X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/e57b146bf49e38b3bb57615110b27de5b4d1ae69..665c88719c108626edee864affe07b085d28b7b4:/librarian/__init__.py?ds=inline diff --git a/librarian/__init__.py b/librarian/__init__.py index c46d5d1..119b6b1 100644 --- a/librarian/__init__.py +++ b/librarian/__init__.py @@ -3,26 +3,28 @@ # 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 +from __future__ import print_function, unicode_literals import os import re import shutil +from tempfile import NamedTemporaryFile import urllib +from lxml import etree +import six +from six.moves.urllib.request import FancyURLopener +from .util import makedirs +@six.python_2_unicode_compatible class UnicodeException(Exception): def __str__(self): - """ Dirty workaround for Python Unicode handling problems. """ - return unicode(self).encode('utf-8') - - def __unicode__(self): """ Dirty workaround for Python Unicode handling problems. """ args = self.args[0] if len(self.args) == 1 else self.args try: - message = unicode(args) + message = six.text_type(args) except UnicodeDecodeError: - message = unicode(args, encoding='utf-8', errors='ignore') + message = six.text_type(args, encoding='utf-8', errors='ignore') return message class ParseError(UnicodeException): @@ -65,26 +67,29 @@ class EmptyNamespace(XMLNamespace): return tag # some common namespaces we use +XMLNS = XMLNamespace('http://www.w3.org/XML/1998/namespace') RDFNS = XMLNamespace('http://www.w3.org/1999/02/22-rdf-syntax-ns#') DCNS = XMLNamespace('http://purl.org/dc/elements/1.1/') XINS = XMLNamespace("http://www.w3.org/2001/XInclude") XHTMLNS = XMLNamespace("http://www.w3.org/1999/xhtml") NCXNS = XMLNamespace("http://www.daisy.org/z3986/2005/ncx/") OPFNS = XMLNamespace("http://www.idpf.org/2007/opf") +PLMETNS = XMLNamespace("http://dl.psnc.pl/schemas/plmet/") WLNS = EmptyNamespace() +@six.python_2_unicode_compatible class WLURI(object): """Represents a WL URI. Extracts slug from it.""" slug = None example = 'http://wolnelektury.pl/katalog/lektura/template/' - _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektura/' + _re_wl_uri = re.compile(r'http://(www\.)?wolnelektury.pl/katalog/lektur[ay]/' '(?P[-a-z0-9]+)/?$') def __init__(self, uri): - uri = unicode(uri) + uri = six.text_type(uri) self.uri = uri self.slug = uri.rstrip('/').rsplit('/', 1)[-1] @@ -100,16 +105,13 @@ class WLURI(object): def from_slug(cls, slug): """Contructs an URI from slug. - >>> WLURI.from_slug('a-slug').uri - u'http://wolnelektury.pl/katalog/lektura/a-slug/' + >>> print(WLURI.from_slug('a-slug').uri) + http://wolnelektury.pl/katalog/lektura/a-slug/ """ uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug return cls(uri) - def __unicode__(self): - return self.uri - def __str__(self): return self.uri @@ -142,11 +144,10 @@ class DirDocProvider(DocProvider): def by_slug(self, slug): fname = slug + '.xml' - return open(os.path.join(self.dir, fname)) + return open(os.path.join(self.dir, fname), 'rb') -import lxml.etree as etree -import dcparser +from . import dcparser DEFAULT_BOOKINFO = dcparser.BookInfo( { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'}, @@ -171,14 +172,14 @@ DEFAULT_BOOKINFO = dcparser.BookInfo( def xinclude_forURI(uri): e = etree.Element(XINS("include")) e.set("href", uri) - return etree.tostring(e, encoding=unicode) + return etree.tostring(e, encoding='unicode') def wrap_text(ocrtext, creation_date, bookinfo=DEFAULT_BOOKINFO): """Wrap the text within the minimal XML structure with a DC template.""" bookinfo.created_at = creation_date dcstring = etree.tostring(bookinfo.to_etree(), \ - method='xml', encoding=unicode, pretty_print=True) + method='xml', encoding='unicode', pretty_print=True) return u'\n' + dcstring + u'\n\n' + ocrtext + \ u'\n\n' @@ -188,7 +189,7 @@ def serialize_raw(element): b = u'' + (element.text or '') for child in element.iterchildren(): - e = etree.tostring(child, method='xml', encoding=unicode, + e = etree.tostring(child, method='xml', encoding='unicode', pretty_print=True) b += e @@ -208,7 +209,7 @@ def get_resource(path): class OutputFile(object): """Represents a file returned by one of the converters.""" - _string = None + _bytes = None _filename = None def __del__(self): @@ -216,14 +217,14 @@ class OutputFile(object): os.unlink(self._filename) def __nonzero__(self): - return self._string is not None or self._filename is not None + return self._bytes is not None or self._filename is not None @classmethod - def from_string(cls, string): + def from_bytes(cls, bytestring): """Converter returns contents of a file as a string.""" instance = cls() - instance._string = string + instance._bytes = bytestring return instance @classmethod @@ -234,33 +235,31 @@ class OutputFile(object): instance._filename = filename return instance - def get_string(self): - """Get file's contents as a string.""" + def get_bytes(self): + """Get file's contents as a bytestring.""" if self._filename is not None: - with open(self._filename) as f: + with open(self._filename, 'rb') as f: return f.read() else: - return self._string + return self._bytes def get_file(self): """Get file as a file-like object.""" - if self._string is not None: - from StringIO import StringIO - return StringIO(self._string) + if self._bytes is not None: + return six.BytesIO(self._bytes) elif self._filename is not None: - return open(self._filename) + return open(self._filename, 'rb') def get_filename(self): """Get file as a fs path.""" if self._filename is not None: return self._filename - elif self._string is not None: - from tempfile import NamedTemporaryFile + elif self._bytes is not None: temp = NamedTemporaryFile(prefix='librarian-', delete=False) - temp.write(self._string) + temp.write(self._bytes) temp.close() self._filename = temp.name return self._filename @@ -271,11 +270,10 @@ class OutputFile(object): """Save file to a path. Create directories, if necessary.""" dirname = os.path.dirname(os.path.abspath(path)) - if not os.path.isdir(dirname): - os.makedirs(dirname) + makedirs(dirname) shutil.copy(self.get_filename(), path) -class URLOpener(urllib.FancyURLopener): +class URLOpener(FancyURLopener): version = 'FNP Librarian (http://github.com/fnp/librarian)' urllib._urlopener = URLOpener()