lp://staging/~pierregm/scipy/numpydocs

Branch merges

Propose for merging

No branches dependent on this one.

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

21. By Pierre GM on 2008-08-10

Docscraper
8 fixed _fix_see_also

Numpydoc:
* Add the numpydoc_default_block_type and numpydoc_default_list_type options
* Fixed mangle_signature

20. By Pierre GM on 2008-08-03

* Defined a .format() method for FunctionDoc and modified __str__ accordingly

19. By Pierre GM on 2008-07-27

moved to ./ext

18. By Pierre GM on 2008-07-27

docscraper: suppressed the example

17. By Pierre GM on 2008-07-27

* moved `docscraper.py` and `numpydoc.py` to ./ext
* add `generate.py` and `summarize.py` from Pauli's branch

docscraper:
* add Docstring.get_paragraphs()
* add Docstring.paragraphs_as_list()

16. By Pierre GM on 2008-07-26

docscraper:
* put .split_sections() in .parse()
* renamed ._parse_xxx() to ._fix_xxx()
* renamed ._str_xxx() to ._format_xxx()
* introduced the .format() method:
  it transforms the parsed docstring to a new list, properly formatted
* .__str__() now just join the list w/ \n as a separator
* fixed the formatting of the signature

15. By Pierre GM on 2008-07-25

numpydoc : moving around some code to improve readability

14. By Pierre GM on 2008-07-25

* Update to Pauli's modification (r50)

13. By Pierre GM on 2008-07-25

Overhaul #2 : The "We're getting there" version
-----------------------------------------------

Following Stefan's and Pauli's examples, the numpydoc file is now split in 2

* `docscraper.py` implements an alternative to the base classes NumpyDocString,
  SphinxDocString, SphinxFunctionDoc,... Details below
* numpydoc is basically Pauli's version, with a few modifications. Note that it
  won't work with Sphinx-0.4.1 but with dev200080721 at least.

Usage should be the same. I tested it with some of the numpy docs and it works
pretty nicely.

The alternative implementations are midway between Stefan's approach of a reader
and David's and my approach of an improved list.
First, the docstring is split in sections.
Some (examples, see also, parameters list) sections are then processed and
modified in place.
Then, the __str__ method deals with the finer details of formatting.

By default, list-like blocks (parameters, returns, ...) are formatted with a
field + list approach, while the others (notes, examples...) are formatted with
a header in bold. However, this behavior can be changed by the user.
At the moment, the default are defined at the module level. Later on, they should
depend on some configuration options.

12. By Pierre GM on 2008-07-20

* force see_also_parser to indent by 3
* force description_parser to insert a new line before each new description.