X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/a0af4cad2e33a60a2baa132938dbd67a470e730f..5175c4cb9704f442c20abffbd575fb588278b3b2:/librarian/__init__.py diff --git a/librarian/__init__.py b/librarian/__init__.py index 09bdcd7..5c145d3 100644 --- a/librarian/__init__.py +++ b/librarian/__init__.py @@ -9,6 +9,7 @@ import os import re import shutil import urllib +import lxml.etree as etree class UnicodeException(Exception): @@ -25,22 +26,27 @@ class UnicodeException(Exception): message = unicode(args, encoding='utf-8', errors='ignore') return message + class ParseError(UnicodeException): pass + class ValidationError(UnicodeException): pass + class NoDublinCore(ValidationError): """There's no DublinCore section, and it's required.""" pass + 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 @@ -57,6 +63,7 @@ class XMLNamespace(object): def __str__(self): return '%s' % self.uri + class EmptyNamespace(XMLNamespace): def __init__(self): super(EmptyNamespace, self).__init__('') @@ -79,9 +86,10 @@ class WLURI(object): """Represents a WL URI. Extracts slug from it.""" slug = None - example = 'http://edukacjamedialna.edu.pl/' - _re_wl_uri = re.compile(r'http://(www\.)?edukacjamedialna.edu.pl/' - '(?P[-a-z0-9]+)/?$') + example = 'http://edukacjamedialna.edu.pl/lekcje/template' + _re_wl_uri = re.compile( + r'http://(www\.)?edukacjamedialna.edu.pl/lekcje/' + '(?P[-a-z0-9]+)/?$') def __init__(self, uri): uri = unicode(uri) @@ -101,10 +109,10 @@ class WLURI(object): """Contructs an URI from slug. >>> WLURI.from_slug('a-slug').uri - u'http://wolnelektury.pl/katalog/lektura/a-slug/' + u'http://edukacjamedialna.edu.pl/lekcje/a-slug/' """ - uri = 'http://prawokultury.pl/publikacje/%s/' % slug + uri = 'http://edukacjamedialna.edu.pl/lekcje/%s/' % slug return cls(uri) def __unicode__(self): @@ -113,6 +121,9 @@ class WLURI(object): def __str__(self): return self.uri + def canonical(self): + return type(self).from_slug(self.slug) + def __eq__(self, other): return self.slug == other.slug @@ -124,11 +135,11 @@ class DocProvider(object): """ def by_slug(self, slug): - """Should return a file-like object with a WL document XML.""" + """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(wluri.slug) @@ -142,46 +153,49 @@ class DirDocProvider(DocProvider): def by_slug(self, slug): fname = slug + '.xml' - return open(os.path.join(self.dir, fname)) + 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.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]"] }) - 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' @@ -191,8 +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 @@ -201,39 +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 @@ -266,6 +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 @@ -278,6 +297,23 @@ class OutputFile(object): 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)'