Merge branch 'universal' into edumed-ofop

[librarian.git] / librarian / __init__.py

diff --git a/librarian/__init__.py b/librarian/__init__.py
old mode 100755 (executable)
new mode 100644 (file)
index d56440d..bf41c7a
--- a/librarian/__init__.py
+++ b/librarian/__init__.py
@@ -1,19 +1,42 @@
 # -*- coding: utf-8 -*-
 # -*- coding: utf-8 -*-

-# exception classes
+#
+# 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

 
 

-class ParseError(Exception):
-    
-    def __init__(self, cause, message=None):
-        self.cause = cause
+import os
+import re
+import shutil
+import urllib
+
+
+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:
         try:

-            self.message = message or self.cause.message
-        except:
-            self.message = "No message."
+            message = unicode(args)
+        except UnicodeDecodeError:
+            message = unicode(args, encoding='utf-8', errors='ignore')
+        return message

 
 

-class ValidationError(Exception):
+class ParseError(UnicodeException):
+    pass
+
+class ValidationError(UnicodeException):

     pass
 
 class NoDublinCore(ValidationError):
     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):
     pass
 
 class XMLNamespace(object):


@@ -26,7 +49,7 @@ class XMLNamespace(object):
         return '{%s}%s' % (self.uri, tag)
 
     def __contains__(self, tag):
         return '{%s}%s' % (self.uri, tag)
 
     def __contains__(self, tag):

-        return tag.startswith('{'+str(self)+'}')
+        return tag.startswith('{' + str(self) + '}')

 
     def __repr__(self):
         return 'XMLNamespace(%r)' % self.uri
 
     def __repr__(self):
         return 'XMLNamespace(%r)' % self.uri


@@ -46,20 +69,97 @@ 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")
 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")

 
 WLNS = EmptyNamespace()
 
 
 WLNS = EmptyNamespace()
 

+
+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<slug>[-a-z0-9]+)/?$')
+
+    def __init__(self, uri):
+        uri = unicode(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.
+
+        >>> WLURI.from_slug('a-slug').uri
+        u'http://wolnelektury.pl/katalog/lektura/a-slug/'
+
+        """
+        uri = 'http://prawokultury.pl/publikacje/%s/' % slug
+        return cls(uri)
+
+    def __unicode__(self):
+        return self.uri
+
+    def __str__(self):
+        return self.uri
+
+    def __eq__(self, other):
+        return self.slug == other.slug
+
+
+class DocProvider(object):
+    """Base class for a repository of XML files.
+
+    Used for generating joined files, like EPUBs.
+    """
+
+    def by_slug(self, slug):
+        """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 <slug>.xml """
+
+    def __init__(self, dir_):
+        self.dir = dir_
+        self.files = {}
+
+    def by_slug(self, slug):
+        fname = slug + '.xml'
+        return open(os.path.join(self.dir, fname))
+
+

 import lxml.etree as etree
 import dcparser
 
 DEFAULT_BOOKINFO = dcparser.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'],
+        { 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('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'):
           # DCNS('date'): [creation_date],
           DCNS('publisher'): [u"Fundacja Nowoczesna Polska"],
           DCNS('description'):


@@ -67,43 +167,139 @@ DEFAULT_BOOKINFO = dcparser.BookInfo(
              Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa
              wykonana przez Bibliotekę Narodową z egzemplarza
              pochodzącego ze zbiorów BN."""],
              Wolne Lektury (http://wolnelektury.pl). Reprodukcja cyfrowa
              wykonana przez Bibliotekę Narodową z egzemplarza
              pochodzącego ze zbiorów BN."""],

-          DCNS('identifier.url'):
-            [u"http://wolnelektury.pl/katalog/lektura/template"],
+          DCNS('identifier.url'): [WLURI.example],

           DCNS('rights'):
             [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] })
 
 def xinclude_forURI(uri):
           DCNS('rights'):
             [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] })
 
 def xinclude_forURI(uri):

-    e = etree.Element( XINS("include") )
+    e = etree.Element(XINS("include"))

     e.set("href", uri)
     return etree.tostring(e, encoding=unicode)
     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
 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(),\
+
+    dcstring = etree.tostring(bookinfo.to_etree(), \

         method='xml', encoding=unicode, pretty_print=True)
 
         method='xml', encoding=unicode, pretty_print=True)
 

-    return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext +\
-        u'\n</plain-text>\n</utwor>';
+    return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext + \
+        u'\n</plain-text>\n</utwor>'

 
 
 def serialize_raw(element):
     b = u'' + (element.text or '')
 
     for child in element.iterchildren():
 
 
 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, method='xml', encoding=unicode,
+                pretty_print=True)

         b += e
 
     return b
 
         b += e
 
     return b
 

-from wl_light import serialize_nl
-
-

 SERIALIZERS = {
     'raw': serialize_raw,
 SERIALIZERS = {
     'raw': serialize_raw,

-    'nl': serialize_nl,

 }
 
 def serialize_children(element, format='raw'):
     return SERIALIZERS[format](element)
 }
 
 def serialize_children(element, format='raw'):
     return SERIALIZERS[format](element)

+
+def get_resource(path):
+    return os.path.join(os.path.dirname(__file__), path)
+
+
+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_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, *args, **kwargs):
+        """Converter returns contents of a file as a string."""
+
+        instance = cls(*args, **kwargs)
+        instance._string = string
+        return instance
+
+    @classmethod
+    def from_filename(cls, filename, *args, **kwargs):
+        """Converter returns contents of a file as a named file."""
+
+        instance = cls(*args, **kwargs)
+        instance._filename = filename
+        return instance
+
+    def get_string(self):
+        """Get file's contents as a string."""
+
+        if self._filename is not None:
+            with open(self._filename) as f:
+                return f.read()
+        else:
+            return self._string
+
+    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)
+        elif self._filename is not None:
+            return open(self._filename)
+
+    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
+            temp = NamedTemporaryFile(prefix='librarian-', delete=False)
+            temp.write(self._string)
+            temp.close()
+            self._filename = temp.name
+            self._filename_tmp = True
+            return self._filename
+        else:
+            return None
+
+    def save_as(self, path):
+        """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)
+        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()