codewalker/marked
1
0
Fork 0
Code Pull Requests Actions Packages Releases Activity

fix: update punctuation regex syntax to fix babel mistaken transpile (#3547)

Browse Source 
Co-authored-by: jinbowen <jinbowen@bytedance.com>
This commit is contained in:
WizardMeow 2024-11-29 13:19:10 +08:00 committed by GitHub
parent 152175615c
commit 9b988c47bd
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 22 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
src/rules.ts
View File

@ -243,26 +243,30 @@ const br = /^( {2,}|\\)\n(?!\s*$)/;
const inlineText = /^(`+|[^`])(?:(?= {2,}\n)|[\s\S]*?(?:(?=[\\<!\[`*_]|\b_|$)|[^ ](?= {2,}\n)))/; const inlineText = /^(`+|[^`])(?:(?= {2,}\n)|[\s\S]*?(?:(?=[\\<!\[`*_]|\b_|$)|[^ ](?= {2,}\n)))/;
// list of unicode punctuation marks, plus any missing characters from CommonMark spec // list of unicode punctuation marks, plus any missing characters from CommonMark spec
const _punctuation = /\p{P}\p{S}/u; const _punctuation = /[\p{P}\p{S}]/u;
const punctuation = edit(/^((?![*_])[\spunctuation])/, 'u') const _punctuationOrSpace = /[\s\p{P}\p{S}]/u;
.replace(/punctuation/g, _punctuation).getRegex(); const _notPunctuationOrSpace = /[^\s\p{P}\p{S}]/u;
const punctuation = edit(/^((?![*_])punctSpace)/, 'u')
.replace(/punctSpace/g, _punctuationOrSpace).getRegex();
// sequences em should skip over [title](link), `code`, <html> // sequences em should skip over [title](link), `code`, <html>
const blockSkip = /\[[^[\]]*?\]\((?:\\.|[^\\\(\)]|\((?:\\.|[^\\\(\)])*\))*\)|`[^`]*?`|<[^<>]*?>/g; const blockSkip = /\[[^[\]]*?\]\((?:\\.|[^\\\(\)]|\((?:\\.|[^\\\(\)])*\))*\)|`[^`]*?`|<[^<>]*?>/g;
const emStrongLDelim = edit(/^(?:\*+(?:((?!\*)[punct])|[^\s*]))|^_+(?:((?!_)[punct])|([^\s_]))/, 'u') const emStrongLDelim = edit(/^(?:\*+(?:((?!\*)punct)|[^\s*]))|^_+(?:((?!_)punct)|([^\s_]))/, 'u')
.replace(/punct/g, _punctuation) .replace(/punct/g, _punctuation)
.getRegex(); .getRegex();
const emStrongRDelimAst = edit( const emStrongRDelimAst = edit(
'^[^_*]*?__[^_*]*?\\*[^_*]*?(?=__)' // Skip orphan inside strong '^[^_*]*?__[^_*]*?\\*[^_*]*?(?=__)' // Skip orphan inside strong
+ '|[^*]+(?=[^*])' // Consume to delim + '|[^*]+(?=[^*])' // Consume to delim
+ '|(?!\\*)[punct](\\*+)(?=[\\s]|$)' // (1) #*** can only be a Right Delimiter + '|(?!\\*)punct(\\*+)(?=[\\s]|$)' // (1) #*** can only be a Right Delimiter
+ '|[^punct\\s](\\*+)(?!\\*)(?=[punct\\s]|$)' // (2) a***#, a*** can only be a Right Delimiter + '|notPunctSpace(\\*+)(?!\\*)(?=punctSpace|$)' // (2) a***#, a*** can only be a Right Delimiter
+ '|(?!\\*)[punct\\s](\\*+)(?=[^punct\\s])' // (3) #***a, ***a can only be Left Delimiter + '|(?!\\*)punctSpace(\\*+)(?=notPunctSpace)' // (3) #***a, ***a can only be Left Delimiter
+ '|[\\s](\\*+)(?!\\*)(?=[punct])' // (4) ***# can only be Left Delimiter + '|[\\s](\\*+)(?!\\*)(?=punct)' // (4) ***# can only be Left Delimiter
+ '|(?!\\*)[punct](\\*+)(?!\\*)(?=[punct])' // (5) #***# can be either Left or Right Delimiter + '|(?!\\*)punct(\\*+)(?!\\*)(?=punct)' // (5) #***# can be either Left or Right Delimiter
+ '|[^punct\\s](\\*+)(?=[^punct\\s])', 'gu') // (6) a***a can be either Left or Right Delimiter + '|notPunctSpace(\\*+)(?=notPunctSpace)', 'gu') // (6) a***a can be either Left or Right Delimiter
.replace(/notPunctSpace/g, _notPunctuationOrSpace)
.replace(/punctSpace/g, _punctuationOrSpace)
.replace(/punct/g, _punctuation) .replace(/punct/g, _punctuation)
.getRegex(); .getRegex();
@ -270,15 +274,17 @@ const emStrongRDelimAst = edit(
const emStrongRDelimUnd = edit( const emStrongRDelimUnd = edit(
'^[^_*]*?\\*\\*[^_*]*?_[^_*]*?(?=\\*\\*)' // Skip orphan inside strong '^[^_*]*?\\*\\*[^_*]*?_[^_*]*?(?=\\*\\*)' // Skip orphan inside strong
+ '|[^_]+(?=[^_])' // Consume to delim + '|[^_]+(?=[^_])' // Consume to delim
+ '|(?!_)[punct](_+)(?=[\\s]|$)' // (1) #___ can only be a Right Delimiter + '|(?!_)punct(_+)(?=[\\s]|$)' // (1) #___ can only be a Right Delimiter
+ '|[^punct\\s](_+)(?!_)(?=[punct\\s]|$)' // (2) a___#, a___ can only be a Right Delimiter + '|notPunctSpace(_+)(?!_)(?=punctSpace|$)' // (2) a___#, a___ can only be a Right Delimiter
+ '|(?!_)[punct\\s](_+)(?=[^punct\\s])' // (3) #___a, ___a can only be Left Delimiter + '|(?!_)punctSpace(_+)(?=notPunctSpace)' // (3) #___a, ___a can only be Left Delimiter
+ '|[\\s](_+)(?!_)(?=[punct])' // (4) ___# can only be Left Delimiter + '|[\\s](_+)(?!_)(?=punct)' // (4) ___# can only be Left Delimiter
+ '|(?!_)[punct](_+)(?!_)(?=[punct])', 'gu') // (5) #___# can be either Left or Right Delimiter + '|(?!_)punct(_+)(?!_)(?=punct)', 'gu') // (5) #___# can be either Left or Right Delimiter
.replace(/notPunctSpace/g, _notPunctuationOrSpace)
.replace(/punctSpace/g, _punctuationOrSpace)
.replace(/punct/g, _punctuation) .replace(/punct/g, _punctuation)
.getRegex(); .getRegex();
const anyPunctuation = edit(/\\([punct])/, 'gu') const anyPunctuation = edit(/\\(punct)/, 'gu')
.replace(/punct/g, _punctuation) .replace(/punct/g, _punctuation)
.getRegex(); .getRegex();