knio / dominate Goto Github PK

pyy_doc up some documentation

Convert 4-spaced source files to use 2 spaces.

pyy_cgi depends on parts of pyy_httpserver which import stackless. Factor so dependence on stackless python is removed unless explicitly using pyy_httpserver

Move readme text to the module's init.doc and generate the README automatically. Use whichever markup format will look best in both locations.

input_, map_, object_, del_

Add request keyword argument that defaults to None which is stored in self.request. This will allow deriving classes to have access to the request object in their __init__ methods (assuming they've called response.__init__ first).

make pyy_httpserver have configurable log files and statistics

Mark as experimental and outline briefly.

dom1core.py does not explicitly assign to the name "basestring" as the other files do, so dom1core.getElementsByTagName() will raise

NameError: name 'basestring' is not defined

at line 49 in py3k.

Use these methods rather than a single main method.

HTML5 will need one written (likely)

Create a barebones website which it is easy to get started with.

• arguments to functions

MathML, SVG, etc.

fr looks a bit ugly, reactjs had a similar problem, they are using htmlFor.

For cls, they are using className.

I think it might make sense to adapt these names, maybe with underscores.

regex parsing in web.py

so /folder/ will look for Index in pages/folder/__init__.py

We should make sure pyy conforms to http://www.w3.org/TR/2009/NOTE-xhtml-media-types-20090116/

NOT DONE
A.1: Processing Instructions and the XML Declaration
A.9: Character Encoding

PARTIALLY DONE
A.2: Elements that can never have content (elements render as but will accept children)
A.7: The lang and xml:lang Attributes (xml:lang is required)
A.12: Using Ampersands (converted in attribute values, not content)

DONE
A.3: Elements that have no content
A.5: Line Breaks within Attribute Values
A.8: Fragment Identifiers
A.10: Boolean Attributes
A.14: Referencing Style Elements when serving as XML
A.17: The XML DTD Internal Subset
A.18: CDATA Sections
A.19: Explicit Elements
A.20: base vs. xml:base

NOT APPLICABLE
A.4: Embedded Style Sheets and Scripts
A.6: Deleted
A.11: Document Object Model and XHTML
A.13: Cascading Style Sheets (CSS) and XHTML
A.15: Formfeed Character in HTML vs. XML
A.16: The Named Character Reference '
A.21: document.write
A.22: Updating a document using innerHTML
A.23: Scripts and missing tbody elements
A.24: Styling the entire document
A.25: noscript Element
A.26: iframe Element
A.27: Creating elements in JavaScript

FUTURE
A.8: Fragment Identifiers (limit id value with regex)
A.10: Boolean Attributes (limit with regex)

Change to pygments and fail back to fileserver if module isn't installed?

I'm not sure what it is exactly, but it's a popular buzzword and our competitors do it.

python test.py pyy_html runs only the pyy_html module's tests.

@document(title="blah")
def page():
    p('Hello World')

Make a nice way to add/modify head, title, etc to this document

Simple yet powerful session manager for pyy_web.

is_awesome, soylentsoft, etc.

Will provide good real-world examples.

Look into using py.test (http://codespeak.net/py/dist/test/index.html) for testing

Move to examples?

Should host the documentation, examples, faq, links to sites using pyy, links to GitHub source, etc.

pyy_httpserver is not stable when serving large binary files, or possibly many active connections.
find out why

Like jquery etc.

Add sessions.py to aid in the tracking of sessions in pyy_web.

Happens when doing

document += pipe("echo hi")

But I haven't tracked down the cause yet.

Change parse to be just an HTML parser and create an HTML tree to pyy object tree function which uses the ElementTree API allowing other parsers such as lxml and BeautifulSoup to be used.