pdf: headers, table of contents

[librarian.git] / librarian / pdf.py

diff --git a/librarian/pdf.py b/librarian/pdf.py
index c899afa..d67bddf 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
+from itertools import chain
 
 from Texml.processor import process
 from lxml import etree
@@ -25,9 +26,9 @@ from lxml.etree import XMLSyntaxError, XSLTApplyError
 
 from librarian.dcparser import Person
 from librarian.parser import WLDocument
-from librarian import ParseError, DCNS, get_resource, OutputFile
+from librarian import ParseError, DCNS, get_resource, OutputFile, RDFNS
 from librarian import functions
-from librarian.cover import DefaultEbookCover
+from librarian.cover import make_cover
 from .sponsor import sponsor_logo
 
 
@@ -85,7 +86,7 @@ def substitute_hyphens(doc):
     insert_tags(doc,
                 re.compile("(?<=[^-\s])-(?=[^-\s])"),
                 "dywiz",
-                exclude=[DCNS("identifier.url"), DCNS("rights.license")]
+                exclude=[DCNS("identifier.url"), DCNS("rights.license"), "meta"]
                 )
 
 
@@ -102,13 +103,25 @@ def fix_tables(doc):
         if kol.tail is not None:
             if not kol.tail.strip():
                 kol.tail = None
-    for table in doc.iter(tag='tabela'):
+    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 mark_subauthors(doc):
+    root_author = ', '.join(elem.text for elem in doc.findall('./' + RDFNS('RDF') + '//' + DCNS('creator_parsed')))
+    last_author = None
+    # jeśli autor jest inny niż autor całości i niż poprzedni autor
+    # to wstawiamy jakiś znacznik w rdf?
+    for subutwor in doc.xpath('/utwor/utwor'):
+        author = ', '.join(elem.text for elem in subutwor.findall('.//' + DCNS('creator_parsed')))
+        if author not in (last_author, root_author):
+            subutwor.find('.//' + RDFNS('RDF')).append(etree.Element('use_subauthor'))
+        last_author = author
+
+
 def move_motifs_inside(doc):
     """ moves motifs to be into block elements """
     for master in doc.xpath('//powiesc|//opowiadanie|//liryka_l|//liryka_lp|'
@@ -199,7 +212,7 @@ 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='', latex_dir=False):
     """ produces a PDF file with XeLaTeX
 
     wldoc: a WLDocument
@@ -219,7 +232,7 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
 
         if cover:
             if cover is True:
-                cover = DefaultEbookCover
+                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))
@@ -259,6 +272,7 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
         substitute_hyphens(document.edoc)
         fix_hanging(document.edoc)
         fix_tables(document.edoc)
+        mark_subauthors(document.edoc)
 
         # wl -> TeXML
         style_filename = get_stylesheet("wl2tex")
@@ -268,6 +282,9 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
         # 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)
@@ -301,18 +318,24 @@ 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)
 
+        if latex_dir:
+            return 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
+        # (table of contents, [line numbers - disabled])
+        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)