summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2014-12-25Makefile: specify python2 in places where it's needed.John MacFarlane
2014-12-25Updated C parser to handle #252.John MacFarlane
Closes #252.
2014-12-25Added test case for link-in-link-in-image (#252).John MacFarlane
2014-12-25JS: fixed parsing of link-in-link-in-image.John MacFarlane
Partially addresses #252. Still need to: - update C parser. - put an example in the spec.
2014-12-25Fixed infinite loop in JS parser for link-in-link-in-image.John MacFarlane
Partially addresses #252. This fixes the infinite loop, and brings the JS parser into agreement with cmark, but both still have bad output in this case, so more work is needed.
2014-12-25Improved rules for emphasis and strong emphasis.John MacFarlane
This improves parsing of emphasis around punctuation. Background: http://talk.commonmark.org/t/emphasis-inside-strong-broken-in-js-implementation-when-parenthesis-involved/903/6 The basic idea of the change is that if the delimiter is part of a delimiter clump that has punctuation to the left and a normal character (non-space, non-punctuation) to the right, it can only be an opener. If it has punctuation to the right and a normal character (non-space, non-punctuation) to the left, it can only be a closer. This handles cases like **Gomphocarpus (*Gomphocarpus physocarpus*, syn. *Asclepias physocarpa*)** and **foo "*bar*" foo** better than before. The spec section on Emphasis and Strong Emphasis has been extensively revised. The C and JS implementations have been brought up to date, and all tests pass.
2014-12-24Use python2 for man, test scripts.John MacFarlane
Closes #251.
2014-12-24Fixed make_man_page.py so it works with both python2 and python3.John MacFarlane
Closes #251.
2014-12-23Added definitions of whitespace and other character classes.John MacFarlane
Closes #108.
2014-12-23Made spec_tests.py more DRY.John MacFarlane
Factored test selection logic out of do_tests. Eliminated do_tests, putting the (now very short) code into main. Improves 1ba4822 @cirosantilli
2014-12-22Added HTML normalization tests to test/CMakeLists.txt.John MacFarlane
Removed from .travis.yml. This should be a standard part of the test suite. See #245.
2014-12-22Fixed normalization bug, added more doctests for normalization.John MacFarlane
* The tests test for removal of whitespace around block-level tags. * Previously whitespace wasn't removed before an initial block-level tag; this commit fixes that. * Also revised wording so it's clear that whitespace is removed on both sides of block-level tags. Closes #246 in a slightly different way.
2014-12-22spec_tests.py: make --dump-tests sensitive to --pattern, --number.John MacFarlane
Closes #249 in a slightly different way.
2014-12-22Merge pull request #248 from cirosantilli/short-optionsJohn MacFarlane
spec_tests add short options for common parameters
2014-12-22Merge pull request #247 from cirosantilli/run-single-test-by-numberJohn MacFarlane
Add --number option to run a single test by its id
2014-12-22Slight improvement to #250.John MacFarlane
Indicate which is expected and which is actual output.
2014-12-22Merge pull request #250 from cirosantilli/deal-invalid-unicodeJohn MacFarlane
Don't raise exception on invalid UTF-8 output
2014-12-21Don't raise exception on invalid UTF-8 outputCiro Santilli
This could be reproduced with: ./test/spec_tests.py --program kramdown --pattern Enti with Kramdown version 1.5.0. The main problem is that it would stop future tests from being carried out. After this commit it is just counted as a failure. The error message printed is of the form: Unicode error: 'utf8' codec can't decode byte 0xfc in position 8: invalid start byte '<p># \xd3\x92 \xcf\xa0 \xef\xbf\xbd</p>\n' '<p># \xd3\x92 \xcf\xa0 \xfc\x85\xb8\xb0\xa9\xb8</p>\n'
2014-12-21spec_tests add short options for common parametersCiro Santilli
2014-12-21Add --number option to run a single test by its idCiro Santilli
Do not check input validity, e.g. input `>` number of tests, because other inputs don't check it either, e.g. `--pattern`. If given, the summary line will be like: 1 passed, 0 failed, 0 errored, 509 skipped
2014-12-19Added a few more doctests for HTML normalization (#245).John MacFarlane
2014-12-19Fixed whitespace normalization (#245).John MacFarlane
2014-12-19Merge pull request #245 from cirosantilli/normalize-whitespace-failJohn MacFarlane
Expose failure to normalize whitespaces
2014-12-19Expose failure to normalize whitespacesCiro Santilli
2014-12-18Merge pull request #244 from petere/patch-1John MacFarlane
Remove stray character
2014-12-18Remove stray characterPeter Eisentraut
2014-12-17Merge pull request #243 from maxlinc/patch-1John MacFarlane
Fix links to the spec
2014-12-17Fix links to the specMax Lincoln
All the links the spec were broken. I changed `http://jgm.github.io/CommonMark/spec.html` to `http://spec.commonmark.org/0.13/`. It'd be nice if there was a `http://spec.commonmark.org/latest/` so you don't need to change the README every time a new version (unless there's major changes so the URL fragments aren't valid anymore. The spec itself has some broken fragment links, e.g. `#image` instead of `#images`. Unfortunately I don't have a good tool that checks fragments so I don't have the full list.
2014-12-17More Makefile adjustments.John MacFarlane
2014-12-17Updated Makefile for changes in man page generation.John MacFarlane
2014-12-17Create cmark.3 in cmake build process.John MacFarlane
Remove it from the repository. Closes #241.
2014-12-17Merge pull request #240 from cirosantilli/ignore-pycJohn MacFarlane
Ignore pyc files
2014-12-17Ignore pyc filesCiro Santilli
They may be generated when you run the tests.
2014-12-16scanners.re: use cmark_chunk rather than chunk.John MacFarlane
2014-12-16Added 'literal' field to 'code' struct.John MacFarlane
In the last few commits we were using as.code.fenced and as.literal at the same time for NODE_CODE_BLOCK, which obviously led to problems.
2014-12-16HTML writer: fixed bug in detection of fenced code blocks.John MacFarlane
2014-12-16print: print fenced and fence_length for code blocks.John MacFarlane
2014-12-15Re-added cmark_ prefix to strbuf and chunk.John MacFarlane
Reverts 225d720.
2014-12-14Clarify that unicode whitespace counts as whitespace in emph rules.John MacFarlane
Added a test case with a unicode nonbreaking space. See #108, though "whitespace" should still be defined more systematically. This is a step forward.
2014-12-14Added comment.John MacFarlane
2014-12-14Use UTF-8 sensitive functions to check for spaces around emph markers.John MacFarlane
2014-12-14Rename get/set_string_content -> get/set_literal.John MacFarlane
2014-12-14Use cmark_iter to avoid stack allocation in process_inlines.John MacFarlane
2014-12-14man writer: use chunk literal in code blocks.John MacFarlane
2014-12-14Added GNU profiler flags to debug build.John MacFarlane
2014-12-14Use chunk for fenced code info, instead of strbuf.John MacFarlane
2014-12-14Use as.literal instead of string_content for HTML and code blocks.John MacFarlane
This is for consistency with the other types of nodes that have literal strings as contents.
2014-12-14Added empty clauses for DOCUMENT node to html and man writers.John MacFarlane
For completeness.
2014-12-14Rename CMARK_NODE_INLINE_CODE -> CMARK_NODE_CODE.John MacFarlane
2014-12-13Clarified spec on indented code blocks.John MacFarlane
Previously the spec said, wrongly, that a blank line was needed between a paragraph and a following code block. It is only needed between a code block and a following paragraph (due to lazy continuations). Thanks to textnut for pointing this out.