# This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
#
+"""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 with_statement
import os
import os.path
import shutil
+from distutils.dir_util import copy_tree # shutil.copytree is so uncapable.
from StringIO import StringIO
from tempfile import mkdtemp, NamedTemporaryFile
import re
from librarian.parser import WLDocument
from librarian import ParseError, DCNS, get_resource, OutputFile
from librarian import functions
+from librarian.cover import WLCover
functions.reg_substitute_entities()
'wl2tex': 'pdf/wl2tex.xslt',
}
-CUSTOMIZATIONS = [
- 'nofootnotes',
- 'nothemes',
- '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
insert_tags(doc,
re.compile("(?<=[^-\s])-(?=[^-\s])"),
"dywiz",
- exclude=[DCNS("identifier.url"), DCNS("rights.license")]
+ exclude=[DCNS("identifier.url"), DCNS("rights.license"), "www"]
)
def parse_creator(doc):
- """ find all dc:creator and dc.contributor tags and add *_parsed versions with forenames first """
+ """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', 'contributor.editor', 'contributor.technical_editor')),
+ 'creator', 'contributor.translator')),
namespaces = {'dc': str(DCNS)})[::-1]:
if not person.text:
continue
return p == 0
-def transform(wldoc, verbose=False, save_tex=None, morefloats=None,
- cover=None, flags=None, customizations=None):
+def transform(wldoc, verbose=False, save_tex=None, save_texml=None, morefloats=None,
+ cover=None, flags=None, customizations=None, documentclass='wl', resources=None):
""" 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
+ save_texml: path to save the intermediary TeXML 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)
+ documentclass: LaTeX document class, defaults to wl
+ resources: a directory with resources, copied to place where LaTeX compilation is made
"""
# Parse XSLT
try:
+ book_info = wldoc.book_info
document = load_including_children(wldoc)
+ root = document.edoc.getroot()
if cover:
- document.edoc.getroot().set('data-cover-width', str(cover.width))
- document.edoc.getroot().set('data-cover-height', str(cover.height))
+ if cover is True:
+ cover = WLCover
+ bound_cover = cover(book_info)
+ 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:
- document.edoc.getroot().set('flag-' + flag, 'yes')
+ root.set('flag-' + flag, 'yes')
# check for LaTeX packages
if morefloats:
- document.edoc.getroot().set('morefloats', morefloats.lower())
+ root.set('morefloats', morefloats.lower())
elif package_available('morefloats', 'maxfloats=19'):
- document.edoc.getroot().set('morefloats', 'new')
+ root.set('morefloats', 'new')
# add customizations
if customizations is not None:
- document.edoc.getroot().set('customizations', u','.join(customizations))
+ root.set('customizations', u','.join(customizations))
+
+ root.set('documentclass', documentclass or 'wl')
+
+ # add editors info
+ root.set('editors', u', '.join(sorted(
+ editor.readable() for editor in document.editors())))
# hack the tree
move_motifs_inside(document.edoc)
texml = document.transform(style)
+ if save_texml:
+ texml.write(save_texml)
+
# 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)
+ bound_cover.save(f)
del document # no longer needed large object :)
# LaTeX -> PDF
shutil.copy(get_resource('pdf/wl.cls'), temp)
+ shutil.copy(get_resource('pdf/wlpub.cls'), temp)
shutil.copy(get_resource('res/wl-logo.png'), temp)
-
- cwd = os.getcwd()
+ shutil.copy(get_resource('res/cover.jpg'), temp)
+ if resources:
+ copy_tree(resources, temp)
+
+ try:
+ cwd = os.getcwd()
+ except OSError:
+ cwd = None
os.chdir(temp)
if verbose:
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')
text = re.sub(ur"([\u0400-\u04ff]+)", ur"<alien>\1</alien>", text)
- document = WLDocument.from_string(text, parse_dublincore=True)
+ document = WLDocument.from_string(text,
+ parse_dublincore=True, provider=provider)
document.swap_endlines()
for child_uri in document.book_info.parts:
fnp / librarian.git / blobdiff