apache / age-website Goto Github PK

The query provided under the terminal CREATE clauses here gives error instead of outputting '0 rows returned'

Upon executing the following query:
SELECT * FROM cypher('graph_name', $$ CREATE /* Create clause here, no following clause */ $$) as (a agtype);

I get:

ERROR: syntax error at end of input
LINE 4: $$) as (a agtype)

Issue :

Ref:
Ref package file

Desc:
wrong name, auther, discription

Sol:
fix apahce -> apache

The SET clause documentation says that with SET clause, we can update the labels of nodes but any example is not provided to support that. Besides this, I tried updating the label of the node but that resulted in an error.

Current Syntax:

SELECT * 
FROM cypher('graph_name', $$
    CREATE (a {name: 'Andres')
    RETURN a
$$) as (a agtype);

'}' was missing.

Correct Syntax:

SELECT * 
FROM cypher('graph_name', $$
    CREATE (a {name: 'Andres'})
    RETURN a
$$) as (a agtype);

There are some warnings regarding broken links and broken highlighting in the documentation build output (e.g., https://github.com/apache/age-website/actions/runs/3195377462/jobs/5215965276 -> section Run sphinx-multiversion docs build/html).

There is a statement under the heading Create full path

v1 When you use CREATE and a pattern, all parts of the pattern that are not already in scope at this time will be created.

Here v1 is confusing me. Is it a typo?

In the documentation of the MERGE clause, there is an example as shown below, which I personally think is problematic.

In the following sql statement
'''
SELECT * FROM cypher('graph_name', $$
MERGE (charlie {name: 'Charlie Sheen', age: 10})
RETURN charlie
$$) as (v agtype);
'''
No label name is given.
So the sentence "If there exists a vertex with the label 'Critic', that vertex will be returned. Otherwise, the vertex will be created and returned." should be deleted.
And output
{id: 0; label: ‘Actor’: properties:{name: 'Charlie Sheen', age: 10}}::vertex
should be
{id: 0; label: ‘’: properties:{name: 'Charlie Sheen', age: 10}}::vertex

The syntax structure and usage documentation of CALL ... YIELD has yet to be added.

A basic example:

SELECT *
FROM cypher('test', $$
    MATCH (v)        
    CALL label(v) YIELD label AS l_v                                           
    RETURN l_v
$$) as (labels agtype);

The following example is not working properly which is at https://age.apache.org/age-manual/master/functions/user_functions.html
It has multiple issues:

1. Missing ')' closing parentheses
2. Missing typecasting
3. Result is incorrect (sqrt(25) is 5 while the result is mentioned as 25)

Test environment:

PG: 11
AGE: 1.2.0
OS: Ubuntu 18

The output after having getting the closing parentheses fixed:

rrr=# SELECT *
rrr-# FROM cypher('test', $$
rrr$# RETURN pg_catalog.sqrt(25)
rrr$# $$) as (result agtype);
ERROR:  function pg_catalog.sqrt(agtype) does not exist
LINE 3: RETURN pg_catalog.sqrt(25)
                              ^
HINT:  No function matches the given name and argument types. You might need to add explicit type casts.

The repository for the documentation can be found here:

https://github.com/apache/age-website

Please clone a version of the repository to your own machine and make a new branch to make changes.

** Note that the repository for the documentation is actually separate from the age repo, but discussion for this issue can happen here in this thread.

Something helpful to get started would be going through the sections of the documentation and checking for any grammar or spelling errors, or any area that needs revisions and making those changes. If you feel like there are improvements that can be made, feel free to suggest them as well. The files can be found here:

https://github.com/apache/age-website/tree/master/docs

After you have made changes to the docs, you can make a pull request on that repo and someone will review those changes. We will give feedback on the changes made, and after the changes have been reviewed and approved, they will be added to the master branch.

Is there a way to create constraints for node/vertex properties? I want to use concurrent loading of AGE but need constraints to avoid multiple threads from creating duplicate nodes/vertex. I tried this:

ON pubmed."Article" 
((properties->>'pmid'));

but I get this error:

operator does not exist: ag_catalog.agtype ->> unknown
Hint: No operator matches the given name and argument types. You might need to add explicit type casts.

Are constraints on a Node property possible?

The documentation states that example files can be found in regress/age_load/data. What location is being referenced here? I can't seem to find it. Although a sample is always helpful, I do think the documentation is already clear enough in its description.

https://age.apache.org/age-doap.rdf is incomplete.

It is missing the following
bug-database
repository

Issue :

Ref:
Doc Link Here

Desc:
i'm trying to search AGE JIRA but getting no result it will so nice if we provide link here instead of bold text

Sol:
need to convert AGE JIRA into the link

Current Syntax:

SELECT * 
FROM cypher('graph_name', $$
    CREATE (:Person {name: 'Andres', title: 'Developer')
$$) as (n agtype);

'}' is missing

Correct Syntax:

SELECT * 
FROM cypher('graph_name', $$
    CREATE (:Person {name: 'Andres', title: 'Developer'})
$$) as (n agtype);

The text on https://age.apache.org/age-manual/master/intro/comparability.html says: -

The ordering of different Agtype, when using <, <=, >, >= from smallest value to largest value is:

1 Edge
2 Path
3 Map
4 Vertex
5 Edge
6 Array
7 String
8 Bool
9 Numeric, Integer, Float
10 NULL

'Edge' is repeated on 2 different positions, position 1 and position 5.

Is there an explanation for this, or is it an error?

Hi, just a question. I noticed https://github.com/apache/age-website/blob/new-web/static/admin/config.yml and wondering if you're using any kind of git/headless CMS with your website? If so, how can you advise?

/cc @pdpotter @JoshInnis

-Rohit (I'm from the Apache CloudStack community)

There is a heaading named 'Agtype NULL vs Postgres NULL' under 'Null' in 'Data Types - An Introduction to agtype' which is neither linked nor filled.
https://age.apache.org/age-manual/master/intro/types.html#agtype-null-vs-postgres-null.

A SET clause syntax (SET a = {..}) is implemented and merged recently (apache/age#468). It may be useful to add it in the documentation.

The following code from the test files shows how the syntax is used. It is similar to https://neo4j.com/docs/cypher-manual/current/clauses/set/#set-copying-properties-between-nodes-and-relationships.

...
SELECT * FROM cypher('cypher_set_1', $$ CREATE (a:Andy {name:'Andy', age:36, hungry:true}) $$) AS (a agtype);
SELECT * FROM cypher('cypher_set_1', $$ CREATE (a:Peter {name:'Peter', age:34}) $$) AS (a agtype);
SELECT * FROM cypher('cypher_set_1', $$ CREATE (a:Kevin {name:'Kevin', age:32, hungry:false}) $$) AS (a agtype);
SELECT * FROM cypher('cypher_set_1', $$ CREATE (a:Matt {name:'Matt', city:'Toronto'}) $$) AS (a agtype);
SELECT * FROM cypher('cypher_set_1', $$ CREATE (a:Juan {name:'Juan', role:'admin'}) $$) AS (a agtype);

-- test copying properties between entities
SELECT * FROM cypher('cypher_set_1', $$
    MATCH (at {name: 'Andy'}), (pn {name: 'Peter'})
    SET at = properties(pn)
    RETURN at, pn
$$) AS (at agtype, pn agtype);

SELECT * FROM cypher('cypher_set_1', $$
    MATCH (at {name: 'Kevin'}), (pn {name: 'Matt'})
    SET at = pn
    RETURN at, pn
$$) AS (at agtype, pn agtype);

-- test replacing all properties using a map and =
SELECT * FROM cypher('cypher_set_1', $$
    MATCH (m {name: 'Matt'})
    SET m = {name: 'Peter Smith', position: 'Entrepreneur', city:NULL}
    RETURN m
$$) AS (m agtype);

-- test removing all properties using an empty map and =
SELECT * FROM cypher('cypher_set_1', $$
    MATCH (p {name: 'Juan'})
    SET p = {}
    RETURN p
$$) AS (p agtype);

-- test assigning non-map to an enitity
SELECT * FROM cypher('cypher_set_1', $$
    MATCH (p {name: 'Peter'})
    SET p = "Peter"
    RETURN p
$$) AS (p agtype);

SELECT * FROM cypher('cypher_set_1', $$
    MATCH (p {name: 'Peter'})
    SET p = sqrt(4)
    RETURN p
$$) AS (p agtype);
...

Aside from that, two other SET syntaxes are being worked on. I will post update in this issue.
Thanks!

https://age.apache.org/age-doap.rdf contains

<asfext:pmc rdf:resource="https://comdev.apache.org" />

This should be

<asfext:pmc rdf:resource="https://age.apache.org" />

To install requirements for building the documentation locally , the documentation gives steps to follow but the command
sudo apt install python3 virtualenv texlive-latex-base texlive-latex-extra texlive-extra-utils returns an error on MacOS.

It is because apt is not available in Mac OS. However, there is no mention of the OS requirement for running the above command in the README.md file.

The solution is to use a Package manager like Homebrew to install the relevant packages for Mac OS.

I ran the following commands to install the requirements.

brew install python3 virtualenv
brew install --cask mactex-no-gui

hello ~ @apache/age-committers

We are preparing to reorganize the age website to recruit commiter and encourage open source participation Please check the new-web branch and give us your opinion.

The file at https://age.apache.org/age-doap.rdf contains the following:

<asfext:pmc rdf:resource="https://comdev.apache.org" />

This is wrong, as AGE is not part of comdev.

The entry should be:

<asfext:pmc rdf:resource="https://age.apache.org" />

Apache AGE via docker: require some extra arguments to make it work.

Release notes are outdated https://age.apache.org/release-notes. Currently the latest release is v1.3.0. It would be great if the website could fetch the releases from github using github's api. This would automatatically update the website whenever there is a new release and will also provide correct release notes as listed on github.

I see that there is a python packages requirement defined in requirements.txt which also contains the ones mentioned in requirements_min.txt.

Is there a reason for requirements_min.txt?

After installing age extension, I followed the documentation to create the graph.

website link: Create a Graph

create_graph(graph_name); <-- this is the command, to run for creating a new graph, as per the documentation.

If I run command as:

postgres=# create_graph(graph_name);

Output

ERROR:  syntax error at or near "create_graph"
LINE 1: create_graph(graph_name);

or

postgres=# create_graph("graph_name");

Output

ERROR:  syntax error at or near "create_graph"
LINE 1: create_graph("graph_name");

or

postgres=# create_graph('graph_name');

Output

ERROR:  syntax error at or near "create_graph"
LINE 1: create_graph('graph_name');

The result is same.

What worked for me:

I will have to use the command as:

postgres=# SELECT create_graph('graph_name');

Output

NOTICE:  graph "graph_name" has been created
 create_graph
--------------

(1 row)

Question:

Just want to ask if docs can be updated with working command.

@pdpotter
@shinhanbyeol

There seems to be a rollback issue on the AGE Website.
All the spelling and grammar errors I've requested David(shinhanbyeol) to apply and the Github star/contributor tracking feature Andrew(aked21) requested disappeared from the website.

Upon checking with David, the rollback seemed to have occurred due to renewed AGE Documentation.
Is there some sort of setting that does not affect the details on the website while renewing AGE documentation?

It would be a hassle to recover or edit contents on the website every time the document is edited and applied.

deleted

In the documentation of age here the toUpper() function contains the wrong definition.

The definition states 'toUpper() returns the original string in lowercase.' whereas it should be 'toUpper() returns the original string in lowercase.'

I have also made a PR related to this issue here

In the documentation, the List Functions section lacks the description of the nodes function, and there are some minor typos, such as mentioning edge entities instead of integer elements and differencebetween instead of difference between in the range function part.

@dehowef could you please review these pull requests addressing the issue? #167 and #166

I was going through the documentation and found one typo in the function "load_edges_from_file" at URL missing character "l".

LTrim function in documentation here contains some misleading information.

The ltrim function in the String functions is used to remove the leading white spaces and it is mentioned as trailing white spaces in the documentation.

Provide users with the ability to switch between which graph the metadata information is displayed for.

In the example heading of the arctangent function, it shows it returns the arccosine of 0.5, which should be arctangent instead.

How a Cypher query can return varchar(50), is this some typo in the documentation or is it possible that Cypher queries can return other datatypes?

But on the other hand, the documentation says:

link to page in documentation:
https://age.apache.org/age-manual/master/advanced/advanced.html

Copyright is still set to 2023 at the bottom of the website.