Closed SrAceves closed 4 years ago
Is the source UTF-8 encoded? We are enabling UTF-8 mode for the regex engine. [I guess if it weren't, pandoc would have thrown an error.]
Is the source UTF-8 encoded? We are enabling UTF-8 mode for the regex engine.
Yes
Trace output:
Trying rule Rule {rMatcher = RegExpr (RE {reString = "\\\\(?:[\"\\\\/bfnrt]|u[0-9a-fA-f]{4})", reCaseSensitive = True}), rAttribute = CharTok, rIncludeAttribute = False, rDynamic = False, rCaseSensitive = True, rChildren = [], rLookahead = False, rFirstNonspace = False, rColumn = Nothing, rContextSwitch = []}
FALLTHROUGH Just (DataTypeTok,"descripci\243n\"")
Compare without the accent:
Trying rule Rule {rMatcher = RegExpr (RE {reString = "\\\\(?:[\"\\\\/bfnrt]|u[0-9a-fA-f]{4})", reCaseSensitive = True}), rAttribute = CharTok, rIncludeAttribute = False, rDynamic = False, rCaseSensitive = True, rChildren = [], rLookahead = False, rFirstNonspace = False, rColumn = Nothing, rContextSwitch = []}
FALLTHROUGH Just (DataTypeTok,"descripcion")
So when the accent is present, the fallthrough case is capturing the final "
, and that's what causes the problem. No idea why.
Running
on
example.md
with the following fencedjson
code: