| Age | Commit message (Collapse) | Author |
|---|
|
We need to cast value passed to isspace(3) to unsigned char to explicitly
prevent possibly undefined behavior.
/tmp/pkgsrc-tmp/wip/cmark/work/cmark-0.24.1/src/commonmark.c: In function 'S_render_node':
/tmp/pkgsrc-tmp/wip/cmark/work/cmark-0.24.1/src/commonmark.c:273:9: warning: array subscript has type 'char' [-Wchar-subscripts]
(code_len > 2 && !isspace(code[0]) &&
^
/tmp/pkgsrc-tmp/wip/cmark/work/cmark-0.24.1/src/commonmark.c:274:10: warning: array subscript has type 'char' [-Wchar-subscripts]
!(isspace(code[code_len - 1]) && isspace(code[code_len - 2]))) &&
^
/tmp/pkgsrc-tmp/wip/cmark/work/cmark-0.24.1/src/commonmark.c:274:10: warning: array subscript has type 'char' [-Wchar-subscripts]
CTYPE(3) Library Functions Manual CTYPE(3)
NAME
isalpha, isupper, islower, isdigit, isxdigit, isalnum, isspace, ispunct,
isprint, isgraph, iscntrl, isblank, toupper, tolower, - character
classification and mapping functions
LIBRARY
Standard C Library (libc, -lc)
CAVEATS
The first argument of these functions is of type int, but only a very
restricted subset of values are actually valid. The argument must either
be the value of the macro EOF (which has a negative value), or must be a
non-negative value within the range representable as unsigned char.
Passing invalid values leads to undefined behavior.
NetBSD 7.99 February 25, 2015 NetBSD 7.99
|
|
|
|
|
|
|
|
We try not to escape punctuation unless we absolutely have
to. So, `)` and `.` are no longer escaped whenever they
occur after digits; now they are only escaped if they are
geuninely in a position where they'd cause a list item.
This required a couple changes to render.c.
- `renderer->begin_content` is only set to false AFTER a
string of digits at the beginning of the line. (This is
slightly unprincipled.)
- We never break before a numeral (also slightly unprincipled).
|
|
following list or code block.
This has several advantages. First, the two blank lines breaks
out of list syntax is still controversial in CommonMark.
And it isn't used in other implementations. HTML comments
will always work.
Second, two blank lines breaks out of all lists; an HTML
comment can be used to break out of just one level of nesting.
|
|
This makes the output compatible with more implementations.
|
|
This is more portable.
Closes #90.
|
|
This did not allow for the possibility that a node
might have no containing block, causing the commonmark
renderer to segfault if passed an inline node with no
block parent.
|
|
when they're at the beginning of a block, e.g.
> \- foo
|
|
API change. Sorry, but this is the time to break things,
before 1.0 is released. This matches the recent changes to
CommonMark.dtd.
|
|
CMARK_NODE_HRULE -> CMARK_NODE_THEMATIC_BREAK.
However we've defined the former as the latter to keep
backwards compatibility.
See jgm/CommonMark 8fa94cb460f5e516b0e57adca33f50a669d51f6c
|
|
Defined CMARK_NODE_HEADER to CMARK_NODE_HEADING to ease
the transition.
|
|
See jgm/CommonMark commit 0cdbcee4e840abd0ac7db93797b2b75ca4104314
Note that we have defined
cmark_node_get_header_level = cmark_node_get_heading_level
and
cmark_node_set_header_level = camrk_node_set_heading_level
for backwards compatibility in the API.
|
|
Otherwise we get failures of roundtrip tests.
|
|
Instead of using their `as.literal` content, we now
give each custom node *two* literal fields, one to
be printed on entering the node (before rendering
the children, if any), the other on exiting (after
rendering children).
This gives us the flexibility to have custom nodes
with children.
|
|
|
|
These are passed through verbatim by all writers, with no
escaping.
They are never generated by the parser, and do not correspond
to CommonMark elements. They are designed to be inserted by
filters that postprocess the AST. For example, a filter might
convert specially marked code blocks to svg diagrams in HTML
and tikz diagrams in LaTeX, passing these through to the renderer
as a RAW_BLOCK.
|
|
|
|
|
|
|
|
This fixes an MSVC warning "conversion from 'size_t' to 'int', possible loss of data"
|
|
* Reformatted all source files.
* Added 'format' target to Makefile.
* Removed 'astyle' target.
* Updated .editorconfig.
|
|
|
|
This avoids an allocation and use of strbuf_printf.
|
|
|
|
Now all round-trip tests pass.
|
|
This is needed for #s at the end of atx headers.
|
|
|
|
Moved begin_line setting into render.c, so you don't need to
worry about it in outc.
|
|
|
|
Added options argument to render_node function, and rearrange
argument order.
|
|
This allows us to remove direct manipulation of buffer from
the latex and commonmark renderers.
|
|
|
|
|
|
* Added functions for cr, blankline, out to renderer object.
* Removed lit (we'll handle this with a macro).
* Changed type of out so it takes a regular string instead of
a chunk.
* Use macros LIT, OUT, BLANKLINE, CR in renderers to simplify code.
(Not sure about this, but `renderer->out(renderer, ...)` sure is
verbose.)
|
|
|
|
* Added render.c, render.h.
* Moved common functions and definitions from latex.c and commonmark.c
to render.c, render.h.
* Added a wrapper, cmark_render, that creates a renderer given a
character-escaper and a node renderer.
Closes #63.
|
|
|
|
|
|
|
|
There are probably a couple of places I missed. But this will only
be a problem if we use a 64-bit bufsize_t at some point. Then, we'll
get warnings from -Wshorten-64-to-32.
|
|
|
|
Previously *any* link with an absolute URL was treated as an
autolink. Closes #50. See also jgm/pandoc#2203.
|
|
|
|
If unicode parsing gives an error condition, we just skip the
rest of the string without rendering anything and proceed.
I'm not sure if that's the best way to handle this, but
garbage in, garbage out.
Note: this bug was found using american fuzzy lop.
|
|
snprintf is not available on some platforms.
|
|
instead.
|
|
strlen should be safe here, as we use it on strings generated
by cmark_chunk_to_cstr and these should be null terminated.
|
|
|
