md4c: upgrade 0.5.2 -> 0.5.3

Changes:
==========
- Avoid repeated prefix language- in code block language specification if the
  input already explicitly includes the prefix.
- Permissive autolink extensions (MD_FLAG_PERMISSIVExxxAUTOLINKS) are now tiny
  bit more permissive, allowing + and - characters to be anywhere in the path
  portion of the URL. This also improves compatibility with GFM.
- Make Unicode-specific code compliant to Unicode 18.0.

Fixes: ===========
- #236: Fix quadratic time behavior caused by one-by-one walking over block
  lines instead of calling md_lookup_line().
- #238: Fix quadratic time and output size behavior caused by malicious misuse
  of link reference definitions.
- #242: The strike-through extension (with flag MD_FLAG_STRIKETHROUGH) now
  follows same logic as other emphasis spans in respect to punctuation
  character and word boundaries.
- #248: Fix handling tab when removing trailing whitespace, especially in
  connection with ATX headers.
- #266: We now correctly abort the parser when a callback returns non-zero.
  (Previously it worked correctly only for negative values, values greater than
  zero were causing strange and inconsistent behavior.)
- #271: Fix handling a code span whose closer is on the next line and yet
  another text follows. In the case we erroneously outputted the closer code
  span mark as part of the text.
- #275: Fix md_decode_utf16le_before__(). (Only affected MD4C builds built with
  -MD4C_USE_UTF16 on Windows.)
- #278, #294: Do not try to interpret characters in a link URL as Markdown
  syntax characters.
- #292: Fix detection of closing code block fence if it has a trailing
  tabulator.
- #299: Fix invalid free() in an error path.

Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>

0 files changed, 0 insertions, 0 deletions