Conversion: When converting markdown documents recognize basic metadata in the markdown document

src/calibre/ebooks/conversion/plugins/txt_input.py

@@ -77,7 +77,7 @@ class TXTInput(InputFormatPlugin):
         from calibre.ebooks.chardet import detect
         from calibre.utils.zipfile import ZipFile
         from calibre.ebooks.txt.processor import (convert_basic,
-                convert_markdown, separate_paragraphs_single_line,
+                convert_markdown_with_metadata, separate_paragraphs_single_line,
                 separate_paragraphs_print_formatted, preserve_spaces,
                 detect_paragraph_type, detect_formatting_type,
                 normalize_line_endings, convert_textile, remove_indents,
@@ -195,10 +195,11 @@ class TXTInput(InputFormatPlugin):
 
         # Process the text using the appropriate text processor.
         html = ''
+        input_mi = None
         if options.formatting_type == 'markdown':
             log.debug('Running text through markdown conversion...')
             try:
-                html = convert_markdown(txt, extensions=[x.strip() for x in options.markdown_extensions.split(',') if x.strip()])
+                input_mi, html = convert_markdown_with_metadata(txt, extensions=[x.strip() for x in options.markdown_extensions.split(',') if x.strip()])
             except RuntimeError:
                 raise ValueError('This txt file has malformed markup, it cannot be'
                     ' converted by calibre. See http://daringfireball.net/projects/markdown/syntax')
@@ -236,11 +237,12 @@ class TXTInput(InputFormatPlugin):
         os.remove(htmlfile.name)
 
         # Set metadata from file.
+        if input_mi is None:
             from calibre.customize.ui import get_file_type_metadata
+            input_mi = get_file_type_metadata(stream, file_ext)
         from calibre.ebooks.oeb.transforms.metadata import meta_info_to_oeb_metadata
-        mi = get_file_type_metadata(stream, file_ext)
+        meta_info_to_oeb_metadata(input_mi, oeb.metadata, log)
-        meta_info_to_oeb_metadata(mi, oeb.metadata, log)
+        self.html_postprocess_title = input_mi.title
-        self.html_postprocess_title = mi.title
 
         return oeb
 
@@ -250,4 +252,3 @@ class TXTInput(InputFormatPlugin):
                 for title in item.data.xpath('//*[local-name()="title"]'):
                     if title.text == _('Unknown'):
                         title.text = self.html_postprocess_title
-

src/calibre/ebooks/oeb/transforms/cover.py

@@ -96,7 +96,10 @@ class CoverManager(object):
             from calibre.ebooks.covers import create_cover
             series = series_index = None
             if m.series:
+                try:
                     series, series_index = unicode(m.series[0]), m.series_index[0]
+                except IndexError:
+                    pass
             img_data = create_cover(title, authors, series, series_index)
             id, href = self.oeb.manifest.generate('cover',
                     u'cover_image.jpg')

src/calibre/ebooks/txt/processor.py

@@ -99,7 +99,10 @@ def convert_basic(txt, title='', epub_split_size_kb=0):
     return HTML_TEMPLATE % (title, u'\n'.join(lines))
 
 
-def convert_markdown(txt, title='', extensions=('footnotes', 'tables', 'toc')):
+DEFAULT_MD_EXTENSIONS = ('footnotes', 'tables', 'toc')
+
+
+def convert_markdown(txt, title='', extensions=DEFAULT_MD_EXTENSIONS):
     from calibre.ebooks.conversion.plugins.txt_input import MD_EXTENSIONS
     from calibre.ebooks.markdown import Markdown
     extensions = ['calibre.ebooks.markdown.extensions.' + x.lower() for x in extensions if x.lower() in MD_EXTENSIONS]
@@ -107,6 +110,46 @@ def convert_markdown(txt, title='', extensions=('footnotes', 'tables', 'toc')):
     return HTML_TEMPLATE % (title, md.convert(txt))
 
 
+def convert_markdown_with_metadata(txt, title='', extensions=DEFAULT_MD_EXTENSIONS):
+    from calibre.ebooks.conversion.plugins.txt_input import MD_EXTENSIONS
+    from calibre.ebooks.markdown import Markdown
+    from calibre.ebooks.metadata.book.base import Metadata
+    from calibre.utils.date import parse_only_date
+    from calibre.db.write import get_series_values
+    extensions = ['calibre.ebooks.markdown.extensions.' + x.lower() for x in extensions if x.lower() in MD_EXTENSIONS]
+    meta_ext = 'calibre.ebooks.markdown.extensions.meta'
+    if meta_ext not in extensions:
+        extensions.append(meta_ext)
+    md = Markdown(extensions=extensions)
+    html = md.convert(txt)
+    mi = Metadata(title or _('Unknown'))
+    m = md.Meta
+    for k, v in {'date':'pubdate', 'summary':'comments'}.iteritems():
+        if v not in m and k in m:
+            m[v] = m.pop(k)
+    for k in 'title authors series tags pubdate comments publisher rating'.split():
+        val = m.get(k)
+        if val:
+            mf = mi.metadata_for_field(k)
+            if not mf.get('is_multiple'):
+                val = val[0]
+            if k == 'series':
+                val, si = get_series_values(val)
+                mi.series_index = 1 if si is None else si
+            if k == 'rating':
+                try:
+                    val = max(0, min(int(float(val)), 10))
+                except Exception:
+                    continue
+            if mf.get('datatype') == 'datetime':
+                try:
+                    val = parse_only_date(val, assume_utc=False)
+                except Exception:
+                    continue
+            setattr(mi, k, val)
+    return mi, HTML_TEMPLATE % (mi.title, html)
+
+
 def convert_textile(txt, title=''):
     from calibre.ebooks.textile import textile
     html = textile(txt, encoding='utf-8')