X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/b8e34e0e6730ef76a353a15ff653faa9e8c88a77..2c3ff563f5db80c36871f58c670d3df197df29f6:/librarian/pdf.py?ds=inline diff --git a/librarian/pdf.py b/librarian/pdf.py index cb65311..9af6b50 100644 --- a/librarian/pdf.py +++ b/librarian/pdf.py @@ -3,6 +3,12 @@ # 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 @@ -29,6 +35,8 @@ functions.reg_strip() functions.reg_starts_white() functions.reg_ends_white() functions.reg_texcommand() +functions.reg_urlquote() +functions.reg_breakurl() STYLESHEETS = { 'wl2tex': 'pdf/wl2tex.xslt', @@ -77,7 +85,7 @@ 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"), 'www'] ) @@ -88,6 +96,18 @@ def fix_hanging(doc): exclude=[DCNS("identifier.url"), DCNS("rights.license")] ) +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 """ @@ -135,9 +155,14 @@ def hack_motifs(doc): 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', + 'contributor.editor', 'contributor.technical_editor')), namespaces = {'dc': str(DCNS)})[::-1]: if not person.text: continue @@ -174,45 +199,53 @@ 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 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: + book_info = wldoc.book_info document = load_including_children(wldoc) + root = document.edoc.getroot() 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) + bound_cover = cover(book_info, width=2400) + 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)) + + # add editors info + root.set('editors', u', '.join(sorted( + editor.readable() for editor in document.editors()))) # hack the tree move_motifs_inside(document.edoc) @@ -220,6 +253,7 @@ 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") @@ -230,10 +264,12 @@ 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) + if cover: - c = cover(document.book_info) 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 :) @@ -249,8 +285,18 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None, # LaTeX -> PDF shutil.copy(get_resource('pdf/wl.cls'), temp) shutil.copy(get_resource('res/wl-logo.png'), temp) - - cwd = os.getcwd() + # shutil.copy(get_resource('res/prawokultury-logo.png'), temp) + # shutil.copy(get_resource('res/trust-logo.eps'), temp) + shutil.copy(get_resource('res/fnp-logo.eps'), temp) + shutil.copy(get_resource('res/koed-logo.eps'), temp) + + if latex_dir: + return temp + + try: + cwd = os.getcwd() + except OSError: + cwd = None os.chdir(temp) if verbose: @@ -260,7 +306,8 @@ def transform(wldoc, verbose=False, save_tex=None, morefloats=None, 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') @@ -290,7 +337,8 @@ def load_including_children(wldoc=None, provider=None, uri=None): text = re.sub(ur"([\u0400-\u04ff]+)", ur"\1", 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: