X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/05b19dad8ca44136e064abcc6beff5dac4489c60..565aa32ff31bdbe6e2aaf9f47b124ac1f883a990:/librarian/__init__.py diff --git a/librarian/__init__.py b/librarian/__init__.py index feb9974..5c145d3 100644 --- a/librarian/__init__.py +++ b/librarian/__init__.py @@ -8,29 +8,45 @@ from __future__ import with_statement import os import re import shutil +import urllib +import lxml.etree as etree + -class ParseError(Exception): +class UnicodeException(Exception): def __str__(self): """ Dirty workaround for Python Unicode handling problems. """ - return self.message + return unicode(self).encode('utf-8') def __unicode__(self): """ Dirty workaround for Python Unicode handling problems. """ - return self.message + args = self.args[0] if len(self.args) == 1 else self.args + try: + message = unicode(args) + except UnicodeDecodeError: + message = unicode(args, encoding='utf-8', errors='ignore') + return message + + +class ParseError(UnicodeException): + pass -class ValidationError(Exception): + +class ValidationError(UnicodeException): pass + class NoDublinCore(ValidationError): """There's no DublinCore section, and it's required.""" pass -class NoProvider(Exception): + +class NoProvider(UnicodeException): """There's no DocProvider specified, and it's needed.""" pass + class XMLNamespace(object): - '''A handy structure to repsent names in an XML namespace.''' + """A handy structure to repsent names in an XML namespace.""" def __init__(self, uri): self.uri = uri @@ -47,6 +63,7 @@ class XMLNamespace(object): def __str__(self): return '%s' % self.uri + class EmptyNamespace(XMLNamespace): def __init__(self): super(EmptyNamespace, self).__init__('') @@ -66,42 +83,36 @@ WLNS = EmptyNamespace() class WLURI(object): - """Represents a WL URI. Extracts slug and language from it.""" - DEFAULT_LANGUAGE = u'pol' - + """Represents a WL URI. Extracts slug from it.""" slug = None - language = None - - example = 'http://wolnelektury.pl/katalog/lektura/template/' - _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/lektura/' - '(?P[-a-z0-9]+)(/(?P[a-z]{3}))?/?$') - - def __init__(self, uri=None): - if uri is not None: - uri = unicode(uri) - self.uri = uri - match = self._re_wl_uri.match(uri) - if not match: - raise ValueError('Supplied URI (%s) does not match ' - 'the WL document URI template.' % uri) - self.slug = match.group('slug') - self.language = match.group('lang') or self.DEFAULT_LANGUAGE + + example = 'http://edukacjamedialna.edu.pl/lekcje/template' + _re_wl_uri = re.compile( + r'http://(www\.)?edukacjamedialna.edu.pl/lekcje/' + '(?P[-a-z0-9]+)/?$') + + def __init__(self, uri): + uri = unicode(uri) + self.uri = uri + self.slug = uri.rstrip('/').rsplit('/', 1)[-1] @classmethod - def from_slug_and_lang(cls, slug, lang): - """Contructs an URI from slug and language code. + def strict(cls, uri): + match = cls._re_wl_uri.match(uri) + if not match: + raise ValidationError(u'Invalid URI (%s). Should match: %s' % ( + uri, cls._re_wl_uri.pattern)) + return cls(uri) + + @classmethod + def from_slug(cls, slug): + """Contructs an URI from slug. - >>> WLURI.from_slug_and_lang('a-slug', WLURI.DEFAULT_LANGUAGE).uri - u'http://wolnelektury.pl/katalog/lektura/a-slug/' - >>> WLURI.from_slug_and_lang('a-slug', 'deu').uri - u'http://wolnelektury.pl/katalog/lektura/a-slug/deu/' + >>> WLURI.from_slug('a-slug').uri + u'http://edukacjamedialna.edu.pl/lekcje/a-slug/' """ - if lang is None: - lang = cls.DEFAULT_LANGUAGE - uri = 'http://wolnelektury.pl/katalog/lektura/%s/' % slug - if lang is not None and lang != cls.DEFAULT_LANGUAGE: - uri += lang + '/' + uri = 'http://edukacjamedialna.edu.pl/lekcje/%s/' % slug return cls(uri) def __unicode__(self): @@ -110,18 +121,11 @@ class WLURI(object): def __str__(self): return self.uri - def __eq__(self, other): - return self.slug, self.language == other.slug, other.language - - def filename_stem(self): - stem = self.slug - if self.language != self.DEFAULT_LANGUAGE: - stem += '_' + self.language - return stem + def canonical(self): + return type(self).from_slug(self.slug) - def validate_language(self, language): - if language != self.language: - raise ValidationError("Incorrect language definition in URI") + def __eq__(self, other): + return self.slug == other.slug class DocProvider(object): @@ -130,18 +134,14 @@ class DocProvider(object): Used for generating joined files, like EPUBs. """ - def by_slug_and_lang(self, slug, lang=None): - """Should return a file-like object with a WL document XML.""" - raise NotImplementedError - def by_slug(self, slug): - """Should return a file-like object with a WL document XML.""" - return self.by_slug_and_lang(slug) + """Should return an IOFile object with a WL document XML.""" + raise NotImplementedError def by_uri(self, uri, wluri=WLURI): - """Should return a file-like object with a WL document XML.""" + """Should return an IOFile object with a WL document XML.""" wluri = wluri(uri) - return self.by_slug_and_lang(wluri.slug, wluri.language) + return self.by_slug(wluri.slug) class DirDocProvider(DocProvider): @@ -151,45 +151,51 @@ class DirDocProvider(DocProvider): self.dir = dir_ self.files = {} - def by_slug_and_lang(self, slug, lang=None): - fname = WLURI.from_slug_and_lang(slug, lang).filename_stem() + '.xml' - return open(os.path.join(self.dir, fname)) - + def by_slug(self, slug): + fname = slug + '.xml' + return IOFile.from_filename(os.path.join(self.dir, fname)) + + +def get_default_bookinfo(): + import dcparser + dcparser.BookInfo( + {RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'}, + { + DCNS('creator.expert'): [u'Some, Author'], + DCNS('creator.scenario'): [u'Some, Author'], + DCNS('creator.textbook'): [u'Some, Author'], + DCNS('title'): [u'Some Title'], + DCNS('subject.period'): [u'Unknown'], + DCNS('subject.type'): [u'Unknown'], + DCNS('subject.genre'): [u'Unknown'], + DCNS('date'): ['1970-01-01'], + DCNS('language'): [u'pol'], + # DCNS('date'): [creation_date], + DCNS('publisher'): [u"Fundacja Nowoczesna Polska"], + DCNS('description'): + [u"""Publikacja zrealizowana w ramach projektu + Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa + wykonana przez Bibliotekę Narodową z egzemplarza + pochodzącego ze zbiorów BN."""], + DCNS('identifier.url'): [WLURI.example], + DCNS('rights'): + [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"], + }) + +DEFAULT_BOOKINFO = get_default_bookinfo() -import lxml.etree as etree -import dcparser - -DEFAULT_BOOKINFO = dcparser.BookInfo( - { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'}, - { DCNS('creator'): [u'Some, Author'], - DCNS('title'): [u'Some Title'], - DCNS('subject.period'): [u'Unknown'], - DCNS('subject.type'): [u'Unknown'], - DCNS('subject.genre'): [u'Unknown'], - DCNS('date'): ['1970-01-01'], - DCNS('language'): [WLURI.DEFAULT_LANGUAGE], - # DCNS('date'): [creation_date], - DCNS('publisher'): [u"Fundacja Nowoczesna Polska"], - DCNS('description'): - [u"""Publikacja zrealizowana w ramach projektu - Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa - wykonana przez Bibliotekę Narodową z egzemplarza - pochodzącego ze zbiorów BN."""], - DCNS('identifier.url'): [WLURI.example], - DCNS('rights'): - [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] }) def xinclude_forURI(uri): e = etree.Element(XINS("include")) e.set("href", uri) 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) + dcstring = etree.tostring(bookinfo.to_etree(), encoding=unicode, pretty_print=True) return u'\n' + dcstring + u'\n\n' + ocrtext + \ u'\n\n' @@ -199,8 +205,7 @@ def serialize_raw(element): b = u'' + (element.text or '') for child in element.iterchildren(): - e = etree.tostring(child, method='xml', encoding=unicode, - pretty_print=True) + e = etree.tostring(child, encoding=unicode, pretty_print=True) b += e return b @@ -209,39 +214,44 @@ SERIALIZERS = { 'raw': serialize_raw, } + def serialize_children(element, format='raw'): return SERIALIZERS[format](element) + def get_resource(path): return os.path.join(os.path.dirname(__file__), path) -class OutputFile(object): - """Represents a file returned by one of the converters.""" - +class IOFile(object): + """ Represents a file fed as input or returned as a result. """ _string = None _filename = None + _filename_tmp = False + + def __init__(self, attachments=None): + self.attachments = attachments or {} def __del__(self): - if self._filename: + if self._filename_tmp: os.unlink(self._filename) def __nonzero__(self): return self._string is not None or self._filename is not None @classmethod - def from_string(cls, string): + def from_string(cls, string, *args, **kwargs): """Converter returns contents of a file as a string.""" - instance = cls() + instance = cls(*args, **kwargs) instance._string = string return instance @classmethod - def from_filename(cls, filename): + def from_filename(cls, filename, *args, **kwargs): """Converter returns contents of a file as a named file.""" - instance = cls() + instance = cls(*args, **kwargs) instance._filename = filename return instance @@ -274,6 +284,7 @@ class OutputFile(object): temp.write(self._string) temp.close() self._filename = temp.name + self._filename_tmp = True return self._filename else: return None @@ -285,3 +296,25 @@ class OutputFile(object): if not os.path.isdir(dirname): os.makedirs(dirname) shutil.copy(self.get_filename(), path) + + def dump_to(self, path, directory=None): + """ Path should be name for main file. """ + self.save_as(path) + dirname = os.path.dirname(os.path.abspath(path)) + for filename, attachment in self.attachments.items(): + attachment.save_as(os.path.join(dirname, filename)) + + +class Format(object): + """ Generic format class. """ + def __init__(self, wldoc, **kwargs): + self.wldoc = wldoc + self.customization = kwargs + + def build(self): + raise NotImplementedError + + +class URLOpener(urllib.FancyURLopener): + version = 'FNP Librarian (http://github.com/fnp/librarian)' +urllib._urlopener = URLOpener()