X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/223fd8f247b4a588d263afaf798dca4cb9ffa639..5b407667ca47cf4d9752821fd49e5611737146d2:/librarian/parser.py?ds=sidebyside diff --git a/librarian/parser.py b/librarian/parser.py index b470957..5ae06e2 100644 --- a/librarian/parser.py +++ b/librarian/parser.py @@ -3,7 +3,7 @@ # This file is part of Librarian, licensed under GNU Affero GPLv3 or later. # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. # -from librarian import ValidationError, NoDublinCore, ParseError +from librarian import ValidationError, NoDublinCore, ParseError, NoProvider from librarian import RDFNS from librarian import dcparser @@ -11,14 +11,17 @@ from xml.parsers.expat import ExpatError from lxml import etree from lxml.etree import XMLSyntaxError, XSLTApplyError +import os import re from StringIO import StringIO class WLDocument(object): - LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE); + LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE) + provider = None - def __init__(self, edoc, parse_dublincore=True): + def __init__(self, edoc, parse_dublincore=True, provider=None): self.edoc = edoc + self.provider = provider root_elem = edoc.getroot() @@ -42,7 +45,7 @@ class WLDocument(object): return cls.from_file(StringIO(xml), *args, **kwargs) @classmethod - def from_file(cls, xmlfile, swap_endlines=False, parse_dublincore=True, preserve_lines=True): + def from_file(cls, xmlfile, parse_dublincore=True, provider=None): # first, prepare for parsing if isinstance(xmlfile, basestring): @@ -57,18 +60,46 @@ class WLDocument(object): if not isinstance(data, unicode): data = data.decode('utf-8') - if swap_endlines: - sub = u'
' - if preserve_lines: - sub += u'\n' - data = cls.LINE_SWAP_EXPR.sub(sub, data) + data = data.replace(u'\ufeff', '') try: parser = etree.XMLParser(remove_blank_text=False) - return cls(etree.parse(StringIO(data), parser), parse_dublincore=parse_dublincore) + tree = etree.parse(StringIO(data.encode('utf-8')), parser) + + return cls(tree, parse_dublincore=parse_dublincore, provider=provider) except (ExpatError, XMLSyntaxError, XSLTApplyError), e: raise ParseError(e) + def swap_endlines(self): + """Converts line breaks in stanzas into
tags.""" + # only swap inside stanzas + for elem in self.edoc.iter('strofa'): + for child in list(elem): + if child.tail: + chunks = self.LINE_SWAP_EXPR.split(child.tail) + ins_index = elem.index(child) + 1 + while len(chunks) > 1: + ins = etree.Element('br') + ins.tail = chunks.pop() + elem.insert(ins_index, ins) + child.tail = chunks.pop(0) + if elem.text: + chunks = self.LINE_SWAP_EXPR.split(elem.text) + while len(chunks) > 1: + ins = etree.Element('br') + ins.tail = chunks.pop() + elem.insert(0, ins) + elem.text = chunks.pop(0) + + def parts(self): + if self.provider is None: + raise NoProvider('No document provider supplied.') + if self.book_info is None: + raise NoDublinCore('No Dublin Core in document.') + for part_uri in self.book_info.parts: + yield self.from_file(self.provider.by_uri(part_uri), + provider=self.provider) + def chunk(self, path): # convert the path to XPath expr = self.path_to_xpath(path) @@ -119,4 +150,52 @@ class WLDocument(object): except Exception, e: unmerged.append( repr( (key, xpath, e) ) ) - return unmerged \ No newline at end of file + return unmerged + + def clean_ed_note(self): + """ deletes forbidden tags from nota_red """ + + for node in self.edoc.xpath('|'.join('//nota_red//%s' % tag for tag in + ('pa', 'pe', 'pr', 'pt', 'begin', 'end', 'motyw'))): + tail = node.tail + node.clear() + node.tag = 'span' + node.tail = tail + + # Converters + + def as_html(self, *args, **kwargs): + from librarian import html + return html.transform(self, *args, **kwargs) + + def as_text(self, *args, **kwargs): + from librarian import text + return text.transform(self, *args, **kwargs) + + def as_epub(self, *args, **kwargs): + from librarian import epub + return epub.transform(self, *args, **kwargs) + + def as_pdf(self, *args, **kwargs): + from librarian import pdf + return pdf.transform(self, *args, **kwargs) + + def as_mobi(self, *args, **kwargs): + from librarian import mobi + return mobi.transform(self, *args, **kwargs) + + def save_output_file(self, output_file, output_path=None, + output_dir_path=None, make_author_dir=False, ext=None): + if output_dir_path: + save_path = output_dir_path + if make_author_dir: + save_path = os.path.join(save_path, + unicode(self.book_info.author).encode('utf-8')) + save_path = os.path.join(save_path, + self.book_info.uri.filename_stem()) + if ext: + save_path += '.%s' % ext + else: + save_path = output_path + + output_file.save_as(save_path)