X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/a91e41e489588ebad550cd9e22d157062effa2ff..47cd67ea0e820d6805c8ed23233e5e408b98b803:/librarian/pdf.py

diff --git a/librarian/pdf.py b/librarian/pdf.py
index 12c07ea..c8af03b 100644
--- a/librarian/pdf.py
+++ b/librarian/pdf.py
@@ -41,20 +41,21 @@ STYLESHEETS = {
     'wl2tex': 'pdf/wl2tex.xslt',
 }
 
-#CUSTOMIZATIONS = [
-#    'nofootnotes',
-#    'nothemes',
-#    'defaultleading',
-#    'onehalfleading',
-#    'doubleleading',
-#    'nowlfont',
-#    ]
+# CUSTOMIZATIONS = [
+#     'nofootnotes',
+#     'nothemes',
+#     'defaultleading',
+#     'onehalfleading',
+#     'doubleleading',
+#     'nowlfont',
+# ]
+
 
 def insert_tags(doc, split_re, tagname, exclude=None):
     """ 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');
+    >>> t = etree.fromstring('<a><b>A-B-C</b>X-Y-Z</a>')
+    >>> insert_tags(t, re.compile('-'), 'd')
     >>> print etree.tostring(t)
     <a><b>A<d/>B<d/>C</b>X<d/>Y<d/>Z</a>
     """
@@ -96,12 +97,26 @@ def fix_hanging(doc):
                 )
 
 
+def fix_tables(doc):
+    for kol in doc.iter(tag='kol'):
+        if kol.tail is not None:
+            if not kol.tail.strip():
+                kol.tail = None
+    for table in doc.iter(tag='tabela'):
+        if table.get('ramka') == '1' or table.get('ramki') == '1':
+            table.set('_format', '|' + 'X|' * len(table[0]))
+        else:
+            table.set('_format', 'X' * len(table[0]))
+
+
 def move_motifs_inside(doc):
     """ moves motifs to be into block elements """
-    for master in doc.xpath('//powiesc|//opowiadanie|//liryka_l|//liryka_lp|//dramat_wierszowany_l|//dramat_wierszowany_lp|//dramat_wspolczesny'):
+    for master in doc.xpath('//powiesc|//opowiadanie|//liryka_l|//liryka_lp|'
+                            '//dramat_wierszowany_l|//dramat_wierszowany_lp|//dramat_wspolczesny'):
         for motif in master.xpath('motyw'):
             for sib in motif.itersiblings():
-                if sib.tag not in ('sekcja_swiatlo', 'sekcja_asterysk', 'separator_linia', 'begin', 'end', 'motyw', 'extra', 'uwaga'):
+                if sib.tag not in ('sekcja_swiatlo', 'sekcja_asterysk', 'separator_linia',
+                                   'begin', 'end', 'motyw', 'extra', 'uwaga'):
                     # motif shouldn't have a tail - it would be untagged text
                     motif.tail = None
                     motif.getparent().remove(motif)
@@ -147,9 +162,8 @@ def parse_creator(doc):
     Finds all dc:creator and dc.contributor.translator tags
     and adds *_parsed versions with forenames first.
     """
-    for person in doc.xpath("|".join('//dc:'+(tag) for tag in (
-                    'creator', 'contributor.translator')),
-                    namespaces = {'dc': str(DCNS)})[::-1]:
+    for person in doc.xpath("|".join('//dc:' + tag for tag in ('creator', 'contributor.translator')),
+                            namespaces={'dc': str(DCNS)})[::-1]:
         if not person.text:
             continue
         p = Person.from_text(person.text)
@@ -185,7 +199,7 @@ def package_available(package, args='', verbose=False):
 
 
 def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
-              cover=None, flags=None, customizations=None):
+              cover=None, flags=None, customizations=None, ilustr_path=''):
     """ produces a PDF file with XeLaTeX
 
     wldoc: a WLDocument
@@ -213,8 +227,7 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
                 if book_info.cover_by:
                     root.set('data-cover-by', book_info.cover_by)
                 if book_info.cover_source:
-                    root.set('data-cover-source',
-                            book_info.cover_source)
+                    root.set('data-cover-source', book_info.cover_source)
         if flags:
             for flag in flags:
                 root.set('flag-' + flag, 'yes')
@@ -245,14 +258,19 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
         parse_creator(document.edoc)
         substitute_hyphens(document.edoc)
         fix_hanging(document.edoc)
+        fix_tables(document.edoc)
 
         # wl -> TeXML
         style_filename = get_stylesheet("wl2tex")
         style = etree.parse(style_filename)
+        functions.reg_mathml_latex()
 
         # TeXML -> LaTeX
         temp = mkdtemp('-wl2pdf')
 
+        for ilustr in document.edoc.findall("//ilustr"):
+            shutil.copy(os.path.join(ilustr_path, ilustr.get("src")), temp)
+
         for sponsor in book_info.sponsors:
             ins = etree.Element("data-sponsor", name=sponsor)
             logo = sponsor_logo(sponsor)
@@ -271,7 +289,7 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
             with open(os.path.join(temp, 'cover.png'), 'w') as f:
                 bound_cover.save(f, quality=80)
 
-        del document # no longer needed large object :)
+        del document  # no longer needed large object :)
 
         tex_path = os.path.join(temp, 'doc.tex')
         fout = open(tex_path, 'w')
@@ -292,12 +310,14 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
             cwd = None
         os.chdir(temp)
 
-        if verbose:
-            p = call(['xelatex', tex_path])
-        else:
-            p = call(['xelatex', '-interaction=batchmode', tex_path], stdout=PIPE, stderr=PIPE)
-        if p:
-            raise ParseError("Error parsing .tex file")
+        # some things work better when compiled twice
+        for run in xrange(2):
+            if verbose:
+                p = call(['xelatex', tex_path])
+            else:
+                p = call(['xelatex', '-interaction=batchmode', tex_path], stdout=PIPE, stderr=PIPE)
+            if p:
+                raise ParseError("Error parsing .tex file")
 
         if cwd is not None:
             os.chdir(cwd)
@@ -330,8 +350,7 @@ def load_including_children(wldoc=None, provider=None, uri=None):
 
     text = re.sub(ur"([\u0400-\u04ff]+)", ur"<alien>\1</alien>", text)
 
-    document = WLDocument.from_string(text,
-                parse_dublincore=True, provider=provider)
+    document = WLDocument.from_string(text, parse_dublincore=True, provider=provider)
     document.swap_endlines()
 
     for child_uri in document.book_info.parts: