X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/fefdce4e24f9e397df5538fe6e7f54b5ece4d841..c072a663ad2d9d8871bb76a7e27b3580b7d92222:/src/librarian/__init__.py diff --git a/src/librarian/__init__.py b/src/librarian/__init__.py index 119b6b1..68afe74 100644 --- a/src/librarian/__init__.py +++ b/src/librarian/__init__.py @@ -27,20 +27,25 @@ class UnicodeException(Exception): message = six.text_type(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.''' @@ -59,6 +64,7 @@ class XMLNamespace(object): def __str__(self): return '%s' % self.uri + class EmptyNamespace(XMLNamespace): def __init__(self): super(EmptyNamespace, self).__init__('') @@ -66,6 +72,7 @@ class EmptyNamespace(XMLNamespace): def __call__(self, tag): 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#') @@ -79,44 +86,6 @@ 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/lektur[ay]/' - '(?P[-a-z0-9]+)/?$') - - def __init__(self, uri): - uri = six.text_type(uri) - self.uri = uri - self.slug = uri.rstrip('/').rsplit('/', 1)[-1] - - @classmethod - 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. - - >>> 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 __str__(self): - return self.uri - - def __eq__(self, other): - return self.slug == other.slug class DocProvider(object): @@ -129,11 +98,6 @@ class DocProvider(object): """Should return a file-like 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.""" - wluri = wluri(uri) - return self.by_slug(wluri.slug) - class DirDocProvider(DocProvider): """ Serve docs from a directory of files in form .xml """ @@ -148,38 +112,49 @@ class DirDocProvider(DocProvider): from . import dcparser +from .meta.types.wluri import WLURI + 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'): [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]"] }) + { + 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'): [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(), method='xml', encoding='unicode', + pretty_print=True + ) return u'\n' + dcstring + u'\n\n' + ocrtext + \ u'\n\n' @@ -190,18 +165,21 @@ def serialize_raw(element): for child in element.iterchildren(): e = etree.tostring(child, method='xml', encoding='unicode', - pretty_print=True) + pretty_print=True) b += e return b + 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) @@ -276,4 +254,6 @@ class OutputFile(object): class URLOpener(FancyURLopener): version = 'FNP Librarian (http://github.com/fnp/librarian)' + + urllib._urlopener = URLOpener()