From: martin f. krafft <madduck@madduck.net>
Date: Thu, 21 Dec 2023 13:43:12 +0000 (+0100)
Subject: preserve format=flowed with nl2br
X-Git-Url: https://git.madduck.net/etc/neomutt.git/commitdiff_plain/e75c1b39e98e2f4b4062620ec8ea1e8eede23c65?ds=sidebyside;hp=aa5817ed858f5dc37dd282f2329e2c5f752019c4

preserve format=flowed with nl2br
---

diff --git a/.config/neomutt/buildmimetree.py b/.config/neomutt/buildmimetree.py
index 1ff0cb0..f10158f 100755
--- a/.config/neomutt/buildmimetree.py
+++ b/.config/neomutt/buildmimetree.py
@@ -54,7 +54,11 @@ from contextlib import contextmanager
 from collections import namedtuple, OrderedDict
 from markdown.extensions import Extension
 from markdown.blockprocessors import BlockProcessor
-from markdown.inlinepatterns import ImageInlineProcessor, IMAGE_LINK_RE
+from markdown.inlinepatterns import (
+    SimpleTextInlineProcessor,
+    ImageInlineProcessor,
+    IMAGE_LINK_RE,
+)
 from email.utils import make_msgid
 from urllib import request
 
@@ -176,7 +180,6 @@ def parse_cli_args(*args, **kwargs):
 
 
 class File:
-
     class Op(enum.Enum):
         R = enum.auto()
         W = enum.auto()
@@ -195,10 +198,7 @@ class File:
         if content and not re.search(r"[r+]", mode):
             raise RuntimeError("Cannot specify content without read mode")
 
-        self._cache = {
-            File.Op.R: [content] if content else [],
-            File.Op.W: []
-        }
+        self._cache = {File.Op.R: [content] if content else [], File.Op.W: []}
         self._lastop = None
         self._mode = mode
         self._kwargs = kwargs
@@ -250,7 +250,6 @@ class File:
             return self._file.read()
 
     def write(self, s, *, cache=True):
-
         if self._lastop == File.Op.R:
             try:
                 self._file.seek(0)
@@ -476,11 +475,22 @@ def apply_styling(html, css):
     )
 
 
+# [ FORMAT=FLOWED HANDLING ] ##################################################
+
+
+class FormatFlowedNewlineExtension(Extension):
+    FFNL_RE = r"(?!\S)(\s)\n"
+
+    def extendMarkdown(self, md):
+        ffnl = SimpleTextInlineProcessor(self.FFNL_RE)
+        md.inlinePatterns.register(ffnl, "ffnl", 125)
+
+
 # [ QUOTE HANDLING ] ##########################################################
 
 
 class QuoteToAdmonitionExtension(Extension):
-    class EmailQuoteBlockProcessor(BlockProcessor):
+    class BlockProcessor(BlockProcessor):
         RE = re.compile(r"(?:^|\n)>\s*(.*)")
 
         def __init__(self, parser):
@@ -528,9 +538,7 @@ class QuoteToAdmonitionExtension(Extension):
 
             admonition[0].set("class", "admonition-title")
             with self.disable():
-                self.parser.parseChunk(
-                    admonition, "\n".join(quotelines)
-                )
+                self.parser.parseChunk(admonition, "\n".join(quotelines))
 
         @contextmanager
         def disable(self):
@@ -545,7 +553,7 @@ class QuoteToAdmonitionExtension(Extension):
 
     def extendMarkdown(self, md):
         md.registerExtension(self)
-        email_quote_proc = self.EmailQuoteBlockProcessor(md.parser)
+        email_quote_proc = self.BlockProcessor(md.parser)
         md.parser.blockprocessors.register(email_quote_proc, "emailquote", 25)
 
 
@@ -700,6 +708,7 @@ def convert_markdown_to_html(
     ] = _CODEHILITE_CLASS
 
     extensions = extensions or []
+    extensions.append(FormatFlowedNewlineExtension())
     extensions.append(QuoteToAdmonitionExtension())
 
     draft = draft_f.read()
@@ -1768,7 +1777,7 @@ try:
                 with (
                     File() as draft_f,
                     File(mode="w") as cmd_f,
-                    File(content=css) as css_f
+                    File(content=css) as css_f,
                 ):
                     do_massage(
                         draft_f=draft_f,
@@ -1916,9 +1925,7 @@ try:
                 "This is the plain-text version",
             )
             htmlsig = "HTML Signature from {path} but as a string"
-            html = (
-                f'<div id="signature"><p>{htmlsig.format(path=fakepath2)}</p></div>'
-            )
+            html = f'<div id="signature"><p>{htmlsig.format(path=fakepath2)}</p></div>'
 
             sig_f = fakefilefactory(fakepath2, content=html)
 
@@ -2117,6 +2124,46 @@ try:
                 == mailparts[-2]
             )
 
+        @pytest.mark.converter
+        def test_converter_format_flowed_with_nl2br(
+            self, fakepath, fakefilefactory
+        ):
+            mailparts = (
+                "This is format=flowed text ",
+                "with spaces at the end ",
+                "and there ought be no newlines.",
+                "",
+                "[link](https://example.org) ",
+                "and text.",
+                "",
+                "[link text ",
+                "broken up](https://example.org).",
+                "",
+                "This is on a new line with a hard break  ",
+                "due to the double space",
+            )
+            with fakefilefactory(
+                fakepath, content="\n".join(mailparts)
+            ) as draft_f:
+                convert_markdown_to_html(
+                    draft_f, extensions=["nl2br"], filefactory=fakefilefactory
+                )
+
+            soup = bs4.BeautifulSoup(
+                fakefilefactory[fakepath.with_suffix(".html")].read(),
+                "html.parser",
+            )
+            import ipdb
+
+            p = soup.p.extract().text
+            assert "".join(mailparts[0:3]) == p
+            p = ''.join(map(str, soup.p.extract().contents))
+            assert p == '<a href="https://example.org">link</a> and text.'
+            p = ''.join(map(str, soup.p.extract().contents))
+            assert (
+                p == '<a href="https://example.org">link text broken up</a>.'
+            )
+
         @pytest.mark.fileio
         def test_file_class_contextmanager(self, const1, monkeypatch):
             state = dict(o=False, c=False)