X-Git-Url: https://git.madduck.net/etc/mutt.git/blobdiff_plain/45e0ac7684ad3e7992531506c5136e339c95cee4..a83d50d5a64488deb3796bbd0ca0a03b0516cc77:/.mutt/markdown2html
diff --git a/.mutt/markdown2html b/.mutt/markdown2html
index 72d0830..487d3db 100755
--- a/.mutt/markdown2html
+++ b/.mutt/markdown2html
@@ -95,6 +95,7 @@ h4 { font-size: 107%; }
h5 { font-size: 103%; }
h6 { font-size: 100%; }
p { padding: 0 0.5em; }
+pre { padding: 0 1em; }
'''
STYLESHEET = os.path.join(os.path.expanduser('~/.mutt'),
@@ -102,20 +103,16 @@ STYLESHEET = os.path.join(os.path.expanduser('~/.mutt'),
if os.path.exists(STYLESHEET):
DEFAULT_CSS += open(STYLESHEET).read()
-HTML_DOCUMENT = '''
-
-
-
-
-
-{htmlbody}
-'''
-
-
SIGNATURE_HTML = \
'-- {sig}
'
+def _preprocess_signature(sig):
+ '''
+ Preprocess the signature before markdown processing.
+ '''
+ return sig
+
def _preprocess_markdown(mdwn):
'''
Preprocess Markdown for handling by the converter.
@@ -133,6 +130,13 @@ def _preprocess_markdown(mdwn):
# text-mode HTML2text converters, and so it's left commented for now.
#ret = re.sub(r'\n>', r' \n>[>]{.quotechar}', ret, flags=re.MULTILINE)
+ # With the autolink_bare_uris extension, we do not need to put links into
+ # angle brackets to have them converted, so let's conserve the brackets
+ # when used around email addresses. Note that this needs a postprocessing
+ # hack because the pandoc autolink converted includes the ambersand
+ # (https://github.com/jgm/pandoc/issues/7398).
+ ret = re.sub(r'<([^@]+@.+\.[^>]+)>', r'<\g<1> -PANDOC_BUG_7398->', ret)
+
return ret
@@ -174,7 +178,7 @@ def _identify_quotes_for_later(mdwn):
# The lead-in to a quote is a single line immediately preceding the
# quote, and ending with ':'. Note that there could be multiple of
# these:
- if re.match(r'^.+:\s*$', cur) and nxt.startswith('>'):
+ if re.match(r'^[^>]+.*:\s*$', cur) and nxt.startswith('>'):
ret.append(f'{{.quotelead}}{cur.strip()}')
# pandoc needs an empty line before the blockquote, so
# we enter one for the purpose of HTML rendition:
@@ -183,13 +187,13 @@ def _identify_quotes_for_later(mdwn):
# The first blockquote after such a lead-in gets marked as the
# "initial" quote:
- elif prev and re.match(r'^.+:\s*$', prev) and cur.startswith('>'):
+ elif prev and re.match(r'^[^>]+.*:\s*$', prev) and cur.startswith('>'):
ret.append(re.sub(r'^(\s*>\s*)+(.+)',
r'\g<1>{.quoteinitial}\g<2>',
cur, flags=re.MULTILINE))
# All other occurrences of blockquotes get the "subsequent" marker:
- elif cur.startswith('>') and prev and not prev.startswith('>'):
+ elif cur.startswith('>') and prev is not None and not prev.startswith('>'):
ret.append(re.sub(r'^((?:\s*>\s*)+)(.+)',
r'\g<1>{.quotesubsequent}\g<2>',
cur, flags=re.MULTILINE))
@@ -214,7 +218,7 @@ def _reformat_quotes(html):
def _convert_with_pandoc(mdwn, inputfmt='markdown', outputfmt='html5',
ext_enabled=None, ext_disabled=None,
- standalone=True, title="HTML E-Mail"):
+ standalone=True, selfcontained=True, title=None):
'''
Invoke pandoc to do the actual conversion of Markdown to HTML5.
'''
@@ -257,6 +261,8 @@ def _convert_with_pandoc(mdwn, inputfmt='markdown', outputfmt='html5',
args = []
if standalone:
args.append('--standalone')
+ if selfcontained:
+ args.append('--self-contained')
if title:
args.append(f'--metadata=pagetitle:"{title}"')
@@ -275,6 +281,10 @@ def _postprocess_html(html):
'''
Postprocess the generated and styled HTML.
'''
+
+ # Preprocessing leaves a sentinel to work around
+ # https://github.com/jgm/pandoc/issues/7398, and so we need to remove it:
+ html = html.replace(' -PANDOC_BUG_7398->', '>')
return html
@@ -294,14 +304,19 @@ def convert_markdown_to_html(mdwn):
if body:
body = _preprocess_markdown(body)
body = _identify_quotes_for_later(body)
- html = _convert_with_pandoc(body, standalone=False)
+ html = _convert_with_pandoc(body, standalone=True, selfcontained=True,
+ title=None)
+ html = html.replace('Untitled\n','')
html = _reformat_quotes(html)
if sig:
+ sig = _preprocess_signature(sig)
sig = _preprocess_markdown(sig)
- html += SIGNATURE_HTML.format(sig='
'.join(sig.splitlines()))
+ sig = _convert_with_pandoc(sig, standalone=False, selfcontained=False)
+ sig = SIGNATURE_HTML.format(sig='
'.join(sig.splitlines()))
+ eob = html.find(' ')
+ html = f'{html[:eob]}{sig}\n{html[eob:]}'
- html = HTML_DOCUMENT.format(htmlbody=html)
html = _apply_styling(html)
html = _postprocess_html(html)