TXT Output: Add option to generate Markdown output. Turn <br> tags into spaces.

src/calibre/ebooks/txt/markdownml.py

@@ -0,0 +1,63 @@
+# -*- coding: utf-8 -*-
+
+__license__ = 'GPL 3'
+__copyright__ = '2009, John Schember <john@nachtimwald.com>'
+__docformat__ = 'restructuredtext en'
+
+'''
+Transform OEB content into Markdown formatted plain text
+'''
+
+import re
+
+from lxml import etree
+
+from calibre.utils.html2text import html2text
+
+class MarkdownMLizer(object):
+
+    def __init__(self, log):
+        self.log = log
+
+    def extract_content(self, oeb_book, opts):
+        self.log.info('Converting XHTML to Markdown formatted TXT...')
+        self.oeb_book = oeb_book
+        self.opts = opts
+
+        return self.mlize_spine()
+
+    def mlize_spine(self):
+        output = [u'']
+        
+        for item in self.oeb_book.spine:
+            self.log.debug('Converting %s to Markdown formatted TXT...' % item.href)
+            
+            html = unicode(etree.tostring(item.data, encoding=unicode))
+            
+            if not self.opts.keep_links:
+                html = re.sub(r'<\s*a[^>]*>', '', html)
+                html = re.sub(r'<\s*/\s*a\s*>', '', html)
+            if not self.opts.keep_image_references:
+                html = re.sub(r'<\s*img[^>]*>', '', html)
+                html = re.sub(r'<\s*img\s*>', '', html)
+            
+            text = html2text(html)
+        
+            # Ensure the section ends with at least two new line characters.
+            # This is to prevent the last paragraph from a section being
+            # combined into the fist paragraph of the next.
+            end_chars = text[-4:]
+            # Convert all newlines to \n
+            end_chars = end_chars.replace('\r\n', '\n')
+            end_chars = end_chars.replace('\r', '\n')
+            end_chars = end_chars[-2:]
+            if not end_chars[1] == '\n':
+                text += '\n\n'
+            if end_chars[1] == '\n' and not end_chars[0] == '\n':
+                text += '\n'
+            
+            output += text
+            
+        output = u''.join(output)
+
+        return output

src/calibre/ebooks/txt/output.py

@@ -8,6 +8,7 @@ import os
 
 from calibre.customize.conversion import OutputFormatPlugin, \
     OptionRecommendation
+from calibre.ebooks.txt.markdownml import MarkdownMLizer
 from calibre.ebooks.txt.txtml import TXTMLizer
 from calibre.ebooks.txt.newlines import TxtNewlines, specified_newlines
 
@@ -44,10 +45,27 @@ class TXTOutput(OutputFormatPlugin):
             recommended_value=False, level=OptionRecommendation.LOW,
             help=_('Force splitting on the max-line-length value when no space '
             'is present. Also allows max-line-length to be below the minimum')),
+        OptionRecommendation(name='markdown_format',
+            recommended_value=False, level=OptionRecommendation.LOW,
+            help=_('Produce Markdown formatted text.')),
+        OptionRecommendation(name='keep_links',
+            recommended_value=False, level=OptionRecommendation.LOW,
+            help=_('Do not remove links within the document. This is only ' \
+            'useful when paired with the markdown-format option because' \
+            'links are always removed with plain text output.')),
+        OptionRecommendation(name='keep_image_references',
+            recommended_value=False, level=OptionRecommendation.LOW,
+            help=_('Do not remove image references within the document. This is only ' \
+            'useful when paired with the markdown-format option because' \
+            'image references are always removed with plain text output.')),
      ])
 
     def convert(self, oeb_book, output_path, input_plugin, opts, log):
-        writer = TXTMLizer(log)
+        if opts.markdown_format:
+            writer = MarkdownMLizer(log)
+        else:
+            writer = TXTMLizer(log)
+        
         txt = writer.extract_content(oeb_book, opts)
 
         log.debug('\tReplacing newlines with selected type...')

src/calibre/ebooks/txt/txtml.py

@@ -35,6 +35,7 @@ BLOCK_STYLES = [
 
 SPACE_TAGS = [
     'td',
+    'br',
 ]
 
 class TXTMLizer(object):

src/calibre/gui2/convert/txt_output.py

@@ -21,7 +21,7 @@ class PluginWidget(Widget, Ui_Form):
     def __init__(self, parent, get_option, get_help, db=None, book_id=None):
         Widget.__init__(self, parent,
         ['newline', 'max_line_length', 'force_max_line_length',
-        'inline_toc'])
+        'inline_toc', 'markdown_format', 'keep_links', 'keep_image_references'])
         self.db, self.book_id = db, book_id
         self.initialize_options(get_option, get_help, db, book_id)
 

src/calibre/gui2/convert/txt_output.ui

@@ -6,7 +6,7 @@
    <rect>
     <x>0</x>
     <y>0</y>
-    <width>400</width>
+    <width>477</width>
     <height>300</height>
    </rect>
   </property>
@@ -27,7 +27,7 @@
    <item row="0" column="1">
     <widget class="QComboBox" name="opt_newline"/>
    </item>
-   <item row="4" column="0">
+   <item row="7" column="0">
     <spacer name="verticalSpacer">
      <property name="orientation">
       <enum>Qt::Vertical</enum>
@@ -67,6 +67,27 @@
      </property>
     </widget>
    </item>
+   <item row="4" column="0">
+    <widget class="QCheckBox" name="opt_markdown_format">
+     <property name="text">
+      <string>Apply Markdown formatting to text</string>
+     </property>
+    </widget>
+   </item>
+   <item row="5" column="0">
+    <widget class="QCheckBox" name="opt_keep_links">
+     <property name="text">
+      <string>Do not remove links (&lt;a&gt; tags) before processing</string>
+     </property>
+    </widget>
+   </item>
+   <item row="6" column="0">
+    <widget class="QCheckBox" name="opt_keep_image_references">
+     <property name="text">
+      <string>Do not remove image references before processing</string>
+     </property>
+    </widget>
+   </item>
   </layout>
  </widget>
  <resources/>