Commit Graph
13 Commits
Author SHA1 Message Date
Christophe Simonis cfe0523714 [MERGE] forward port branch 12.0 up to 8f21148e1a 2019-05-31 14:37:38 +02:00
Xavier Morel 7b00776c6c [FIX] doc: error & potential warnings
* properly handle 3-args version of odoo.define(): just ignore the
second argument, pretty much the only use-case is dynamic
dependencies (otherwise it should be using static require() calls)
so there's little the extractor can do
* fix an issue with sys.path ordering meaning sphinx would pick up an
odoo installed into the current virtualenv (editable or not) over
the local copy it's part of

closes odoo/odoo#33538

Signed-off-by: Xavier Morel (xmo) <xmo@odoo.com>
2019-05-21 14:19:13 +00:00
Xavier Morel cca26f7952 [FIX] doc: workarounds for weirdo code & refs
Make UnknownNS more capable (of giving no fucks) so the documentation
builder blows up less on weird-ass code at module toplevels.

UnknownNS should probably be split out from NSDoc though, that requires
odd workarounds in `__getitem__`.
2019-02-18 08:55:57 +00:00
Christophe Simonis 5e055a2afd [MERGE] forward port branch saas-11.4 up to f6ca72b3ce 2018-11-02 10:52:55 +01:00
Xavier Morel 35bb8304ab [FIX] doc: building on Python 3.6.7, 3.7.1, possibly 2.7.???
https://bugs.python.org/issue33899

tokenize.generate_tokens was altered to match the C tokenizer,
previously it would end the tokenization with just an ENDMARKER, in the
titled releases it adds a NEWLINE before the ENDMARKER if none is
present, this broke the parsing of jsdoc type specifications as Python's
tokenizer is used under the cover.

closes odoo/odoo#28056
2018-10-23 10:39:22 +00:00
Xavier Morel 00eefae98d [FIX] doc: implement delegation/inheritance in JS documenter
Else website_sale_wishlist.wishlist blows up the documenter:
``events`` is set to sAnimations.Class.events but that is not an "own
property" of sAnimation.Class, it is instead inherited from Widget.

If delegation is not implemented, the value resolves to <nothing>,
which blows up when trying to set its name to the "property name" (by
calling ``set_name``).
2018-10-01 17:31:41 +02:00
Xavier Morel 0e307f8280 [FIX] doc building and warnings
doc:

* fix incorrect doc comments (documenting params which don't exist)
* correctly quote non-refs
* fix role label syntax (backticks must be escaped)
* add newline to fix warning

JS doc parser:

* fix handling of already-resolved objects
* fix handling of non-string properties (e.g. foo[0] at module toplevel)
* don't blow up if subject of .include call can't be resolved (just ignore)

Others:

* add translator support for ``problematic`` node (apparently used for
  some refs by recent Sphinx)
2018-08-27 12:55:21 +02:00
Xavier Morel a56372a4ca [FIX] various broken JS docstrings
Resulting in sphinx warnings or ill-formatted doc all the way to the
doc not building at all.
2018-07-24 12:06:13 +02:00
Xavier Morel f1143b4073 [FIX] doc: tokenization of nested generics in JS docstrings
With more than 1 level of generic type, the "closure" of generic
parameters would get tokenized by python as a single right shift (>>)
rather than two gt (>, >) leading to our own tokenizing pass getting
lost and blowing up.

Fix by explicitly recognizing and splitting (>>) into (>, >). We could
consider writing the entire tokenizer ourselves instead of
post-processing Python's but given the entire post-processing function
is currently shorter than just Pythons' tokenizer's regexes...

Anyway issue found by vsc while trying to write better docstrings.
2018-02-19 13:22:58 +01:00
Xavier Morel f7323dbe23 [FIX] doc: informal sub-params
Revert 6cb12273e2 which reverted
08d48b0fff.

Issue was the support for sub-parameters (aka @param {type} foo.bar)
where the "parent" parameter was not formal (only present in a
function's docstring, but not in the function's parameters list),
subtype creation would fail to find a root parameter to attach the
sub-parameter to.

This has not actually been fixed, but for the specific case where the
formal parameters list is *completely empty* we fall back on a list of
the docstring's parameters, which works well enough for this case.
2018-01-10 12:36:58 +01:00
Xavier Morel e8dc72fd69 [FIX] doc: handle unknown browser globals better
Before this, if a new global was added to JS code (module toplevel,
function bodies are not considered) it would blow up when attempting
to resolve variable refs & uses.

Make the toplevel scope a defaultdict such that if e.g. a "foo" is
looked up in a module but not a local variable of that module it's
assumed to be an UnknownNS object (from which further UnknownNS can be
deref'd).
2017-10-30 16:16:41 +01:00
Xavier Morel 40ca8557e9 [FIX] building documentation with P3
* fix P3 incompatibilities in doc/_extensions
* vendor pyjsdoc, strip out everything we're not using and make what's
  left P3-compatible (upstream is not P3 compliant)
* drop undeclared dependency on attrs as it turns out not to be much
  used
2017-10-10 13:29:11 +02:00
Xavier Morel 313f69e948 [ADD] doc: JS document extraction system
* Fix a bunch of ill-documented/incomplete/incorrect method docs
* add start of Sphinx extension to extract & integrate jsdoc into
  Sphinx documentation:

  - parse JS files (and don't blow up), uses a fork of pyjsparser as
    the project currently does not parse comments
  - extract cross-module dependency information
  - parse JsDoc comments using pyjsdoc and infer structure from code &
    jsdoc
  - ``ast`` CLI printing a simplified AST of the input files
  - ``dependencies`` creating a dependency graph of either all modules
    in the provided input files or the modules matching the specified
    filters (warning: will not work if missing dependencies),
    generates a .dot file
  - ``extractor`` generating a plain text module documentation (mix of
    rst and markdown styles, not anything formal)
* sphinx extension with an "automodule" directive taking a module name
  and generating the documentation for it
2017-09-18 11:54:36 +02:00