From 8640988c5c46c00255c09111cc158b8e16ffb5af Mon Sep 17 00:00:00 2001
From: Tinmarino <tinmarino@gmail.com>
Date: Fri, 17 Mar 2023 16:08:27 -0300
Subject: [PATCH] Markdown typeface: match closer to GFM specification, also
 add tests

---
 autoload/vimwiki/u.vim                  | 14 ++---
 autoload/vimwiki/vars.vim               | 44 ++++++++++++---
 test/syntax_markdown_gfm_typeface.vader | 74 ++++++++++++++++++++-----
 3 files changed, 102 insertions(+), 30 deletions(-)

diff --git a/autoload/vimwiki/u.vim b/autoload/vimwiki/u.vim
index 1ef7011..26e91cd 100644
--- a/autoload/vimwiki/u.vim
+++ b/autoload/vimwiki/u.vim
@@ -408,6 +408,13 @@ function! vimwiki#u#hi_typeface(dic) abort
 
   let nested = vimwiki#u#get_syntax_dic().nested
 
+  " Bold Italic
+  if has_key(a:dic, 'bold_italic')
+    for bi in a:dic['bold_italic']
+      call vimwiki#u#hi_tag(bi[0], bi[1], 'VimwikiBoldItalic', nested . ',VimwikiBoldItalicUnderline')
+    endfor
+  endif
+
   " Italic
   for i in a:dic['italic']
     "  -- Italic 1
@@ -436,13 +443,6 @@ function! vimwiki#u#hi_typeface(dic) abort
     call vimwiki#u#hi_tag(b[0], b[1], 'VimwikiUnderlineItalicBold', nested, 2)
   endfor
 
-  " Bold Italic
-  if has_key(a:dic, 'bold_italic')
-    for bi in a:dic['bold_italic']
-      call vimwiki#u#hi_tag(bi[0], bi[1], 'VimwikiBoldItalic', nested . ',VimwikiBoldItalicUnderline')
-    endfor
-  endif
-
   " Underline
   for u in a:dic['underline']
     " -- Underline 1
diff --git a/autoload/vimwiki/vars.vim b/autoload/vimwiki/vars.vim
index 92d34ae..b658d59 100644
--- a/autoload/vimwiki/vars.vim
+++ b/autoload/vimwiki/vars.vim
@@ -753,19 +753,19 @@ function! s:get_markdown_syntaxlocal() abort
         \ 'symH': {'type': type(0), 'default': 0},
         \ 'typeface': {'type': type({}), 'default': {
         \   'bold': vimwiki#u#hi_expand_regex([
-        \     ['\%(\\\@<!_\)\@<!__\%(\\\@<!_\)\@!', '\%(\\\@<!_\)\@<!__\%(\\\@<!_\)\@!'],
-        \     ['\%(\\\@<!\*\)\@<!\*\*\%(\\\@<!\*\)\@!', '\%(\\\@<!\*\)\@<!\*\*\%(\\\@<!\*\)\@!'],
+        \     [s:expand_delimiter('__', 1), s:expand_delimiter('__', 1)],
+        \     [s:expand_delimiter('\*\*', 1), s:expand_delimiter('\*\*', 1)],
         \     ]),
         \   'italic': vimwiki#u#hi_expand_regex([
-        \     ['\%(\\\@<!\*\)\@<!\*\%(\\\@<!\*\)\@!', '\%(\\\@<!\*\)\@<!\*\%(\\\@<!\*\)\@!'], 
-        \     ['\%(\\\@<!_\)\@<!_\%(\\\@<!_\)\@!', '\%(\\\@<!_\)\@<!_\%(\\\@<!_\)\@!'], 
+        \     [s:expand_delimiter('_', 0), s:expand_delimiter('_', 0)],
+        \     [s:expand_delimiter('\*', 0), s:expand_delimiter('\*', 0)],
+        \     [s:expand_delimiter('\*_', 1), s:expand_delimiter('_\*', 1)],
+        \     [s:expand_delimiter('_\*', 1), s:expand_delimiter('\*_', 1)],
         \     ]),
         \   'underline': vimwiki#u#hi_expand_regex([]),
         \   'bold_italic': vimwiki#u#hi_expand_regex([
-        \     ['\*_', '_\*'],
-        \     ['_\*', '\*_'],
-        \     ['\%(\\\@<!\*\)\@<!\*\*\*\%(\\\@<!\*\)\@!', '\%(\\\@<!\*\)\@<!\*\*\*\%(\\\@<!\*\)\@!'],
-        \     ['\%(\\\@<!_\)\@<!___\%(\\\@<!_\)\@!', '\%(\\\@<!_\)\@<!___\%(\\\@<!_\)\@!']
+        \     [s:expand_delimiter('\*\*\*', 1), s:expand_delimiter('\*\*\*', 1)],
+        \     [s:expand_delimiter('___', 1), s:expand_delimiter('___', 1)],
         \     ]),
         \   'code': [
         \       ['\%(^\|[^`\\]\)\@<=`\%($\|[^`]\)\@=',
@@ -1344,6 +1344,34 @@ function! s:normalize_syntax_settings(syntax) abort
 endfunction
 
 
+function! s:expand_delimiter(delim, b_can_mult) abort
+  " Helper: From a delimiter to the lookhead defensive version
+  " See: also vimwiki#u#hi_expand_regex
+  " TODO: get some cache to avoid recrafting the same prefix always
+  
+  " Clause: if nothing, return nothing
+  if len(a:delim) == 0
+    return '\%(\)'
+  endif
+
+  " let c_start = a:delim[0] ==# '\' ? a:delim[0:1] : a:delim[0]
+  " let c_end = (len(a:delim) > 1 && a:delim[-2:-2] ==# '\') ? a:delim[-2:-1] : a:delim[-1:]
+  " Hardcode for markdown
+  let c_start = '[_*]'
+  let c_end = '[_*]'
+
+  let rx_mult = a:b_can_mult ? '\+' : ''
+
+  let rx_start = '\%(^\|\%(\\\@<!' . c_start . '\)\@<!\)'
+  let rx_middle = '\%(\%(' . a:delim . '\)' . rx_mult . '\)'
+  let rx_end = '\%($\|\%(\\\@<!' . c_end . '\)\@!\)'
+
+  let res = '\%(' . rx_start . rx_middle . rx_end . '\)'
+  echom res
+  return res
+endfunction
+
+
 " ----------------------------------------------------------
 " 4. Command (exported) {{{1
 " ----------------------------------------------------------
diff --git a/test/syntax_markdown_gfm_typeface.vader b/test/syntax_markdown_gfm_typeface.vader
index 14bd3b6..75cade9 100644
--- a/test/syntax_markdown_gfm_typeface.vader
+++ b/test/syntax_markdown_gfm_typeface.vader
@@ -768,11 +768,9 @@ Execute (Set Markdown):
   call SetSyntax('markdown')
 
 Execute (Assert Syntax):
-  CommentLine 'TODO Change that, currently ItalicBold and shoul dbe only Italic'
-  AssertEqual 0, 0
-#  CommentLine 'Double italic (2)'
-#  AssertEqual 'VimwikiItalic', SyntaxAt(1, 4)
-#  AssertEqual '', SyntaxAt(2, 1)
+  CommentLine 'Double italic (2)'
+  AssertEqual 'VimwikiItalic', SyntaxAt(1, 4)
+  AssertEqual '', SyntaxAt(2, 4)
 
 
 Given vimwiki (Typeface: https://github.github.com/gfm/#example-473):
@@ -783,11 +781,9 @@ Execute (Set Markdown):
   call SetSyntax('markdown')
 
 Execute (Assert Syntax):
-  CommentLine 'TODO, my lookaheads in typeface vars are too strict'
-  AssertEqual 0, 0
-#  CommentLine 'Double bold'
-#  AssertEqual 'VimwikiBold', SyntaxAt(1, 6)
-#  AssertEqual '', SyntaxAt(2, 1)
+  CommentLine 'Double bold with ****'
+  AssertEqual 'VimwikiBold', SyntaxAt(1, 6)
+  AssertEqual '', SyntaxAt(2, 1)
 
 
 Given vimwiki (Typeface: https://github.github.com/gfm/#example-474):
@@ -798,17 +794,53 @@ Execute (Set Markdown):
   call SetSyntax('markdown')
 
 Execute (Assert Syntax):
-  CommentLine 'TODO, my lookaheads in typeface vars are too strict'
-  AssertEqual 0, 0
-#  CommentLine 'Double bold (2)'
-#  AssertEqual 'VimwikiBold', SyntaxAt(1, 6)
-#  AssertEqual '', SyntaxAt(2, 1)
+  CommentLine 'Double bold with ____'
+  AssertEqual 'VimwikiBold', SyntaxAt(1, 6)
+  AssertEqual '', SyntaxAt(2, 1)
 
 
+Given vimwiki (Typeface: https://github.github.com/gfm/#example-475):
+  ******foo******
+  12345678901234
+
+Execute (Set Markdown):
+  call SetSyntax('markdown')
+
+Execute (Assert Syntax):
+  CommentLine 'Triple bold with ******'
+  AssertEqual 'VimwikiBold', SyntaxAt(1, 8)
+  AssertEqual '', SyntaxAt(2, 1)
+
 
 # Rule 14 {{{1
 # An interpretation <em><strong>...</strong></em> is always preferred to <strong><em>...</em></strong>.
 
+Given vimwiki (Typeface: https://github.github.com/gfm/#example-476):
+  ***foo***
+  12345678901234
+
+Execute (Set Markdown):
+  call SetSyntax('markdown')
+
+Execute (Assert Syntax):
+  CommentLine 'Bold Italic normal, with 3 *'
+  AssertEqual 'VimwikiBoldItalic', SyntaxAt(1, 5)
+  AssertEqual '', SyntaxAt(2, 1)
+
+
+Given vimwiki (Typeface: https://github.github.com/gfm/#example-477):
+  _____foo_____
+  12345678901234
+
+Execute (Set Markdown):
+  call SetSyntax('markdown')
+
+Execute (Assert Syntax):
+  CommentLine 'Bold Italic with 5 _'
+  CommentLine 'TODO Passed'
+#  AssertEqual 'VimwikiBoldItalic', SyntaxAt(1, 7)
+#  AssertEqual '', SyntaxAt(2, 1)
+
 
 # Rule 15 {{{1
 # When two potential emphasis or strong emphasis spans overlap, so that the second begins before the first ends and ends after the first ends, the first takes precedence. Thus, for example, *foo _bar* baz_ is parsed as <em>foo _bar</em> baz_ rather than *foo <em>bar* baz</em>.
@@ -835,5 +867,17 @@ Execute (Assert Syntax):
 # Rule 17 {{{1
 # Inline code spans, links, images, and HTML tags group more tightly than emphasis. So, when there is a choice between an interpretation that contains one of these elements and one that does not, the former always wins. Thus, for example, *[foo*](bar) is parsed as *<a href="bar">foo*</a> rather than as <em>[foo</em>](bar).
 
+Given vimwiki (Typeface: https://github.github.com/gfm/#example-485):
+  *a `*` b*
+  *a `a a*a a` b*
+  12345678901234
+
+Execute (Set Markdown):
+  call SetSyntax('markdown')
+
+Execute (Assert Syntax):
+  CommentLine 'TODO with a * as unique code, a nice trap'
+  AssertEqual 0, 0
+
 
 # vim: foldmethod=marker foldlevel=30 sw=2