
LET_AS_INIT_WITH(1)
— Defined if the macroM_LET
shall always initialize the object withINIT_WITH
regardless of the given input. The value of the property is 1 (enabled) or 0 (disabled/default).NOCLEAR(1)
— Defined if the objectCLEAR
operator can be omitted (like for basic types or POD data). The value of the property is 1 (enabled) or 0 (disabled/default).
Note
The properties names listed above shall not be defined as macro.
More properties are expected.
Let's take the interface of the MPZ library:
void mpz_init(mpz_t z); // Constructor of the object z
void mpz_init_set(mpz_t z, const mpz_t s); // Copy Constructor of the object z
void mpz_set(mpz_t z, const mpz_t s); // Copy operator of the object z
void mpz_clear(mpz_t z); // Destructor of the object z
A basic oplist will be:
(INIT(mpz_init),SET(mpz_set),INIT_SET(mpz_init_set),CLEAR(mpz_clear),TYPE(mpz_t))
Much more complete oplist can be built for this type however, enabling much more powerful generation: See in the example
Oplist can be registered globally by defining, for the type type
, a macro named
M_OPL_ ## type ()
that expands to the oplist of the type.
Only type without any space in their name can be registered. A typedef of the type
can be used instead, but this typedef shall be used everywhere.
Example:
#define M_OPL_mpz_t() (INIT(mpz_init),SET(mpz_set), \
INIT_SET(mpz_init_set),CLEAR(mpz_clear),TYPE(mpz_t))
This can simplify a lot OPLIST
usage and it is recommended.
Then each times a macro expects an oplist, you can give instead its type. This make the code much easier to read and understand.
There is one exception however: the macros that are used to build oplist
(like ARRAY_OPLIST
) don't perform this simplification and the oplist of
the basic type shall be given instead
(This is due to limitation in the C preprocessing).
Within an OPLIST
, you can also specify the needed low-level transformation to perform for calling your method.
This is called API Interface Adaptation: it enables to transform the API requirements of the selected operator
(which is very basic in general) to the API provided by the given method.
Assuming that the method to call is called 'method' and the first argument of the operator is 'output',
which interface is OPERATOR(output, ...)
then the following predefined adaptation are available:
API | Method | Description |
---|---|---|
API_0 | method(output, ...) |
No adaptation |
API_1 | method(oplist, output, ...) |
No adaptation but gives the oplist to the method |
API_2 | method(&output, ...) |
Pass by address the first argument (like with M_IPTR ) |
API_3 | method(oplist, &output, ...) |
Pass by address the first argument (like with M_IPTR ) and give the oplist of the type |
API_4 | output = method(...) |
Pass by return value the first argument |
API_5 | output = method(oplist, ...) |
Pass by return value the first argument and give the oplist of the type |
API_6 | method(&output, &...) |
Pass by address the two first arguments |
API_7 | method(oplist, &output, &...) |
Pass by address the two first argument and give the oplist of the type |
The API Adaptation to use shall be embedded in the OPLIST
definition.
For example:
(INIT(API_0(mpz_init)), SET(API_0(mpz_set)), INIT_SET(API_0(mpz_init_set)), CLEAR(API_0(mpz_clear)))
The default adaptation is API_0 (i.e. no adaptation between operator interface and method interface). If an adaptation gives an oplist to the method, the method shall be implemented as macro.
Let's take the interface of a pseudo library:
typedef struct { ... } obj_t;
obj_t *obj_init(void); // Constructor of the object z
obj_t *obj_init_str(const char *str); // Constructor of the object z
obj_t *obj_clone(const obj_t *s); // Copy Constructor of the object z
void obj_clear(obj_t *z); // Destructor of the object z
The library returns a pointer to the object, so we need API_4 for these methods.
There is no method for the SET
operator available. However, we can use the macro M_SET_THROUGH_INIT_SET
to emulate a SET semantics by using a combination of CLEAR
+ INIT_SET
. This enables to support
the type for array containers in particular. Or we can avoid this definition if we don't need it.
A basic oplist will be:
(INIT(API_4(obj_init)),SET(API_1(M_SET_THROUGH_INIT_SET)),INIT_SET(API_4(obj_clone)),CLEAR(obj_clear),TYPE(obj_t *))
You can also describe the exact transformation to perform for calling the method: this is called Generic API Interface Adaptation (or GAIA). With this, you can add constant values to parameter of the method, reorder the parameters as you wish, pass then by pointers, or even change the return value.
The API adaptation is also described in the operator mapping with the method name by using the API keyword. Usage in oplist:
, OPERATOR( API( method, returncode, args...) ) ,
Within the API
keyword,
- method is the pure method name (as like any other oplist)
returncode
is the transformation to perform of the return code.- args are the list of the arguments of the function. It can be:
returncode
can be
NONE
— no transformation,VOID
— cast to void,NEG
— inverse the result,EQ(val)
/NEQ(val)
/LT(val
)/GT(val)
/LE(val)
/GE(val)
— compare the return code to the given valueARG[1-9]
— set the associated argument number of the operator to the return code
An argument can be:
- a constant,
- a variable name — probably global,
ID( constant or variable)
— if the constant or variable is not a valid token,ARG[1-9]
— the associated argument number of the operator,ARGPTR[1-9]
— the pointer of the associated argument number of the operator,OPLIST
— the oplist
Therefore, it supports at most 9 arguments.
Example:
, EQUAL( API(mpz_cmp, EQ(0), ARG1, ARG2) ) ,
This will transform a return value of 0 by the mpz_cmp method into a boolean for the EQUAL
operator.
Another Example:
, OUT_STR( API(mpz_out_str, VOID, ARG1, 10, ARG2) ) ,
This will serialize the mpz_t value in base 10 using the mpz_out_str method, and discarding the return value.
An operator OP
can be defined, omitted or disabled:
( OP(f) )
: the function f is the method of the operator OP( OP(API_N(f)) )
: the function f is the method of the operator OP with the API transformation number N,( )
: the operator OP is omitted, and the default global operation for OP is used (if it exists).( OP(0) )
: the operator OP is disabled, and it can never be used.
This can be useful to disable an operator in an inherited oplist.
My type is:
- a C Boolean:
M_BOOL_OPLIST
(M_BASIC_OPLIST
also works partially), - a C integer or a C float:
M_BASIC_OPLIST
(it can also be omitted), - a C enumerate:
M_ENUM_OPLIST
, - a pointer to something (the container do nothing on the pointed object):
M_PTR_OPLIST
, - a plain structure that can be init/copy/compare with memset/memcpy/memcmp:
M_POD_OPLIST
, - a plain structure that is passed by reference using [1] and can be init,copy,compare with
memset
,memcpy
,memcmp
:M_A1_OPLIST
, - a type that offers
name_init
,name_init_set
,name_set
,name_clear
methods:M_CLASSIC_OPLIST
, - a const string (
const char *
) that is neither freed nor moved:M_CSTR_OPLIST
, - a M*LIB string_t:
STRING_OPLIST
, - a M*LIB container: the
OPLIST
of the used container, - other things: you need to provide a custom
OPLIST
to your type.
Note
The precise exported methods of the OPLIST depend on the version
of the used C language. Typically, in C11 mode, the M_BASIC_OPLIST
exports all needed methods to handle generic input/output of int/floats
(using _Generic
keyword) whereas it is not possible in C99 mode.
This explains why JSON import/export is only available in C11 mode (See below chapter).
Basic usage of oplist is available in the example
Oplist can inherit from another one.
This is useful when you want to customize some specific operators while keeping other ones by default.
For example, internally M_BOOL_OPLIST
inherits from M_BASIC_OPLIST
.
A typical example is if you want to provide the OOR_SET
and OOR_EQUAL
operators to a type
so that it can be used in an OA
dict.
To do it, you use the M_OPEXTEND
macro. It takes as first argument the oplist you want to inherit with,
and then you provide the additional associations between operators to methods you want to add
or override in the inherited oplist. For example:
int my_int_oor_set(char c) { return INT_MIN + c; }
bool my_int_oor_equal(int i1, int i2) { return i1 == i2; }
#define MY_INT_OPLIST \
M_OPEXTEND(M_BASIC_OPLIST, OOR_SET(API_4(my_int_oor_set)), \
OOR_EQUAL(my_int_oor_equal))
You can even inherit from another oplist to disable some operators in your new oplist. For example:
#define MY_INT_OPLIST \
M_OPEXTEND(M_BASIC_OPLIST, HASH(0), CMP(0), EQUAL(0))
MY_INT_OPLIST
is a new oplist that handles integers but has disabled the operators HASH
, CMP
and EQUAL
.
The main interest is to disable the generation of optional methods of a container (since they are only
expanded if the oplist provides some methods).
Usage of inheritance and oplist is available in the int example and the cstr example
Let's take a look at the interface of the FILE*
interface:
FILE *fopen(const char *filename, const char *mode);
fclose(FILE *f);
There is no INIT
operator (an argument is mandatory), no INIT_SET
operator.
It is only possible to open a file from a filename.
FILE*
contains some space, so an alias is needed.
There is an optional mode argument, which is a constant string, and isn't a valid preprocessing token.
An oplist can therefore be:
typedef FILE *m_file_t;
#define M_OPL_m_file_t() (INIT_WITH(API(fopen, ARG1, ARG2, ID("wt"))), \
SET(0),INIT_SET(0),CLEAR(fclose),TYPE(m_file_t))
Since there is no INIT_SET
operator available, pretty much no container can work.
However, you'll be able to use a writing text FILE*
using a M_LET
:
M_LET( (f, ("tmp.txt")), m_file_t) {
fprintf(f, "This is a tmp file.");
}
This is pretty useless, except if you enable exceptions... In which case, this enables you to close the file even if an exception is thrown.
Memory Allocation functions can be globally set by overriding the following macros before using the definition _DEF
macros:
M_MEMORY_ALLOC (type)
-->ptr
: return a pointer to a new object of typetype
.M_MEMORY_DEL (ptr)
: free the single object pointed byptr
.M_MEMORY_REALLOC (type, ptr, number)
-->ptr
: return a pointer to an array of 'number' objects of typetype
, reusing the old array pointed byptr
.ptr
can be NULL, in which case the array will be created.M_MEMORY_FREE (ptr)
: free the array of objects pointed byptr
.
ALLOC
and DEL
operators are supposed to allocate fixed size single element object (no array).
These objects are not expected to grow.
REALLOC
and FREE
operators deal with allocated memory for growing objects.
Do not mix pointers between both: a pointer allocated by ALLOC
(resp. REALLOC
) is supposed
to be only freed by DEL
(resp. FREE
). There are separated 'free' operators to enable
specialization in the allocator (a good allocator can take this hint into account).
M_MEMORY_ALLOC
and M_MEMORY_REALLOC
are supposed to return NULL in case of memory allocation failure.
The defaults are malloc
, free
, realloc
and free
.
You can also override the methods NEW
, DEL
, REALLOC
and DEL
in the oplist given to a container
so that only the container will use these memory allocation functions instead of the global ones.
When a memory exhaustion is reached, the global macro M_MEMORY_FULL
is called
and the function returns immediately after.
The object remains in a valid (if previously valid) and unchanged state in this case.
By default, the macro prints an error message and aborts the program: handling non-trivial memory errors can be hard, testing them is even harder but still mandatory to avoid security holes. So the default behavior is rather conservative.
It can however be overloaded to handle other policy for error handling like:
- throwing an error (which is automatically done by including header m-try ),
- set a global error and handle it when the function returns [planned, not possible yet],
- other policy.
This function takes the size in bytes of the memory that has been tried to be allocated.
If needed, this macro shall be defined prior to instantiate the structure.
Note
A good design should handle a process entire failure (using for examples multiple processes for doing the job) so that even if a process stops, it should be recovered. See here for more information about why abandonment is good software practice.
In M*LIB, we classify the kind of errors according to this classification:
- logical error: the expectations of the function are not met (null pointer passed as argument, negative argument, invalid object state, ...). In which case, the sanction is abnormal halt of the program, if it is detected, regardless of the configuration of M*LIB. Normally, debug build will detect such errors.
- abnormal error: errors that are unlikely to be expected during the execution of the program (like no more memory). In which case, the sanction is either abnormal halt of the program or throwing an exception.
- normal error: errors that can be expected in the execution of the program (all I/O errors like file not found or invalid file format, parsing of invalid user input, no solution found, etc). In which case, the error is reported by the return code of the function or by polling for error (See
ferror
) in the data structure.
For M*LIB, 'emplace' means pushing a new object in the container,
while not giving it a copy of the new object, but the parameters needed
to construct this object.
This is a shortcut to the pattern of creating the object with the arguments,
pushing it in the container, and deleting the created object
(even if using PUSH_MOVE
could simplify the design).
The containers defining an emplace like method generate the emplace functions
based on the provided EMPLACE_TYPE
of the oplist. If EMPLACE_TYPE
doesn't exist or is disabled, no emplace function is generated. Otherwise EMPLACE_TYPE
identifies
which types can be used to construct the object and which methods to use to construct then:
EMPLACE_TYPE( typeA )
, means that the object can be constructed fromtypeA
using the method of theINIT_WITH
operator. An emplace function without suffix will be generated.EMPLACE_TYPE( (typeA, typeB, ...) )
, means that the object can be constructed from the lists of typestypeA
,typeB
,...
using the method of theINIT_WITH
operator. An emplace function without suffix will be generated.EMPLACE_TYPE( LIST( (suffix, function, typeA, typeB, ...)
, (suffix
,function
,typeA
,typeB
,...
) means that the object can be constructed from all the provided lists of typestypeA
,typeB
,...
using the provided methodfunction
. Thefunction
is the method to call to construct the object from the list of types. It supports API transformation if needed. As many emplace function will be generated as there are constructor function. Each generated function will be generated by suffixing it with the providedsuffix
(if suffix is empty, no suffix will be added).
For example, for an ARRAY
definition named vec, if there is such a definition of EMPLACE_TYPE(const char *)
, it will generate a function vec_emplace_back(const char *)
. This function will take a const char*
parameter, construct the object from it (for example a string_t) then push the result back on the array.
Another example, for an ARRAY
definition named vec, if there is such a definition of EMPLACE_TYPE( LIST( (_ui, mpz_init_set_ui, unsigned int), (_si, mpz_init_set_si, int) ) )
, it will generate two functions vec_emplace_back_ui(unsigned int)
and vec_emplace_back_si(int)
. These functions will take the (unsigned) int parameter, construct the object from it then push the result back on the array.
If the container is an associative array, the name will be constructed as follows:
name_emplace[_key_keysuffix][_val_valsuffix]
where keysuffix
(resp. valsuffix
) is the emplace suffix of the key (resp. value
) oplist.
If you take once again the example of the FILE*
, a more complete oplist can be:
typedef FILE *m_file_t;
#define M_OPL_m_file_t() (INIT_WITH(API_1(M_INIT_WITH_THROUGH_EMPLACE_TYPE)), \
SET(0),INIT_SET(0),CLEAR(fclose),TYPE(m_file_t), \
EMPLACE_TYPE(LIST((_str, API(fopen, ARG1, ARG2, ID("wb")), char *))))
The INIT_WITH
operator will use the provided init methods in the EMPLACE_TYPE
.
EMPLACE_TYPE
defines a _str
suffix method with a GAIA for fopen
, and accepts a char*
as argument.
The GAIA specifies that the output (ARG1) is set as return value,
ARG2 is given as the first argument, and a third constant argument is used.
M*LIB implements internally some controls to reduce the list of errors/warnings generated by a compiler when it detects some violation in the use of oplist by use of static assertion. It can also transform some type warnings into proper errors. In C99 mode, it will produce illegal code with the name of the error as attribute. In C11 mode, it will use static assert and produce a detailed error message.
The list of errors it can generate:
M_LIB_NOT_AN_OPLIST
: something has been given (directly or indirectly) and it doesn't reduce as a proper oplist. You need to give an oplist for this definition.M_LIB_ERROR(ARGUMENT_OF_*_OPLIST_IS_NOT_AN_OPLIST, name, oplist)
: sub error of the previous error one, identifying the root cause. The error is in the oplist construction of the given macro. You need to give an oplist for this oplist construction.M_LIB_MISSING_METHOD
: a required operator doesn't define any method in the given oplist. You need to complete the oplist with the missing method.M_LIB_TYPE_MISMATCH
: the given oplist and the type do not match each other. You need to give the right oplist for this type.M_LIB_NOT_A_BASIC_TYPE
: The oplistM_BASIC_OPLIST
(directly or indirectly) has been used with the given type, but the given type is not a basic C type (int/float). You need to give the right oplist for this type.
You should focus mainly on the first reported error/warning even if the link between what the compiler report and what the error is is not immediate. The error is always in one of the oplist definition.
Examples of typical errors:
- lack of inclusion of an header,
- overriding locally operator names by macros (like
NEW
,DEL
,INIT
,OPLIST
,...
), - lack of
( )
or double level of( )
around the oplist, - an unknown variable (example using
BASIC_OPLIST
instead ofM_BASIC_OPLIST
), - the name given to the oplist is not the same as the one used to define the methods,
- use of a type instead of an oplist in the
OPLIST
definition, - a missing sub oplist in the
OPLIST
definition.
A good way to avoid these errors is to register the oplist globally as soon as you define the container.
In case of difficulties, debugging can be done by generating the preprocessed file
(by usually calling the compiler with the -E
option instead of -c
)
and check what's wrong in the preprocessed file:
cc -std=c99 -E test-file.c |grep -v '^#' > test-file.i
perl -i -e 's/;/;\n/g' test-file.i
cc -std=c99 -c -Wall test-file.i
If there is a warning reported by the compiler in the generated code, then there is definitely an error you should fix (except if it reports shadowed variables), in particular cast evolving pointers.
You should also turn off the macro expansion of the errors reported by your compiler. There are often completely useless and misleading:
- For GCC, uses
-ftrack-macro-expansion=0
- For CLANG, uses
-fmacro-backtrace-limit=1
Due to the unfortunate weak nature of the C language for pointers,
you should turn incompatible pointer type warning into an error in your compiler.
For GCC / CLANG, uses -Werror=incompatible-pointer-types
For MS Visual C++ compiler , you need the following options:
/Zc:__cplusplus /Zc:preprocessor /D__STDC_WANT_LIB_EXT1__ /EHsc
Many other implementation of generic container libraries in C exist. For example, a non exhaustive list is:
- BKTHOMPS/CONTAINERS
- BSD tree.h
- CC
- CDSA
- CELLO
- C-Macro-Collections
- COLLECTIONS-C
- CONCURRENCY KIT
- CTL by glouw or by rurban
- GDB internal library
- GLIB
- KLIB
- LIBCHASTE
- LIBCOLLECTION
- LIBDICT
- LIBDYNAMIC
- LIBLFDS
- LIBSRT: Safe Real-Time library for the C programming language
- NEDTRIES
- POTTERY
- QLIBC
- SC
- SGLIB
- Smart pointer for GNUC
- STB stretchy_buffer
- STC - Smart Template Container for C
- TommyDS
- UTHASH
Each can be classified into one of the following concept:
- Each object is handled through a pointer to void (with potential registered callbacks to handle the contained objects for the specialized methods). From a user point of view, this makes the code harder to use (as you don't have any help from the compiler) and type unsafe with a lot of cast (so no formal proof of the code is possible). This also generally generate slower code (even if using link time optimization, this penalty can be reduced). Properly used, it can yet be the most space efficient for the code, but can consume a lot more for the data due to the obligation of using pointers. This is however the easiest to design & code.
- Macros are used to access structures in a generic way (using known fields of a structure — typically size, number, etc.). From a user point of view, this can create subtitle bug in the use of the library (as everything is done through macro expansion in the user defined code) or hard to understand warnings. This can generates fully efficient code. From a library developer point of view, this can be quite limiting in what you can offer.
- Macros detect the type of the argument passed as parameter using _Generics, and calls the associated methods of the switch. The difficulty is how to add pure user types in this generic switch.
- A known structure is put in an intrusive way in the type of all the objects you wan to handle. From a user point of view, he needs to modify its structure and has to perform all the allocation & deallocation code itself (which is good or bad depending on the context). This can generate efficient code (both in speed and size). From a library developer point of view, this is easy to design & code. You need internally a cast to go from a pointer to the known structure to the pointed object (a reverse of offsetof) that is generally type unsafe (except if mixed with the macro generating concept). This is quite limitation in what you can do: you can't move your objects so any container that has to move some objects is out-of-question (which means that you cannot use the most efficient container).
- Header files are included multiple times with different contexts (some different values given to defined macros) in order to generate different code for each type. From a user point of view, this creates a new step before using the container: an instantiating stage that has to be done once per type and per compilation unit (The user is responsible to create only one instance of the container, which can be troublesome if the library doesn't handle proper prefix for its naming convention). The debug of the library is generally easy and can generate fully specialized & efficient code. Incorrectly used, this can generate a lot of code bloat. Properly used, this can even create smaller code than the void pointer variant. The interface used to configure the library can be quite tiresome in case of a lot of specialized methods to configure: it doesn't enable to chain the configuration from a container to another one easily. It also cannot have heavy customization of the code.
- Macros are used to generate context-dependent C code enabling to generate code for different type. This is pretty much like the headers solution but with added flexibility. From a user point of view, this creates a new step before using the container: an instantiating stage that has to be done once per type and per compilation unit (The user is responsible to create only one instance of the container, which can be troublesome if the library doesn't handle proper prefix for its naming convention). This can generate fully specialized & efficient code. Incorrectly used, this can generate a lot of code bloat. Properly used, this can even create smaller code than the void pointer variant. From a library developer point of view, the library is harder to design and to debug: everything being expanded in one line, you can't step in the library (there is however a solution to overcome this limitation by adding another stage to the compilation process). You can however see the generated code by looking at the preprocessed file. You can perform heavy context-dependent customization of the code (transforming the macro preprocessing step into its own language). Properly done, you can also chain the methods from a container to another one easily, enabling expansion of the library. Errors within the macro expansion are generally hard to decipher, but errors in code using containers are easy to read and natural.
M*LIB category is mainly the last one. Some macros are also defined to access structure in a generic way, but they are optional. There are also intrusive containers.
M*LIB main added value compared to other libraries is its oplist feature enabling it to chain the containers and/or use complex types in containers: list of array of dictionary of C++ objects are perfectly supported by M*LIB.
For the macro-preprocessing part, other libraries and reference also exist. For example:
- Boost preprocessor
- C99 Lambda
- C Preprocessor Tricks, Tips and Idioms
- CPP MAGIC
- MAP MACRO
- metalang99
- OrderPP
- P99
- Zlang
You can also consult awesome-c-preprocessor for a more comprehensive list of libraries.
For the string part, there is the following libraries for example:
- POTTERY STRING
- SDS
- STC - C99 Standard Container library
- STR -yet another string library for C language
- The Better String Library with a page that lists a lot of other string libraries.
- VSTR with a page that lists a lot of other string libraries.
The M*LIB reference card is available here.
The generated containers tries to generate and provide a consistent interface: their methods would behave the same for all generated containers. This chapter will explain the generic interface. In case of difference, it will be explained in the specific container.
In the following description:
name
is the prefix used for the container generation,name_t
refers to the type of the container,name_it_t
refers to the iterator type of the container,type_t
refers to the type of the object stored in the container,key_type_t
refers to the type of the key object used to associate an element to,value_type_t
refers to the type of the value object used to associate an element to.name_itref_t
refers to a pair of key and value for associative arrays.
An object shall be initialized (aka constructor) before being used by other methods. It shall be cleared (aka destructor) after being used and before program termination. An iterator has not destructor but shall be set before being used.
A container takes as input the
name
— it is a mandatory argument that is the prefix used to generate all functions and types,type
— it is a mandatory argument that the basic element of the container,oplist
— it is an optional argument that defines the methods associated to the type. The provided oplist (if provided) shall be the one that is associated to the type, otherwise it won't generate compilable code. If there is no oplist parameter provided, a globally registered oplist associated to the type is used if possible, or the basic oplist for basic C types is used. This oplist will be used to handle internally the objects of the container.
The type
given to any templating macro can be either
an integer, a float, a boolean, an enum, a named structure, a named union, a pointer to such types,
or a typedef alias of any C type:
in fact, the only constraint is that the preprocessing concatenation between type
and variable
into
'type variable' shall be a valid C expression.
Therefore the type
cannot be a C array or a function pointer
and you shall use a typedef as an intermediary named type for such types.
The output parameters are listed fist, then the input/output parameters and finally the input parameters.
This generic interface is specified as follow:
Initialize the container container
(aka constructor) to an empty container.
Also called the default constructor of the container.
Initialize the container container
(aka constructor) and set it to a copy of ref
.
This method is only created if the INIT_SET
& SET
methods are provided.
Also called the copy constructor of the container.
Set the container container
to the copy of ref
.
This method is only created if the INIT_SET
and SET
methods are provided.
Initialize the container container
(aka constructor)
by stealing as many resources from ref
as possible.
After-wise ref
is cleared and can no longer be used (aka destructor).
Set the container container
(aka constructor)
by stealing as many resources from ref
as possible.
After-wise ref
is cleared and can no longer be used (aka destructor).
Clear the container container
(aka destructor),
calling the CLEAR
method of all the objects of the container and freeing memory.
The object can't be used anymore, except to be reinitialized with a constructor.
Reset the container clearing and freeing all objects stored in it. The container becomes empty but remains initialized.
Return a pointer to the data stored in the back of the container. This pointer should not be stored in a global variable.
Return a pointer to the data stored in the front of the container. This pointer should not be stored in a global variable.
Push a new element in the container container
as a copy of the object value
.
This method is created only if the INIT_SET
operator is provided.
The _back
suffixed method will push it in the back of the container.
The _front
suffixed method will push it in the front of the container.
Push a new element in the container container
without initializing it and returns a pointer to the non-initialized data.
The first thing to do after calling this function shall be to initialize the data
using the proper constructor of the object of type type_t
.
This enables using more specialized constructor than the generic copy one.
The user should use other _push
function if possible rather than this one
as it is low level and error prone.
This pointer should not be stored in a global variable.
The _back
suffixed method will push it in the back of the container.
The _front
suffixed method will push it in the front of the container.
Push a new element in the container container
and initialize it with the default constructor associated to the type type_t
.
Return a pointer to the initialized object.
This pointer should not be stored in a global variable.
This method is only created if the INIT
method is provided.
The _back
suffixed method will push it in the back of the container.
The _front
suffixed method will push it in the front of the container.
Push a new element in the container container
as a move from the object *value
:
it will therefore steal as much resources from *value
as possible.
Afterward *value
is cleared and cannot longer be used (aka. destructor).
The _back
suffixed method will push it in the back of the container.
The _front
suffixed method will push it in the front of the container.
Push a new element in the container container
by initializing it with the provided arguments.
The provided arguments shall therefore match one of the constructor provided
by the EMPLACE_TYPE
operator.
There is one generated method per suffix defined in the EMPLACE_TYPE
operator,
and the suffix
in the generated method name corresponds to the suffix defined
in the EMPLACE_TYPE
operator (it can be empty).
This method is created only if the EMPLACE_TYPE
operator is provided.
See emplace chapter for more details.
The _back
suffixed method will push it in the back of the container.
The _front
suffixed method will push it in the front of the container.
Pop a element from the container container
;
and set *data
to this value if data is not the NULL pointer,
otherwise it only pops the data.
This method is created if the SET
or INIT_MOVE
operator is provided.
The _back
suffixed method will pop it from the back of the container.
The _front
suffixed method will pop it from the front of the container.
Pop a element from the container container
and initialize and set *data
to this value (aka. constructor)
by stealing as much resources from the container as possible.
data shall not be a NULL pointer.
The _back
suffixed method will pop it from the back of the container.
The _front
suffixed method will pop it from the front of the container.
Return true if the container is empty, false otherwise.
Swap the container container1
and container2
.
Set the element associated to key
of the container container
to value
.
If the container is sequence-like (like array), the key is an integer.
The selected element is the key
-th element of the container,
The index key
shall be within the size of the container.
This method is created if the INIT_SET
operator is provided.
If the container is associative-array like,
the selected element is the value
object associated to the key
object in the container.
It is created if it doesn't exist, overwritten otherwise.
void name_emplace_key[keysuffix](name_t container, keyargs..., value_type_t value)
[for associative array]
void name_emplace_val[valsuffix](name_t container, key_type_t key, valargs...)
[for associative array]
void name_emplace_key[keysuffix]_val[valsuffix](name_t container, keyargs..., valargs...)
[for associative array]
Set the element associated to key
of the container container
to value
.
key
(resp. value) is constructed from the provided args keyargs
(resp. valargs
) if not provided.
keyargs
(resp. valargs
) shall therefore match one of the constructor provided
by the EMPLACE_TYPE
operator of the key (resp. the value).
There is:
- one generated method per suffix defined in the
EMPLACE_TYPE
operator of the key, - one generated method per suffix defined in the
EMPLACE_TYPE
operator of the value, - one generated method per pair of suffix defined in the
EMPLACE_TYPE
operators of the key and value,
The suffix
in the generated method name corresponds to the suffix defined in
in the EMPLACE_TYPE
operator (it can be empty).
This method is created only if one EMPLACE_TYPE
operator is provided.
See emplace chapter for more details.
Return a modifiable (resp. constant) pointer to
the element associated to key
in the container.
If the container is sequence-like, the key is an integer.
The selected element is the key
-th element of the container,
the front being at the index 0, the last at the index (size-1
).
The index key
shall be within the size of the container.
Therefore, it will never return NULL in this case.
If the container is associative array like,
the selected element is the value
object associated to the key
object in the container.
If the key is not found, it returns NULL.
If the container is set-like,
the selected element is the value
object which is equal to the key
object in the container.
If the key is not found, it returns NULL.
This pointer remains valid until the container is modified by another method. This pointer should not be stored in a global variable.
Return a modifiable (resp. constant) pointer to
the element associated to the key constructed from args
in the container.
The selected element is the value
object associated to the key
object in the container
for an associative array, or the element itself for a set.
If the key is not found, it returns NULL.
This pointer remains valid until the container is modified by another method. This pointer should not be stored in a global variable.
Return a modifiable pointer to
the element associated to key
in the container,
creating a new element if it doesn't exist (ensuring therefore a safe get
operation).
If the container is sequence-like, key_type_t is an alias for size_t and key an integer,
the selected element is the key
-th element of the container.
If the element doesn't exist, the container size will be increased
if needed (creating new elements with the INIT
operator),
which might increase the container to much in some cases.
The returned pointer cannot be NULL.
This method is only created if the INIT
method is provided.
This pointer remains valid until the array is modified by another method.
This pointer should not be stored in a global variable.
Remove the element referenced by the key key
in the container container
.
Do nothing if key
is no present in the container.
Return true if the key was present and erased, false otherwise.
Return the number elements of the container (its size). Return 0 if there is no element.
Return the capacity of the container, i.e. the maximum number of elements supported by the container before a reserve operation is needed to accommodate new elements.
Resize the container container
to the size size
,
increasing or decreasing the size of the container
by initializing new elements or clearing existing ones.
This method is only created if the INIT method is provided.
Extend or reduce the capacity of the container
to a rounded value based on capacity
.
If the given capacity is below the current size of the container,
the capacity is set to a rounded value based on the size of the container.
Therefore a capacity of 0 is used to perform a shrink-to-fit operation on the container,
i.e. reducing the container usage to the minimum possible allocation.
Set the iterator it
to the first element of the container container
.
There is no destructor associated to this initialization.
Set the iterator it
to the iterator ref
.
There is no destructor associated to this initialization.
Set the iterator it
to a non valid element of the container.
There is no destructor associated to this initialization.
Return true if the iterator doesn't reference a valid element anymore.
Return true if the iterator references the last element of the container or if the iterator doesn't reference a valid element anymore.
Return true if the iterator it1
references the same element than the iterator it2
.
Move the iterator to the next element of the container.
Move the iterator it
to the previous element of the container.
Return a modifiable (resp. constant) pointer to the element pointed by the iterator.
For associative-array like container, this element is the pair composed of
the key (key
field) and the associated value (value
field);
otherwise this element is simply the basic type of the container (type_t
).
This pointer should not be stored in a global variable. This pointer remains valid until the container is modified by another method.
Insert the object x
after the position referenced by the iterator it
in the container container
or if the iterator it
doesn't reference anymore to a valid element of the container,
it is added as the first element of the container.
it
shall be an iterator of the container container
.
Remove the element referenced by the iterator it
from the container container
.
it
shall be an iterator of the container container
.
Afterwards, it
references the next element of the container if it exists,
or not a valid element otherwise.
Return true if both containers container1
and container2
are considered equal.
This method is only created if the EQUAL
method is provided.
Return a fast hash value of the container container
,
suitable to be used by a dictionary.
This method is only created if the HASH
method is provided.
Set str
to the formatted string representation of the container container
if append
is false
or append str
with this representation if append
is true.
This method is only created if the GET_STR
method is provided.
Parse the formatted string str
,
that is assumed to be a string representation of a container,
and set container
to this representation.
It returns true if success, false otherwise.
If endp
is not NULL, it sets *endp
to the pointer of the first character not
decoded by the function (or where the parsing fails).
This method is only created if the PARSE_STR
& INIT
methods are provided.
It is ensured that the container gets from parsing a formatted string representation
gets from name_get_str
and the original container are equal.
Generate a formatted string representation of the container container
and outputs it into the file FILE*
.
The file FILE*
shall be opened in write text mode and without error.
This method is only created if the OUT_STR
methods is provided.
Read from the file FILE*
a formatted string representation of a container
and set container
to this representation.
It returns true if success, false otherwise.
This method is only created if the IN_STR
& INIT
methods are provided.
It is ensured that the container gets from parsing a formatted string representation gets from name_out_str and the original container are equal.
Output the container container
into the serializing object serial
.
How and where it is output depends on the serializing object.
It returns M_SERIAL_OK_DONE
in case of success,
or M_SERIAL_FAILURE
in case of failure.
In case of failure, the serializing object is in an unspecified but valid state.
This method is only created if the OUT_SERIAL
methods is provided.
Read from the serializing object serial
a representation of a container
and set container
to this representation.
It returns M_SERIAL_OK_DONE
in case of success,
or M_SERIAL_FAILURE
in case of failure.
In case of failure, the serializing object is in an unspecified but valid state.
This method is only created if the IN_SERIAL
& INIT
methods are provided.
It is ensured that the container gets from parsing a representation gets from name_out_serial and the original container are equal (using the same type of serialization object).
This header is for creating singly linked list.
A linked list is a linear collection of elements, in which each element points to the next, all of then representing a sequence.
A fundamental property of a list is that the objects created within the list will remain at their initialized address, and won't moved due to operations done on the list (except if it is removed). Therefore a returned pointer to an element of the container remains valid until this element is destroyed in the container.
LIST_DEF
defines the singly linked list named name_t
that contains objects of type type
and their associated methods as static inline
functions.
name
shall be a C identifier that will be used to identify the list.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The oplist shall have at least the following operators (INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
For this container, the back is always the first element, and the front is the last element: the list grows from the back. Therefore, the iteration of this container using iterator objects will go from the back element to the front element (contrary to an array).
Even if it provides random access functions, theses access are slow and should be avoided: it iterates linearly over all the elements of the container until it reaches the requested element. The size method has the same drawback.
The push/pop methods of the container always operate on the back of the container.
LIST_DEF_AS
is the same as LIST_DEF
except the name of the types name_t
, name_it_t
are provided by the user,
and not computed from the name
prefix.
Example:
#include <stdio.h>
#include <gmp.h>
#include "m-list.h"
#define MPZ_OUT_STR(stream, x) mpz_out_str(stream, 0, x)
LIST_DEF(list_mpz, mpz_t, \
(INIT(mpz_init), INIT_SET(mpz_init_set), SET(mpz_set), \
CLEAR(mpz_clear), OUT_STR(MPZ_OUT_STR)))
int main(void) {
list_mpz_t a;
list_mpz_init(a);
mpz_t x;
mpz_init_set_ui(x, 16);
list_mpz_push_back (a, x);
mpz_set_ui(x, 45);
list_mpz_push_back (a, x);
mpz_clear(x);
printf ("LIST is: ");
list_mpz_out_str(stdout, a);
printf ("\n");
printf ("First element is: ");
mpz_out_str(stdout, 0, *list_mpz_back(a));
printf ("\n");
list_mpz_clear(a);
}
If the given oplist contain the method MEMPOOL
, then LIST_DEF
macro will create
a dedicated mempool that is named with the given value of the method MEMPOOL
.
This mempool (see mempool chapter) is optimized for this kind of list:
- it creates a mempool named by the concatenation of
name
and_mempool
, - it creates a variable named by the value of the method
MEMPOOL
with the linkage defined by the value of the methodMEMPOOL_LINKAGE
(can be extern, static or none). This variable is shared by all lists of the same type. - it links the memory allocation of the list to use this mempool with this variable.
mempool
create heavily efficient list. However, it is only worth the
effort in some heavy performance context.
Using mempool should be therefore avoided except in performance critical code.
The created mempool has to be explicitly initialized before using any
methods of the created list by calling mempool_list_name_init(variable)
and cleared by calling mempool_list_name_clear(variable)
.
Example:
LIST_DEF(list_uint, unsigned int, (MEMPOOL(list_mpool), MEMPOOL_LINKAGE(static)))
static void test_list (size_t n)
{
list_uint_mempool_init(list_mpool);
M_LET(a1, LIST_OPLIST(uint)) {
for(size_t i = 0; i < n; i++)
list_uint_push_back(a1, rand_get() );
}
list_uint_mempool_clear(list_mpool);
}
Return the oplist of the list defined by calling LIST_DEF
and LIST_DUAL_PUSH_DEF
with name & oplist.
If there is no given oplist, the basic oplist for basic C types is used.
There is no globally registered oplist support.
Define an initial value that is suitable to initialize global variable(s)
of type list
as created by LIST_DEF
or LIST_DEF_AS
.
It enables to create a list as a global variable and to initialize it.
The list shall still be cleared manually to avoid leaking memory.
Example:
LIST_DEF(list_int, int)
list_int_t my_list = LIST_INIT_VALUE();
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the list of type
objects.
Type of an iterator over this list.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t list)
void name_init_set(name_t list, const name_t ref)
void name_set(name_t list, const name_t ref)
void name_init_move(name_t list, name_t ref)
void name_move(name_t list, name_t ref)
void name_clear(name_t list)
void name_reset(name_t list)
type *name_back(const name_t list)
void name_push_back(name_t list, const type value)
type *name_push_raw(name_t list)
type *name_push_new(name_t list)
void name_push_move(name_t list, type *value)
void name_emplace_back[suffix](name_t list, args...)
void name_pop_back(type *data, name_t list)
void name_pop_move(type *data, name_t list)
bool name_empty_p(const name_t list)
void name_swap(name_t list1, name_t list2)
void name_it(name_it_t it, const name_t list)
void name_it_set(name_it_t it, const name_it_t ref)
void name_it_end(name_it_t it, const name_t list)
bool name_end_p(const name_it_t it)
bool name_last_p(const name_it_t it)
bool name_it_equal_p(const name_it_t it1, const name_it_t it2)
void name_next(name_it_t it)
type *name_ref(name_it_t it)
const type *name_cref(const name_it_t it)
type *name_get(const name_t list, size_t i)
const type *name_cget(const name_t list, size_t i)
size_t name_size(const name_t list)
void name_insert(name_t list, const name_it_t it, const type x)
void name_remove(name_t list, name_it_t it)
void name_get_str(string_t str, const name_t list, bool append)
bool name_parse_str(name_t list, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t list)
bool name_in_str(name_t list, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
bool name_equal_p(const name_t list1, const name_t list2)
size_t name_hash(const name_t list)
The following specialized methods are automatically created by the previous definition macro:
Move the element referenced by the iterator it
from the list list2
to the back position of the list list1
.
it
shall be an iterator of list2
.
Afterwards, it
references the next element of the list if it exists,
or not a valid element otherwise.
Move the element referenced by the iterator it2
from the list list2
to the position just after it1
in the list list1
.
(If it1
is not a valid position, it inserts it at the back just like name_insert
).
it1
shall be an iterator of list1
.
it2
shall be an iterator of list2
.
Afterwards, it2
references the next element of the list if it exists,
or not a valid element otherwise,
and it1
references the inserted element in list1
.
Move all the element of the list list2
into the list list1
,
moving the last element of list2
after the first element of list1
.
Afterwards, list2
remains initialized but is emptied.
list1
and list2
shall reference different objects.
Reverse the order of the list.
LIST_DUAL_PUSH_DEF
defines the singly linked list named name_t
that contains the objects of type type
and their associated methods as static inline
functions.
The only difference with the list defined by LIST_DEF
is
the support of the method for PUSH_FRONT
in addition to the one for PUSH_BACK
(so the DUAL_PUSH
name).
However, there is still only POP
method (associated to POP_BACK
).
The list is a bit bigger to be able to handle such method to work, but not the nodes.
This list is therefore able to represent:
- either a stack (
PUSH_BACK
+POP_BACK
) - or a queue (
PUSH_FRONT
+POP_BACK
).
LIST_DUAL_PUSH_DEF_AS
is the same as LIST_DUAL_PUSH_DEF
except the name of the types name_t
, name_it_t
are provided by the user.
See LIST_DEF
for more details and constraints.
Example:
#include <stdio.h>
#include <gmp.h>
#include "m-list.h"
#define MPZ_OUT_STR(stream, x) mpz_out_str(stream, 0, x)
LIST_DUAL_PUSH_DEF(list_mpz, mpz_t, \
(INIT(mpz_init), INIT_SET(mpz_init_set), SET(mpz_set), \
CLEAR(mpz_clear), OUT_STR(MPZ_OUT_STR)))
int main(void) {
list_mpz_t a;
list_mpz_init(a);
mpz_t x;
mpz_init_set_ui(x, 16);
list_mpz_push_back (a, x);
mpz_set_ui(x, 45);
list_mpz_push_back (a, x);
mpz_clear(x);
printf ("LIST is: ");
list_mpz_out_str(stdout, a);
printf ("\n");
printf ("First element is: ");
mpz_out_str(stdout, 0, *list_mpz_back(a));
printf ("\n");
list_mpz_clear(a);
}
The methods follow closely the methods defined by LIST_DEF
.
Define an initial value that is suitable to initialize global variable(s)
of type list
as created by LIST_DUAL_PUSH_DEF
or LIST_DUAL_PUSH_DEF_AS
.
It enables to create a list as a global variable and to initialize it.
The list should still be cleared manually to avoid leaking memory.
Example:
LIST_DUAL_PUSH_DEF(list_int, int)
list_int_t my_list = LIST_DUAL_PUSH_INIT_VALUE();
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the list of type
.
Type of an iterator over this list.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t list)
void name_init_set(name_t list, const name_t ref)
void name_set(name_t list, const name_t ref)
void name_init_move(name_t list, name_t ref)
void name_move(name_t list, name_t ref)
void name_clear(name_t list)
void name_reset(name_t list)
type *name_back(const name_t list)
void name_push_back(name_t list, type value)
type *name_push_back_raw(name_t list)
type *name_push_back_new(name_t list)
void name_push_back_move(name_t list, type *value)
void name_emplace_back[suffix](name_t list, args...)
const type *name_front(const name_t list)
void name_push_front(name_t list, type value)
type *name_push_front_raw(name_t list)
type *name_push_front_new(name_t list)
void name_push_front_move(name_t list, type *value)
void name_emplace_front[suffix](name_t list, args...)
void name_pop_back(type *data, name_t list)
void name_pop_move(type *data, name_t list)
bool name_empty_p(const name_t list)
void name_swap(name_t list1, name_t list2)
void name_it(name_it_t it, name_t list)
void name_it_set(name_it_t it, const name_it_t ref)
void name_it_end(name_it_t it, const name_t list)
bool name_end_p(const name_it_t it)
bool name_last_p(const name_it_t it)
bool name_it_equal_p(const name_it_t it1, const name_it_t it2)
void name_next(name_it_t it)
type *name_ref(name_it_t it)
const type *name_cref(const name_it_t it)
size_t name_size(const name_t list)
void name_insert(name_t list, const name_it_t it, const type x)
void name_remove(name_t list, name_it_t it)
void name_get_str(string_t str, const name_t list, bool append)
bool name_parse_str(name_t list, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t list)
bool name_in_str(name_t list, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
bool name_equal_p(const name_t list1, const name_t list2)
size_t name_hash(const name_t list)
The following specialized methods are automatically created by the previous definition macro:
Move the element pointed by it
from the list list2
to the back position of the list list1
.
it
shall be an iterator of list2
.
Afterwards, it
points to the next element of list2
.
Move all the element of the list list2
into the list list1
,
moving the last element of list2
after the first element of list1
.
Afterwards, list2
is emptied.
list1
and list2
shall reference different objects.
Reverse the order of the list.
An array is a growable collection of element that are individually indexable.
ARRAY_DEF
defines the array name_t
that contains the objects of type type
in a sequence
and its associated methods as static inline
functions.
Compared to C arrays, the created methods handle automatically the size (aka growable array).
name
shall be a C identifier that will be used to identify the container.
It will be used to create all the types (including the iterator)
and functions to handle the container.
The provided oplist shall have at least the following operators (CLEAR
).
It should also provide the following operators (INIT
, INIT_SET
, SET
),
so that at least some methods are defined for the array!
The push / pop methods of the container always operate on the back of the container, acting like a stack-like container.
ARRAY_DEF_AS
is the same as ARRAY_DEF
except the name of the types name_t
, name_it_t
are provided by the user.
Example:
#include <stdio.h>
#include <gmp.h>
#include "m-array.h"
#define MPZ_OUT_STR(stream, x) mpz_out_str(stream, 0, x)
ARRAY_DEF(array_mpz, mpz_t, \
(INIT(mpz_init), INIT_SET(mpz_init_set), SET(mpz_set), \
CLEAR(mpz_clear), OUT_STR(MPZ_OUT_STR)))
int main(void) {
array_mpz_t a;
array_mpz_init(a);
mpz_t x;
mpz_init_set_ui(x, 16);
array_mpz_push_back (a, x);
mpz_set_ui(x, 45);
array_mpz_push_back (a, x);
mpz_clear(x);
printf ("ARRAY is: ");
array_mpz_out_str(stdout, a);
printf ("\n");
printf ("First element is: ");
mpz_out_str(stdout, 0, *array_mpz_get(a, 0));
printf ("\n");
array_mpz_clear(a);
}
Return the oplist of the array defined by calling ARRAY_DEF
with name & oplist.
If there is no given oplist, the basic oplist for basic C types is used.
Define an initial value that is suitable to initialize global variable(s)
of type array
as created by ARRAY_DEF
or ARRAY_DEF_AS
.
It enables to create an array as a global variable and to initialize it.
The array should still be cleared manually to avoid leaking memory.
Example:
ARRAY_DEF(array_int_t, int)
array_int_t my_array = ARRAY_INIT_VALUE();
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the array of type
.
Type of an iterator over this array.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t array)
void name_init_set(name_t array, const name_t ref)
void name_set(name_t array, const name_t ref)
void name_init_move(name_t array, name_t ref)
void name_move(name_t array, name_t ref)
void name_clear(name_t array)
void name_reset(name_t array)
type *name_push_raw(name_t array)
void name_push_back(name_t array, const type value)
type *name_push_new(name_t array)
void name_push_move(name_t array, type *val)
void name_emplace_back[suffix](name_t array, args...)
void name_pop_back(type *data, name_t array)
void name_pop_move(type *data, name_t array)
type *name_front(const name_t array)
type *name_back(const name_t array)
void name_set_at(name_t array, size_t i, type value)
type *name_get(name_t array, size_t i)
const type *name_cget(const name_t it, size_t i)
type *name_safe_get(name_t array, size_t i)
bool name_empty_p(const name_t array)
size_t name_size(const name_t array)
size_t name_capacity(const name_t array)
void name_resize(name_t array, size_t size)
void name_reserve(name_t array, size_t capacity)
void name_swap(name_t array1, name_t array2)
void name_it(name_it_t it, name_t array)
void name_it_last(name_it_t it, name_t array)
void name_it_end(name_it_t it, name_t array)
void name_it_set(name_it_t it1, name_it_t it2)
bool name_end_p(name_it_t it)
bool name_last_p(name_it_t it)
bool name_it_equal_p(const name_it_t it1, const name_it_t it2)
void name_next(name_it_t it)
void name_previous(name_it_t it)
type *name_ref(name_it_t it)
const type *name_cref(const name_it_t it)
void name_remove(name_t array, name_it_t it)
void name_insert(name_t array, name_it_t it, const type x)
void name_get_str(string_t str, const name_t array, bool append)
bool name_parse_str(name_t array, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t array)
bool name_in_str(name_t array, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
bool name_equal_p(const name_t array1, const name_t array2)
size_t name_hash(const name_t array)
The following specialized methods are automatically created by the previous definition macro:
Push a new element into the position key
of the array array
with the value value
.
All elements after the position key
(included) will be moved in the array towards the back,
and the array will have one more element.
key
shall be a valid position of the array: from 0 to the size of array (included).
This method is created only if the INIT_SET
operator is provided.
Pop all elements of the array array
from 'position' to the back of the array,
while clearing them.
This method is created only if the INIT
operator is provided.
Set *dest
to the value the element key
if dest is not NULL,
then remove the element key
from the array (decreasing the array size).
key
shall be within the size of the array.
This method is created only if the SET or INIT_MOVE operator is provided.
Remove the element i
(included) to the element j
(excluded)
from the array.
i
and j
shall be within the size of the array, and i < j
.
Insert from the element i
(included) to the element j
(excluded)
new empty elements to the array.
i
and j
shall be within the size of the array, and i < j
.
This method is only defined if the type of the element defines an INIT
method.
Swap the elements i
and j
of the array array
.
i
and j
shall reference valid elements of the array.
This method is created if the INIT_SET
or INIT_MOVE
operator is provided.
Sort the array array
.
This method is defined if the type of the element defines CMP
method.
This method uses the qsort function of the C library.
Sort the array array
using a stable sort.
This method is defined if the type of the element defines CMP
and SWAP
and SET
methods.
This method provides an ad-hoc implementation of the stable sort.
In practice, it is faster than the _sort method for small types and fast
comparisons.
Move all the elements of the array array2
to the end of the array array1
.
Afterwards, array2
is empty.
array1
and array2
shall reference different objects.
This header is for creating double-ended queue (or deque). A deque is an abstract data type that generalizes a queue, for that elements can be added to or removed from either the front (head) or back (tail)
DEQUE_DEF
defines the deque name_t
that contains the objects of type type
and its associated methods as static inline
functions.
name
shall be a C identifier that will be used to identify the deque.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The oplist shall have at least the following operators (INIT
, INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
The algorithm complexity to access random elements is in O(ln(n)). Removing an element may unbalance the deque, which breaks the promise of algorithm complexity for the _get method.
DEQUE_DEF_AS
is the same as DEQUE_DEF
except the name of the types name_t
, name_it_t
are provided.
Example:
#include <stdio.h>
#include <gmp.h>
#include "m-deque.h"
DEQUE_DEF(deque_mpz, mpz_t, \
(INIT(mpz_init), INIT_SET(mpz_init_set), SET(mpz_set), CLEAR(mpz_clear)))
int main(void) {
deque_mpz_t a;
deque_mpz_init(a);
mpz_t x;
mpz_init_set_ui(x, 16);
deque_mpz_push_back (a, x);
mpz_set_ui(x, 45);
deque_mpz_push_front (a, x);
deque_mpz_pop_back(&x, a);
mpz_set_ui(x, 5);
deque_mpz_push_back (a, x);
mpz_clear(x);
printf ("First element is: ");
mpz_out_str(stdout, 0, *deque_mpz_front(a));
printf ("\n");
printf ("Last element is: ");
mpz_out_str(stdout, 0, *deque_mpz_back(a));
printf ("\n");
deque_mpz_clear(a);
}
Return the oplist of the deque defined by calling DEQUE_DEF
with name oplist
.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the deque of type
.
Type of an iterator over this deque.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t deque)
void name_init_set(name_t deque, const name_t ref)
void name_set(name_t deque, const name_t ref)
void name_init_move(name_t deque, name_t ref)
void name_move(name_t deque, name_t ref)
void name_clear(name_t deque)
void name_reset(name_t deque)
type *name_back(const name_t deque)
void name_push_back(name_t deque, type value)
type *name_push_back_raw(name_t deque)
type *name_push_back_new(name_t deque)
void name_emplace_back[suffix](name_t list, args...)
void name_pop_back(type *data, name_t deque)
type *name_front(const name_t deque)
void name_push_front(name_t deque, type value)
type *name_push_front_raw(name_t deque)
type *name_push_front_new(name_t deque)
void name_emplace_front[suffix](name_t list, args...)
void name_pop_front(type *data, name_t deque)
bool name_empty_p(const name_t deque)
void name_swap(name_t deque1, name_t deque2)
void name_it(name_it_t it, name_t deque)
void name_it_set(name_it_t it, const name_it_t ref)
bool name_end_p(const name_it_t it)
bool name_last_p(const name_it_t it)
bool name_it_equal_p(const name_it_t it1, const name_it_t it2)
void name_next(name_it_t it)
void name_previous(name_it_t it)
type *name_ref(name_it_t it)
const type *name_cref(const name_it_t it)
void name_remove(name_t deque, name_it_t it)
type *name_get(const name_t deque, size_t i)
const type *name_cget(const name_t deque, size_t i)
size_t name_size(const name_t deque)
void name_get_str(string_t str, const name_t deque, bool append)
bool name_parse_str(name_t deque, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t deque)
bool name_in_str(name_t deque, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
bool name_equal_p(const name_t deque1, const name_t deque2)
size_t name_hash(const name_t deque)
void name_swap_at(name_t deque, size_t i, size_t j)
A dictionary (or associative array, map, symbol table) is an abstract data type composed of a collection of (key, value) pairs, such that each possible key appears at most once in the collection, and is associated to only one value. It is possible to search for a key in the dictionary and get back its value.
Several dictionaries are proposed. The "best" to use depends on the data type
and in particular the size of the data.
For very small, fast types (integer, or floats, or pair of such types),
DICT_OA_DEF2
may be the best to use (but slightly more complex to instantiate).
However DICT_DEF2
should be good enough for all scenarios.
DICT_DEF2_AS(name, name_t, name_it_t, name_itref_t, key_type[, key_oplist], value_type[, value_oplist])
DICT_DEF2
defines the dictionary name_t
and its associated methods as static inline
functions as an associative array of key_type
to value_type
.
name
shall be a C identifier that will be used to identify the dictionary.
It will be used to create all the types (including the iterator and the iterated object type)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
Both oplist shall have at least the following operators (INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
The key_oplist
shall also define the additional operators (HASH
and EQUAL
).
Elements in the dictionary are unordered. On insertion of a new element, contained objects may moved. Maximum number of elements of this dictionary is 3'006'477'107 elements.
The _set_at
method overwrites the already existing value if key
is already present in the dictionary
(contrary to C++).
The iterated object type name##_itref_t
is a pair of key_type and value_type.
What is exactly the "first" element for the iteration is not specified. It is only ensured that all elements of the dictionary are explored by going from "first" to "end".
DICT_DEF2_AS
is the same as DICT_DEF2
except the name of the types name_t
, name_it_t
, name_itref_t
are provided.
Example:
#include <stdio.h>
#include "m-string.h"
#include "m-dict.h"
DICT_DEF2(dict_string, string_t, unsigned)
int main(void) {
dict_string_t a;
dict_string_init(a);
string_t x;
string_init_set_str(x, "This is an example");
dict_string_set_at (a, x, 1);
string_set_str(x, "This is an another example");
dict_string_set_at (a, x, 2);
string_set_str(x, "This is an example");
unsigned *val = dict_string_get(a, x);
printf ("Value of %s is %u\n", string_get_cstr(x), *val);
string_clear(x);
printf ("Dictionary is: ");
dict_string_out_str(stdout, a);
printf ("\n");
dict_string_clear(a);
}
DICT_OA_DEF2_AS(name, name_t, name_it_t, name_itref_t, key_type[, key_oplist], value_type[, value_oplist])
DICT_OA_DEF2
defines the dictionary name_t
and its associated methods
as static inline
functions much like DICT_DEF2
.
The difference is that it uses an Open Addressing Hash-Table as container
that stores the data with the table.
The key_oplist
shall also define the additional operators:
OOR_EQUAL
and OOR_SET
The Out-Of-Range operators (OOR_EQUAL
and OOR_SET
) are used to store uninitialized keys
in the dictionary and be able to detect it. This enables avoiding a separate bit-field
to know the state of the entry in the dictionary (which increases memory
usage and is cache unfriendly).
The elements may move when inserting / deleting other elements (and not just the iterators).
This implementation is in general faster for small types of keys (like integer or float) but are slower for larger types.
DICT_OA_DEF2_AS
is the same as DICT_OA_DEF2
except the name of the types name_t
, name_it_t
, name_itref_t
are provided.
Example:
#include <stdio.h>
#include "m-string.h"
#include "m-dict.h"
// Define an Out Of Range equal function
static inline bool oor_equal_p(unsigned k, unsigned char n) {
return k == (unsigned)(-n-1);
}
// Define an Out Of Range function
static inline void oor_set(unsigned *k, unsigned char n) {
*k = (unsigned)(-n-1);
}
DICT_OA_DEF2(dict_unsigned, unsigned, M_OPEXTEND(M_BASIC_OPLIST,
OOR_EQUAL(oor_equal_p), OOR_SET(API_2(oor_set))), long long, M_BASIC_OPLIST)
unsigned main(void) {
dict_unsigned_t a;
dict_unsigned_init(a);
dict_unsigned_set_at (a, 13566, 14890943049);
dict_unsigned_set_at (a, 656, -2);
long long *val = dict_unsigned_get(a, 458);
printf ("Value of %d is %p\n", 458, val); // Not found value
val = dict_unsigned_get(a, 656);
printf ("Value of %d is %lld\n", 656, *val);
dict_unsigned_clear(a);
}
Return the oplist of the dictionary defined by calling any DICT_*_DEF2
with name
, key_oplist
, value_oplist
.
DICT_SET_DEF
defines the dictionary set name_t
and its associated methods as static inline
functions.
A dictionary set is a specialized version of a dictionary with no value (only keys).
name
shall be a C identifier that will be used to identify the dictionary.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The oplist shall have at least the following operators (INIT_SET
, SET
, CLEAR
, HASH
and EQUAL
),
otherwise it won't generate compilable code.
The _push
method will overwrite the already existing value if key
is already present in the dictionary (contrary to C++).
What is exactly the "first" element for the iteration is not specified. It is only ensured that all elements of the dictionary are explored by going from "first" to "end".
DICT_SET_DEF_AS
is the same as DICT_SET_DEF
except the name of the types name_t
, name_it_t
are provided.
Example:
#include <stdio.h>
#include "m-string.h"
#include "m-dict.h"
DICT_SET_DEF(set_string, double)
unsigned main(void) {
set_string_t a;
set_string_init(a);
set_string_push (a, 13566);
set_string_push (a, 656);
double *val = set_string_get(a, 458);
printf ("Value of %d is %p\n", 458, val); // Not found value
val = set_string_get(a, 656);
printf ("Value of %d is %f\n", 656, *val);
printf("Set is ");
set_string_out_str(stdout, a);
printf("\n");
set_string_clear(a);
}
DICT_OASET_DEF
defines the dictionary set name_t
and its associated methods as static inline
functions just like DICT_SET_DEF
.
The difference is that it uses an Open Addressing Hash-Table as container.
The key_oplist
shall also define the additional operators:
OOR_EQUAL
and OOR_SET
The elements may move when inserting / deleting other elements (and not just the iterators).
This implementation is in general faster for small types of keys (like integer) but slower for larger types.
DICT_OASET_DEF_AS
is the same as DICT_OASET_DEF
except the name of the types name_t
, name_it_t
are provided.
Return the oplist of the set defined by calling DICT_SET_DEF
(or DICT_OASET_DEF
) with name
and key_oplist
.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the dictionary which
- either associate
key_type
tovalue_type
, - or store unique element
key_type
.
Type of an iterator over this dictionary.
Type of one item referenced in the dictionary for associative array.
It is a structure composed of the key (field key
) and the value (field value
).
This type is created only for associative arrays (_DEF2
suffix) and not for sets.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t dict)
void name_clear(name_t dict)
void name_init_set(name_t dict, const name_t ref)
void name_set(name_t dict, const name_t ref)
void name_init_move(name_t dict, name_t ref)
void name_move(name_t dict, name_t ref)
void name_reset(name_t dict)
size_t name_size(const name_t dict)
bool name_empty_p(const name_t dict)
value_type *name_get(const name_t dict, const key_type key)
value_type_t * name_get_emplace[suffix](name_t container, args...)
value_type *name_safe_get(name_t dict, const key_type key)
void name_set_at(name_t dict, const key_type key, const value_type value) /* for associative array */
void name_push(name_t dict, const key_type key) /* for dictionary set */
bool name_erase(name_t dict, const key_type key)
void name_it(name_it_t it, name_t dict)
void name_it_set(name_it_t it, const name_it_t ref)
bool name_end_p(const name_it_t it)
bool name_last_p(const name_it_t it)
void name_next(name_it_t it)
name_itref_t *name_ref(name_it_t it) /* for associative array */
key_type *name_ref(name_it_t it) /* for dictionary set */
const name_itref_t *name_cref(name_it_t it) /* for associative array */
const key_type *name_cref(name_it_t it) /* for dictionary set */
void name_get_str(string_t str, const name_t dict, bool append)
bool name_parse_str(name_t dict, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t dict)
bool name_in_str(name_t dict, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
bool name_equal_p(const name_t dict1, const name_t dict2)
void name_emplace[suffix](name_t container, keyargs...) [for dictionary set */
void name_emplace_key[keysuffix]_val[valsuffix](name_t container, keyargs..., valargs...) /* for associative array */
void name_reserve(name_t dict, size_t capacity)
The following specialized methods are automatically created by the previous definition macro:
Move all items from dict2
into dict1
.
If there is the same key between dict2
into dict1
,
then their values are added (as per the ADD
method of the value type).
Afterward dict2
is reset (i.e. empty).
This method is only defined if the value type defines an ADD
method.
dict1
and dict2
shall reference different objects.
A tuple is a finite ordered list of elements of different types.
TUPLE_DEF2
defines the tuple name_t
and its associated methods as static inline
functions.
Each parameter of the macro is expected to be an element of the tuple.
Each element is defined by three parameters within parenthesis:
- the element name (the field name of the structure)
- the element type (the associated type)
- and the optional element oplist associated to this type (see generic interface for the behavior if it is absent)
name
and element
shall be C identifiers that will be used to identify the container and the fields.
It will be used to create all the types (including the iterator) and functions to handle the container. This definition shall be done once per name and per compilation unit.
This is more or less a C structure. The main added value compared to using a C struct is that it generates also all the basic methods to handle it which is quite handy.
The oplist shall have at least the following operators (INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
In general, an optional method of the tuple will only be created if all oplists define the needed optional methods for the underlying type.
The _hash
(resp. _equal_p
and _cmp
) method is an exception.
This method is created only if at least one oplist of the tuple defines the HASH
(resp. EQUAL
) method.
You can disable the use of a specific field for the hash computation of the tuple
by disabling the HASH
operator of such field (with HASH(0)
in its oplist),
in which case it is coherent to also disable the EQUAL
operator too.
Resp., you can disable the use of a field for the equality of the tuple
by disabling the EQUAL
operator of such field (with EQUAL(0)
in its oplist)
The comparison of two tuples uses lexicographic order of the fields defining the CMP
method.
It is created only if at least one oplist of the tuple define CMP
method.
You can disable the use of a field for the comparison of the tuple
by disabling the CMP
operator of such field (with CMP(0)
in its oplist)
TUPLE_DEF2_AS
is the same as TUPLE_DEF2
except the name of the type name_t
is provided.
Example:
#include <stdio.h>
#include <gmp.h>
#include "m-string.h"
#include "m-tuple.h"
#define MPZ_OUT_STR(stream, x) mpz_out_str(stream, 0, x)
TUPLE_DEF2(pair, (key_field, string_t, STRING_OPLIST), \
(value_field, mpz_t, M_OPEXTEND(M_CLASSIC_OPLIST(mpz), \
OUT_STR(MPZ_OUT_STR))))
int main(void) {
pair_t p1;
pair_init (p1);
string_set_str(p1->key_field, "HELLO");
mpz_set_str(p1->value_field, "1742", 10);
printf("The pair is ");
pair_out_str(stdout, p1);
printf("\n");
pair_clear(p1);
}
Return the oplist of the tuple defined by calling TUPLE_DEF2
with the given name
and the oplist
.
The following type is automatically defined by the previous definition macro if not provided by the user:
Type of the defined tuple.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t tuple)
void name_init_set(name_t tuple, const name_t ref)
void name_set(name_t tuple, const name_t ref)
void name_init_move(name_t tuple, name_t ref)
void name_move(name_t tuple, name_t ref)
void name_clear(name_t tuple)
void name_reset(name_t tuple)
void name_get_str(string_t str, const name_t tuple, bool append)
bool name_parse_str(name_t tuple, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t tuple)
bool name_in_str(name_t tuple, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
size_t name_hash(const name_t tuple)
int name_equal_p(const name_t tuple1, const name_t tuple2)
int name_cmp(const name_t tuple1, const name_t tuple2)
The following specialized methods are automatically created by the previous definition macro:
Initialize the tuple tuple
(aka constructor) and set it to the value of the constructed tuple (element1[, ...]
).
Set the tuple tuple
to the value of the tuple constructed with (element1[, ...]
).
Return a constant pointer to the element element1
of the tuple.
There is as many methods as there are elements.
Return a pointer to the element element1
of the tuple.
There is as many methods as there are elements.
Set the element of the tuple to element1
There is as many methods as there are elements.
Compare tuple1
to tuple2
using the given order.
order
is a NULL terminated array of int that defines the order of comparison:
an order of {1, 4, 2, 0} indicates to compare first the first field,
if it is equal, to compare the fourth and so on. The third field is not
compared. A negative value indicates to revert the comparison.
This method is created only if all oplist of the tuple define CMP
method.
Compare tuple1
to tuple2
using only the element element1
as reference.
This method is created only if the oplist of element1 defines the CMP
method.
A variant is a finite exclusive list of elements of different types: the variant can only be equal to one element at a time.
VARIANT_DEF2
defines the variant name_t
and its associated methods as static inline
functions.
Each parameter of the macro is expected to be an element of the variant.
Each element is defined by three parameters within parenthesis:
- the mandatory element name,
- the mandatory element type
- and the optional element oplist.
If an OPLIST
is given, it shall be the one matching the associated type.
name
and element<n>
shall be C identifiers that will be used to identify the variant.
name
will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
This is like a C union. The main added value compared to using a union
is that it generates all the basic methods to handle and it dynamically
identifies which element is stored within.
It is also able to store an EMPTY
state for the variant, contrary to an union
(this is the state when default constructing it).
The oplists shall have at least the following operators (INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
In general, an optional method of the variant will only be created
if all oplists define the needed optional methods for the underlying type.
VARIANT_DEF2_AS
is the same as VARIANT_DEF2
except the name of the type name_t
is provided.
name_parse_str
and name_in_str
depend also on the INIT
operator.
Example:
#include <stdio.h>
#include "m-string.h"
#include "m-variant.h"
VARIANT_DEF2(item,
(name, string_t),
(age, long))
int main(void) {
item_t p1;
item_init (p1);
item_set_name(p1, STRING_CTE("HELLO"));
printf("The variant is ");
item_out_str(stdout, p1);
printf("\n");
item_set_age(p1, 43);
printf("The variant is now ");
item_out_str(stdout, p1);
printf("\n");
item_clear(p1);
}
Return the oplist of the variant defined by calling VARIANT_DEF2
with the given name
and the oplists
used to generate it.
The following type is automatically defined by the previous definition macro if not provided by the user:
Type of the defined variant.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t variant)
void name_init_set(name_t variant, const name_t ref)
void name_set(name_t variant, const name_t ref)
void name_init_move(name_t variant, name_t ref)
void name_move(name_t variant, name_t ref)
void name_clear(name_t variant)
void name_reset(name_t variant)
bool name_empty_p(const name_t variant)
size_t name_hash(const name_t variant)
bool name_equal_p(const name_t variant1, const name_t variant2)
void name_swap(name_t variant1, name_t variant2)
void name_get_str(string_t str, name_t variant, bool append)
bool name_parse_str(name_t variant, const char str[], const char **endp)
void name_out_str(FILE *file, name_t variant)
bool name_in_str(name_t variant, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
The following specialized methods are automatically created by the previous definition macro:
Initialize the variant variant
to the type of element1
[constructor]
using the default constructor of this element.
This method is defined if all methods define an INIT
method.
Initialize the variant variant
and set it to the type and value of elementN
[constructor]
Set the variant variant
by stealing as many resources from ref
as possible.
Afterwards ref
is cleared (destructor)
This method is created only if all oplist of the variant define MOVE
method.
Set the variant variant
to the type and value of elementN
.
Return a pointer to the variant
viewed as of type typeN
.
If the variant isn't an object of such type, it returns NULL.
Return a pointer to the variant
viewed as of type typeN
.
If the variant isn't an object of such type, it returns NULL.
Return true if the variant is of the type of elementN
.
A binary tree is a tree data structure in which each node has at most two children, which are referred to as the left child and the right child. A node without any child is called a leaf. It can be seen as an ordered set.
A R-B Tree is a tree where all elements are also totally ordered, and the worst-case of any operation is in logarithm of the number of elements in the tree. The current implementation is RED-BLACK TREE which provides performance guarantee for both insertion and lockup operations. It has not to be confused with a B-TREE.
RBTREE_DEF defines the binary ordered tree name_t
and its associated methods as static inline
functions.
name
shall be a C identifier that will be used to identify the R-B Tree.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The CMP
operator is used to perform the total ordering of the elements.
The oplist shall have at least the following operators (INIT
, INIT_SET
, SET
, CLEAR
& CMP
),
otherwise it won't generate compilable code.
Some methods may return a modifiable pointer to the found element
(for example, _get
). In this case, the user shall not modify the
key order of the element, as there is no reordering of the tree
in this case.
A push method on the tree will put the given key
in its right place in the tree
by keeping the tree ordered.
It overwrites the already existing value if the key is already present in the dictionary (contrary to C++).
RBTREE_DEF_AS
is the same as RBTREE_DEF2
except the name of the types name_t
, name_it_t
are provided by the user.
Example:
RBTREE_DEF(rbtree_uint, unsigned int)
void f(unsigned int num) {
rbtree_uint_t tree;
rbtree_uint_init(tree);
for(unsigned int i = 0; i < num; i++)
rbtree_uint_push(tree, i);
rbtree_uint_clear(tree);
}
Return the oplist of the Red-Black tree defined by calling RBTREE_DEF
with name
and oplist
.
If there is no given oplist, the basic oplist for basic C types is used.
The following types are automatically defined by the previous definition macro with name
if not provided by the user:
Type of the Red Black Tree.
Type of an iterator over this Red Black Tree.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t rbtree)
void name_clear(name_t rbtree)
void name_init_set(name_t rbtree, const name_t ref)
void name_set(name_t rbtree, const name_t ref)
void name_init_move(name_t rbtree, name_t ref)
void name_move(name_t rbtree, name_t ref)
void name_reset(name_t rbtree)
size_t name_size(const name_t rbtree)
void name_push(name_t rbtree, const type data)
void name_emplace[suffix](name_t rbtree, args...)
type *name_get(const name_t rbtree, const type data)
const type *name_cget(const name_t rbtree, const type data)
void name_swap(name_t rbtree1, name_t rbtree2)
bool name_empty_p(const name_t rbtree)
void name_it(name_it_t it, name_t rbtree)
void name_it_set(name_it_t it, const name_it_t ref)
void name_it_last(name_it_t it, name_t rbtree)
void name_it_end(name_it_t it, name_t rbtree)
bool name_end_p(const name_it_t it)
bool name_last_p(const name_it_t it)
void name_it_remove(name_t rbtree, name_it_t it)
void name_next(name_it_t it)
void name_previous(name_it_t it)
type *name_ref(name_it_t it)
const type *name_ref(name_it_t it)
void name_get_str(string_t str, const name_t rbtree, bool append)
bool name_parse_str(name_t tree, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t rbtree)
bool name_in_str(name_t rbtree, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
bool name_equal_p(const name_t rbtree1, const name_t rbtree2)
size_t name_hash(const name_t rbtree)
The following specialized methods are automatically created by the previous definition macro:
Pop data
from the Red Black Tree rbtree
and save the popped value into dest
if the pointer is not null
while keeping the tree balanced.
Do nothing if data
is no present in the Red Black Tree.
Return a pointer to the minimum element of the tree or NULL if there is no element.
Return a pointer to the maximum element of the tree or NULL if there is no element.
Set the iterator it
to
the lowest element of the tree rbtree
greater or equal than data
or an iterator to no element is there is none.
Return true if it
references an element that is greater or equal than data
or if it references no longer a valid element, false otherwise.
Return true if it
references an element that is lower or equal than data
.
Otherwise (or if it references no longer a valid element) it returns false.
A B+TREE is a variant of BTREE that itself is a generalization of Binary Tree.
A B+TREE is an N-ary tree with a variable but often large number of children per node. It is mostly used for handling slow media by file system and database.
It provides a fully sorted container enabling fast access to individual item or range of items, and as such is concurrent to Red-Black Tree. On modern architecture, a B+TREE is typically faster than a red-black tree due to being more cache friendly (The RAM itself can be considered as a slow media nowadays!)
When defining a B+TREE it is necessary to give the type of the item within, but also
the maximum number of child per node (N
). The best maximum number of child per node
depends on the type itself (its size, its compare cost) and the cache of the
processor.
BPTREE_DEF2_AS(name, name_t, name_it_t, name_itref_t, N, key_type, key_oplist, value_type, value_oplist)
Define the B+TREE tree of rank N
name_t
and its associated methods as
static inline
functions. This B+TREE will be created as an associative
array of the key_type
to the value_type
.
The CMP
operator is used to perform the total ordering of the key elements.
N
is the number of items per node and shall be greater or equal than 2.
It shall be done once per type and per compilation unit.
It also define the iterator name##_it_t
and its associated methods as static inline
functions.
The object oplist shall have at least the operators (INIT
, INIT_SET
, SET
, CLEAR
and CMP
).
BPTREE_DEF2_AS
is the same as BPTREE_DEF2
except the name of the
container type name_t
, the iterator type name_it_t
, and the
iterated object type name_itref_t
are provided by the user.
Example:
BPTREE_DEF2(tree_uint, 4, unsigned int, (), float, ())
void f(unsigned int num) {
tree_uint_t tree;
tree_uint_init(tree);
for(unsigned int i = 0; i < num; i++)
tree_uint_set_at(tree, i, (float) i);
tree_uint_clear(tree);
}
Return the oplist of the BPTREE
defined by calling BPTREE_DEF2
with name
,
key_oplist
and value_oplist
.
Define the B+TREE tree of rank N
name_t
and its associated methods as
static inline
functions. This B+TREE will be created as an ordered set
of key_type
.
The CMP
operator is used to perform the total ordering of the key elements.
N
is the number of items per node and shall be greater or equal than 2.
It shall be done once per type and per compilation unit.
It also define the iterator name##_it_t
and its associated methods as static inline
functions.
The object oplist shall have at least the operators (INIT
, INIT_SET
, SET
, CLEAR
and CMP
).
BPTREE_DEF_AS
is the same as BPTREE_DEF
except the name of
the container type name_t
, the iterator type name_it_t
and the iterated object type name_itref_t
are provided by the user.
Example:
BPTREE_DEF(tree_uint, unsigned int)
void f(unsigned int num) {
tree_uint_t tree;
tree_uint_init(tree);
for(unsigned int i = 0; i < num; i++)
tree_uint_push(tree, i);
tree_uint_clear(tree);
}
Return the oplist of the BPTREE
defined by calling BPTREE_DEF
with name
and key_oplist
.
If there is no given oplist, the basic oplist for basic C types is used.
BPTREE_MULTI_DEF2_AS(name, name_t, name_it_t, name_itref_t, N, key_type, key_oplist, value_type, value_oplist)
Define the B+TREE tree of rank N
name_t
and its associated methods as
static inline
functions. This B+TREE will be created as an associative
array of the key_type
to the value_type
and allows multiple instance of
the same key in the tree (aka it is a multi-map: re-adding the same key in
the tree will add a new instance of the key in the tree rather than update
the value associated to the key).
See BPTREE_DEF2
for additional details and example.
BPTREE_MULTI_DEF2_AS
is the same as BPTREE_MULTI_DEF2
except the name of the types name_t
, name_it_t
, name_itref_t
are provided by the user.
Define the B+TREE tree of rank N
name_t
and its associated methods as
static inline
functions. This B+TREE will be created as an ordered set
of key_type
and allows multiple instance of
the same key in the tree (aka it is a multi set: re-adding the same key in
the tree will add a new instance of the key in the tree rather than update
the key value).
See BPTREE_DEF
for additional details and example.
BPTREE_MULTI_DEF_AS
is the same as BPTREE_MULTI_DEF
except the name of the types name_t
, name_it_t
, name_itref_t
are provided by the user.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the B+Tree of type
.
Type of an iterator over this B+Tree.
Type of one item referenced in the B+Tree. It is either:
- a structure composed of a pointer to the key (field
key_ptr
) and a pointer to the value (fieldvalue_ptr
) if the B+Tree is an associative array, - or the basic type of the container if the B+Tree is a set.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t tree)
void name_clear(name_t tree)
void name_init_set(name_t tree, const name_t ref)
void name_set(name_t tree, const name_t ref)
void name_init_move(name_t tree, name_t ref)
void name_move(name_t tree, name_t ref)
void name_reset(name_t tree)
size_t name_size(const name_t tree)
void name_push(name_t tree, const key_type data) [for set definition only]
void name_set_at(name_t tree, const key_type data, const value_type val) /* for associative array definition only */
bool name_erase(name_t tree, const key_type data)
value_type *name_get(const name_t tree, const key_type data)
const value_type *name_cget(const name_t tree, const key_type data)
value_type *name_safe_get(name_t tree, const key_type data)
void name_swap(name_t tree1, name_t tree2)
bool name_empty_p(const name_t tree)
void name_it(name_it_t it, name_t tree)
void name_it_set(name_it_t it, const name_it_t ref)
void name_it_end(name_it_t it, name_t tree)
bool name_end_p(const name_it_t it)
bool name_it_equal_p(const name_it_t it1, const name_it_t it1)
void name_next(name_it_t it)
name_itref_t *name_ref(name_it_t it)
const name_itref_t *name_cref(name_it_t it)
void name_get_str(string_t str, const name_t tree, bool append)
bool name_parse_str(name_t tree, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t tree)
bool name_in_str(name_t tree, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
bool name_equal_p(const name_t tree1, const name_t tree2)
size_t name_hash(const name_t tree)
void name_emplace[suffix](name_t container, keyargs...) [for dictionary set]
void name_emplace_key[keysuffix]_val[valsuffix](name_t container, keyargs..., valargs...) /* for associative array */
The following specialized methods are automatically created by the previous definition macro:
Pop data
from the B+Tree tree
and save the popped value into dest
if the pointer is not NULL
while keeping the tree balanced.
Do nothing if data
is no present in the B+Tree.
Return true if data
has been erased, false otherwise.
Return a pointer to the minimum element of the tree or NULL if there is no element in the B+Tree.
Return a pointer to the maximum element of the tree or NULL if there is no element in the B+Tree.
Set the iterator it
to the lowest element of tree
greater or equal than data
or the end iterator is there is none.
Return true if it
references an element that is greater or equal than data
or if it references no longer a valid element, false otherwise.
Return true if it
references an element that is lower or equal than data
.
Otherwise (or if it references no longer a valid element) it returns false.
A tree is an abstract data type to represent the hierarchic nature of a structure with a set of connected nodes. Each node in the tree can be connected to many children, but must be connected to exactly one parent, except for the root node, which has no parent.
Define the tree name_t
and its associated methods as static inline
functions.
The tree will be composed of object of type type
, connected each other.
name
shall be a C identifier that will be used to identify the tree.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
Any insert (resp. remove) in the tree shall specific the insertion point
(resp. deleting point). To construct a tree, you start by creating the
root node (using the _set_root
method) and then insert new nodes from there.
Each insertion of node in the tree will return an iterator of the inserted
node. This can be used to construct quickly a full tree.
The oplist shall have at least the following operators (INIT_SET
& CLEAR
),
otherwise it won't generate compilable code.
The tree handles its own pool of nodes for the nodes.
It is called the capacity of the tree.
This pool of nodes will increase when needed by default.
However, in case of capacity increased, all the nodes of the tree may move in memory to accommodate the new need.
You may also request to reserve more capacity to avoid moving the items, and disable this auto-expand feature (in which a MEMORY_FAILURE
is raised).
There are several way to iterate over this container:
- Scan all nodes: first the parent then the children (pre-order walk).
- Scan all nodes: first the children then the parent (post-order walk).
- Scan the nodes of a sub-tree: first the parent then the children (pre-order walk of a sub-tree).
- Scan the nodes of a sub-tree: first the children then the parent (post-order walk of a sub-tree).
On insertion, all iterators remain valid. Except if it says otherwise, all functions accepting iterators expect a valid iterator (i.e. it references an existing node).
TRREE_DEF_AS
is the same as TREE_DEF
except the name of the types name_t
, name_it_t
are provided.
Define the oplist of the generic tree defined with name
and potentially oplist
.
If there is no given oplist, the basic oplist for basic C types is used.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the generic tree of type
.
Type of an iterator over this generic tree.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t tree)
void name_init_set(name_t tree, const name_t ref)
void name_init_move(name_t tree, name_t ref)
void name_set(name_t tree, const name_t ref)
void name_move(name_t tree, name_t ref)
void name_clear(name_t tree)
void name_reset(name_t tree)
size_t name_size(const name_t tree)
size_t name_capacity(const name_t tree)
void name_reserve(name_t tree, size_t capacity)
bool name_empty_p(const name_t tree)
void name_swap(name_t tree1, name_t tree2)
bool name_equal_p(const name_t tree1, const name_t tree2)
bool name_end_p(const name_it_t it)
bool name_it_equal_p(const name_it_t it1, const name_it_t it2)
const type *name_cref(name_it_t it)
type *name_ref(name_it_t it)
bool name_remove(it_t it)
void name_get_str(string_t str, const name_t tree, bool append)
bool name_parse_str(name_t tree, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t tree)
bool name_in_str(name_t tree, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
The following specialized methods are automatically created by the previous definition macro:
Disable the auto-resize feature of the tree (if lock is true), or enable it otherwise. By default, the feature is enabled. Locking a tree shall be done after reserving the maximum number of nodes that can be added by your tree, so that the returned pointers to the internal types won't move on inserting a new node.
Set the tree to a single root node and set this node to value
.
Set the tree to a single root node and set this node to the value
initialized with the given args.
The provided arguments shall therefore match one of the constructor provided
by the EMPLACE_TYPE
operator.
See emplace chapter for more details.
Insert a node up (resp. left, right, down and down) the given referenced iterator without initializing it.
It returns an iterator to the inserted node with non-initialized data.
The first thing to do after calling this function shall be to initialize the data
using the proper constructor of the object of type type
(the pointer can be get through name_ref
)
This enables using more specialized constructor than the generic copy one.
The user should use other the non _raw
function if possible rather than this one
as it is low level and error prone.
name_insert_down_raw
will move all children of the referenced node as children of the inserted children,
whereas name_insert_child_raw
will insert the node as the new first child of the referenced node.
Insert a node up (resp. left, right, down and down) the given referenced iterator and initialize it with a copy of value
.
It returns an iterator to the inserted node.
name_insert_down
will move all children of the referenced node as children of the inserted children,
whereas name_insert_child
will insert the node as the new first child of the referenced node.
Insert a node up (resp. left, right, down and down) the given referenced iterator and initialize it with value
by stealing as much resource from value
as possible (and destroy *value
)
It returns an iterator to the inserted node.
name_move_down
will move all children of the referenced node as children of the inserted children,
whereas name_move_child
will insert the node as the new first child of the referenced node.
Insert a node up (resp. left, right, down and down) the given referenced iterator
and initialize this node to the value initialized with the given args.
The provided arguments shall therefore match one of the constructor provided
by the EMPLACE_TYPE
operator.
See emplace chapter for more details.
It returns an iterator to the inserted node.
name_emplace_down
will move all children of the referenced node as children of the inserted children,
whereas name_emplace_child
will insert the node as the new first child of the referenced node.
Return a pointer to the type of the node which is
- up the given iterator,
- down the given iterator (i.e. the first child of the node)
- left the given iterator,
- right the given iterator,
respectively. It returns NULL if there is no such node.
Update the iterator to point to the node which is up (resp. down, left and right) the given iterator. Return true in case of success, false otherwise (as such node doesn't exist, the iterator remains unchanged).
Return true if it
references the root node, false otherwise.
Return true if it
references a non-leaf node, false otherwise.
Return true if it
references a leaf node, false otherwise.
Return the degree of the referenced node (aka the number of children). A leaf node has a degree of 0. This function is linear in the number of children of the node.
Return the depth of the referenced node (aka the number of nodes until reaching the root node). The root node has a depth of 0. This function is linear in the depth of the node.
Unlink the referenced node from the tree, so that the node is removed from the tree. All children of the removed node become children of the parent node. If the removed node is the root node, than the root node shall have only one child.
Return a reference to the internal type and give back ownership of the type:
you shall destroy the type (using CLEAR
or MOVE
methods) before calling any other tree functions
(as the memory area used by the node may be removed on inserting a new node)
You should use the remove service instead as it has the same semantics but it is safer as it perform the clear of the data.
Remove the referenced node including all its children.
See name_remove
for more details.
Return an iterator that references no valid node.
Set the iterator *it
to ref
.
Note
You can use the =
affectation operator of the C language to copy tree iterators too.
Return an iterator of the tree that will iterator on the tree in global pre-order walk (the root).
Update the iterator of the tree so that it references the next node in a global pre-order walk, or set it to invalid if the walk is finished.
Return an iterator of the tree that will iterator on the tree in global post-order walk
Update the iterator of the tree so that it references the next node in a global post-order walk, or set it to invalid if the walk is finished.
Return an iterator of the tree that will iterator on the tree in pre-order walk of the child nodes of the referenced one.
Update the iterator of the tree so that it references the next node in a local pre-order walk of the child nodes of the referenced one, or set it to invalid if the walk is finished (the sub tree is fully scanned).
The referenced iterator shall be the same as the one used to create the updated iterator (with name_it_subpre
).
Return an iterator of the tree that will iterator on the tree in post-order walk of the child nodes of the referenced one.
Update the iterator of the tree so that it references the next node in a local post-order walk of the child nodes of the referenced one, or set it to invalid if the walk is finished (the sub-tree is fully scanned).
The referenced iterator shall be the same as the one used to create the updated iterator (with name_it_subpost
).
Compute the Lowest Common Ancestor of the two iterators. Both iterators shall belong to the same tree.
Swap the node referenced by it1
and the node referenced by it2
in the tree.
If swapChild
is true, the children nodes perform the swap with their parent.
Otherwise, only the referenced nodes are swapped.
Sort the child of the node referenced by it1
in the order of the type.
This method is constructed if the basic type exports the CMP
type.
A priority queue is a queue where each element has a "priority" associated with it: an element with high priority is served before an element with low priority. It is currently implemented as a heap.
Define the priority queue name_t
and its associated methods
as static inline
functions.
The queue will be composed of object of type type
.
name
shall be a C identifier that will be used to identify the queue.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The CMP
operator is used to sort the queue so that the highest priority is the minimum.
The EQUAL
operator is used to identify an item on update or remove operations.
It is uncorrelated with the CMP
operator from the point of view of this container.
(i.e. EQUAL() == TRUE
is not equivalent to CMP() == 0
for this container)
This queue will push the object at the right place in the queue in function of their priority. A pop from this queue will always return the minimum of all objects within the queue (contrary to C++ which returns the maximum), and the front method will reference this object.
The oplist shall have at least the following operators (INIT
, INIT_SET
, SET
, CLEAR
, CMP
),
otherwise it won't generate compilable code.
The equal_p
, update
& erase
methods have a complexity of O(n)
due to the linear
search of the data and are only created if the EQUAL
method is defined.
Iteration over this container won't iterate from minimum to maximum but in an implementation define way that ensures that all items are accessed.
PRIOQUEUE_DEF_AS
is the same as PRIOQUEUE_DEF
except the name of the types name_t
, name_it_t
are provided.
Define the oplist of the PRIOQUEUE
defined with name
and potentially oplist
.
If there is no given oplist, the basic oplist for basic C types is used.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the priority queue of type
.
Type of an iterator over this priority queue.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t queue)
void name_clear(name_t queue)
void name_init_set(name_t queue, const name_t ref)
void name_set(name_t queue, const name_t ref)
void name_init_move(name_t queue, name_t ref)
void name_move(name_t queue, name_t ref)
void name_reset(name_t queue)
size_t name_size(const name_t queue)
bool name_empty_p(const name_t queue)
void name_swap(name_t queue1, name_t queue2)
void name_push(name_t queue, const type x)
const type *name_front(name_t queue)
void name_pop(type *dest, name_t queue)
bool name_equal_p(const name_t queue1, const name_t queue2)
bool name_erase(name_t queue, const type_t val)
void name_it(name_it_t it, name_t queue)
void name_it_last(name_it_t it, name_t queue)
void name_it_set(name_it_t it, const name_it_t ref)
void name_it_end(name_it_t it, name_t queue)
bool name_end_p(const name_it_t it)
bool name_last_p(const name_it_t it)
bool name_it_equal_p(const name_it_t it1, const name_it_t it2)
void name_next(name_it_t it)
void name_previous(name_it_t it)
const type *name_cref(name_it_t it)
void name_get_str(string_t str, const name_t queue, bool append)
bool name_parse_str(name_t queue, const char str[], const char **endp)
void name_out_str(FILE *file, const name_t queue)
bool name_in_str(name_t queue, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
void name_emplace[suffix](name_t queue, args...)
The following specialized methods are automatically created by the previous definition macro:
Change the priority of the data of the priority equals to old_val
(in EQUAL
sense)
to new_val
(increase or decrease priority).
This method has a complexity of O(n)
(due to linear search of the data).
This method is defined only if the EQUAL
method is defined.
This header implements different kind of fixed circular buffer.
A circular buffer (or ring buffer or circular queue) is a data structure using a single, bounded buffer as if its head was connected to its tail.
Define the buffer name_t
and its associated methods as static inline
functions.
A buffer is a fixed circular queue implementing a queue (or stack) interface.
It can be used to transfer message from multiple producer threads to multiple consumer threads.
This is done internally using a mutex and conditional waits
(if it is built with the BUFFER_THREAD_SAFE
option — default)
name
shall be a C identifier that will be used to identify the buffer.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The size
parameter defined the fixed size of the queue.
It can be 0. In this case, the fixed size is defined at initialization time only
and the needed objects to handle the buffer are allocated at initialization time too.
Otherwise the needed objects are embedded within the structure, preventing
any other allocations.
The size of the buffer shall be lower or equal than the maximum of the type int
.
Multiple additional policy can be applied to the buffer by performing a logical or of the following properties:
BUFFER_QUEUE
— define a FIFO queue (default),BUFFER_STACK
— define a stack (exclusive withBUFFER_QUEUE
),BUFFER_THREAD_SAFE
— define thread safe functions (default),BUFFER_THREAD_UNSAFE
— define thread unsafe functions (exclusive withBUFFER_THREAD_SAFE
),BUFFER_PUSH_INIT_POP_MOVE
— change the behavior ofPUSH
to push a new initialized object, andPOP
as moving this new object into the new emplacement (this is mostly used for performance reasons or to handle properly a shared_ptr semantic). In practice, it works as ifPOP
performs the initialization of the object.BUFFER_PUSH_OVERWRITE
—PUSH
overwrites the last entry if the queue is full instead of blocking,BUFFER_DEFERRED_POP
— do not consider the object to be fully popped from the buffer by calling the pop method until the call topop_deferred
; this enables to handle object that are in-progress of being consumed by the thread.
This container is designed to be used for synchronization inter-threads of data
(and the buffer variable should be a global shared one). A function tagged "thread safe"
is thread safe only if the container has been generated with the THREAD_SAFE
option.
The oplist shall have at least the following operators (INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
The PUSH
and POP
methods operate like push_blocking
and pop_blocking
when the blocking parameter
is true.
BUFFER_DEF_AS
is the same as BUFFER_DEF
except the name of the type name_t
is provided.
Example:
BUFFER_DEF(buffer_uint, unsigned int, 10, BUFFER_QUEUE|BUFFER_BLOCKING)
buffer_uint_t g_buff;
void f(unsigned int i) {
buffer_uint_init(g_buff, 10);
buffer_uint_push(g_buff, i);
buffer_uint_pop(&i, g_buff);
buffer_uint_clear(g_buff);
}
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the buffer.
The following methods of the generic interface are defined (See generic interface for details):
void name_clear(buffer_t buffer) /* Not thread safe */
void name_reset(buffer_t buffer) /* Thread safe */
bool name_empty_p(const buffer_t buffer) /* Thread safe */
size_t name_size(const buffer_t buffer) /* Thread safe */
size_t name_capacity(const buffer_t buffer) /* Thread safe */
bool name_push(buffer_t buffer, const type data)
bool name_pop(type *data, buffer_t buffer)
The following specialized methods are automatically created by the previous definition macro:
Initialize the buffer buffer
for size
elements.
The size
argument shall be the same as the one used to create the buffer
or the one used to create the buffer was 0
.
The size of the buffer shall be lower or equal than UINT_MAX
.
This function is not thread safe.
Return true if the buffer is full, false otherwise. This function is thread safe if the buffer was built thread safe.
If the buffer is built with the BUFFER_PUSH_OVERWRITE
option,
this function returns the number of elements that have been overwritten
and thus discarded.
If the buffer was not built with the BUFFER_PUSH_OVERWRITE
option,
it returns 0.
Push the object data
in the buffer buffer
,
waiting for an empty room if blocking
is true (performing a blocking wait)
Returns true if the data was pushed, false otherwise.
Always return true if the buffer is blocking.
This function is thread safe if the buffer was built thread safe.
Pop from the buffer buffer
into the object *data
,
waiting for a data if blocking
is true.
If the buffer is built with the BUFFER_PUSH_INIT_POP_MOVE
option,
the object pointed by data
shall be uninitialized
as the pop function will perform a quick initialization of the object
(using an INIT_MOVE
operator)
, otherwise it shall be an initialized object (the pop function will
perform a SET
operator).
If the buffer is built with the BUFFER_DEFERRED_POP
option,
the object is still considered being present in the queue until
a call to name_pop_release.
Returns true if a data was popped, false otherwise. Always return true if the buffer is blocking. This function is thread safe if the buffer was built thread safe.
If the buffer is built with the BUFFER_DEFERRED_POP
option,
the object being popped is considered fully release (freeing a
space in the queue).
Otherwise it does nothing.
Define the MPMC queue name_t
and its associated methods as static inline
functions.
A MPMC queue is a fixed circular queue implementing a queue (or stack) interface.
It can be used to transfer message from Multiple Producer threads to Multiple Consumer threads.
This queue is not strictly lock free but has
a lot of the properties of such algorithms.
The size is specified only at run-time and shall be a power of 2.
name
shall be a C identifier that will be used to identify the queue.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
An additional policy can be applied to the buffer by performing a logical or of the following properties:
BUFFER_QUEUE
— define a FIFO queue (default),BUFFER_PUSH_INIT_POP_MOVE
— change the behavior ofPUSH
to push a new initialized object, andPOP
as moving this new object into the new emplacement (this is mostly used for performance reasons or to handle properly a shared_ptr semantic). In practice, it works as ifPOP
performs the initialization of the object.
This container is designed to be used for easy synchronization inter-threads in a context of very fast communication (the variable should be a global shared one). There should not have much more threads using this queue than they are available hardware cores due to the only partial protection on Context-switch Immunity of this structure (This can happen only if you abuse massively the number of threads vs the number of cores).
The oplist shall have at least the following operators (INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
The size
method is thread safe but may return a size greater than
the size of the queue in some race condition.
QUEUE_MPMC_DEF_AS
is the same as QUEUE_MPMC_DEF
except the name of the type name_t
is provided.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the circular queue.
The following methods of the generic interface are defined (See generic interface for details):
void name_clear(buffer_t buffer)
bool name_empty_p(const buffer_t buffer) /* Thread safe */
size_t name_size(const buffer_t buffer) /* Thread safe */
size_t name_capacity(const buffer_t buffer) /* Thread safe */
The following specialized methods are automatically created by the previous definition macro:
Initialize the buffer buffer
with size
elements.
The size
argument shall be a power of two greater than 0, and less than UINT_MAX
.
This function is not thread safe.
##### bool name_full_p(const buffer_t buffer)
Return true if the buffer is full, false otherwise. This function is thread safe.
Push the object data
in the buffer buffer
if possible.
Returns true if the data was pushed, false otherwise
(buffer full or unlikely data race).
This function is thread safe.
Pop from the buffer buffer
into the object *data
if possible.
If the buffer is built with the BUFFER_PUSH_INIT_POP_MOVE
option,
the object pointed by data
shall be uninitialized
as the pop function will perform a quick initialization of the object
(using an INIT_MOVE
operator)
, otherwise it shall be an initialized object (the pop function will
perform a SET
operator).
Returns true if a data was popped, false otherwise (buffer empty or unlikely data race). This function is thread safe.
Define the SPSC queue name_t
and its associated methods as static inline
functions.
A SPSC queue is a fixed circular queue implementing a queue (or stack) interface.
It can be used to transfer message from a Single Producer thread to a Single Consumer thread.
This is done internally using lock-free objects.
It is more specialized than QUEUE_MPMC_DEF
and as such, is faster.
The size
is specified only at run-time and shall be a power of 2.
name
shall be a C identifier that will be used to identify the queue.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
An additional policy can be applied to the buffer by performing a logical or of the following properties:
BUFFER_QUEUE
— define a FIFO queue (default),BUFFER_PUSH_INIT_POP_MOVE
— change the behavior ofPUSH
to push a new initialized object, andPOP
as moving this new object into the new emplacement (this is mostly used for performance reasons or to handle properly a shared_ptr semantic). In practice, it works as ifPOP
performs the initialization of the object.
This container is designed to be used for easy synchronization inter-threads in a context of very fast communication (the variable should be a global shared one).
The oplist shall have at least the following operators (INIT
, INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
QUEUE_SPSC_DEF_AS
is the same as QUEUE_MPMC_DEF
except the name of the type name_t
is provided.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the circular queue.
The following methods of the generic interface are defined (See generic interface for details):
void name_clear(buffer_t buffer)
bool name_empty_p(const buffer_t buffer) /* Thread safe */
size_t name_size(const buffer_t buffer) /* Thread safe */
size_t name_capacity(const buffer_t buffer) /* Thread safe */
The following specialized methods are automatically created by the previous definition macro:
Initialize the buffer buffer
with size
elements.
The size
argument shall be a power of two greater than 0, and less than UINT_MAX
.
This function is not thread safe.
Return true if the buffer is full, false otherwise. This function is thread safe.
Push the object data
in the buffer buffer
if possible.
Returns true if the data was pushed, false otherwise (buffer full).
This function is thread safe.
Push & move the object *data
in the buffer buffer
if possible.
Returns true if the data was pushed, false otherwise (buffer full).
Afterwards *data
is cleared (destructor) if true is returned.
This function is thread safe.
Push the object data
in the buffer buffer
discarding the oldest data if needed.
This function is thread safe.
Push as much objects from the array data
in the buffer buffer
as possible,
starting from the object at index 0 to the object at index n-1
.
Returns the number of objects effectively pushed (it depends on the free size of the queue)
This function is thread safe.
Pop from the buffer buffer
into the object *data
if possible.
If the buffer is built with the BUFFER_PUSH_INIT_POP_MOVE
option,
the object pointed by data
shall be uninitialized
as the pop function will perform a quick initialization of the object
(using an INIT_MOVE
operator)
, otherwise it shall be an initialized object (the pop function will
perform a SET
operator).
Returns true if a data was popped, false otherwise (buffer empty or unlikely data race). This function is thread safe.
Pop from the buffer buffer
as many objects as possible to fill in tab
and at most n
.
If the buffer is built with the BUFFER_PUSH_INIT_POP_MOVE
option,
the object pointed by data
shall be uninitialized
as the pop function will perform a quick initialization of the object
(using an INIT_MOVE
operator)
, otherwise it shall be an initialized object (the pop function will
perform a SET
operator).
It returns the number of objects popped. This function is thread safe.
This header is for created snapshots.
A snapshot is a mechanism enabling a reader thread and a writer thread, working at different speeds, to exchange messages in a fast, reliable and thread safe way (the message is always passed atomically from a thread point of view) without waiting for synchronization. The consumer thread always accesses to the latest published data of the producer thread.
It is implemented in a fast way as the writer directly writes the message in the buffer that will be passed to the reader (avoiding copy of the buffer) and a simple exchange of index is sufficient to handle the switch.
This container is designed to be used for easy synchronization inter-threads.
This is linked to shared atomic register in the literature and snapshot names vector of such registers where each element of the vector can be updated separately. They can be classified according to the number of producers/consumers:
- SPSC (Single Producer, Single Consumer),
- MPSC (Multiple Producer, Single Consumer),
- SPMC (Single Producer, Multiple Consumer),
- MPMC (Multiple Producer, Multiple Consumer),
The provided containers by the library are designed to handle huge
structure efficiently and as such deal with the memory reclamation needed to handle them.
If the data you are sharing are supported by the atomic header (like bool or integer),
using atomic_load
and atomic_store
is a much more efficient and simple way
to do even in the case of MPMC.
Define the snapshot name ## _t
(or name_t
) and its associated methods as static inline
functions.
Only a single reader thread and a single writer thread are supported.
It is a SPSC atomic shared register.
In practice, it is implemented using a triple buffer (lock-free).
name
shall be a C identifier that will be used to identify the snapshot.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The oplist shall have at least the following operators (INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
Example:
SNAPSHOT_DEF(snapshot_uint, unsigned int)
snapshot_uint_t message;
void f(unsigned int i) {
unsigned *p = snapshot_uint_get_write_buffer(message);
*p = i;
snapshot_uint_write(message);
}
unsigned int g(void) {
unsigned *p = snapshot_uint_read(message);
return *p;
}
SNAPSHOT_SPSC_DEF_AS
is the same as SNAPSHOT_SPSC_DEF
except the name of the type name_t
is provided.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the circular queue.
The following methods of the generic interface are defined (See generic interface for details) but none is thread safe:
void name_init(snapshot_t snapshot)
void name_clear(snapshot_t snapshot)
void name_init_set(snapshot_t snapshot, const snapshot_t org)
void name_set(snapshot_t snapshot, const snapshot_t org)
void name_init_move(snapshot_t snapshot, snapshot_t org)
void name_move(snapshot_t snapshot, snapshot_t org)
The following specialized methods are automatically created by the previous definition macro:
Publish the "in-progress" data of the snapshot snap
so that the read
thread can have access to the data.
It returns the pointer to the new "in-progress" data buffer
of the snapshot (which is not yet published but will be published
for the next call of name_write).
This function is thread-safe and performs atomic operation on the snapshot.
Get access to the last published data of the snapshot snap
.
It returns the pointer to the data.
If a publication has been performed since the last call to name_read
a new pointer to the data is returned.
Otherwise the previous pointer is returned.
This function is thread-safe and performs atomic operation on the snapshot.
Return true if a new publication is available since the last time it was read. This function is thread-safe and performs atomic operation on the snapshot.
Return a pointer to the active "in-progress" data of the snapshot snap
.
It is the same as the last return from name_write.
This function is thread-safe and performs atomic operation on the snapshot.
Return a pointer to the last already read published data of the snapshot snap
.
It is the same as the last return from name_read.
It doesn't perform any switch of the data that has to be read.
This function is thread-safe and performs atomic operation on the snapshot.
Define the snapshot name ## _t
(or name_t
) and its associated methods as static inline
functions.
A snapshot is an atomic shared register where only the latest state is
important and accessible: it is like a global variable of type type
but ensuring integrity and coherency of the data across multiple threads.
One single writer and multiple (=N
) readers are supported.
In practice, it is implemented using a N+2
buffer (lock-free).
name
shall be a C identifier that will be used to identify the snapshot.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The oplist shall have at least the following operators (INIT
, INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
SNAPSHOT_SPMC_DEF_AS
is the same as SNAPSHOT_SPMC_DEF
except the name of the type name_t is provided.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the circular queue.
The following methods of the generic interface are defined (See generic interface for details):
void name_clear(snapshot_t snapshot)
The following specialized methods are automatically created by the previous definition macro:
Initialize the communication snapshot snapshot
with numReaders
reader threads.
numReaders
shall be less than 2046
.
This function is not thread safe.
Publish the "in-progress" data of the snapshot snap
so that the read
threads can have access to the data.
It returns the pointer to the new "in-progress" data buffer
of the snapshot (which is not yet published but will be published
for the next call of name_write).
This function is thread-safe and performs atomic operation on the snapshot.
Get access to the last published data of the snapshot snap
.
It returns the pointer to the data.
If a publication has been performed since the last call to name_read_start
a new pointer to the data is returned.
Otherwise the previous pointer is returned.
It marks the data has being reserved by the thread, so afterwards, using the pointer is safe until the end of the reservation. This function is thread-safe and performs atomic operation on the snapshot.
For each call to name_read_start
a matching call to
name_read_end
shall be performed by the same thread before
any new call to name_read_start.
End the reservation of the data old
.
This function is thread-safe and performs atomic operation on the snapshot.
Return a pointer to the active "in-progress" data of the snapshot snap
.
It is the same as the last return from name_write.
This function is thread-safe and performs atomic operation on the snapshot.
Define the snapshot name ## _t
(or name_t
) and its associated methods as static inline
functions.
A snapshot is an atomic shared register where only the latest state is
important and accessible: it is like a global variable of type type
but ensuring integrity and coherency of the data across multiple threads.
Multiple (=M
) writers and multiple (=N
) readers are supported.
In practice, it is implemented using a M+N+2
buffer (lock-free)
by avoiding copying the data.
name
shall be a C identifier that will be used to identify the snapshot.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The oplist shall have at least the following operators (INIT
, INIT_SET
, SET
and CLEAR
),
otherwise it won't generate compilable code.
SNAPSHOT_MPMC_DEF_AS
is the same as SNAPSHOT_MPMC_DEF
except the name of the type name_t is provided.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the circular queue.
The following methods of the generic interface are defined (See generic interface for details):
void name_clear(snapshot_t snapshot)
Initialize the communication snapshot snapshot
with numReaders
reader threads
and numWriters
writer threads.
The sum of numReaders
and numWriters
shall be less than 2046
.
This function is not thread safe.
Return the current "in-progress" data of the snapshot snap
so that the writer thread can update this data.
This function is thread-safe and performs atomic operation on the snapshot.
Mark the provided "in-progress" data
of the snapshot snap
as available for the reader threads: this will be the new seen data.
This function is thread-safe and performs atomic operation on the snapshot.
Get access to the last published data of the snapshot snap
.
It returns the pointer to the data.
If a publication has been performed since the last call to name_read_start
a new pointer to the data is returned.
Otherwise, the previous pointer is returned.
It marks the data has being reserved by the thread, so afterwards, using the pointer is safe until the end of the reservation. This function is thread-safe and performs atomic operation on the snapshot.
For each call to name_read_start
a matching call to
name_read_end
shall be performed by the same thread before
any new call to name_read_start
.
End the reservation of the data old
.
This function is thread-safe and performs atomic operation on the snapshot.
This header is for creating shared pointer. A shared pointer is a smart pointer that retains shared ownership of an object. Several shared pointers may own the same object, sharing ownership of this object. It is a mechanism to keep tracks of all registered users of an object and performs an automatic destruction of the object only when all users release their need on this object. As shared pointer owns no object if it is a null constant pointer.
In addition, M*LIB shared pointer provides methods that protects concurrent access of the object by different threads,
transforming a standard container (LIST
, ARRAY
, DICT
, DEQUE
, ...)
into an equivalent container but compatible with concurrent access by different threads.
In practice, it puts a mutex lock to access the container.
As such it is quite generic. However, it is less efficient than containers
specially tuned for multiple threads.
Finally, M*LIB shared pointer provides abstraction between declaration and definition, enabling to decouple then.
There are two kinds of shared pointer:
- weak shared pointer (don't support thread concurrency)
- shared pointer (support for multiple threads both for the counter and lock for the data)
with 3 flavors:
_DECL
: for header files, declare only the functions and opaque type._DEF_EXTERN
: for source files, define the functions (to be used with DECL)_DEF
: for header/source files, define the function as static inline.
You can provide or not the name of the shared pointer (_AS
). Therefore we got 12 macros to generate the shared pointers.
There are also two oplists for this object, depending on how you want to handle it:
- one to handle the shared pointer itself (ie. the copy method creates a new owner of the same shared data),
- one to handle the data behind the shared pointer (ie. the copy method create a new shared data)
There are designed to work with buffers created with policy BUFFER_PUSH_INIT_POP_MOVE
so that we can send a shared pointer across multiple threads, and destroying the object
only when all threads have terminated their job.
Two level of API are created:
- the public one, created by
_DECL
generation macro to be used by everyone, - the private one, created by
_DEF
and_DEF_EXTERN
generation macro to be used by the implementation only.
The only mandatory operator is CLEAR
.
Declare the shared pointer name_t *
as a C pointer to an opaque structure and
declare also the associated public methods.
name
shall be a C identifier that will be used to identify the shared pointer.
It will be used to create the type and functions to handle the container.
This declaration shall be done once per name and per compilation unit.
A corresponding definition SHARED_PTR_DEF_EXTERN
(resp. SHARED_PTR_DEF_EXTERN_AS
, SHARED_WEAK_PTR_DEF_EXTERN
and SHARED_WEAK_PTR_DEF_EXTERN_AS
)
shall be done once in one compilation unit.
SHARED_PTR_DECL
and SHARED_PTR_DECL_AS
create a shared pointer qualified as strong since the tracking of ownership is atomic and the destruction of the object is thread safe. In addition, there is also a lock protecting the integrity of the data.
They shall be used to track ownership of an object in multi thread program.
SHARED_WEAK_PTR_DECL
and SHARED_WEAK_PTR_DECL_AS
create a shared pointer qualified as weak since the tracking of ownership is non atomic and the destruction of the object is not thread safe. There is also no lock protecting the integrity of the data.
They shall be used to track ownership of an object in a single thread program.
The given oplist
is mandatory but it might be a simplified oplist (even if a real oplist works).
This parameter is only used for two reasons:
- to identify if a method of a shared pointer exists: for example, the method
name_new()
is created only if theINIT
operator is provided. - to provides the sub-types of the type: the key type, the value type, the subtype, the emplace types, the oplist of the subtype, the emplace types of the subtype, ...
As such, a simplified oplist like (INIT(1), INIT_SET(1))
is valid for the declaration only.
Theses macros should be put typically in header files.
SHARED_PTR_DECL_AS
(resp. SHARED_WEAK_PTR_DECL_AS
) is the same as SHARED_PTR_DECL
(resp. SHARED_WEAK_PTR_DECL
)
except the type of the shared pointer name_type
is given as parameter instead of defined as name_t
.
Define the shared pointer name_t *
as a C pointer to an opaque structure that encapsulate the access to the C type
type
which oplist is oplist
and the associated public and private methods as external linkage.
oplist
parameter is optional. If not present, it will look for a globaly registered oplist.
It shall match with the oplist
argument given to the corresponding _DECL
macro with the same name
,
meaning it shall provide at least all operators of the oplist provided to the corresponding _DECL
macro
and all operator types shall be the same.
It shall be done once in one compilation unit. Theses macros should be put typically in one source file.
See SHARED_PTR_DECL
for additional information.
Define the shared pointer name_t *
as a C pointer to an opaque structure that encapsulate the access to the C
object (aka the shared object) of type
type
which oplist is oplist
and the associated public and private methods as static inline.
oplist
parameter is optional. If not present, it will look for a globaly registered oplist.
This definition shall be done once per name and per compilation unit.
See SHARED_PTR_DECL
for additional information.
Define the oplist of the shared pointer name
view as a pointer (i.e. share ownership).
The parameter oplist
can be a simplified oplist of the encapsulated type (See _DECL
macros).
Define the oplist of the shared pointer name
view as an object (i.e. copy object).
The parameter oplist
can be a simplified oplist of the encapsulated type (See _DECL
macros).
Example:
#include "m-shared-ptr.h"
#include "m-buffer.h"
// To put in one header
// Define a shared data that supports pushing/popping integer
#define OPL (INIT(1),INIT_SET(1),SET(1),CLEAR(1),SUBTYPE(int),PUSH(1),POP(1))
SHARED_PTR_DECL(shared_data, OPL)
// Define a communication buffer of this shared data that supports sending/receiving the shared pointer
#define OPL_COMM (INIT(1),INIT_SET(1),SET(1),CLEAR(1),SUBTYPE(shared_data_t*),PUSH(1),POP(1))
SHARED_PTR_DECL(comm_buffer, OPL_COMM)
extern comm_buffer_t *comm1, *comm2;
// A C file using the header
void f(void) {
// Create a new shared data
shared_data_t *p = shared_data_new();
// Push some data in it
shared_data_push(p, 23);
shared_data_push(p, 32);
// And send it through 2 channels of communication
comm_buffer_push(comm1, p); // shared_data_acquire is done by the buffer to acquite ownership of the data
comm_buffer_push(comm2, p);
// Release our ownership of the data
shared_data_clear(p);
}
// To put in one C file
#include "m-array.h"
// Define the shared_data as an encapsulation of an array of int
ARRAY_DEF(array, int)
SHARED_PTR_DEF_EXTERN(shared_data, array_t, ARRAY_OPLIST(array, M_BASIC_OPLIST))
// Define the communication buffer as an encapsulation of a 10-size circular buffer over the shared_data
BUFFER_DEF(buffer, shared_data_t *, 10, BUFFER_QUEUE|BUFFER_PUSH_INIT_POP_MOVE, SHARED_PTR_OPLIST(shared_data, OPL))
SHARED_PTR_DEF_EXTERN(comm_buffer, buffer_t, BUFFER_OPLIST(buffer, SHARED_PTR_OPLIST(shared_data, OPL)))
comm_buffer_t *comm1, *comm2;
The following types are automatically defined by the previous definition macro if not provided by the user:
Opaque structure of the shared object, for which name_t *
is a shared pointer.
The public interface is declared and defined to be used by the user of the shared object / shared pointer.
Create a new shared object initialized with its INIT
operator and return a shared pointer to it.
This method is created only if the INIT
operator is defined.
Create a new shared object initialized with the content of *src
, creating effectively a clone.
This method is created only if the INIT_SET
operator is defined.
Create a new shared object initialized with the content of <emplace_args>
.
This method is created only if the EMPLACE_TYPE
operator is defined.
Copy into the shared object *out
the content of *src
, creating effectively a copy if the shared pointers are not the same.
This method is created only if the SET
operator is defined.
Acquire a new ownership of the pointer, returning the same pointer but with one more registered owner.
Release the ownership of the pointer.
If there is no longer any owner of the shared data, it is destroyed using its CLEAR
method and the allocated memory freed.
Alias of name_release
Release the current ownership of *dst
and set *dst
as a new owner of *out
(Copy of pointer)
Swap the content of the shared objects *a
and *b
This method is created only if the SWAP
operator is defined.
Reset the content of the shared object *a
.
This method is created only if the RESET
operator is defined.
Test if the shared object *a
is empty (return true) or not (return false).
This method is created only if the EMPTY_P
operator is defined.
Test if the shared object *a
is full (return true) or not (return false).
This method is created only if the FULL_P
operator is defined.
Return the number of elements of the shared object *a
This method is created only if the GET_SIZE
operator is defined.
Test if the shared objects *a
and *b
are equal (return true) or not (return false).
This method is created only if the EQUAL
operator is defined.
Compare the order of the shared objects *a
and *b
, returning their relative order.
This method is created only if the CMP
operator is defined.
Return the hash of the shared object *a
.
This method is created only if the HASH
operator is defined.
Compute in *a
the sum of *b
and *c
This method is created only if the ADD
operator is defined.
Compute in *a
the difference of *b
and *c
This method is created only if the SUB
operator is defined.
Compute in *a
the product of *b
and *c
This method is created only if the MUL
operator is defined.
Compute in *a
the dividend of *b
and *c
This method is created only if the DIV
operator is defined.
Append in *a
all elements of *b
and reset *b
.
This method is created only if the SPLICE
operator is defined.
a
and b
shall reference different objects.
Set *val
to the value associated to the key key
in the shared object *a
Return true in this case. If there is no association for key
, return false
if the called GET_KEY
operator supports returning NULL for the encapsulated container.
This method is created only if the GET_KEY
operator is defined.
Set *val
to the value associated to the key key
in the shared object *a
,
creating a new entry for key
if needed.
This method is created only if the GET_SAFE_KEY
operator is defined.
Set the association of key
to val
in *a
This method is created only if the SET_KEY
operator is defined.
Erase the association of key
in *a
if it exists (return true in this case).
Otherwise return false.
This method is created only if the ERASE_KEY
operator is defined.
Push in *a
the element el
, waiting forever for the container to be not full if needed.
This method is created only if the PUSH
operator is defined.
Push in *a
the element *el
, stealing as much resource from it as possible, waiting forever for the container to be not full if needed.
Afterwards *el
is cleared.
This method is created only if the PUSH_MOVE
operator is defined.
Try to push in *a
the element constructed from the arguments args
if possible, waiting forever for the container to be not full if needed.
This method is created only if the PUSH_MOVE
operator and the EMPLACE_TYPE
of the sub_type (within the sub-oplist operator OPLIST
) are defined.
Try to push in *a
the element el
,if it is not full (return true in this case).
Return false otherwise (cannot push element)
This method is created only if the PUSH
operator is defined.
Try to push in *a
the element *el
,if the container is not full and clear *el
(return true in this case).
Return false otherwise (cannot push element) and *el
is still initializated.
This method is created only if the PUSH_MOVE
operator is defined.
Try to push in *a
the element constructed from the arguments args
, if the container is not full (return true in this case).
Return false otherwise (cannot push element).
This method is created only if the PUSH_MOVE
operator and the EMPLACE_TYPE
of the sub_type (within the sub OPLIST
) are defined.
Pop the element from *a
and set *el
with it, waiting forever for an element to be available.
This method is created only if the POP
operator is defined.
Pop the element from *a
and initialize *el
with it, stealing as much resource as possible, waiting forever for an element to be available.
This method is created only if the POP_MOVE
operator is defined.
Pop the element from *a
and set *el
with it if an element to be available (return true in this case).
Otherwise return false.
This method is created only if the POP
operator is defined.
Pop the element from *a
and initialize *el
with it, stealing as much resource as possible, if an element is available (return true in this case).
Otherwise return false.
This method is created only if the POP_MOVE
operator is defined.
Apply the callback callback
to all elements of the container *a
from front to back.
The callback may modify the given element if possible.
data
is a user parameter given to the callback at user convenience.
If the callback returns a non null argument, the function stops and returns immediately with this error code.
This method is created only if the IT_FIRST
, IT_NEXT
and IT_REF
operators are defined.
Apply the callback callback
to all elements of the container *a
from front to back.
The callback shall not modify the given element.
data
is a user parameter given to the callback at user convenience.
If the callback returns a non null argument, the function stops and returns immediately with this error code.
This method is created only if the IT_FIRST
, IT_NEXT
and IT_CREF
operators are defined.
Apply the callback callback
to all elements of the container *a
from back to front.
The callback may modify the given element if possible.
data
is a user parameter given to the callback at user convenience.
If the callback returns a non null argument, the function stops and returns immediately with this error code.
This method is created only if the IT_LAST
, IT_PREVIOUS
and IT_REF
operators are defined.
Apply the callback callback
to all elements of the container *a
from back to front.
The callback shall not modify the given element.
data
is a user parameter given to the callback at user convenience.
If the callback returns a non null argument, the function stops and returns immediately with this error code.
This method is created only if the IT_LAST
, IT_PREVIOUS
and IT_CREF
operators are defined.
Output *a
into the FILE *f
This method is created only if the OUT_STR
operator is defined.
Read *a
from the FILE f
This method is created only if the IN_STR
operator is defined.
Output *a
into the string str
, appending it if append
is true.
This method is created only if the GET_STR
operator is defined.
Set *a
to the value read from the string str
.
*endptr
is set to the end of the parsing in the string if endptr
is not null.
This method is created only if the PARSE_STR
operator is defined.
Output *a
into the serial object serial
.
This method is created only if the OUT_SERIAL
operator is defined.
Set *a
to the value read from the serial object serial
.
This method is created only if the IN_SERIAL
operator is defined.
The private interface is only availale for the implementation. It is used by the implementator of the shared object to provide additional, more specialized, functions to its user for the shared object.
This function shall be called from a custom constructor to initialize the part of the shared object not linked to the encapsulated type. It initialize the internal locks and counters of the opaque structure of the object.
Enter the lock of the object for reading its state (The object shall hold no lock). This lock is non recursive. An object shall not be at the same time in read lock and in write lock (it is exclusive).
Wait for the object to receive new data. It shall be called within the read lock.
Leave the lock of the object for reading its state.
It shall be called after name_read_lock
Enter the lock of the object for writing its state (The object shall hold no lock). This lock is non recursive. An object shall not be at the same time in read lock and in write lock (it is exclusive).
Wait for the object to create some more space to store more data. It shall be called within the write lock.
Signal to the threads that some new data have been appended in the object. It shall be called within the write lock.
Signal to the threads that some space to store more data have been created. It shall be called within the write lock.
Leave the lock of the object for writing its state.
It shall be called after name_write_lock
Return a pointer to the encapsulated object.
Return a constant pointer to the encapsulated object.
Create a new shared object initialized with its INIT_SET
operator on the given src
and return a shared pointer to it.
This method is created only if the INIT_SET
operator is defined.
Note
This header is obsolete: M-SHARED-PTR should be used instead.
This header is for creating shared pointer. A shared pointer is a smart pointer that retains shared ownership of an object. Several shared pointers may own the same object, sharing ownership of an object.
Define the shared pointer name_t
and its associated methods as static inline
functions.
A shared pointer is a mechanism to keep tracks of all registered users of an object
and performs an automatic destruction of the object only when all users release
their need on this object.
name
shall be a C identifier that will be used to identify the shared pointer.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The tracking of ownership is atomic and the destruction of the object is thread safe.
The object oplist is expected to have at least the following operators (CLEAR
to clear the object and DEL
to free the allocated memory).
There are designed to work with buffers with policy BUFFER_PUSH_INIT_POP_MOVE
to send a shared pointer across multiple threads.
If the type is an incomplete type, you need to disable the INIT
operator and not defined the EMPLACE_TYPE
operator in the given oplist like this:
(INIT(0), CLEAR(API_2(incomplete_clear))) )
SHARED_PTR_DEF_AS
is the same as SHARED_PTR_DEF
except the name of the type name_t
is provided.
Example:
SHARED_PTR_DEF(shared_mpz, mpz_t, (CLEAR(mpz_clear)))
void f(void) {
shared_mpz_t p;
mpz_t z;
mpz_init(z);
shared_mpz_init2 (p, z);
buffer_uint_push(g_buff1, p);
buffer_uint_push(g_buff2, p);
shared_mpz_clear(p);
}
Theses are the same as SHARED_PTR_DEF
/ SHARED_PTR_DEF_AS
except that they are not thread safe.
See SHARED_PTR_DEF
for other details.
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the shared pointer.
The following specialized methods are automatically created by the previous definition macro:
Initialize the shared pointer shared
to represent the NULL pointer
(no object is therefore referenced).
Initialize the shared pointer shared
to reference the object *data
and takes ownership of this object.
User code shall not use *data
(or any pointer to it) anymore
as the shared pointer gets the exclusive ownership of the object.
Initialize the shared pointer shared
to a new object of type type
.
The default constructor of type is used to initialize the object.
This method is only created if the INIT
method is provided.
Initialize the shared pointer shared
to the same object than the one
pointed by src
(sharing ownership).
This function is thread safe from src
point of view.
Initialize the shared pointer shared
to the new object initialized with args.
The provided arguments shall therefore match one of the constructor provided
by the EMPLACE_TYPE
operator.
There is one generated method per suffix defined in the EMPLACE_TYPE
operator,
and the suffix
in the generated method name corresponds to the suffix defined
in the EMPLACE_TYPE
operator (it can be empty).
This method is created only if the EMPLACE_TYPE
operator is provided.
See emplace chapter for more details.
It is equivalent to the C++ make_shared.
Return true if shared doesn't reference any object (i.e. is the NULL pointer).
Clear the shared pointer (destructor): the shared pointer loses its ownership of the object and it destroys the shared object if no longer any other shared pointers own it. This function is thread safe.
shared
loses ownership of its shared object and destroys it
if no longer any other shared pointers own it.
Then it makes the shared pointer shared
references no object (NULL)
(it doesn't reference its object any-longer and loses its ownership of it).
This function is thread safe.
shared
loses ownership of its object and destroy it
if no longer any other shared pointers own it.
Then it sets the shared pointer shared
to the same object
than the one pointed by src
(sharing ownership).
This function is thread safe.
Move the shared pointer from the initialized src
to shared
.
shared
loses ownership of its object and destroy it
if no longer any other shared pointers own it.
Then it moves the shared pointer from the initialized src
to shared
.
Swap the references of the objects owned by the shared pointers shared1
and shared2
.
Return true if both shared pointers own the same object.
Return a constant pointer to the shared object owned by the shared pointer. The pointer shall be kept only until another use of shared pointer method. Keeping the pointer otherwise is undefined behavior.
Return a pointer to the shared object pointed by the shared pointer. The pointer shall be kept only until another use of shared pointer method. Keeping the pointer otherwise is undefined behavior.
TODO: Document shared resource
Note
This header is obsolete: M-SHARED-PTR should be used instead.
This header is for creating intrusive shared pointer.
Extend the definition of the structure of an object of type type
by adding the
necessary interface to handle it as a shared pointer named name
.
It shall be put within the structure definition of the object (See example).
Provide the static initialization value of an object defined with a
ISHARED_PTR_INTERFACE
extra fields. It shall be used only for global
variables with the _init_once function.
Usage (provided that the interface is used as the first element of the structure):
struct mystruct variable = { ISHARED_PTR_STATIC_INIT(ishared_double, struct mystruct) };
Provide the static initialization value of an object defined with a
ISHARED_PTR_INTERFACE
extra fields. It shall be used only for global
variables with the _init_once function.
It uses designated initializers to set the right fields.
Usage:
struct mystruct variable = {ISHARED_PTR_STATIC_DESIGNATED_INIT(ishared_double, struct mystruct) };
Define the associated methods to handle the shared pointer named name
as static inline
functions.
A shared pointer is a mechanism to keep tracks of all users
of an object
and performs an automatic destruction of the object whenever all users release
their need on this object.
The destruction of the object is thread safe and occurs when the last user of the object releases it. The destruction of the object implies:
- calling the
CLEAR
operator to clear the object, - calling the
DEL
operator to release the memory used by the object (if the method has not been disabled).
The object oplist is expected to have the following operators (CLEAR
and DEL
),
otherwise default methods are used. If there is no given oplist, the default
operators are also used. The created methods will use the operators to init, set
and clear the contained object.
There are designed to work with buffers with policy BUFFER_PUSH_INIT_POP_MOVE
to send a shared pointer across multiple threads.
It is recommended to use the intrusive shared pointer over the standard one if possible. They are faster and cleaner.
The default is to use heap allocated entities, which are allocated by NEW
and
freed by DEL
.
It can be used for statically allocated entities. However, in this case,
you shall disable the operator NEW
and DEL
when expanding the oplist
so that the destruction doesn't try to free the objects, like this:
(NEW(0), DEL(0))
NEW
and DEL
operators shall be either both defined, or both disabled.
ISHARED_PTR_DEF_AS
is the same as ISHARED_PTR_DEF
except the name of the type name_t
is provided.
Example (dynamic):
typedef struct mystruct_s {
ISHARED_PTR_INTERFACE(ishared_mystruct, struct mystruct_s);
char *message;
} mystruct_t;
static inline void mystruct_init(mystruct_t *p) { p->message = NULL; }
static inline void mystruct_clear(mystruct_t *p) { free(p->message); }
ISHARED_PTR_DEF(ishared_mystruct, mystruct_t, (INIT(mystruct_init), CLEAR(mystruct_clear M_IPTR)))
void f(void) {
mystruct_t *p = ishared_mystruct_init_new();
p->message = strdup ("Hello");
buffer_mystruct_push(g_buff1, p);
buffer_mystruct_push(g_buff2, p);
ishared_mystruct_clear(p);
}
The following types are automatically defined by the previous definition macro if not provided by the user:
It will define name_t as a pointer to shared counted object. This is a synonymous to a pointer to the object.
The following specialized methods are automatically created by the previous definition macro:
Return a shared pointer to object
which owns object
.
It initializes the private fields of object
handling the shared pointer,
returning the same pointer to the object from a value point of view,
but with the shared pointer field initialized.
As a consequence, the shared pointer part of object
shall not have been initialized yet.
The other part of object
may or may not be initialized before calling this method.
Return a new shared pointer to the same object than the one pointed by shared
,
incrementing the ownership of the object.
This function is thread safe.
Allocate a new object, initialize it and return an initialized shared pointer to it.
The used allocation function is the NEW
operator.
This method is only created if the INIT
and NEW
methods are provided and not disabled.
Initialize the new object object
and return an initialized shared pointer to it.
The INIT
operator of object
is ensured to be called only once,
even if multiple threads try to initialize it at the same time.
Once the object is fully cleared, the initialization function may occur once again.
object
shall be a global variable initialized with the
ISHARED_PTR_STATIC_INIT
macro.
This method is only created if the INIT
and NEW
methods are provided and not disabled.
Clear the shared
pointer, releasing ownership of the object
and destroying the shared object if no longer
any other shared
pointers own it.
This function is thread safe.
Clear the shared pointer *shared
, releasing ownership of the object
and destroying the shared object if no longer
any other shared
pointers own it.
This function is thread safe.
Afterwards, *shared
is set to NULL.
Update the shared pointer *shared1
to point to the same object than
the shared pointer shared2
.
Destroy the shared object pointed by *shared1
if no longer any other shared
pointers own it, set the shared pointer shared
to the same object
than the one pointed by src
.
This function is thread safe.
This header is for creating intrusive doubly-linked list.
Extend an object by adding the necessary interface to handle it within
an intrusive doubly-linked list.
This is the intrusive part.
It shall be put within the structure of the object to link, at the top
level of the structure.
See example of ILIST_DEF
.
Initialize the fields added by ILIST_INTERFACE
of the object.
Define the intrusive doubly-linked list
and define the associated methods to handle it as static inline
functions.
name
shall be a C identifier that will be used to identify the list.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
The oplist shall have at least the following operators (CLEAR
),
otherwise it won't generate compilable code.
An object is expected to be part of only one list of a kind in the entire program at a time.
An intrusive list enables to move from an object to the next object without needing to go through the entire list,
or to remove an object from a list in O(1)
.
It may, or may not, be better than standard list. It depends on the context.
The given interface won't allocate anything to handle the objects as
all allocations and initialization are let to the user.
However, the objects within the list can be automatically be cleared
(by calling the CLEAR
method to destruct the object) on list destruction.
Then the memory allocation, performed by the user program, can also be reclaimed
by defining a DEL
operator to free the used memory in the object oplist.
If there is no DEL
operator, it is up to the user to free the used memory.
The list iterates from back to front.
ILIST_DEF_AS
is the same as ILIST_DEF
except the name of the types name_t
, name_it_t
are provided.
Example:
typedef struct test_s {
int n;
ILIST_INTERFACE (ilist_tname, struct test_s);
} test_t;
ILIST_DEF(ilist_tname, test_t)
void f(void) {
test_t x1, x2, x3;
ilist_tname_t list;
x1.n = 1;
x2.n = 2;
x3.n = 3;
ilist_tname_init(list);
ilist_tname_push_back (list, &x3);
ilist_tname_push_front (list, &x1);
ilist_tname_push_after (&x1, &x2);
int n = 1;
for M_EACH(item, list, ILIST_OPLIST(ilist_tname)) {
assert (n == item->n);
n++;
}
ilist_tname_clear(list);
}
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the list of type
.
Type of an iterator over this list.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t list)
void name_clear(name_t list)
void name_reset(name_t list)
type *name_back(const name_t list)
type *name_front(const name_t list)
bool name_empty_p(const name_t list)
void name_swap(name_t list1, name_t list2)
void name_it(name_it_t it, name_t list)
void name_it_set(name_it_t it, const name_it_t ref)
void name_it_last(name_it_t it, name_t list)
void name_it_end(name_it_t it, name_t list)
bool name_end_p(const name_it_t it)
bool name_last_p(const name_it_t it)
bool name_it_equal_p(const name_it_t it1, const name_it_t it2)
void name_next(name_it_t it)
void name_previous(name_it_t it)
type *name_ref(name_it_t it)
const type *name_cref(const name_it_t it)
size_t name_size(const name_t list)
void name_remove(name_t list, name_it_t it)
The following specialized methods are automatically created by the previous definition macro:
Initialize the additional fields of the object *obj
handling the list.
This function shall be used in the object constructor.
Push the object *obj
itself (and not a copy) at the back of the list list
.
Push the object *obj
itself (and not a copy) at the front of the list list
.
Push the object *obj
itself (and not a copy) after the object *position
.
Pop the object from the back of the list list
and return a pointer to the popped object,
given back the ownership of the object to the caller program
(which becomes responsible to calling the destructor of this object).
Pop the object from the front of the list list
and return a pointer to the popped object,
given back the ownership of the object to the caller program
(which becomes responsible to calling the destructor of this object).
Remove the object *obj
from the list.
It gives back the ownership of the object to the caller program
which becomes responsible to calling the destructor of this object.
Return the object that is after the object *obj
in the list
or NULL if there is no more object.
Return the object that is before the object *obj
in the list
or NULL if there is no more object.
This method is the same as the generic one, except it is really x
that is added in the list, not a copy.
Move the element pointed by it
from the list list2
to the back position of list1
.
it
shall be an iterator of list2
.
Afterwards, it
points to the next element of list2
.
Move all the element of list2
into list1
, moving the last element
of list2
after the first element of list1
.
Afterwards, list2
is emptied.
list1
and list2
shall reference different objects.
Note
This header is obsolete: M-SHARED-PTR should be used instead.
This header is for transforming a standard container (LIST
, ARRAY
, DICT
, DEQUE
, ...)
into an equivalent container but compatible with concurrent access by different threads.
In practice, it puts a mutex lock to access the container.
As such it is quite generic. However, it is less efficient than containers specially tuned for multiple threads. There is also no iterators.
Define the concurrent container name
based on container type
of oplist oplist
,
and define the associated methods to handle it as static inline
functions.
name
shall be a C identifier that will be used to identify the concurrent container.
It will be used to create all the types (including the iterator)
and functions to handle the container.
This definition shall be done once per name and per compilation unit.
It scans the oplist
of the type to create equivalent function,
so the exact generated methods depend on explicitly the exported methods in the oplist.
The init method is only defined if the base container exports the INIT
operator,
same for the clear method and the CLEAR
operator,
and so on for all created methods.
In the description below,
subtype_t
is the type of the element within the given container type
(if it exists),
key_t
is the key type of the element within the given container type
(if it exists),
value_t
is the value type of the element within the given container type
(if it exists).
CONCURRENT_DEF_AS
is the same as CONCURRENT_DEF
except the name of the type name_t
is provided.
Example:
/* Define a stack container (STACK)*/
ARRAY_DEF(array1, int)
CONCURRENT_DEF(parray1, array1_t, ARRAY_OPLIST(array1))
/* Define a queue container (FIFO) */
DEQUE_DEF(deque_uint, unsigned int)
CONCURRENT_DEF(cdeque_uint, deque_uint_t, M_OPEXTEND(DEQUE_OPLIST(deque_uint, M_BASIC_OPLIST), PUSH(deque_uint_push_front)))
extern parray1_t x1;
extern cdeque_uint_t x2;
void f(void) {
parray1_push (x1, 17);
cdeque_uint_push (x2, 17);
}
The following types are automatically defined by the previous definition macro if not provided by the user:
Type of the concurrent container of type
.
The following methods of the generic interface are defined (See generic interface for details):
void name_init(name_t concurrent)
void name_init_set(name_t concurrent, const name_t src)
void name_init_move(name_t concurrent, name_t src)
void name_set(name_t concurrent, const name_t src)
void name_move(name_t concurrent, name_t src)
void name_reset(name_t concurrent)
void name_clear(name_t concurrent)
void name_swap(name_t concurrent1, name_t concurrent2)
bool name_empty_p(const name_t concurrent)
void name_set_at(name_t concurrent, key_t key, value_t value)
bool name_erase(name_t concurrent, const key_t key)
void name_push(name_t concurrent, const subtype_t data)
void name_push_move(name_t concurrent, subtype_t *data)
void name_pop_move(subtype_t *data, name_t concurrent)
void name_get_str(string_t str, name_t concurrent, bool append)
void name_out_str(FILE *file, name_t concurrent)
bool name_parse_str(name_t concurrent, const char str[], const char **end)
bool name_in_str(name_t concurrent, FILE *file)
m_serial_return_code_t name_out_serial(m_serial_write_t serial, const name_t container)
m_serial_return_code_t name_in_str(name_t container, m_serial_read_t serial)
bool name_equal_p(name_t concurrent1, name_t concurrent2)
size_t name_hash(name_t concurrent)
Returns true in case of success, false otherwise.
The following specialized methods are automatically created by the previous definition macro:
Read the value associated to the key key
.
If it exists, it sets *value
to it and returns true.
Otherwise it returns false (*value
is unchanged).
This method is only defined if the base container exports the GET_KEY
operator.
Read the value associated to the key key
.
If it exists, it sets *value
to it.
Otherwise, it creates a new value (default constructor) and sets *value
to it.
This method is only defined if the base container exports the SAFE_GET_KEY
operator.
Pop data from the container and set it in *data
.
There shall be at least one data to pop.
Testing with the operator EMPTY_P
before calling this function is not enough
as there can be some concurrent scenario where another thread pop the last value.
name_pop_blocking
should be used instead.
This method is only defined if the base container exports the POP
operator.
Read the value associated to the key key
.
If it exists, it sets *value
to it and returns true.
Otherwise, if blocking is true, it waits for the data to be filled.
After the wait, it sets *value
to it and returns true.
Otherwise, if blocking is false, it returns false.
This method is only defined if the base container exports the GET_KEY
operator.
Pop a value from the container and set *data
with it.
If the container is not empty, it sets *data
and return true.
Otherwise, if blocking is true, it waits for the data to be pushed.
After the wait, it sets *data
to it and returns true.
Otherwise, if blocking is false, it returns false.
This method is only defined if the base container exports the POP
and EMPTY_P
operators.
Pop a value from the container and initialize and set *data
with it.
If the container is not empty, it initializes and sets *data
and return true.
Otherwise if blocking is true, it waits for the data to be pushed.
After the wait, it initializes and sets *data
to it and returns true.
Otherwise, if blocking is false, it returns false.
Warning
*data
remains uninitialized!
This method is only defined if the base container exports the POP_MOVE
and EMPTY_P
operators.
This header is for using bitset.
A bitset can be seen as a specialized version of an array of bool, where each item takes only 1 bit. It enables for compact representation of such array.
Example:
void f(void) {
bitset_t set;
bitset_init(set);
for(int i = 0; i < 100; i ++)
bitset_push_back(set, i%2);
bitset_clear(set);
}
The following methods are available:
This type defines a dynamic array of bit and is the primary type of the module.
This type defines an iterator over the bitset.
Initialize the bitset array
(aka constructor) to an empty array.
Initialize the bitset array
(aka constructor) and set it to the value of ref
.
Set the bitset array
to the value of ref
.
Initialize the bitset array
(aka constructor) by stealing as many
resources from ref
as possible. Afterwards ref
is cleared.
Set the bitset array
by stealing as many resources from ref
as possible.
Afterwards ref
is cleared.
Clear the bitset array
(aka destructor).
Reset the bitset. The bitset becomes empty but remains initialized.
Push a new element into the back of the bitset array
with the value value
.
Push a new element into the position key
of the bitset array
with the value value
.
key
shall be a valid position of the array:
from 0 to the size of array (included).
Pop a element from the back of the bitset array
and set *data
to this value
if data is not NULL (if data is NULL, the popped data is cleared).
Set *dest
to the value the element key
if dest is not NULL,
then remove the element key
from the bitset.
key
shall be within the size of the bitset.
Return the first element of the bitset. The bitset shall have at least one element.
Return the last element of the bitset. The bitset shall have at least one element.
Set the element i
of bitset array
to value
.
i
shall be within 0 to the size of the array (excluded).
Flip the element i
of bitset array
'.
i
shall be within 0 to the size of the array (excluded).
Return the element i
of the bitset.
i
shall be within 0 to the size of the array (excluded).
Return true if the bitset is empty, false otherwise.
Return the size of the bitset.
Return the capacity of the bitset.
Resize the bitset array
to the size size
(initializing or clearing elements).
Extend or reduce the capacity of the array
to a rounded value based on capacity
.
If the given capacity is below the current size of the bitset,
the capacity is set to the size of the bitset.
Swap the bitsets array1
and array2
.
Swap the elements i
and j
of the bitset array
.
i
and j
shall reference valid elements of the array.
Set the iterator it
to the first element of array
.
Set the iterator it
to the last element of array
.
Set the iterator it
to the end of array
.
Set the iterator it1
to it2
.
Return true if the iterator doesn't reference a valid element anymore.
Return true if the iterator references the last element of the array, or doesn't reference a valid element.
Return true if both iterators reference the same element.
Move the iterator it
to the next element of the array.
Move the iterator it
to the previous element of the array.
Return a constant pointer to the element pointed by the iterator. This pointer remains valid until the array or the iterator is modified by another method.
Generate a formatted string representation of the bitset array
and set str
to this
representation (if append
is false) or append str
with this representation
(if append
is true).
This method is only defined if the header m-string.h was included before
including m-bitset.h
Parse the formatted string str
that is assumed to be a string representation of a bitset
and set array
to this representation.
It returns true if success, false otherwise.
If endp
is not NULL, it sets *endp
to the pointer of the first character not
decoded by the function.
Generate a formatted string representation of the bitset array
and outputs it into the FILE file
.
Read from the file file
a formatted string representation of a bitset
and set array
to this representation.
Return true if both bitsets array1
and array2
are equal.
Return a hash value of array
.
Perform a bitwise AND
operation in dst
between dst
and src
(effectively performing an intersection of the sets).
Perform a bitwise OR
operation in dst
between dst
and src
(effectively performing an union of the sets).
Perform a bitwise XOR
operation in dst
between dst
and src
.