HTML E-Mail

From d60226ed4be8d76d12ee1db8f46ee49de9c3a567 Mon Sep 17 00:00:00 2001 From: "martin f. krafft" Date: Fri, 1 Nov 2019 20:42:24 +1300 Subject: [PATCH 1/1] Enable multipart/alternative sending if supported --- .gitignore.d/mutt | 5 +- .mutt/markdown2html | 288 ++++++++++++++++++++++++++++++++++++ .mutt/multipart-alternative | 4 + .mutt/muttrc | 2 + 4 files changed, 298 insertions(+), 1 deletion(-) create mode 100755 .mutt/markdown2html create mode 100644 .mutt/multipart-alternative diff --git a/.gitignore.d/mutt b/.gitignore.d/mutt index 17a9235..a6f0ff5 100644 --- a/.gitignore.d/mutt +++ b/.gitignore.d/mutt @@ -7,7 +7,9 @@ !/.mutt/append-header !/.mutt/batch-subject-editor !/.mutt/bgrun -!/.mutt/bgrun-fifo +!/.mutt/bgview +!/.mutt/bgview-delay +!/.mutt/bgview-fifo !/.mutt/colours !/.mutt/confvars !/.mutt/edit-header @@ -26,6 +28,7 @@ !/.mutt/mailcap.htmldump !/.mutt/mailcap.icalendar !/.mutt/mkconf +!/.mutt/multipart-alternative !/.mutt/muttrc !/.muttprintrc !/.mutt/remove-header diff --git a/.mutt/markdown2html b/.mutt/markdown2html new file mode 100755 index 0000000..44ec07c --- /dev/null +++ b/.mutt/markdown2html @@ -0,0 +1,288 @@ +#!/usr/bin/python3 +# +# markdown2html.py â simple Markdown-to-HTML converter for use with Mutt +# +# Mutt recently learnt [how to compose `multipart/alternative` +# emails][1]. This script assumes a message has been composed using Markdown +# (with a lot of pandoc extensions enabled), and translates it to `text/html` +# for Mutt to tie into such a `multipart/alternative` message. +# +# [1]: https://gitlab.com/muttmua/mutt/commit/0e566a03725b4ad789aa6ac1d17cdf7bf4e7e354) +# +# Configuration: +# muttrc: +# set send_multipart_alternative=yes +# set send_multipart_alternative_filter=/path/to/markdown2html.py +# +# Optionally, Custom CSS styles will be read from `~/.mutt/markdown2html.css`, +# if present. +# +# Requirements: +# - python3 +# - PyPandoc (and pandoc installed, or downloaded) +# - Pynliner +# +# Optional: +# - Pygments, if installed, then syntax highlighting is enabled +# +# Latest version: +# https://git.madduck.net/etc/mutt.git/blob_plain/HEAD:/.mutt/markdown2html +# +# Copyright Â© 2019 martin f. krafft +# Released under the GPL-2+ licence, just like Mutt itself. +# + +import pypandoc +import pynliner +import re +import os +import sys + +try: + from pygments.formatters import get_formatter_by_name + formatter = get_formatter_by_name('html', style='default') + DEFAULT_CSS = formatter.get_style_defs('.sourceCode') + +except ImportError: + DEFAULT_CSS = "" + + +DEFAULT_CSS += ''' +.quote { + padding: 0 0.5em; + margin: 0; + font-style: italic; + border-left: 2px solid #ccc; + color: #999; + font-size: 80%; +} +.quotelead { + font-style: italic; + margin-bottom: -1em; + color: #999; + font-size: 80%; +} +.footnote-ref, .footnote-back { text-decoration: none;} +.signature { + color: #999; + font-family: monospace; + white-space: pre; + margin: 1em 0 0 0; + font-size: 80%; +}''' + +STYLESHEET = os.path.join(os.path.expanduser('~/.mutt'), + 'markdown2html.css') +if os.path.exists(STYLESHEET): + DEFAULT_CSS += open(STYLESHEET).read() + +HTML_DOCUMENT = ''' + + + + +HTML E-Mail + +{htmlbody} +''' + + +SIGNATURE_HTML = \ + '

-- {sig}

' + + +def _preprocess_markdown(mdwn): + ''' + Preprocess Markdown for handling by the converter. + ''' + # convert hard line breaks within paragraphs to 2 trailing spaces, which + # is the markdown way of representing hard line breaks. Note how the + # regexp will not match between paragraphs. + ret = re.sub(r'(\S)\n(\s*\S)', r'\g<1> \n\g<2>', mdwn, re.MULTILINE) + + return ret + + +def _identify_quotes_for_later(mdwn): + ''' + Email quoting such as: + + ``` + On 1970-01-01, you said: + > The Flat Earth Society has members all around the globe. + ``` + + isn't really properly handled by Markdown, so let's do our best to + identify the individual elements, and mark them, using a syntax similar to + what pandoc uses already in some cases. As pandoc won't actually use these + data (yet?), we call `self._reformat_quotes` later to use these markers + to slap the appropriate classes on the HTML tags. + ''' + + def generate_lines_with_context(mdwn): + ''' + Iterates the input string line-wise, returning a triplet of + previous, current, and next line, the first and last of which + will be None on the first and last line of the input data + respectively. + ''' + prev = cur = nxt = None + lines = iter(mdwn.splitlines()) + cur = next(lines) + for nxt in lines: + yield prev, cur, nxt + prev = cur + cur = nxt + yield prev, cur, None + + ret = [] + for prev, cur, nxt in generate_lines_with_context(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*$', cur) and nxt.startswith('>'): + ret.append(f'{{.quotelead}}{cur.strip()}') + # pandoc needs an empty line before the blockquote, so + # we enter one for the purpose of HTML rendition: + ret.append('') + continue + + # The first blockquote after such a lead-in gets marked as the + # "initial" quote: + elif prev and re.match(r'^.+:\s*$', prev) and cur.startswith('>'): + ret.append(re.sub(r'^(\s*>\s*)+(.+)', + r'\g<1>{.quoteinitial}\g<2>', + cur, re.MULTILINE)) + + # All other occurrences of blockquotes get the "subsequent" marker: + elif cur.startswith('>') and not prev.startswith('>'): + ret.append(re.sub(r'^((?:\s*>\s*)+)(.+)', + r'\g<1>{.quotesubsequent}\g<2>', + cur, re.MULTILINE)) + + else: # pass through everything else. + ret.append(cur) + + return '\n'.join(ret) + + +def _reformat_quotes(html): + ''' + Earlier in the pipeline, we marked email quoting, using markers, which we + now need to turn into HTML classes, so that we can use CSS to style them. + ''' + ret = html.replace('

{.quotelead}', '

') + ret = re.sub(r'

\n((?:
\n)*)
(?:\{\.quote(\w+)\})', + r'
\n\g<1>
', ret, re.MULTILINE) + return ret + + + +def _convert_with_pandoc(mdwn, inputfmt='markdown', outputfmt='html5', + ext_enabled=None, ext_disabled=None, + standalone=True, title="HTML E-Mail"): + ''' + Invoke pandoc to do the actual conversion of Markdown to HTML5. + ''' + if not ext_enabled: + ext_enabled = [ 'backtick_code_blocks', + 'line_blocks', + 'fancy_lists', + 'startnum', + 'definition_lists', + 'example_lists', + 'table_captions', + 'simple_tables', + 'multiline_tables', + 'grid_tables', + 'pipe_tables', + 'all_symbols_escapable', + 'intraword_underscores', + 'strikeout', + 'superscript', + 'subscript', + 'fenced_divs', + 'bracketed_spans', + 'footnotes', + 'inline_notes', + 'emoji', + 'tex_math_double_backslash', + ] + if not ext_disabled: + ext_disabled = [ 'tex_math_single_backslash', + 'tex_math_dollars', + 'raw_html' + ] + + enabled = '+'.join(ext_enabled) + disabled = '-'.join(ext_disabled) + inputfmt = f'{inputfmt}+{enabled}-{disabled}' + + args = [] + if standalone: + args.append('--standalone') + if title: + args.append(f'--metadata=pagetitle:"{title}"') + + return pypandoc.convert_text(mdwn, format=inputfmt, to=outputfmt, + extra_args=args) + + +def _apply_styling(html): + ''' + Inline all styles defined and used into the individual HTML tags. + ''' + return pynliner.Pynliner().from_string(html).with_cssString(DEFAULT_CSS).run() + + +def _postprocess_html(html): + ''' + Postprocess the generated and styled HTML. + ''' + return html + + +def convert_markdown_to_html(mdwn): + ''' + Converts the input Markdown to HTML, handling separately the body, as well + as an optional signature. + ''' + parts = re.split(r'^-- $', mdwn, 1, flags=re.MULTILINE) + body = parts[0] + if len(parts) == 2: + sig = parts[1] + else: + sig = None + + html='' + if body: + body = _preprocess_markdown(body) + body = _identify_quotes_for_later(body) + html = _convert_with_pandoc(body, standalone=False) + html = _reformat_quotes(html) + + if sig: + sig = _preprocess_markdown(sig) + html += SIGNATURE_HTML.format(sig='
'.join(sig.splitlines())) + + html = HTML_DOCUMENT.format(htmlbody=html) + html = _apply_styling(html) + html = _postprocess_html(html) + + return html + + +def main(): + ''' + Convert text on stdin to HTML, and print it to stdout, like mutt would + expect. + ''' + html = convert_markdown_to_html(sys.stdin.read()) + if html: + # mutt expects the content type in the first line, so: + print(f'text/html\n\n{html}') + + +if __name__ == '__main__': + main() diff --git a/.mutt/multipart-alternative b/.mutt/multipart-alternative new file mode 100644 index 0000000..ea54506 --- /dev/null +++ b/.mutt/multipart-alternative @@ -0,0 +1,4 @@ +set send_multipart_alternative=yes +set send_multipart_alternative_filter=$my_confdir/markdown2html + +# vim:ft=mutt diff --git a/.mutt/muttrc b/.mutt/muttrc index 7eea8cc..5ef2b95 100644 --- a/.mutt/muttrc +++ b/.mutt/muttrc @@ -29,3 +29,5 @@ set my_mutt_mailboxes = "$VARDIR/mutt/mailboxes" source "test -f $my_mutt_mailboxes && cat $my_mutt_mailboxes|" source "test -f $alias_file && cat $alias_file 2>/dev/null || echo unset alias_file|" + +source "mutt -DF/dev/null | grep -q send_multipart_alternative && cat $my_confdir/multipart-alternative|" -- 2.39.5