fix Ó in <naglowekosoba> (hack)

[librarian.git] / librarian / pdf.py

diff --git a/librarian/pdf.py b/librarian/pdf.py
index af258fc..652d4bf 100644 (file)
--- a/librarian/pdf.py
+++ b/librarian/pdf.py
@@ -18,6 +18,7 @@ from tempfile import mkdtemp, NamedTemporaryFile
 import re
 from copy import deepcopy
 from subprocess import call, PIPE
 import re
 from copy import deepcopy
 from subprocess import call, PIPE

+from itertools import chain

 
 from Texml.processor import process
 from lxml import etree
 
 from Texml.processor import process
 from lxml import etree


@@ -27,7 +28,7 @@ from librarian.dcparser import Person
 from librarian.parser import WLDocument
 from librarian import ParseError, DCNS, get_resource, OutputFile
 from librarian import functions
 from librarian.parser import WLDocument
 from librarian import ParseError, DCNS, get_resource, OutputFile
 from librarian import functions

-from librarian.cover import DefaultEbookCover
+from librarian.cover import make_cover

 from .sponsor import sponsor_logo
 
 
 from .sponsor import sponsor_logo
 
 


@@ -41,20 +42,21 @@ STYLESHEETS = {
     'wl2tex': 'pdf/wl2tex.xslt',
 }
 
     '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
 
 
 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>
     """
     >>> print etree.tostring(t)
     <a><b>A<d/>B<d/>C</b>X<d/>Y<d/>Z</a>
     """


@@ -84,7 +86,7 @@ def substitute_hyphens(doc):
     insert_tags(doc,
                 re.compile("(?<=[^-\s])-(?=[^-\s])"),
                 "dywiz",
     insert_tags(doc,
                 re.compile("(?<=[^-\s])-(?=[^-\s])"),
                 "dywiz",

-                exclude=[DCNS("identifier.url"), DCNS("rights.license")]
+                exclude=[DCNS("identifier.url"), DCNS("rights.license"), "meta"]

                 )
 
 
                 )
 
 


@@ -96,12 +98,26 @@ 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 chain(doc.iter(tag='tabela'), doc.iter(tag='tabelka')):
+        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 """
 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():
         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)
                     # motif shouldn't have a tail - it would be untagged text
                     motif.tail = None
                     motif.getparent().remove(motif)


@@ -147,9 +163,8 @@ def parse_creator(doc):
     Finds all dc:creator and dc.contributor.translator tags
     and adds *_parsed versions with forenames first.
     """
     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)
         if not person.text:
             continue
         p = Person.from_text(person.text)


@@ -185,7 +200,7 @@ def package_available(package, args='', verbose=False):
 
 
 def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
 
 
 def transform(wldoc, verbose=False, save_tex=None, morefloats=None,

-              cover=None, flags=None, customizations=None):
+              cover=None, flags=None, customizations=None, ilustr_path='', latex_dir=False):

     """ produces a PDF file with XeLaTeX
 
     wldoc: a WLDocument
     """ produces a PDF file with XeLaTeX
 
     wldoc: a WLDocument


@@ -205,16 +220,15 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
 
         if cover:
             if cover is True:
 
         if cover:
             if cover is True:

-                cover = DefaultEbookCover
-            bound_cover = cover(book_info)
+                cover = make_cover
+            bound_cover = cover(book_info, width=1200)

             root.set('data-cover-width', str(bound_cover.width))
             root.set('data-cover-height', str(bound_cover.height))
             if bound_cover.uses_dc_cover:
                 if book_info.cover_by:
                     root.set('data-cover-by', book_info.cover_by)
                 if book_info.cover_source:
             root.set('data-cover-width', str(bound_cover.width))
             root.set('data-cover-height', str(bound_cover.height))
             if bound_cover.uses_dc_cover:
                 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')
         if flags:
             for flag in flags:
                 root.set('flag-' + flag, 'yes')


@@ -230,8 +244,10 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
             root.set('customizations', u','.join(customizations))
 
         # add editors info
             root.set('customizations', u','.join(customizations))
 
         # add editors info

-        root.set('editors', u', '.join(sorted(
-            editor.readable() for editor in document.editors())))
+        editors = document.editors()
+        if editors:
+            root.set('editors', u', '.join(sorted(
+                editor.readable() for editor in editors)))

         if document.book_info.funders:
             root.set('funders', u', '.join(document.book_info.funders))
         if document.book_info.thanks:
         if document.book_info.funders:
             root.set('funders', u', '.join(document.book_info.funders))
         if document.book_info.thanks:


@@ -243,14 +259,19 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
         parse_creator(document.edoc)
         substitute_hyphens(document.edoc)
         fix_hanging(document.edoc)
         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)
 
         # wl -> TeXML
         style_filename = get_stylesheet("wl2tex")
         style = etree.parse(style_filename)

+        functions.reg_mathml_latex()

 
         # TeXML -> LaTeX
         temp = mkdtemp('-wl2pdf')
 
 
         # 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)
         for sponsor in book_info.sponsors:
             ins = etree.Element("data-sponsor", name=sponsor)
             logo = sponsor_logo(sponsor)


@@ -267,9 +288,9 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
 
         if cover:
             with open(os.path.join(temp, 'cover.png'), 'w') as f:
 
         if cover:
             with open(os.path.join(temp, 'cover.png'), 'w') as f:

-                bound_cover.save(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')
 
         tex_path = os.path.join(temp, 'doc.tex')
         fout = open(tex_path, 'w')


@@ -284,18 +305,24 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
         shutil.copy(get_resource('pdf/wl.cls'), temp)
         shutil.copy(get_resource('res/wl-logo.png'), temp)
 
         shutil.copy(get_resource('pdf/wl.cls'), temp)
         shutil.copy(get_resource('res/wl-logo.png'), temp)
 

+        if latex_dir:
+            return temp
+

         try:
             cwd = os.getcwd()
         except OSError:
             cwd = None
         os.chdir(temp)
 
         try:
             cwd = os.getcwd()
         except OSError:
             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
+        # but they are not enabled now (line numbers)
+        for run in xrange(1):
+            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)
 
         if cwd is not None:
             os.chdir(cwd)


@@ -328,8 +355,7 @@ def load_including_children(wldoc=None, provider=None, uri=None):
 
     text = re.sub(ur"([\u0400-\u04ff]+)", ur"<alien>\1</alien>", text)
 
 
     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:
     document.swap_endlines()
 
     for child_uri in document.book_info.parts: