summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2014-11-24Cast void* for MSVC compatibilityNick Wellnhofer
2014-11-23README: Added note about --dump-tests.John MacFarlane
2014-11-23runtests.py: Added --dump-tests option to dump tests as JSON.John MacFarlane
2014-11-23js: use 'CodeBlock' in AST for all code blocks.John MacFarlane
Don't distinguish fenced, indented.
2014-11-23Added 'fenced' flag to cmark_code struct, renamed from cmark_fenced_code.John MacFarlane
Technically we could do without this, since we can check for cmark_fence_length > 0. But it makes the code clearer and doesn't really increase the size of the node struct (because the size of the union is set by the data for lists).
2014-11-23Added internal flag to distinguish setext from atx headers.John MacFarlane
2014-11-23Do not distinguish btw fenced and indented code in AST.John MacFarlane
Use a single CMARK_NODE_CODE_BLOCK tag for both. Distinguish them when needed for parsing by looking at the fence_length attribute, which is 0 for indented blocks.
2014-11-23Revert "Revert "Remove distinction btw atx and setext header in AST.""John MacFarlane
This reverts commit 4570eb2bff2e1b71fa5b6408abbc69c98ff5ff24.
2014-11-23Merge pull request #215 from nwellnhof/no_trailing_newlineJohn MacFarlane
Fix segfault on docs without trailing newline
2014-11-23Fix segfault on docs without trailing newlineNick Wellnhofer
2014-11-22Revert "Remove distinction btw atx and setext header in AST."John MacFarlane
This reverts commit a71423f6ee1b77d9f79d42599ea00b4ca99f5da0. Not quite sure about this change, so reverting for now. Note that we still have a distinction between fenced and indented code blocks in the AST. These two distinctions seem to stand or fall together.
2014-11-22Remove distinction btw atx and setext header in AST.John MacFarlane
Now we just have 'header' -- Setext and ATX are just two ways of forming these; it's not a semantic difference that should remain in the AST.
2014-11-22Renamed NODE_BQUOTE -> NODE_BLOCK_QUOTE.John MacFarlane
2014-11-22Fixed #192.John MacFarlane
The C and JS implementations were not registering blank lines after atx headers for purposes of tight/loose list calculation. Exmaple: * item * # block1 ## block2
2014-11-22Updated JS and C implementations for #209.John MacFarlane
A setext header was being treated a if it were a blank line for purposes of tight/loose list determination. Closes #209.
2014-11-22Changed Example 196 in spec to reflect the spec's rules.John MacFarlane
This has no blank lines and so should not be a loose list. See #209.
2014-11-22js html renderer: formatting changes to align with changes in spec.John MacFarlane
2014-11-22C html renderer: ensure newline before hr or raw html block.John MacFarlane
2014-11-22Adjusted formatting of test - CR btw `<li>` and `<hr />`.John MacFarlane
This is consistent with treatment of other block elements in list items.
2014-11-22Fixed #214 C and JS implementations.John MacFarlane
They were gobbling whitespace after shortcut reference links, e.g. [foo] bar [foo]: url Closes #214.
2014-11-22Added example to spec of shortcut link with following space.John MacFarlane
See #214.
2014-11-22Added an Authors section to README.John MacFarlane
2014-11-22commonmark.rb: Added 'transform' and demos.John MacFarlane
2014-11-22Fixed typo.John MacFarlane
starting
2014-11-22Revised benchmark procedure.John MacFarlane
Now we take the difference of the time to process the input and the time to run with no input. This compensates for slow startup time in dynamic languages. See comments on 2dcef8a
2014-11-22Merge pull request #210 from nwellnhof/nwellnhof_141122John MacFarlane
C API fixes, additional tests, stackless rendering
2014-11-22Added examples of using walk.John MacFarlane
2014-11-22Make parser accept a char*Nick Wellnhofer
2014-11-22commonmark.rb - added walk, an iterator that walks the AST.John MacFarlane
2014-11-22Make render_html return a char*Nick Wellnhofer
2014-11-22Fix hierarchy testNick Wellnhofer
2014-11-22Stackless HTML renderingNick Wellnhofer
Now that every node has a parent pointer, it's possible to implement the HTML rendering functions without render stacks and any dynamic memory allocations. This commit also adds some minor optimizations that eliminate some strbuf_put* calls for the common case and avoid printf for headers.
2014-11-22Fix debug flagsNick Wellnhofer
2014-11-22Fix and test node_checkNick Wellnhofer
2014-11-22Tests for node_previous, node_parent, node_unlinkNick Wellnhofer
2014-11-22Rearrange get_type testsNick Wellnhofer
2014-11-22Set defaults for new headers and listsNick Wellnhofer
2014-11-22More tree hierarchy checks and testsNick Wellnhofer
2014-11-21commonmark.rb: omit title if empty.John MacFarlane
2014-11-21Added a needed parameter in runtests.py.John MacFarlane
2014-11-21commonmark.rb - support remaining elements in HTML renderer.John MacFarlane
Halt on finding unsupported method.
2014-11-21More commonmark.rb improvements.John MacFarlane
2014-11-21commonmark.rb - more elegant out, blocksep.John MacFarlane
2014-11-21Further improvements to commonmark.rb.John MacFarlane
2014-11-21commonmark.rb - implemented headers.John MacFarlane
2014-11-21commonmark.rb: separated Renderer class.John MacFarlane
2014-11-21html: Simplified render_stack code.John MacFarlane
Now that we have just one node type, it is not necessary to have two separate 'push' functions.
2014-11-20commonmark.rb: more progress.John MacFarlane
2014-11-20commonmark.rb improvements.John MacFarlane
2014-11-20Added commonmark.rb, steps towards a nice ruby wrapper of the parser.John MacFarlane