76 lines
2.3 KiB
Plaintext
76 lines
2.3 KiB
Plaintext
Fixed
|
|
-----
|
|
|
|
[x] Shouldn't the ezcDocumentRstXhtmlVisitorTests tests just be
|
|
ezcDocumentXhtmlVisitorTests ?
|
|
|
|
No, as they are special test for the RstXhtml visitor, which explicitely
|
|
visit the RST AST. We may have other XHtml visitors later, like for
|
|
DocBook
|
|
|
|
[x] Low code coverage in some parts.
|
|
|
|
The parts which are actually implemented do have a >90% coverage. The
|
|
other stuff is just misses test and implementation.
|
|
|
|
The ast node base class also have low coverage, while the uncovered
|
|
methods are jsut debugging helper methods, which are viable for parser
|
|
debugging. But we may want to remove them in the release...
|
|
|
|
[x] WHy is the RstXhtmlVisitor not simple a docbook visitor to generate XHTML?
|
|
|
|
This is what we originally discussed as conversion shortcuts. A Docbook
|
|
XHtml visitor will also be added. The direct conversion has access to more
|
|
semantic information, then after the docbook conversion.
|
|
|
|
[x] Why are the option class names including "Base" as in
|
|
ezcDocumentParserBaseOptions and ezcDocumentConverterBaseOptions,
|
|
ezcDocumentXmlBaseOptions.
|
|
|
|
Fixed, removed "Base" from option class names.
|
|
|
|
[x] Some of the class descriptions in document/rst/nodes/* are incorrect
|
|
(Example is ezcDocumentRstSubstitutionNode).
|
|
|
|
Fixed.
|
|
|
|
[x] ezcDocumentRstToken doesn't have the constants documented - also run
|
|
the doc analysis tool because there's some other stuff as well.
|
|
|
|
Fixed.
|
|
|
|
[x] Docblocks are broken for some files (eg. src/document/rst/nodes): there
|
|
are duplicated @copyright and @license, and the @package is not Document.
|
|
|
|
Fixed.
|
|
|
|
[x] Strings should enclose variables in {}: "Could not find visitor for '{$class}'"
|
|
|
|
Fixed
|
|
|
|
[x] The tutorial should mention on how to extend the RST directives thingy.
|
|
|
|
Fixed
|
|
|
|
[x] A description of how the RST parser actually works should be added.
|
|
|
|
Fixed
|
|
|
|
[x] Why is ezcDocumentManager final?
|
|
|
|
Because it is / will be a class only providing static (conversion)
|
|
methods with a static registry of known documents. Once used in your code
|
|
it makes nearly no sense to extend. But we can remove the final keyword to
|
|
make it possible for users to write (and only use) their own manager on
|
|
this base.
|
|
|
|
Removed final keyword
|
|
|
|
Open
|
|
----
|
|
|
|
[ ] There are some @TODOs in the code.
|
|
|
|
The remaining TODOs will be resolved during further development.
|
|
|