Add podtytul_* tags; refs #4005.

[librarian.git] / librarian / pdf.py

diff --git a/librarian/pdf.py b/librarian/pdf.py
index cb65311..e6d897d 100644 (file)
--- a/librarian/pdf.py
+++ b/librarian/pdf.py
@@ -3,25 +3,34 @@
 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #

-from __future__ import with_statement
+"""PDF creation library.
+
+Creates one big XML from the book and its children, converts it to LaTeX
+with TeXML, then runs it by XeLaTeX.
+
+"""
+from __future__ import print_function, unicode_literals
+

 import os
 import os.path
 import shutil
 import os
 import os.path
 import shutil

-from StringIO import StringIO

 from tempfile import mkdtemp, NamedTemporaryFile
 import re
 from copy import deepcopy
 from subprocess import call, PIPE
 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
 from lxml.etree import XMLSyntaxError, XSLTApplyError
 
 from Texml.processor import process
 from lxml import etree
 from lxml.etree import XMLSyntaxError, XSLTApplyError

+import six

 
 from librarian.dcparser import Person
 from librarian.parser import WLDocument
 
 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 import functions

-from librarian.cover import WLCover
+from librarian.cover import make_cover
+from .sponsor import sponsor_logo

 
 
 functions.reg_substitute_entities()
 
 
 functions.reg_substitute_entities()


@@ -34,21 +43,22 @@ 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');
-    >>> print etree.tostring(t)
+    >>> t = etree.fromstring('<a><b>A-B-C</b>X-Y-Z</a>')
+    >>> insert_tags(t, re.compile('-'), 'd')
+    >>> print(etree.tostring(t, encoding='unicode'))

     <a><b>A<d/>B<d/>C</b>X<d/>Y<d/>Z</a>
     """
 
     <a><b>A<d/>B<d/>C</b>X<d/>Y<d/>Z</a>
     """
 


@@ -77,7 +87,7 @@ 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"]

                 )
 
 
                 )
 
 


@@ -89,12 +99,38 @@ 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 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 """
 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)


@@ -135,10 +171,13 @@ def hack_motifs(doc):
 
 
 def parse_creator(doc):
 
 
 def parse_creator(doc):

-    """ find all dc:creator and dc.contributor tags and add *_parsed versions with forenames first """
-    for person in doc.xpath("|".join('//dc:'+(tag) for tag in (
-                    'creator', 'contributor.translator', 'contributor.editor', 'contributor.technical_editor')),
-                    namespaces = {'dc': str(DCNS)})[::-1]:
+    """Generates readable versions of creator and translator tags.
+
+    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]:

         if not person.text:
             continue
         p = Person.from_text(person.text)
         if not person.text:
             continue
         p = Person.from_text(person.text)


@@ -158,11 +197,11 @@ def package_available(package, args='', verbose=False):
     tempdir = mkdtemp('-wl2pdf-test')
     fpath = os.path.join(tempdir, 'test.tex')
     f = open(fpath, 'w')
     tempdir = mkdtemp('-wl2pdf-test')
     fpath = os.path.join(tempdir, 'test.tex')
     f = open(fpath, 'w')

-    f.write(r"""
-        \documentclass{wl}
-        \usepackage[%s]{%s}
-        \begin{document}
-        \end{document}
+    f.write("""
+        \\documentclass{wl}
+        \\usepackage[%s]{%s}
+        \\begin{document}
+        \\end{document}

         """ % (args, package))
     f.close()
     if verbose:
         """ % (args, package))
     f.close()
     if verbose:


@@ -174,45 +213,58 @@ 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
     verbose: prints all output from LaTeX
     save_tex: path to save the intermediary LaTeX file to
     morefloats (old/new/none): force specific morefloats
     """ produces a PDF file with XeLaTeX
 
     wldoc: a WLDocument
     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
+    cover: a cover.Cover factory or True for default

     flags: less-advertising,
     customizations: user requested customizations regarding various formatting parameters (passed to wl LaTeX class)
     """
 
     # Parse XSLT
     try:
     flags: less-advertising,
     customizations: user requested customizations regarding various formatting parameters (passed to wl LaTeX class)
     """
 
     # Parse XSLT
     try:

+        book_info = wldoc.book_info

         document = load_including_children(wldoc)
         document = load_including_children(wldoc)

+        root = document.edoc.getroot()

 
         if cover:
             if cover is True:
 
         if cover:
             if cover is True:

-                cover = WLCover
-            document.edoc.getroot().set('data-cover-width', str(cover.width))
-            document.edoc.getroot().set('data-cover-height', str(cover.height))
-            if cover.uses_dc_cover:
-                if document.book_info.cover_by:
-                    document.edoc.getroot().set('data-cover-by', document.book_info.cover_by)
-                if document.book_info.cover_source:
-                    document.edoc.getroot().set('data-cover-source', document.book_info.cover_source)
+                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-source', book_info.cover_source)

         if flags:
             for flag in flags:
         if flags:
             for flag in flags:

-                document.edoc.getroot().set('flag-' + flag, 'yes')
+                root.set('flag-' + flag, 'yes')

 
         # check for LaTeX packages
         if morefloats:
 
         # check for LaTeX packages
         if morefloats:

-            document.edoc.getroot().set('morefloats', morefloats.lower())
+            root.set('morefloats', morefloats.lower())

         elif package_available('morefloats', 'maxfloats=19'):
         elif package_available('morefloats', 'maxfloats=19'):

-            document.edoc.getroot().set('morefloats', 'new')
+            root.set('morefloats', 'new')

 
         # add customizations
         if customizations is not None:
 
         # add customizations
         if customizations is not None:

-            document.edoc.getroot().set('customizations', u','.join(customizations))
+            root.set('customizations', u','.join(customizations))
+
+        # add editors info
+        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:
+            root.set('thanks', document.book_info.thanks)

 
         # hack the tree
         move_motifs_inside(document.edoc)
 
         # hack the tree
         move_motifs_inside(document.edoc)


@@ -220,26 +272,43 @@ 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)
+        mark_subauthors(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)

-
-        texml = document.transform(style)
+        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)
+            if logo:
+                fname = 'sponsor-%s' % os.path.basename(logo)
+                shutil.copy(logo, os.path.join(temp, fname))
+                ins.set('src', fname)
+            root.insert(0, ins)
+                
+        if book_info.sponsor_note:
+            root.set("sponsor-note", book_info.sponsor_note)
+
+        texml = document.transform(style)
+

         if cover:
         if cover:

-            c = cover(document.book_info)

             with open(os.path.join(temp, 'cover.png'), 'w') as f:
             with open(os.path.join(temp, 'cover.png'), 'w') as f:

-                c.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')
 
         tex_path = os.path.join(temp, 'doc.tex')

-        fout = open(tex_path, 'w')
-        process(StringIO(texml), fout, 'utf-8')
+        fout = open(tex_path, 'wb')
+        process(six.BytesIO(texml), fout, 'utf-8')

         fout.close()
         del texml
 
         fout.close()
         del texml
 


@@ -250,17 +319,27 @@ 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)
 

-        cwd = os.getcwd()
+        if latex_dir:
+            return temp
+
+        try:
+            cwd = os.getcwd()
+        except OSError:
+            cwd = None

         os.chdir(temp)
 
         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 range(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")

 
 

-        os.chdir(cwd)
+        if cwd is not None:
+            os.chdir(cwd)

 
         output_file = NamedTemporaryFile(prefix='librarian', suffix='.pdf', delete=False)
         pdf_path = os.path.join(temp, 'doc.pdf')
 
         output_file = NamedTemporaryFile(prefix='librarian', suffix='.pdf', delete=False)
         pdf_path = os.path.join(temp, 'doc.pdf')


@@ -268,7 +347,7 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
         shutil.rmtree(temp)
         return OutputFile.from_filename(output_file.name)
 
         shutil.rmtree(temp)
         return OutputFile.from_filename(output_file.name)
 

-    except (XMLSyntaxError, XSLTApplyError), e:
+    except (XMLSyntaxError, XSLTApplyError) as e:

         raise ParseError(e)
 
 
         raise ParseError(e)
 
 


@@ -283,14 +362,14 @@ def load_including_children(wldoc=None, provider=None, uri=None):
         text = f.read().decode('utf-8')
         f.close()
     elif wldoc is not None:
         text = f.read().decode('utf-8')
         f.close()
     elif wldoc is not None:

-        text = etree.tostring(wldoc.edoc, encoding=unicode)
+        text = etree.tostring(wldoc.edoc, encoding='unicode')

         provider = wldoc.provider
     else:
         raise ValueError('Neither a WLDocument, nor provider and URI were provided.')
 
         provider = wldoc.provider
     else:
         raise ValueError('Neither a WLDocument, nor provider and URI were provided.')
 

-    text = re.sub(ur"([\u0400-\u04ff]+)", ur"<alien>\1</alien>", text)
+    text = re.sub(r"([\u0400-\u04ff]+)", r"<alien>\1</alien>", text)

 
 

-    document = WLDocument.from_string(text, parse_dublincore=True)
+    document = WLDocument.from_bytes(text.encode('utf-8'), 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: