From 282477bcfaeda180477878b18034bace8fd419bb Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Tue, 31 Aug 2021 20:56:44 +0530
Subject: [PATCH] When adding markdown (.md) or textile (.textile) files that
 contain references to images, automatically add them as txtz with the images

---
 src/calibre/customize/builtins.py             | 10 +++----
 .../ebooks/conversion/plugins/txt_input.py    | 13 +++++++--
 src/calibre/ebooks/txt/processor.py           | 28 ++++++++++---------
 3 files changed, 30 insertions(+), 21 deletions(-)

diff --git a/src/calibre/customize/builtins.py b/src/calibre/customize/builtins.py
index d152107f83..0bf4038596 100644
--- a/src/calibre/customize/builtins.py
+++ b/src/calibre/customize/builtins.py
@@ -54,21 +54,19 @@ class TXT2TXTZ(FileTypePlugin):
         'containing Markdown or Textile references to images. The referenced '
         'images as well as the TXT file are added to the archive.')
     version = numeric_version
-    file_types = {'txt', 'text'}
+    file_types = {'txt', 'text', 'md', 'markdown', 'textile'}
     supported_platforms = ['windows', 'osx', 'linux']
     on_import = True
 
-    def _get_image_references(self, txt, base_dir):
-        from calibre.ebooks.txt.processor import get_images_from_polyglot_text
-        return get_images_from_polyglot_text(txt, base_dir)
-
     def run(self, path_to_ebook):
         from calibre.ebooks.metadata.opf2 import metadata_to_opf
+        from calibre.ebooks.txt.processor import get_images_from_polyglot_text
 
         with open(path_to_ebook, 'rb') as ebf:
             txt = ebf.read().decode('utf-8', 'replace')
         base_dir = os.path.dirname(path_to_ebook)
-        images = self._get_image_references(txt, base_dir)
+        ext = path_to_ebook.rpartition('.')[-1].lower()
+        images = get_images_from_polyglot_text(txt, base_dir, ext)
 
         if images:
             # Create TXTZ and put file plus images inside of it.
diff --git a/src/calibre/ebooks/conversion/plugins/txt_input.py b/src/calibre/ebooks/conversion/plugins/txt_input.py
index 682364d7c0..2a1f0c0759 100644
--- a/src/calibre/ebooks/conversion/plugins/txt_input.py
+++ b/src/calibre/ebooks/conversion/plugins/txt_input.py
@@ -152,7 +152,9 @@ class TXTInput(InputFormatPlugin):
             zf.extractall('.')
 
             for x in walk('.'):
-                if os.path.splitext(x)[1].lower() in ('.txt', '.text'):
+                ext = os.path.splitext(x)[1].lower()
+                if ext in ('.txt', '.text', '.textile', '.md', '.markdown'):
+                    file_ext = ext
                     with open(x, 'rb') as tf:
                         txt += tf.read() + b'\n\n'
             if os.path.exists('metadata.opf'):
@@ -168,10 +170,17 @@ class TXTInput(InputFormatPlugin):
                     if txt_formatting is not None and txt_formatting.text:
                         txt_formatting = txt_formatting.text.strip()
                         if txt_formatting in ('plain', 'textile', 'markdown') and options.formatting_type == 'auto':
-                            log.info(f'Using metadata from TXTZ archive to set text formmating type to: {txt_formatting}')
+                            log.info(f'Using metadata from TXTZ archive to set text formating type to: {txt_formatting}')
                             options.formatting_type = txt_formatting
                             if txt_formatting != 'plain':
                                 options.paragraph_type = 'off'
+            if options.formatting_type == 'auto':
+                if file_ext == 'textile':
+                    options.formatting_type = txt_formatting
+                    options.paragraph_type = 'off'
+                elif file_ext in ('md', 'markdown'):
+                    options.formatting_type = txt_formatting
+                    options.paragraph_type = 'off'
         else:
             if getattr(stream, 'name', None):
                 base_dir = os.path.dirname(stream.name)
diff --git a/src/calibre/ebooks/txt/processor.py b/src/calibre/ebooks/txt/processor.py
index 5a84cbc965..fe46e39b1d 100644
--- a/src/calibre/ebooks/txt/processor.py
+++ b/src/calibre/ebooks/txt/processor.py
@@ -343,7 +343,7 @@ def detect_formatting_type(txt):
     return 'heuristic'
 
 
-def get_images_from_polyglot_text(txt: str, base_dir: str = '') -> set:
+def get_images_from_polyglot_text(txt: str, base_dir: str = '', file_ext: str = 'txt') -> set:
     from calibre.ebooks.oeb.base import OEB_IMAGES
     from calibre import guess_type
     if not base_dir:
@@ -354,17 +354,19 @@ def get_images_from_polyglot_text(txt: str, base_dir: str = '') -> set:
         if path and not os.path.isabs(path) and guess_type(path)[0] in OEB_IMAGES and os.path.exists(os.path.join(base_dir, path)):
             images.add(path)
 
-    # Textile
-    for m in re.finditer(r'(?mu)(?:[\[{])?\!(?:\. )?(?P<path>[^\s(!]+)\s?(?:\(([^\)]+)\))?\!(?::(\S+))?(?:[\]}]|(?=\s|$))', txt):
-        path = m.group('path')
-        check_path(path)
+    if file_ext in ('txt', 'text', 'textile'):
+        # Textile
+        for m in re.finditer(r'(?mu)(?:[\[{])?\!(?:\. )?(?P<path>[^\s(!]+)\s?(?:\(([^\)]+)\))?\!(?::(\S+))?(?:[\]}]|(?=\s|$))', txt):
+            path = m.group('path')
+            check_path(path)
 
-    # Markdown
-    from markdown import Markdown
-    html = HTML_TEMPLATE % ('', Markdown().convert(txt))
-    from html5_parser import parse
-    root = parse(html)
-    for img in root.iterdescendants('img'):
-        path = img.get('src')
-        check_path(path)
+    if file_ext in ('txt', 'text', 'md', 'markdown'):
+        # Markdown
+        from markdown import Markdown
+        html = HTML_TEMPLATE % ('', Markdown().convert(txt))
+        from html5_parser import parse
+        root = parse(html)
+        for img in root.iterdescendants('img'):
+            path = img.get('src')
+            check_path(path)
     return images