epub/mobi fixes

[librarian.git] / librarian / pdf.py

diff --git a/librarian/pdf.py b/librarian/pdf.py
index f8a4d73..476fbee 100644 (file)
--- a/librarian/pdf.py
+++ b/librarian/pdf.py
@@ -37,7 +37,7 @@ STYLESHEETS = {
 
 
 def insert_tags(doc, split_re, tagname, exclude=None):
-    """ inserts <tagname> for every occurence of `split_re' in text nodes in the `doc' tree 
+    """ inserts <tagname> for every occurence of `split_re' in text nodes in the `doc' tree
 
     >>> t = etree.fromstring('<a><b>A-B-C</b>X-Y-Z</a>');
     >>> insert_tags(t, re.compile('-'), 'd');
@@ -67,7 +67,7 @@ def insert_tags(doc, split_re, tagname, exclude=None):
 
 
 def substitute_hyphens(doc):
-    insert_tags(doc, 
+    insert_tags(doc,
                 re.compile("(?<=[^-\s])-(?=[^-\s])"),
                 "dywiz",
                 exclude=[DCNS("identifier.url"), DCNS("rights.license")]
@@ -75,7 +75,7 @@ def substitute_hyphens(doc):
 
 
 def fix_hanging(doc):
-    insert_tags(doc, 
+    insert_tags(doc,
                 re.compile("(?<=\s\w)\s+"),
                 "nbsp",
                 exclude=[DCNS("identifier.url"), DCNS("rights.license")]
@@ -132,9 +132,12 @@ def parse_creator(doc):
     for person in doc.xpath("|".join('//dc:'+(tag) for tag in (
                     'creator', 'contributor.translator', 'contributor.editor', 'contributor.technical_editor')),
                     namespaces = {'dc': str(DCNS)})[::-1]:
+        if not person.text:
+            continue
         p = Person.from_text(person.text)
         person_parsed = deepcopy(person)
         person_parsed.tag = person.tag + '_parsed'
+        person_parsed.set('sortkey', person.text)
         person_parsed.text = p.readable()
         person.getparent().insert(0, person_parsed)
 
@@ -144,7 +147,7 @@ def get_stylesheet(name):
 
 
 def package_available(package, args='', verbose=False):
-    """ check if a verion of a latex package accepting given args is available """  
+    """ check if a verion of a latex package accepting given args is available """
     tempdir = mkdtemp('-wl2pdf-test')
     fpath = os.path.join(tempdir, 'test.tex')
     f = open(fpath, 'w')
@@ -163,8 +166,9 @@ def package_available(package, args='', verbose=False):
     return p == 0
 
 
-def transform(provider, slug=None, file_path=None, 
-              output_file=None, output_dir=None, make_dir=False, verbose=False, save_tex=None, morefloats=None):
+def transform(provider, slug=None, file_path=None,
+              output_file=None, output_dir=None, make_dir=False, verbose=False, save_tex=None, morefloats=None,
+              cover=None, flags=None):
     """ produces a PDF file with XeLaTeX
 
     provider: a DocProvider
@@ -176,6 +180,8 @@ def transform(provider, slug=None, file_path=None,
     verbose: prints all output from LaTeX
     save_tex: path to save the intermediary LaTeX file to
     morefloats (old/new/none): force specific morefloats
+    cover: a cover.Cover object
+    flags: less-advertising,
     """
 
     # Parse XSLT
@@ -189,6 +195,13 @@ def transform(provider, slug=None, file_path=None,
                 raise ValueError('either slug or file_path should be specified')
             document = load_including_children(provider, slug=slug)
 
+        if cover:
+            document.edoc.getroot().set('data-cover-width', str(cover.width))
+            document.edoc.getroot().set('data-cover-height', str(cover.height))
+        if flags:
+            for flag in flags:
+                document.edoc.getroot().set('flag-' + flag, 'yes')
+
         # check for LaTeX packages
         if morefloats:
             document.edoc.getroot().set('morefloats', morefloats.lower())
@@ -211,10 +224,17 @@ def transform(provider, slug=None, file_path=None,
         style_filename = get_stylesheet("wl2tex")
         style = etree.parse(style_filename)
         texml = document.transform(style)
-        del document # no longer needed large object :)
 
         # TeXML -> LaTeX
         temp = mkdtemp('-wl2pdf')
+
+        if cover:
+            c = cover(document.book_info.author.readable(), document.book_info.title)
+            with open(os.path.join(temp, 'cover.png'), 'w') as f:
+                c.save(f)
+
+        del document # no longer needed large object :)
+
         tex_path = os.path.join(temp, 'doc.tex')
         fout = open(tex_path, 'w')
         process(StringIO(texml), fout, 'utf-8')
@@ -268,7 +288,7 @@ def transform(provider, slug=None, file_path=None,
 
 
 def load_including_children(provider, slug=None, uri=None, file_path=None):
-    """ makes one big xml file with children inserted at end 
+    """ makes one big xml file with children inserted at end
     either slug or uri must be provided
     """
 
@@ -288,7 +308,6 @@ def load_including_children(provider, slug=None, uri=None, file_path=None):
         parse_dublincore=True)
 
     f.close()
-    print document.book_info
     for child_uri in document.book_info.parts:
         print child_uri
         child = load_including_children(provider, uri=child_uri)