GwR apple driver wip

resources/content_server/gui.js

@@ -123,7 +123,7 @@ function fetch_library_books(start, num, timeout, sort, order, search) {
 
     current_library_request = $.ajax({
       type: "GET",
-      url: "library",
+      url: "xml",
       data: data,
       cache: false,
       timeout: timeout, //milliseconds

resources/default_tweaks.py

@@ -35,7 +35,7 @@ bool_custom_columns_are_tristate = 'yes'
 
 
 # Provide a set of columns to be sorted on when calibre starts
-#  The argument is None of saved sort history is to be used
+#  The argument is None if saved sort history is to be used
 #  otherwise it is a list of column,order pairs. Column is the
 #  lookup/search name, found using the tooltip for the column
 #  Order is 0 for ascending, 1 for descending

resources/recipes/american_thinker.recipe

@@ -0,0 +1,43 @@
+__license__   = 'GPL v3'
+__copyright__ = '2010, Walt Anthony <workshop.northpole at gmail.com>'
+'''
+www.americanthinker.com
+'''
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class AmericanThinker(BasicNewsRecipe):
+    title          = u'American Thinker'
+    description    = "American Thinker is a daily internet publication devoted to the thoughtful exploration of issues of importance to Americans."
+    __author__     = 'Walt Anthony'
+    publisher             = 'Thomas Lifson'
+    category              = 'news, politics, USA'
+    oldest_article        = 7 #days
+    max_articles_per_feed = 50
+    summary_length        = 150
+    language              = 'en'
+
+    remove_javascript     = True
+    no_stylesheets        = True
+
+
+    conversion_options = {
+                          'comment'   : description
+                        , 'tags'      : category
+                        , 'publisher' : publisher
+                        , 'language'  : language
+                        , 'linearize_tables' : True
+                        }
+
+    remove_tags  = [
+       dict(name=['table', 'iframe', 'embed', 'object'])
+    ]
+
+    remove_tags_after   = dict(name='div', attrs={'class':'article_body'})
+
+
+    feeds = [(u'http://feeds.feedburner.com/americanthinker'),
+		     (u'http://feeds.feedburner.com/AmericanThinkerBlog')
+            ]
+
+    def print_version(self, url):
+        return 'http://www.americanthinker.com/printpage/?url=' + url

resources/recipes/newsweek.recipe

@@ -50,6 +50,7 @@ class Newsweek(BasicNewsRecipe):
         'articlecontent','photoBox', 'article columnist first']}, ]
     recursions = 1
     match_regexps = [r'http://www.newsweek.com/id/\S+/page/\d+']
+    preprocess_regexps = [(re.compile(r'<!--.*?-->', re.DOTALL), lambda m: '')]
 
     def find_title(self, section):
         d = {'scope':'Scope', 'thetake':'The Take', 'features':'Features',

src/calibre/__init__.py

@@ -451,6 +451,20 @@ def prepare_string_for_xml(raw, attribute=False):
 def isbytestring(obj):
     return isinstance(obj, (str, bytes))
 
+def human_readable(size):
+    """ Convert a size in bytes into a human readable form """
+    divisor, suffix = 1, "B"
+    for i, candidate in enumerate(('B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB')):
+        if size < 1024**(i+1):
+            divisor, suffix = 1024**(i), candidate
+            break
+    size = str(float(size)/divisor)
+    if size.find(".") > -1:
+        size = size[:size.find(".")+2]
+    if size.endswith('.0'):
+        size = size[:-2]
+    return size + " " + suffix
+
 if isosx:
     import glob, shutil
     fdir = os.path.expanduser('~/.fonts')

src/calibre/customize/builtins.py

@@ -449,7 +449,7 @@ from calibre.devices.eslick.driver import ESLICK
 from calibre.devices.nuut2.driver import NUUT2
 from calibre.devices.iriver.driver import IRIVER_STORY
 from calibre.devices.binatone.driver import README
-from calibre.devices.hanvon.driver import N516, EB511, ALEX, AZBOOKA
+from calibre.devices.hanvon.driver import N516, EB511, ALEX, AZBOOKA, THEBOOK
 from calibre.devices.edge.driver import EDGE
 from calibre.devices.teclast.driver import TECLAST_K3, NEWSMY, IPAPYRUS
 from calibre.devices.sne.driver import SNE
@@ -529,6 +529,7 @@ plugins += [
     EB600,
     README,
     N516,
+    THEBOOK,
     EB511,
     ELONEX,
     TECLAST_K3,

src/calibre/devices/apple/driver.py

@@ -21,14 +21,20 @@ if iswindows:
     print "running in Windows"
     import win32com.client
 
+class UserInteractionRequired(Exception):
+    print "UserInteractionRequired() exception"
+    #pass
+
 class ITUNES(DevicePlugin):
     name = 'Apple device interface'
     gui_name = 'Apple device'
-    icon = I('devices/iPad.png')
+    icon = I('devices/ipad.png')
     description    = _('Communicate with iBooks through iTunes.')
     supported_platforms = ['windows','osx']
     author = 'GRiker'
 
+    OPEN_FEEDBACK_MESSAGE = _('Apple device detected, launching iTunes')
+
     FORMATS = ['epub']
 
     VENDOR_ID = [0x05ac]
@@ -39,8 +45,10 @@ class ITUNES(DevicePlugin):
     BCD = [0x01]
 
     # Properties
-    cached_paths = {}
+    cached_books = {}
+    ejected = False
     iTunes= None
+    path_template = 'iTunes/%s - %s.epub'
     sources = None
     verbose = True
 
@@ -68,53 +76,69 @@ class ITUNES(DevicePlugin):
                         in main memory of device. If a card is specified and no
                         books are on the card return empty list.
         @return: A BookList.
+
+        Implementation notes:
+        iTunes does not sync purchased books, they are only on the device.  They are visible, but
+        they are not backed up to iTunes.  Since calibre can't manage them, don't show them in the
+        list of device books.
+
         """
         print "ITUNES:books(oncard=%s)" % oncard
+
         if not oncard:
             # Fetch a list of books from iPod device connected to iTunes
             if isosx:
-                '''
-                print "self.sources: %s" % self.sources
-                print "self.sources['library']: %s" % self.sources['library']
+
+                # Fetch Library|Books
                 lib = self.iTunes.sources['library']
-
                 if 'Books' in lib.playlists.name():
-                    booklist = BookList()
-                    it_books = lib.playlists['Books'].file_tracks()
-                    for it_book in it_books:
-                        this_book = Book(it_book.name(), it_book.artist())
-                        this_book.datetime = parse_date(str(it_book.date_added())).timetuple()
-                        this_book.db_id = None
-                        this_book.device_collections = []
-                        this_book.path = 'iTunes/Books/%s.epub' % it_book.name()
-                        this_book.size = it_book.size()
-                        this_book.thumbnail = None
-                        booklist.add_book(this_book, False)
-                    return booklist
+                    lib_books = lib.playlists['Books'].file_tracks()
+                    library_books = {}
+                    for book in lib_books:
+                        path = self.path_template % (book.name(), book.artist())
+                        library_books[path] = book
 
-                else:
-                    return []
-                '''
+                # Fetch iPod|Books, ignore iPod|Purchased books
                 if 'iPod' in self.sources:
                     device = self.sources['iPod']
+                    if 'Purchased' in self.iTunes.sources[device].playlists.name():
+                        purchased_book_ids = [pb.database_ID() for pb in self.iTunes.sources[device].playlists['Purchased'].file_tracks()]
+                    else:
+                        purchased_books_ids = []
+
                     if 'Books' in self.iTunes.sources[device].playlists.name():
                         booklist = BookList()
-                        cached_paths = {}
-                        books = self.iTunes.sources[device].playlists['Books'].file_tracks()
-                        for book in books:
+                        cached_books = {}
+                        device_books = self.iTunes.sources[device].playlists['Books'].file_tracks()
+                        for book in device_books:
+                            if book.database_ID() in purchased_book_ids:
+                                if self.verbose:
+                                    print " skipping purchased book '%s'" % book.name()
+                                continue
                             this_book = Book(book.name(), book.artist())
                             this_book.datetime = parse_date(str(book.date_added())).timetuple()
                             this_book.db_id = None
                             this_book.device_collections = []
-                            this_book.path = 'iTunes/%s - %s.epub' % (book.name(), book.artist())
+                            this_book.path = self.path_template % (book.name(), book.artist())
                             this_book.size = book.size()
                             this_book.thumbnail = None
-                            booklist.add_book(this_book, False)
-                            cached_paths[this_book.path] = { 'title':book.name(),
+                            cached_books[this_book.path] = { 'title':book.name(),
                                                             'author':book.artist(),
-                                                              'book':book}
-                        self.cached_paths = cached_paths
-                        print self.cached_paths
+                                                          'lib_book':library_books[this_book.path] if this_book.path in library_books else None,
+                                                          'dev_book':book,
+                                                          'bl_index':len(booklist)
+                                                           }
+                            booklist.add_book(this_book, False)
+
+                        if self.verbose:
+                            print
+                            print "%-40.40s %-12.12s" % ('Device Books','In Library')
+                            print "%-40.40s %-12.12s" % ('------------','----------')
+
+                            for cp in cached_books.keys():
+                                print "%-40.40s %6.6s" % (cached_books[cp]['title'], 'yes' if cached_books[cp]['lib_book'] else ' no')
+                            print
+                        self.cached_books = cached_books
                         return booklist
                     else:
                         # No books installed on this device
@@ -139,6 +163,10 @@ class ITUNES(DevicePlugin):
         '''
         # print "ITUNES:can_handle()"
         if isosx:
+            if self.ejected:
+                print "ITUNES:can_handle(): device detected, but ejected from iTunes"
+                return False
+
             if self.iTunes:
                 # Check for connected book-capable device
                 names = [s.name() for s in self.iTunes.sources()]
@@ -196,32 +224,19 @@ class ITUNES(DevicePlugin):
     def delete_books(self, paths, end_session=True):
         '''
         Delete books at paths on device.
-        Since we're deleting through iTunes, we'll use the cached handle to the book
+        iTunes doesn't let us directly delete a book on the device.
+        Delete the path(s) from the library, then update iPad
+
         '''
         for path in paths:
-            title = self.cached_paths[path]['title']
-            author = self.cached_paths[path]['author']
-            book = self.cached_paths[path]['book']
-            print "ITUNES.delete_books(): Searching for '%s - %s'" % (title,author)
-            if True:
-                results = self.iTunes.playlists['library'].file_tracks[
-                    (appscript.its.name == title).AND
-                    (appscript.its.artist == author).AND
-                    (appscript.its.kind == 'Book')].get()
-                if len(results) == 1:
-                    book_to_delete = results[0]
-                    print "book_to_delete: %s" % book_to_delete
-                    if self.verbose:
-                        print "ITUNES:delete_books(): Deleting '%s - %s'" % (title, author)
-                    self.iTunes.delete(results[0])
-                elif len(results) > 1:
-                    print "ITUNES.delete_books(): More than one book matches '%s - %s'" % (title, author)
-                else:
-                    print "ITUNES.delete_books(): No book '%s - %s' found in iTunes" % (title, author)
-            else:
-                if self.verbose:
-                    print "ITUNES:delete_books(): Deleting '%s - %s'" % (title, author)
-                self.iTunes.delete(book)
+            title = self.cached_books[path]['title']
+            author = self.cached_books[path]['author']
+            dev_book = self.cached_books[path]['dev_book']
+            lib_book = self.cached_books[path]['lib_book']
+            if self.verbose:
+                print "ITUNES:delete_books(): Deleting '%s' from iTunes library" % (path)
+            self.iTunes.delete(lib_book)
+        self._update_device()
 
     def eject(self):
         '''
@@ -284,6 +299,8 @@ class ITUNES(DevicePlugin):
             # Launch iTunes if not already running
             if self.verbose:
                 print "ITUNES:open(): Instantiating iTunes"
+
+            # Instantiate iTunes
             running_apps = appscript.app('System Events')
             if not 'iTunes' in running_apps.processes.name():
                 if self.verbose:
@@ -297,13 +314,26 @@ class ITUNES(DevicePlugin):
                 if self.verbose:
                     print " %s - %s (already running)" % (self.iTunes.name(), self.iTunes.version())
 
+            # Init the iTunes source list
+            names = [s.name() for s in self.iTunes.sources()]
+            kinds = [str(s.kind()).rpartition('.')[2] for s in self.iTunes.sources()]
+            self.sources = sources = dict(zip(kinds,names))
+
+            # If we're running, but 'iPod' is not a listed source, device was
+            # previously ejected but not physically removed.  can_handle() needs to know this
+            if not 'iPod' in self.sources:
+                self.ejected = True
+
+            else:
+                print "ITUNES:open(): check for presync here ..."
+
     def post_yank_cleanup(self):
         '''
         Called if the user yanks the device without ejecting it first.
         '''
         raise NotImplementedError()
 
-    def remove_books_from_metadata(cls, paths, booklists):
+    def remove_books_from_metadata(self, paths, booklists):
         '''
         Remove books from the metadata list. This function must not communicate
         with the device.
@@ -312,7 +342,14 @@ class ITUNES(DevicePlugin):
                                 (L{books}(oncard=None), L{books}(oncard='carda'),
                                 L{books}(oncard='cardb')).
         '''
-        print "ITUNES.remove_books_from_metadata(): need to implement"
+        print "ITUNES.remove_books_from_metadata():"
+        for path in paths:
+            print " Removing '%s' from calibre booklist, index: %d" % (path, self.cached_books[path]['bl_index'])
+            booklists[0].pop(self.cached_books[path]['bl_index'])
+
+            print " Removing '%s' from self.cached_books" % path
+            self.cached_books.pop(path)
+
 
     def reset(self, key='-1', log_packets=False, report_progress=None,
             detected_device=None) :
@@ -407,6 +444,15 @@ class ITUNES(DevicePlugin):
         '''
         raise NotImplementedError()
 
+    # Private methods
+    def _update_device(self):
+        '''
+        This probably needs a job spinner
+        '''
+        if self.verbose:
+            print "ITUNES:_update_device(): Syncing device with iTunes Library"
+        self.iTunes.update()
+
 class BookList(list):
     '''
     A list of books. Each Book object must have the fields:

src/calibre/devices/hanvon/driver.py

@@ -24,7 +24,7 @@ class N516(USBMS):
 
     VENDOR_ID   = [0x0525]
     PRODUCT_ID  = [0xa4a5]
-    BCD         = [0x323, 0x326, 0x399]
+    BCD         = [0x323, 0x326]
 
     VENDOR_NAME      = 'INGENIC'
     WINDOWS_MAIN_MEM = '_FILE-STOR_GADGE'
@@ -34,6 +34,16 @@ class N516(USBMS):
     EBOOK_DIR_MAIN = 'e_book'
     SUPPORTS_SUB_DIRS = True
 
+class THEBOOK(N516):
+    name = 'The Book driver'
+    gui_name = 'The Book'
+    description    = _('Communicate with The Book reader.')
+    author         = 'Kovid Goyal'
+
+    BCD = [0x399]
+    MAIN_MEMORY_VOLUME_LABEL  = 'The Book Main Memory'
+    EBOOK_DIR_MAIN = 'My books'
+
 class ALEX(N516):
 
     name = 'Alex driver'

src/calibre/devices/misc.py

@@ -17,7 +17,7 @@ class PALMPRE(USBMS):
     supported_platforms = ['windows', 'osx', 'linux']
 
     # Ordered list of supported formats
-    FORMATS     = ['mobi', 'prc', 'pdb', 'txt']
+    FORMATS     = ['epub', 'mobi', 'prc', 'pdb', 'txt']
 
     VENDOR_ID   = [0x0830]
     PRODUCT_ID  = [0x8004, 0x8002, 0x0101]

src/calibre/ebooks/metadata/book/__init__.py

@@ -88,17 +88,28 @@ CALIBRE_METADATA_FIELDS = frozenset([
     ]
 )
 
+CALIBRE_RESERVED_LABELS = frozenset([
+    'search',  # reserved for saved searches
+    'date',
+    'all',
+    'ondevice',
+    'inlibrary',
+    ]
+)
+
 RESERVED_METADATA_FIELDS = SOCIAL_METADATA_FIELDS.union(
                            PUBLICATION_METADATA_FIELDS).union(
                            BOOK_STRUCTURE_FIELDS).union(
                            USER_METADATA_FIELDS).union(
                            DEVICE_METADATA_FIELDS).union(
-                           CALIBRE_METADATA_FIELDS)
+                           CALIBRE_METADATA_FIELDS).union(
+                           CALIBRE_RESERVED_LABELS)
 
 assert len(RESERVED_METADATA_FIELDS) == sum(map(len, (
     SOCIAL_METADATA_FIELDS, PUBLICATION_METADATA_FIELDS,
     BOOK_STRUCTURE_FIELDS, USER_METADATA_FIELDS,
     DEVICE_METADATA_FIELDS, CALIBRE_METADATA_FIELDS,
+    CALIBRE_RESERVED_LABELS
     )))
 
 SERIALIZABLE_FIELDS = SOCIAL_METADATA_FIELDS.union(

src/calibre/ebooks/metadata/opf.py

@@ -1,538 +0,0 @@
-__license__   = 'GPL v3'
-__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
-'''Read/Write metadata from Open Packaging Format (.opf) files.'''
-
-import re, os
-import uuid
-from urllib import unquote, quote
-
-from calibre.constants import __appname__, __version__
-from calibre.ebooks.metadata import MetaInformation, string_to_authors
-from calibre.ebooks.BeautifulSoup import BeautifulStoneSoup, BeautifulSoup
-from calibre.ebooks.lrf import entity_to_unicode
-from calibre.ebooks.metadata import Resource, ResourceCollection
-from calibre.ebooks.metadata.toc import TOC
-
-class OPFSoup(BeautifulStoneSoup):
-
-    def __init__(self, raw):
-        BeautifulStoneSoup.__init__(self, raw,
-                                  convertEntities=BeautifulSoup.HTML_ENTITIES,
-                                  selfClosingTags=['item', 'itemref', 'reference'])
-
-class ManifestItem(Resource):
-
-    @staticmethod
-    def from_opf_manifest_item(item, basedir):
-        if item.has_key('href'):
-            href = item['href']
-            if unquote(href) == href:
-                try:
-                    href = quote(href)
-                except KeyError:
-                    pass
-            res = ManifestItem(href, basedir=basedir, is_path=False)
-            mt = item.get('media-type', '').strip()
-            if mt:
-                res.mime_type = mt
-            return res
-
-    @dynamic_property
-    def media_type(self):
-        def fget(self):
-            return self.mime_type
-        def fset(self, val):
-            self.mime_type = val
-        return property(fget=fget, fset=fset)
-
-
-    def __unicode__(self):
-        return u'<item id="%s" href="%s" media-type="%s" />'%(self.id, self.href(), self.media_type)
-
-    def __str__(self):
-        return unicode(self).encode('utf-8')
-
-    def __repr__(self):
-        return unicode(self)
-
-
-    def __getitem__(self, index):
-        if index == 0:
-            return self.href()
-        if index == 1:
-            return self.media_type
-        raise IndexError('%d out of bounds.'%index)
-
-
-class Manifest(ResourceCollection):
-
-    @staticmethod
-    def from_opf_manifest_element(manifest, dir):
-        m = Manifest()
-        for item in manifest.findAll(re.compile('item')):
-            try:
-                m.append(ManifestItem.from_opf_manifest_item(item, dir))
-                id = item.get('id', '')
-                if not id:
-                    id = 'id%d'%m.next_id
-                m[-1].id = id
-                m.next_id += 1
-            except ValueError:
-                continue
-        return m
-
-    @staticmethod
-    def from_paths(entries):
-        '''
-        `entries`: List of (path, mime-type) If mime-type is None it is autodetected
-        '''
-        m = Manifest()
-        for path, mt in entries:
-            mi = ManifestItem(path, is_path=True)
-            if mt:
-                mi.mime_type = mt
-            mi.id = 'id%d'%m.next_id
-            m.next_id += 1
-            m.append(mi)
-        return m
-
-    def __init__(self):
-        ResourceCollection.__init__(self)
-        self.next_id = 1
-
-
-    def item(self, id):
-        for i in self:
-            if i.id == id:
-                return i
-
-    def id_for_path(self, path):
-        path = os.path.normpath(os.path.abspath(path))
-        for i in self:
-            if i.path and os.path.normpath(i.path) == path:
-                return i.id
-
-    def path_for_id(self, id):
-        for i in self:
-            if i.id == id:
-                return i.path
-
-class Spine(ResourceCollection):
-
-    class Item(Resource):
-
-        def __init__(self, idfunc, *args, **kwargs):
-            Resource.__init__(self, *args, **kwargs)
-            self.is_linear = True
-            self.id = idfunc(self.path)
-
-    @staticmethod
-    def from_opf_spine_element(spine, manifest):
-        s = Spine(manifest)
-        for itemref in spine.findAll(re.compile('itemref')):
-            if itemref.has_key('idref'):
-                r = Spine.Item(s.manifest.id_for_path,
-                               s.manifest.path_for_id(itemref['idref']), is_path=True)
-                r.is_linear = itemref.get('linear', 'yes') == 'yes'
-                s.append(r)
-        return s
-
-    @staticmethod
-    def from_paths(paths, manifest):
-        s = Spine(manifest)
-        for path in paths:
-            try:
-                s.append(Spine.Item(s.manifest.id_for_path, path, is_path=True))
-            except:
-                continue
-        return s
-
-
-
-    def __init__(self, manifest):
-        ResourceCollection.__init__(self)
-        self.manifest = manifest
-
-
-    def linear_items(self):
-        for r in self:
-            if r.is_linear:
-                yield r.path
-
-    def nonlinear_items(self):
-        for r in self:
-            if not r.is_linear:
-                yield r.path
-
-    def items(self):
-        for i in self:
-            yield i.path
-
-
-class Guide(ResourceCollection):
-
-    class Reference(Resource):
-
-        @staticmethod
-        def from_opf_resource_item(ref, basedir):
-            title, href, type = ref.get('title', ''), ref['href'], ref['type']
-            res = Guide.Reference(href, basedir, is_path=False)
-            res.title = title
-            res.type = type
-            return res
-
-        def __repr__(self):
-            ans = '<reference type="%s" href="%s" '%(self.type, self.href())
-            if self.title:
-                ans += 'title="%s" '%self.title
-            return ans + '/>'
-
-
-    @staticmethod
-    def from_opf_guide(guide_elem, base_dir=os.getcwdu()):
-        coll = Guide()
-        for ref in guide_elem.findAll('reference'):
-            try:
-                ref = Guide.Reference.from_opf_resource_item(ref, base_dir)
-                coll.append(ref)
-            except:
-                continue
-        return coll
-
-    def set_cover(self, path):
-        map(self.remove, [i for i in self if 'cover' in i.type.lower()])
-        for type in ('cover', 'other.ms-coverimage-standard', 'other.ms-coverimage'):
-            self.append(Guide.Reference(path, is_path=True))
-            self[-1].type = type
-            self[-1].title = ''
-
-
-class standard_field(object):
-
-    def __init__(self, name):
-        self.name = name
-
-    def __get__(self, obj, typ=None):
-        return getattr(obj, 'get_'+self.name)()
-
-
-class OPF(MetaInformation):
-
-    MIMETYPE = 'application/oebps-package+xml'
-    ENTITY_PATTERN = re.compile(r'&(\S+?);')
-
-    uid            = standard_field('uid')
-    application_id = standard_field('application_id')
-    title          = standard_field('title')
-    authors        = standard_field('authors')
-    language       = standard_field('language')
-    title_sort     = standard_field('title_sort')
-    author_sort    = standard_field('author_sort')
-    comments       = standard_field('comments')
-    category       = standard_field('category')
-    publisher      = standard_field('publisher')
-    isbn           = standard_field('isbn')
-    cover          = standard_field('cover')
-    series         = standard_field('series')
-    series_index   = standard_field('series_index')
-    rating         = standard_field('rating')
-    tags           = standard_field('tags')
-
-    def __init__(self):
-        raise NotImplementedError('Abstract base class')
-
-    @dynamic_property
-    def package(self):
-        def fget(self):
-            return self.soup.find(re.compile('package'))
-        return property(fget=fget)
-
-    @dynamic_property
-    def metadata(self):
-        def fget(self):
-            return self.package.find(re.compile('metadata'))
-        return property(fget=fget)
-
-
-    def get_title(self):
-        title = self.metadata.find('dc:title')
-        if title and title.string:
-            return self.ENTITY_PATTERN.sub(entity_to_unicode, title.string).strip()
-        return self.default_title.strip()
-
-    def get_authors(self):
-        creators = self.metadata.findAll('dc:creator')
-        for elem in creators:
-            role = elem.get('role')
-            if not role:
-                role = elem.get('opf:role')
-            if not role:
-                role = 'aut'
-            if role == 'aut' and elem.string:
-                raw = self.ENTITY_PATTERN.sub(entity_to_unicode, elem.string)
-                return string_to_authors(raw)
-        return []
-
-    def get_author_sort(self):
-        creators = self.metadata.findAll('dc:creator')
-        for elem in creators:
-            role = elem.get('role')
-            if not role:
-                role = elem.get('opf:role')
-            if role == 'aut':
-                fa = elem.get('file-as')
-                return self.ENTITY_PATTERN.sub(entity_to_unicode, fa).strip() if fa else None
-        return None
-
-    def get_title_sort(self):
-        title = self.package.find('dc:title')
-        if title:
-            if title.has_key('file-as'):
-                return title['file-as'].strip()
-        return None
-
-    def get_comments(self):
-        comments = self.soup.find('dc:description')
-        if comments and comments.string:
-            return self.ENTITY_PATTERN.sub(entity_to_unicode, comments.string).strip()
-        return None
-
-    def get_uid(self):
-        package = self.package
-        if package.has_key('unique-identifier'):
-            return package['unique-identifier']
-
-    def get_category(self):
-        category = self.soup.find('dc:type')
-        if category and category.string:
-            return self.ENTITY_PATTERN.sub(entity_to_unicode, category.string).strip()
-        return None
-
-    def get_publisher(self):
-        publisher = self.soup.find('dc:publisher')
-        if publisher and publisher.string:
-            return self.ENTITY_PATTERN.sub(entity_to_unicode, publisher.string).strip()
-        return None
-
-    def get_isbn(self):
-        for item in self.metadata.findAll('dc:identifier'):
-            scheme = item.get('scheme')
-            if not scheme:
-                scheme = item.get('opf:scheme')
-            if scheme is not None and scheme.lower() == 'isbn' and item.string:
-                return str(item.string).strip()
-        return None
-
-    def get_language(self):
-        item = self.metadata.find('dc:language')
-        if not item:
-            return _('Unknown')
-        return ''.join(item.findAll(text=True)).strip()
-
-    def get_application_id(self):
-        for item in self.metadata.findAll('dc:identifier'):
-            scheme = item.get('scheme', None)
-            if scheme is None:
-                scheme = item.get('opf:scheme', None)
-            if scheme in ['libprs500', 'calibre']:
-                return str(item.string).strip()
-        return None
-
-    def get_cover(self):
-        guide = getattr(self, 'guide', [])
-        if not guide:
-            guide = []
-        references = [ref for ref in guide if 'cover' in ref.type.lower()]
-        for candidate in ('cover', 'other.ms-coverimage-standard', 'other.ms-coverimage'):
-            matches = [r for r in references if r.type.lower() == candidate and r.path]
-            if matches:
-                return matches[0].path
-
-    def possible_cover_prefixes(self):
-        isbn, ans = [], []
-        for item in self.metadata.findAll('dc:identifier'):
-            scheme = item.get('scheme')
-            if not scheme:
-                scheme = item.get('opf:scheme')
-            isbn.append((scheme, item.string))
-        for item in isbn:
-            ans.append(item[1].replace('-', ''))
-        return ans
-
-    def get_series(self):
-        s = self.metadata.find('series')
-        if s is not None:
-            return str(s.string).strip()
-        return None
-
-    def get_series_index(self):
-        s = self.metadata.find('series-index')
-        if s and s.string:
-            try:
-                return float(str(s.string).strip())
-            except:
-                return None
-        return None
-
-    def get_rating(self):
-        s = self.metadata.find('rating')
-        if s and s.string:
-            try:
-                return int(str(s.string).strip())
-            except:
-                return None
-        return None
-
-    def get_tags(self):
-        ans = []
-        subs = self.soup.findAll('dc:subject')
-        for sub in subs:
-            val = sub.string
-            if val:
-                ans.append(val)
-        return [unicode(a).strip() for a in ans]
-
-
-class OPFReader(OPF):
-
-    def __init__(self, stream, dir=os.getcwdu()):
-        manage = False
-        if not hasattr(stream, 'read'):
-            manage = True
-            dir = os.path.dirname(stream)
-            stream = open(stream, 'rb')
-        self.default_title = stream.name if hasattr(stream, 'name') else 'Unknown'
-        if hasattr(stream, 'seek'):
-            stream.seek(0)
-        self.soup = OPFSoup(stream.read())
-        if manage:
-            stream.close()
-        self.manifest = Manifest()
-        m = self.soup.find(re.compile('manifest'))
-        if m is not None:
-            self.manifest = Manifest.from_opf_manifest_element(m, dir)
-        self.spine = None
-        spine = self.soup.find(re.compile('spine'))
-        if spine is not None:
-            self.spine = Spine.from_opf_spine_element(spine, self.manifest)
-
-        self.toc = TOC(base_path=dir)
-        self.toc.read_from_opf(self)
-        guide = self.soup.find(re.compile('guide'))
-        if guide is not None:
-            self.guide = Guide.from_opf_guide(guide, dir)
-        self.base_dir = dir
-        self.cover_data = (None, None)
-
-
-class OPFCreator(MetaInformation):
-
-    def __init__(self, base_path, *args, **kwargs):
-        '''
-        Initialize.
-        @param base_path: An absolute path to the directory in which this OPF file
-        will eventually be. This is used by the L{create_manifest} method
-        to convert paths to files into relative paths.
-        '''
-        MetaInformation.__init__(self, *args, **kwargs)
-        self.base_path = os.path.abspath(base_path)
-        if self.application_id is None:
-            self.application_id = str(uuid.uuid4())
-        if not isinstance(self.toc, TOC):
-            self.toc = None
-        if not self.authors:
-            self.authors = [_('Unknown')]
-        if self.guide is None:
-            self.guide = Guide()
-        if self.cover:
-            self.guide.set_cover(self.cover)
-
-
-    def create_manifest(self, entries):
-        '''
-        Create <manifest>
-
-        `entries`: List of (path, mime-type) If mime-type is None it is autodetected
-        '''
-        entries = map(lambda x: x if os.path.isabs(x[0]) else
-                      (os.path.abspath(os.path.join(self.base_path, x[0])), x[1]),
-                      entries)
-        self.manifest = Manifest.from_paths(entries)
-        self.manifest.set_basedir(self.base_path)
-
-    def create_manifest_from_files_in(self, files_and_dirs):
-        entries = []
-
-        def dodir(dir):
-            for spec in os.walk(dir):
-                root, files = spec[0], spec[-1]
-                for name in files:
-                    path = os.path.join(root, name)
-                    if os.path.isfile(path):
-                        entries.append((path, None))
-
-        for i in files_and_dirs:
-            if os.path.isdir(i):
-                dodir(i)
-            else:
-                entries.append((i, None))
-
-        self.create_manifest(entries)
-
-    def create_spine(self, entries):
-        '''
-        Create the <spine> element. Must first call :method:`create_manifest`.
-
-        `entries`: List of paths
-        '''
-        entries = map(lambda x: x if os.path.isabs(x) else
-                      os.path.abspath(os.path.join(self.base_path, x)), entries)
-        self.spine = Spine.from_paths(entries, self.manifest)
-
-    def set_toc(self, toc):
-        '''
-        Set the toc. You must call :method:`create_spine` before calling this
-        method.
-
-        :param toc: A :class:`TOC` object
-        '''
-        self.toc = toc
-
-    def create_guide(self, guide_element):
-        self.guide = Guide.from_opf_guide(guide_element, self.base_path)
-        self.guide.set_basedir(self.base_path)
-
-    def render(self, opf_stream, ncx_stream=None, ncx_manifest_entry=None):
-        from calibre.utils.genshi.template import MarkupTemplate
-        opf_template = open(P('templates/opf.xml'), 'rb').read()
-        template = MarkupTemplate(opf_template)
-        if self.manifest:
-            self.manifest.set_basedir(self.base_path)
-            if ncx_manifest_entry is not None:
-                if not os.path.isabs(ncx_manifest_entry):
-                    ncx_manifest_entry = os.path.join(self.base_path, ncx_manifest_entry)
-                remove = [i for i in self.manifest if i.id == 'ncx']
-                for item in remove:
-                    self.manifest.remove(item)
-                self.manifest.append(ManifestItem(ncx_manifest_entry, self.base_path))
-                self.manifest[-1].id = 'ncx'
-                self.manifest[-1].mime_type = 'application/x-dtbncx+xml'
-        if not self.guide:
-            self.guide = Guide()
-        if self.cover:
-            cover = self.cover
-            if not os.path.isabs(cover):
-                cover = os.path.abspath(os.path.join(self.base_path, cover))
-            self.guide.set_cover(cover)
-        self.guide.set_basedir(self.base_path)
-
-        opf = template.generate(__appname__=__appname__, mi=self, __version__=__version__).render('xml')
-        if not opf.startswith('<?xml '):
-            opf = '<?xml version="1.0"  encoding="UTF-8"?>\n'+opf
-        opf_stream.write(opf)
-        opf_stream.flush()
-        toc = getattr(self, 'toc', None)
-        if toc is not None and ncx_stream is not None:
-            toc.render(ncx_stream, self.application_id)
-            ncx_stream.flush()
-

src/calibre/gui2/__init__.py

@@ -97,8 +97,6 @@ def _config():
             help=_('Overwrite author and title with new metadata'))
     c.add_opt('enforce_cpu_limit', default=True,
             help=_('Limit max simultaneous jobs to number of CPUs'))
-    c.add_opt('user_categories', default={},
-            help=_('User-created tag browser categories'))
 
     return ConfigProxy(c)
 
@@ -229,19 +227,6 @@ def info_dialog(parent, title, msg, det_msg='', show=False):
     return d
 
 
-def human_readable(size):
-    """ Convert a size in bytes into a human readable form """
-    divisor, suffix = 1, "B"
-    for i, candidate in enumerate(('B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB')):
-        if size < 1024**(i+1):
-            divisor, suffix = 1024**(i), candidate
-            break
-    size = str(float(size)/divisor)
-    if size.find(".") > -1:
-        size = size[:size.find(".")+2]
-    if size.endswith('.0'):
-        size = size[:-2]
-    return size + " " + suffix
 
 class Dispatcher(QObject):
     '''Convenience class to ensure that a function call always happens in the

src/calibre/gui2/dialogs/tag_categories.py

@@ -7,7 +7,7 @@ from PyQt4.QtCore import SIGNAL, Qt
 from PyQt4.QtGui import QDialog, QIcon, QListWidgetItem
 
 from calibre.gui2.dialogs.tag_categories_ui import Ui_TagCategories
-from calibre.gui2 import config
+from calibre.utils.config import prefs
 from calibre.gui2.dialogs.confirm_delete import confirm
 from calibre.constants import islinux
 
@@ -22,7 +22,7 @@ class Item:
         return 'name=%s, label=%s, index=%s, exists='%(self.name, self.label, self.index, self.exists)
 
 class TagCategories(QDialog, Ui_TagCategories):
-    category_labels_orig =   ['', 'author', 'series', 'publisher', 'tag']
+    category_labels_orig =   ['', 'authors', 'series', 'publishers', 'tags']
 
     def __init__(self, window, db, index=None):
         QDialog.__init__(self, window)
@@ -64,7 +64,7 @@ class TagCategories(QDialog, Ui_TagCategories):
                 self.all_items.append(t)
                 self.all_items_dict[label+':'+n] = t
 
-        self.categories = dict.copy(config['user_categories'])
+        self.categories = dict.copy(prefs['user_categories'])
         if self.categories is None:
             self.categories = {}
         for cat in self.categories:
@@ -181,7 +181,7 @@ class TagCategories(QDialog, Ui_TagCategories):
 
     def accept(self):
         self.save_category()
-        config['user_categories'] = self.categories
+        prefs['user_categories'] = self.categories
         QDialog.accept(self)
 
     def save_category(self):

src/calibre/gui2/main.ui

@@ -650,7 +650,7 @@
      <normaloff>:/images/merge_books.svg</normaloff>:/images/merge_books.svg</iconset>
    </property>
    <property name="text">
-    <string>Merge books</string>
+    <string>Merge book records</string>
    </property>
    <property name="shortcut">
     <string>M</string>

src/calibre/gui2/tag_view.py

@@ -126,7 +126,7 @@ class TagTreeItem(object): # {{{
     TAG      = 1
     ROOT     = 2
 
-    def __init__(self, data=None, category_icon=None, icon_map=None, parent=None):
+    def __init__(self, data=None, category_icon=None, icon_map=None, parent=None, tooltip=None):
         self.parent = parent
         self.children = []
         if self.parent is not None:
@@ -144,6 +144,7 @@ class TagTreeItem(object): # {{{
         elif self.type == self.TAG:
             icon_map[0] = data.icon
             self.tag, self.icon_state_map = data, list(map(QVariant, icon_map))
+        self.tooltip = tooltip
 
     def __str__(self):
         if self.type == self.ROOT:
@@ -175,6 +176,8 @@ class TagTreeItem(object): # {{{
             return self.icon
         if role == Qt.FontRole:
             return self.bold_font
+        if role == Qt.ToolTipRole and self.tooltip is not None:
+            return QVariant(self.tooltip)
         return NONE
 
     def tag_data(self, role):
@@ -197,33 +200,32 @@ class TagTreeItem(object): # {{{
 
 class TagsModel(QAbstractItemModel): # {{{
 
-    categories_orig = [_('Authors'), _('Series'), _('Formats'), _('Publishers'),
-                       _('Ratings'), _('News'), _('Tags')]
-    row_map_orig    = ['author', 'series', 'format', 'publisher', 'rating',
-                       'news', 'tag']
-    tags_categories_start= 7
-    search_keys=['search', _('Searches')]
-
     def __init__(self, db, parent=None):
         QAbstractItemModel.__init__(self, parent)
-        self.cat_icon_map_orig = list(map(QIcon, [I('user_profile.svg'),
-                I('series.svg'), I('book.svg'), I('publisher.png'), I('star.png'),
-                I('news.svg'), I('tags.svg')]))
+
+        # must do this here because 'QPixmap: Must construct a QApplication
+        # before a QPaintDevice'
+        self.category_icon_map = {'authors': QIcon(I('user_profile.svg')),
+                    'series': QIcon(I('series.svg')),
+                    'formats':QIcon(I('book.svg')),
+                    'publishers': QIcon(I('publisher.png')),
+                    'ratings':QIcon(I('star.png')),
+                    'news':QIcon(I('news.svg')),
+                    'tags':QIcon(I('tags.svg')),
+                    '*custom':QIcon(I('column.svg')),
+                    '*user':QIcon(I('drawer.svg')),
+                    'search':QIcon(I('search.svg'))}
         self.icon_state_map = [None, QIcon(I('plus.svg')), QIcon(I('minus.svg'))]
-        self.custcol_icon = QIcon(I('column.svg'))
-        self.search_icon = QIcon(I('search.svg'))
-        self.usercat_icon = QIcon(I('drawer.svg'))
-        self.label_to_icon_map = dict(map(None, self.row_map_orig, self.cat_icon_map_orig))
-        self.label_to_icon_map['*custom'] = self.custcol_icon
         self.db = db
         self.search_restriction = ''
-        self.user_categories = {}
         self.ignore_next_search = 0
         data = self.get_node_tree(config['sort_by_popularity'])
         self.root_item = TagTreeItem()
         for i, r in enumerate(self.row_map):
             c = TagTreeItem(parent=self.root_item,
-                    data=self.categories[i], category_icon=self.cat_icon_map[i])
+                    data=self.categories[i],
+                    category_icon=self.category_icon_map[r],
+                    tooltip=_('The lookup/search name is "{0}"').format(r))
             for tag in data[r]:
                 TagTreeItem(parent=c, data=tag, icon_map=self.icon_state_map)
 
@@ -233,74 +235,21 @@ class TagsModel(QAbstractItemModel): # {{{
     def get_node_tree(self, sort):
         self.row_map = []
         self.categories = []
-        # strip the icons after the 'standard' categories. We will put them back later
-        if self.tags_categories_start < len(self.row_map_orig):
-            self.cat_icon_map = self.cat_icon_map_orig[:self.tags_categories_start-len(self.row_map_orig)]
-        else:
-            self.cat_icon_map = self.cat_icon_map_orig[:]
 
-        self.user_categories = dict.copy(config['user_categories'])
-        column_map = config['column_map']
-
-        for i in range(0, self.tags_categories_start): # First the standard categories
-            self.row_map.append(self.row_map_orig[i])
-            self.categories.append(self.categories_orig[i])
         if len(self.search_restriction):
-            data = self.db.get_categories(sort_on_count=sort, icon_map=self.label_to_icon_map,
+            data = self.db.get_categories(sort_on_count=sort, icon_map=self.category_icon_map,
                         ids=self.db.search(self.search_restriction, return_matches=True))
         else:
-            data = self.db.get_categories(sort_on_count=sort, icon_map=self.label_to_icon_map)
+            data = self.db.get_categories(sort_on_count=sort, icon_map=self.category_icon_map)
 
-        for c in data:  # now the custom columns
-            if c not in self.row_map_orig and c in column_map:
-                self.row_map.append(c)
-                self.categories.append(self.db.custom_column_label_map[c]['name'])
-                self.cat_icon_map.append(self.custcol_icon)
+        tb_categories = self.db.get_tag_browser_categories()
+        for category in tb_categories.iterkeys():
+            if category in data: # They should always be there, but ...
+                self.row_map.append(category)
+                self.categories.append(tb_categories[category]['name'])
 
-        # Now the rest of the normal tag categories
-        for i in range(self.tags_categories_start, len(self.row_map_orig)):
-            self.row_map.append(self.row_map_orig[i])
-            self.categories.append(self.categories_orig[i])
-            self.cat_icon_map.append(self.cat_icon_map_orig[i])
-
-        # Clean up the author's tags, getting rid of the '|' characters
-        if data['author'] is not None:
-            for t in data['author']:
-                t.name = t.name.replace('|', ',')
-
-        # Now do the user-defined categories. There is a time/space tradeoff here.
-        # By converting the tags into a map, we can do the verification in the category
-        # loop much faster, at the cost of duplicating the categories lists.
-        taglist = {}
-        for c in self.row_map:
-            taglist[c] = dict(map(lambda t:(t.name, t), data[c]))
-
-        for c in self.user_categories:
-            l = []
-            for (name,label,ign) in self.user_categories[c]:
-                if label in taglist and name in taglist[label]: # use same node as the complete category
-                    l.append(taglist[label][name])
-                # else: do nothing, to eliminate nodes that have zero counts
-            if config['sort_by_popularity']:
-                data[c+'*'] = sorted(l, cmp=(lambda x, y: cmp(x.count, y.count)))
-            else:
-                data[c+'*'] = sorted(l, cmp=(lambda x, y: cmp(x.name.lower(), y.name.lower())))
-            self.row_map.append(c+'*')
-            self.categories.append(c)
-            self.cat_icon_map.append(self.usercat_icon)
-
-        data['search'] = self.get_search_nodes(self.search_icon)  # Add the search category
-        self.row_map.append(self.search_keys[0])
-        self.categories.append(self.search_keys[1])
-        self.cat_icon_map.append(self.search_icon)
         return data
 
-    def get_search_nodes(self, icon):
-        l = []
-        for i in saved_searches.names():
-            l.append(Tag(i, tooltip=saved_searches.lookup(i), icon=icon))
-        return l
-
     def refresh(self):
         data = self.get_node_tree(config['sort_by_popularity']) # get category data
         for i, r in enumerate(self.row_map):

src/calibre/gui2/ui.py

@@ -183,7 +183,7 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
                 _('Error communicating with device'), ' ')
         self.device_error_dialog.setModal(Qt.NonModal)
         self.tb_wrapper = textwrap.TextWrapper(width=40)
-        self.device_connected = False
+        self.device_connected = None
         self.viewers = collections.deque()
         self.content_server = None
         self.system_tray_icon = SystemTrayIcon(QIcon(I('library.png')), self)
@@ -675,6 +675,15 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
         self._sync_menu.fetch_annotations.connect(self.fetch_annotations)
         self._sync_menu.connect_to_folder.connect(self.connect_to_folder)
         self._sync_menu.disconnect_from_folder.connect(self.disconnect_from_folder)
+        if self.device_connected:
+            self._sync_menu.connect_to_folder_action.setEnabled(False)
+            if self.device_connected == 'folder':
+                self._sync_menu.disconnect_from_folder_action.setEnabled(True)
+            else:
+                self._sync_menu.disconnect_from_folder_action.setEnabled(False)
+        else:
+            self._sync_menu.connect_to_folder_action.setEnabled(True)
+            self._sync_menu.disconnect_from_folder_action.setEnabled(False)
 
     def add_spare_server(self, *args):
         self.spare_servers.append(Server(limit=int(config['worker_limit']/2.0)))
@@ -944,7 +953,7 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
             self.status_bar.showMessage(_('Device: ')+\
                 self.device_manager.device.__class__.get_gui_name()+\
                         _(' detected.'), 3000)
-            self.device_connected = True
+            self.device_connected = 'device' if not is_folder_device else 'folder'
             self._sync_menu.enable_device_actions(True,
                     self.device_manager.device.card_prefix(),
                     self.device_manager.device)
@@ -955,7 +964,7 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
             self._sync_menu.connect_to_folder_action.setEnabled(True)
             self._sync_menu.disconnect_from_folder_action.setEnabled(False)
             self.save_device_view_settings()
-            self.device_connected = False
+            self.device_connected = None
             self._sync_menu.enable_device_actions(False)
             self.location_view.model().update_devices()
             self.vanity.setText(self.vanity_template%\

src/calibre/gui2/widgets.py

@@ -13,11 +13,10 @@ from PyQt4.Qt import QListView, QIcon, QFont, QLabel, QListWidget, \
                         QAbstractButton, QPainter, QLineEdit, QComboBox, \
                         QMenu, QStringListModel, QCompleter, QStringList
 
-from calibre.gui2 import human_readable, NONE, \
-                         error_dialog, pixmap_to_data, dynamic
+from calibre.gui2 import NONE, error_dialog, pixmap_to_data, dynamic
 
 from calibre.gui2.filename_pattern_ui import Ui_Form
-from calibre import fit_image
+from calibre import fit_image, human_readable
 from calibre.utils.fonts import fontconfig
 from calibre.ebooks import BOOK_EXTENSIONS
 from calibre.ebooks.metadata.meta import metadata_from_filename

src/calibre/gui2/wizard/__init__.py

@@ -100,6 +100,13 @@ class Booq(Device):
     output_format = 'EPUB'
     id = 'booq'
 
+class TheBook(Device):
+    name = 'The Book'
+    manufacturer = 'Augen'
+    output_profile = 'prs505'
+    output_format = 'EPUB'
+    id = 'thebook'
+
 class Avant(Booq):
     name = 'Booq Avant'
 

src/calibre/library/caches.py

@@ -626,20 +626,24 @@ class ResultCache(SearchQueryParser):
         self._map.sort(cmp=fcmp, reverse=not ascending)
         self._map_filtered = [id for id in self._map if id in self._map_filtered]
 
-    def search(self, query, return_matches=False):
+    def search(self, query, return_matches=False,
+            ignore_search_restriction=False):
         if not query or not query.strip():
-            q = self.search_restriction
-        else:
-            q = '%s (%s)' % (self.search_restriction, query)
+            q = ''
+            if not ignore_search_restriction:
+                q = self.search_restriction
+        elif not ignore_search_restriction:
+            q = u'%s (%s)' % (self.search_restriction, query)
         if not q:
             if return_matches:
-                return list(self.map) # when return_matches, do not update the maps!
+                return list(self._map) # when return_matches, do not update the maps!
             self._map_filtered = list(self._map)
             return []
         matches = sorted(self.parse(q))
+        ans = [id for id in self._map if id in matches]
         if return_matches:
-            return [id for id in self._map if id in matches]
-        self._map_filtered = [id for id in self._map if id in matches]
+            return ans
+        self._map_filtered = ans
         return []
 
     def set_search_restriction(self, s):

src/calibre/library/cli.py

@@ -9,99 +9,18 @@ Command line interface to the calibre database.
 
 import sys, os, cStringIO
 from textwrap import TextWrapper
-from urllib import quote
 
 from calibre import terminal_controller, preferred_encoding, prints
 from calibre.utils.config import OptionParser, prefs
 from calibre.ebooks.metadata.meta import get_metadata
 from calibre.library.database2 import LibraryDatabase2
 from calibre.ebooks.metadata.opf2 import OPFCreator, OPF
-from calibre.utils.genshi.template import MarkupTemplate
 from calibre.utils.date import isoformat
 
 FIELDS = set(['title', 'authors', 'author_sort', 'publisher', 'rating',
     'timestamp', 'size', 'tags', 'comments', 'series', 'series_index',
     'formats', 'isbn', 'uuid', 'pubdate', 'cover'])
 
-XML_TEMPLATE = '''\
-<?xml version="1.0"  encoding="UTF-8"?>
-<calibredb xmlns:py="http://genshi.edgewall.org/">
-<py:for each="record in data">
-    <record>
-        <id>${record['id']}</id>
-        <uuid>${record['uuid']}</uuid>
-        <title>${record['title']}</title>
-        <authors sort="${record['author_sort']}">
-        <py:for each="author in record['authors']">
-            <author>$author</author>
-        </py:for>
-        </authors>
-        <publisher>${record['publisher']}</publisher>
-        <rating>${record['rating']}</rating>
-        <date>${record['timestamp'].isoformat()}</date>
-        <pubdate>${record['pubdate'].isoformat()}</pubdate>
-        <size>${record['size']}</size>
-        <tags py:if="record['tags']">
-        <py:for each="tag in record['tags']">
-            <tag>$tag</tag>
-        </py:for>
-        </tags>
-        <comments>${record['comments']}</comments>
-        <series py:if="record['series']" index="${record['series_index']}">${record['series']}</series>
-        <isbn>${record['isbn']}</isbn>
-        <cover py:if="record['cover']">${record['cover'].replace(os.sep, '/')}</cover>
-        <formats py:if="record['formats']">
-        <py:for each="path in record['formats']">
-            <format>${path.replace(os.sep, '/')}</format>
-        </py:for>
-        </formats>
-    </record>
-</py:for>
-</calibredb>
-'''
-
-STANZA_TEMPLATE='''\
-<?xml version="1.0" encoding="utf-8"?>
-<feed xmlns="http://www.w3.org/2005/Atom" xmlns:py="http://genshi.edgewall.org/">
-  <title>calibre Library</title>
-  <author>
-    <name>calibre</name>
-    <uri>http://calibre-ebook.com</uri>
-  </author>
-  <id>$id</id>
-  <updated>${updated.isoformat()}</updated>
-  <subtitle>
-        ${subtitle}
-  </subtitle>
-  <py:for each="record in data">
-  <entry>
-      <title>${record['title']}</title>
-      <id>urn:calibre:${record['uuid']}</id>
-      <author><name>${record['author_sort']}</name></author>
-      <updated>${record['timestamp'].isoformat()}</updated>
-      <link type="application/epub+zip" href="${quote(record['fmt_epub'].replace(sep, '/'))}"/>
-      <link py:if="record['cover']" rel="x-stanza-cover-image" type="image/png" href="${quote(record['cover'].replace(sep, '/'))}"/>
-      <link py:if="record['cover']" rel="x-stanza-cover-image-thumbnail" type="image/png" href="${quote(record['cover'].replace(sep, '/'))}"/>
-      <content type="xhtml">
-          <div xmlns="http://www.w3.org/1999/xhtml">
-              <py:for each="f in ('authors', 'publisher', 'rating', 'tags', 'series', 'isbn')">
-              <py:if test="record[f]">
-              ${f.capitalize()}:${unicode(', '.join(record[f]) if f=='tags' else record[f])}
-              <py:if test="f =='series'"># ${str(record['series_index'])}</py:if>
-              <br/>
-              </py:if>
-              </py:for>
-              <py:if test="record['comments']">
-              <br/>
-              ${record['comments']}
-              </py:if>
-          </div>
-      </content>
-  </entry>
-  </py:for>
-</feed>
-'''
-
 def send_message(msg=''):
     prints('Notifying calibre of the change')
     from calibre.utils.ipc import RC
@@ -130,81 +49,67 @@ def get_db(dbpath, options):
     return LibraryDatabase2(dbpath)
 
 def do_list(db, fields, afields, sort_by, ascending, search_text, line_width, separator,
-            prefix, output_format, subtitle='Books in the calibre database'):
+            prefix, subtitle='Books in the calibre database'):
     if sort_by:
         db.sort(sort_by, ascending)
     if search_text:
         db.search(search_text)
-    authors_to_string = output_format in ['stanza', 'text']
-    data = db.get_data_as_dict(prefix, authors_as_string=authors_to_string)
+    data = db.get_data_as_dict(prefix, authors_as_string=True)
     fields = ['id'] + fields
     title_fields = fields
     fields = [db.custom_column_label_map[x[1:]]['num'] if x[0]=='*'
             else x for x in fields]
-    if output_format == 'text':
-        for f in data:
-            fmts = [x for x in f['formats'] if x is not None]
-            f['formats'] = u'[%s]'%u','.join(fmts)
-        widths = list(map(lambda x : 0, fields))
-        for record in data:
-            for f in record.keys():
-                if hasattr(record[f], 'isoformat'):
-                    record[f] = isoformat(record[f], as_utc=False)
-                else:
-                    record[f] = unicode(record[f])
-                record[f] = record[f].replace('\n', ' ')
-        for i in data:
-            for j, field in enumerate(fields):
-                widths[j] = max(widths[j], len(unicode(i[field])))
 
-        screen_width = terminal_controller.COLS if line_width < 0 else line_width
-        if not screen_width:
-            screen_width = 80
-        field_width = screen_width//len(fields)
-        base_widths = map(lambda x: min(x+1, field_width), widths)
+    for f in data:
+        fmts = [x for x in f['formats'] if x is not None]
+        f['formats'] = u'[%s]'%u','.join(fmts)
+    widths = list(map(lambda x : 0, fields))
+    for record in data:
+        for f in record.keys():
+            if hasattr(record[f], 'isoformat'):
+                record[f] = isoformat(record[f], as_utc=False)
+            else:
+                record[f] = unicode(record[f])
+            record[f] = record[f].replace('\n', ' ')
+    for i in data:
+        for j, field in enumerate(fields):
+            widths[j] = max(widths[j], len(unicode(i[field])))
 
-        while sum(base_widths) < screen_width:
-            adjusted = False
-            for i in range(len(widths)):
-                if base_widths[i] < widths[i]:
-                    base_widths[i] += min(screen_width-sum(base_widths), widths[i]-base_widths[i])
-                    adjusted = True
-                    break
-            if not adjusted:
+    screen_width = terminal_controller.COLS if line_width < 0 else line_width
+    if not screen_width:
+        screen_width = 80
+    field_width = screen_width//len(fields)
+    base_widths = map(lambda x: min(x+1, field_width), widths)
+
+    while sum(base_widths) < screen_width:
+        adjusted = False
+        for i in range(len(widths)):
+            if base_widths[i] < widths[i]:
+                base_widths[i] += min(screen_width-sum(base_widths), widths[i]-base_widths[i])
+                adjusted = True
                 break
+        if not adjusted:
+            break
 
-        widths = list(base_widths)
-        titles = map(lambda x, y: '%-*s%s'%(x-len(separator), y, separator),
-                widths, title_fields)
-        print terminal_controller.GREEN + ''.join(titles)+terminal_controller.NORMAL
+    widths = list(base_widths)
+    titles = map(lambda x, y: '%-*s%s'%(x-len(separator), y, separator),
+            widths, title_fields)
+    print terminal_controller.GREEN + ''.join(titles)+terminal_controller.NORMAL
 
-        wrappers = map(lambda x: TextWrapper(x-1), widths)
-        o = cStringIO.StringIO()
+    wrappers = map(lambda x: TextWrapper(x-1), widths)
+    o = cStringIO.StringIO()
 
-        for record in data:
-            text = [wrappers[i].wrap(unicode(record[field]).encode('utf-8')) for i, field in enumerate(fields)]
-            lines = max(map(len, text))
-            for l in range(lines):
-                for i, field in enumerate(text):
-                    ft = text[i][l] if l < len(text[i]) else ''
-                    filler = '%*s'%(widths[i]-len(ft)-1, '')
-                    o.write(ft)
-                    o.write(filler+separator)
-                print >>o
-        return o.getvalue()
-    elif output_format == 'xml':
-        template = MarkupTemplate(XML_TEMPLATE)
-        return template.generate(data=data, os=os).render('xml')
-    elif output_format == 'stanza':
-        data = [i for i in data if i.has_key('fmt_epub')]
-        for x in data:
-            if isinstance(x['fmt_epub'], unicode):
-                x['fmt_epub'] = x['fmt_epub'].encode('utf-8')
-            if isinstance(x['cover'], unicode):
-                x['cover'] = x['cover'].encode('utf-8')
-        template = MarkupTemplate(STANZA_TEMPLATE)
-        return template.generate(id="urn:calibre:main", data=data, subtitle=subtitle,
-                sep=os.sep, quote=quote, updated=db.last_modified()).render('xml')
+    for record in data:
+        text = [wrappers[i].wrap(unicode(record[field]).encode('utf-8')) for i, field in enumerate(fields)]
+        lines = max(map(len, text))
+        for l in range(lines):
+            for i, field in enumerate(text):
+                ft = text[i][l] if l < len(text[i]) else ''
+                filler = '%*s'%(widths[i]-len(ft)-1, '')
+                o.write(ft)
+                o.write(filler+separator)
+            print >>o
+    return o.getvalue()
 
 def list_option_parser(db=None):
     fields = set(FIELDS)
@@ -236,9 +141,6 @@ List the books available in the calibre database.
                       help=_('The maximum width of a single line in the output. Defaults to detecting screen size.'))
     parser.add_option('--separator', default=' ', help=_('The string used to separate fields. Default is a space.'))
     parser.add_option('--prefix', default=None, help=_('The prefix for all file paths. Default is the absolute path to the library folder.'))
-    of = ['text', 'xml', 'stanza']
-    parser.add_option('--output-format', choices=of, default='text',
-                      help=_('The format in which to output the data. Available choices: %s. Defaults is text.')%of)
     return parser
 
 
@@ -272,7 +174,7 @@ def command_list(args, dbpath):
         return 1
 
     print do_list(db, fields, afields, opts.sort_by, opts.ascending, opts.search, opts.line_width, opts.separator,
-            opts.prefix, opts.output_format)
+            opts.prefix)
     return 0
 
 

src/calibre/library/custom_columns.py

@@ -141,11 +141,15 @@ class CustomColumns(object):
         }
 
         # Create Tag Browser categories for custom columns
-        for i, v in self.custom_column_num_map.items():
+        for k in sorted(self.custom_column_label_map.keys()):
+            v = self.custom_column_label_map[k]
             if v['normalized']:
-                tn = 'custom_column_{0}'.format(i)
-                self.tag_browser_categories[tn] = [v['label'], 'value']
-                self.tag_browser_datatype[v['label']] = v['datatype']
+                tn = 'custom_column_{0}'.format(v['num'])
+                self.tag_browser_categories[v['label']] = {
+                        'table':tn, 'column':'value',
+                        'type':v['datatype'], 'is_multiple':v['is_multiple'],
+                        'kind':'custom', 'name':v['name']
+                    }
 
     def get_custom(self, idx, label=None, num=None, index_is_id=False):
         if label is not None:

src/calibre/library/database2.py

@@ -33,6 +33,9 @@ from calibre.customize.ui import run_plugins_on_import
 
 from calibre.utils.filenames import ascii_filename
 from calibre.utils.date import utcnow, now as nowf, utcfromtimestamp
+from calibre.utils.ordered_dict import OrderedDict
+from calibre.utils.config import prefs
+from calibre.utils.search_query_parser import saved_searches
 from calibre.ebooks import BOOK_EXTENSIONS, check_ebook_format
 
 if iswindows:
@@ -123,24 +126,33 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
         if isinstance(self.dbpath, unicode):
             self.dbpath = self.dbpath.encode(filesystem_encoding)
 
-        self.tag_browser_categories = {
-                'tags'      : ['tag', 'name'],
-                'series'    : ['series', 'name'],
-                'publishers': ['publisher', 'name'],
-                'authors'   : ['author', 'name'],
-                'news'      : ['news', 'name'],
-                'ratings'   : ['rating', 'rating']
-        }
-        self.tag_browser_datatype = {
-                'tag'       : 'textmult',
-                'series'    : None,
-                'publisher' : 'text',
-                'author'    : 'text',
-                'news'      : None,
-                'rating'    : 'rating',
-        }
-
-        self.tag_browser_formatters = {'rating': lambda x:u'\u2605'*int(round(x/2.))}
+        # Order as has been customary in the tags pane.
+        tag_browser_categories_items = [
+                ('authors',   {'table':'authors', 'column':'name',
+                               'type':'text', 'is_multiple':False,
+                               'kind':'standard', 'name':_('Authors')}),
+                ('series',    {'table':'series', 'column':'name',
+                               'type':None, 'is_multiple':False,
+                               'kind':'standard', 'name':_('Series')}),
+                ('formats',   {'table':None, 'column':None,
+                               'type':None, 'is_multiple':False,
+                               'kind':'standard', 'name':_('Formats')}),
+                ('publishers',{'table':'publishers', 'column':'name',
+                               'type':'text', 'is_multiple':False,
+                               'kind':'standard', 'name':_('Publishers')}),
+                ('ratings',   {'table':'ratings', 'column':'rating',
+                               'type':'rating', 'is_multiple':False,
+                               'kind':'standard', 'name':_('Ratings')}),
+                ('news',      {'table':'news', 'column':'name',
+                               'type':None, 'is_multiple':False,
+                               'kind':'standard', 'name':_('News')}),
+                ('tags',      {'table':'tags', 'column':'name',
+                               'type':'text', 'is_multiple':True,
+                               'kind':'standard', 'name':_('Tags')}),
+        ]
+        self.tag_browser_categories = OrderedDict()
+        for k,v in tag_browser_categories_items:
+            self.tag_browser_categories[k] = v
 
         self.connect()
         self.is_case_sensitive = not iswindows and not isosx and \
@@ -649,36 +661,65 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
     def get_recipe(self, id):
         return self.conn.get('SELECT script FROM feeds WHERE id=?', (id,), all=False)
 
+    def get_tag_browser_categories(self):
+        return self.tag_browser_categories
+
     def get_categories(self, sort_on_count=False, ids=None, icon_map=None):
         self.books_list_filter.change([] if not ids else ids)
 
         categories = {}
-        for tn, cn in self.tag_browser_categories.items():
+
+        #### First, build the standard and custom-column categories ####
+        for category in self.tag_browser_categories.keys():
+            tn = self.tag_browser_categories[category]['table']
+            categories[category] = []   #reserve the position in the ordered list
+            if tn is None:              # Nothing to do for the moment
+                continue
+            cn = self.tag_browser_categories[category]['column']
             if ids is None:
-                query = 'SELECT id, {0}, count FROM tag_browser_{1}'.format(cn[1], tn)
+                query = 'SELECT id, {0}, count FROM tag_browser_{1}'.format(cn, tn)
             else:
-                query = 'SELECT id, {0}, count FROM tag_browser_filtered_{1}'.format(cn[1], tn)
+                query = 'SELECT id, {0}, count FROM tag_browser_filtered_{1}'.format(cn, tn)
             if sort_on_count:
                 query += ' ORDER BY count DESC'
             else:
-                query += ' ORDER BY {0} ASC'.format(cn[1])
+                query += ' ORDER BY {0} ASC'.format(cn)
             data = self.conn.get(query)
-            category = cn[0]
+
+            # icon_map is not None if get_categories is to store an icon and
+            # possibly a tooltip in the tag structure.
             icon, tooltip = None, ''
             if icon_map:
-                if category in icon_map:
-                    icon = icon_map[category]
-                else:
+                if self.tag_browser_categories[category]['kind'] == 'standard':
+                    if category in icon_map:
+                        icon = icon_map[category]
+                elif self.tag_browser_categories[category]['kind'] == 'custom':
                     icon = icon_map['*custom']
+                    icon_map[category] = icon_map['*custom']
                     tooltip = self.custom_column_label_map[category]['name']
-            datatype = self.tag_browser_datatype[category]
-            formatter = self.tag_browser_formatters.get(datatype, lambda x: x)
+
+            datatype = self.tag_browser_categories[category]['type']
+            if datatype == 'rating':
+                item_zero_func = (lambda x: len(formatter(r[1])) > 0)
+                formatter = (lambda x:u'\u2605'*int(round(x/2.)))
+            elif category == 'authors':
+                item_zero_func = (lambda x: x[2] > 0)
+                # Clean up the authors strings to human-readable form
+                formatter = (lambda x: x.replace('|', ','))
+            else:
+                item_zero_func = (lambda x: x[2] > 0)
+                formatter = (lambda x:x)
+
             categories[category] = [Tag(formatter(r[1]), count=r[2], id=r[0],
                                         icon=icon, tooltip = tooltip)
-                                    for r in data
-                                        if r[2] > 0 and
-                                          (datatype != 'rating' or len(formatter(r[1])) > 0)]
-        categories['format'] = []
+                                    for r in data if item_zero_func(r)]
+
+        # We delayed computing the standard formats category because it does not
+        # use a view, but is computed dynamically
+        categories['formats'] = []
+        icon = None
+        if icon_map and 'formats' in icon_map:
+                icon = icon_map['formats']
         for fmt in self.conn.get('SELECT DISTINCT format FROM data'):
             fmt = fmt[0]
             if ids is not None:
@@ -693,13 +734,70 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
                                        WHERE format="%s"'''%fmt,
                                        all=False)
             if count > 0:
-                categories['format'].append(Tag(fmt, count=count))
+                categories['formats'].append(Tag(fmt, count=count, icon=icon))
 
         if sort_on_count:
-            categories['format'].sort(cmp=lambda x,y:cmp(x.count, y.count),
+            categories['formats'].sort(cmp=lambda x,y:cmp(x.count, y.count),
                     reverse=True)
         else:
-            categories['format'].sort(cmp=lambda x,y:cmp(x.name, y.name))
+            categories['formats'].sort(cmp=lambda x,y:cmp(x.name, y.name))
+
+        #### Now do the user-defined categories. ####
+        user_categories = prefs['user_categories']
+
+        # remove all user categories from tag_browser_categories. They can
+        # easily come and go. We will add all the existing ones in below.
+        for k in self.tag_browser_categories.keys():
+            if self.tag_browser_categories[k]['kind'] in ['user', 'search']:
+                del self.tag_browser_categories[k]
+
+        # We want to use same node in the user category as in the source
+        # category. To do that, we need to find the original Tag node. There is
+        # a time/space tradeoff here. By converting the tags into a map, we can
+        # do the verification in the category loop much faster, at the cost of
+        # temporarily duplicating the categories lists.
+        taglist = {}
+        for c in categories.keys():
+            taglist[c] = dict(map(lambda t:(t.name, t), categories[c]))
+
+        for user_cat in sorted(user_categories.keys()):
+            items = []
+            for (name,label,ign) in user_categories[user_cat]:
+                if label in taglist and name in taglist[label]:
+                    items.append(taglist[label][name])
+                # else: do nothing, to not include nodes w zero counts
+            if len(items):
+                cat_name = user_cat+'*' # add the * to avoid name collision
+                self.tag_browser_categories[cat_name] = {
+                                'table':None,  'column':None,
+                                'type':None,   'is_multiple':False,
+                                'kind':'user', 'name':user_cat}
+                # Not a problem if we accumulate entries in the icon map
+                if icon_map is not None:
+                    icon_map[cat_name] = icon_map['*user']
+                if sort_on_count:
+                    categories[cat_name] = \
+                        sorted(items, cmp=(lambda x, y: cmp(y.count, x.count)))
+                else:
+                    categories[cat_name] = \
+                        sorted(items, cmp=(lambda x, y: cmp(x.name.lower(), y.name.lower())))
+
+        #### Finally, the saved searches category ####
+        items = []
+        icon = None
+        if icon_map and 'search' in icon_map:
+                icon = icon_map['search']
+        for srch in saved_searches.names():
+            items.append(Tag(srch, tooltip=saved_searches.lookup(srch), icon=icon))
+        if len(items):
+            self.tag_browser_categories['search'] = {
+                            'table':None,    'column':None,
+                            'type':None,     'is_multiple':False,
+                            'kind':'search', 'name':_('Searches')}
+            if icon_map is not None:
+                icon_map['search'] = icon_map['search']
+            categories['search'] = items
+
         return categories
 
     def tags_older_than(self, tag, delta):

src/calibre/library/server/base.py

@@ -14,14 +14,46 @@ import cherrypy
 from calibre.constants import __appname__, __version__
 from calibre.utils.date import fromtimestamp
 from calibre.library.server import listen_on, log_access_file, log_error_file
+from calibre.library.server.utils import expose
 from calibre.utils.mdns import publish as publish_zeroconf, \
             stop_server as stop_zeroconf, get_external_ip
 from calibre.library.server.content import ContentServer
 from calibre.library.server.mobile import MobileServer
 from calibre.library.server.xml import XMLServer
 from calibre.library.server.opds import OPDSServer
+from calibre.library.server.cache import Cache
 
-class LibraryServer(ContentServer, MobileServer, XMLServer, OPDSServer):
+
+class DispatchController(object): # {{{
+
+    def __init__(self):
+        self.dispatcher = cherrypy.dispatch.RoutesDispatcher()
+        self.funcs = []
+        self.seen = set([])
+
+    def __call__(self, name, route, func, **kwargs):
+        if name in self.seen:
+            raise NameError('Route name: '+ repr(name) + ' already used')
+        self.seen.add(name)
+        kwargs['action'] = 'f_%d'%len(self.funcs)
+        self.dispatcher.connect(name, route, self, **kwargs)
+        self.funcs.append(expose(func))
+
+    def __getattr__(self, attr):
+        if not attr.startswith('f_'):
+            raise AttributeError(attr + ' not found')
+        num = attr.rpartition('_')[-1]
+        try:
+            num = int(num)
+        except:
+            raise AttributeError(attr + ' not found')
+        if num < 0 or num >= len(self.funcs):
+            raise AttributeError(attr + ' not found')
+        return self.funcs[num]
+
+# }}}
+
+class LibraryServer(ContentServer, MobileServer, XMLServer, OPDSServer, Cache):
 
     server_name = __appname__ + '/' + __version__
 
@@ -88,8 +120,16 @@ class LibraryServer(ContentServer, MobileServer, XMLServer, OPDSServer):
 
     def start(self):
         self.is_running = False
+        d = DispatchController()
+        for x in self.__class__.__bases__:
+            if hasattr(x, 'add_routes'):
+                x.add_routes(self, d)
+        root_conf = self.config.get('/', {})
+        root_conf['request.dispatch'] = d.dispatcher
+        self.config['/'] = root_conf
+
         self.setup_loggers()
-        cherrypy.tree.mount(self, '', config=self.config)
+        cherrypy.tree.mount(root=None, config=self.config)
         try:
             try:
                 cherrypy.engine.start()

src/calibre/library/server/cache.py

@@ -0,0 +1,33 @@
+#!/usr/bin/env python
+# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
+
+__license__   = 'GPL v3'
+__copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+from calibre.utils.date import utcnow
+from calibre.utils.ordered_dict import OrderedDict
+
+class Cache(object):
+
+    def add_routes(self, c):
+        self._category_cache = OrderedDict()
+        self._search_cache = OrderedDict()
+
+    def search_cache(self, search):
+        old = self._search_cache.get(search, None)
+        if old is None or old[0] <= self.db.last_modified():
+            matches = self.db.data.search(search)
+            self._search_cache[search] = frozenset(matches)
+            if len(self._search_cache) > 10:
+                self._search_cache.popitem(last=False)
+
+
+    def categories_cache(self, restrict_to=frozenset([])):
+        old = self._category_cache.get(frozenset(restrict_to), None)
+        if old is None or old[0] <= self.db.last_modified():
+            categories = self.db.get_categories(ids=restrict_to)
+            self._category_cache[restrict_to] = (utcnow(), categories)
+            if len(self._category_cache) > 10:
+                self._category_cache.popitem(last=False)
+        return self._category_cache[restrict_to][1]

src/calibre/library/server/content.py

@@ -16,7 +16,7 @@ except ImportError:
 
 from calibre import fit_image, guess_type
 from calibre.utils.date import fromtimestamp
-from calibre.library.server.utils import expose
+
 
 class ContentServer(object):
 
@@ -25,6 +25,13 @@ class ContentServer(object):
     a few utility methods.
     '''
 
+    def add_routes(self, connect):
+        connect('root', '/', self.index)
+        connect('get', '/get/{what}/{id}', self.get,
+                conditions=dict(method=["GET", "HEAD"]))
+        connect('static', '/static/{name}', self.static,
+                conditions=dict(method=["GET", "HEAD"]))
+
     # Utility methods {{{
     def last_modified(self, updated):
         '''
@@ -68,8 +75,7 @@ class ContentServer(object):
     # }}}
 
 
-    @expose
-    def get(self, what, id, *args, **kwargs):
+    def get(self, what, id):
         'Serves files, covers, thumbnails from the calibre database'
         try:
             id = int(id)
@@ -87,7 +93,6 @@ class ContentServer(object):
             return self.get_cover(id)
         return self.get_format(id, what)
 
-    @expose
     def static(self, name):
         'Serves static content'
         name = name.lower()
@@ -108,7 +113,6 @@ class ContentServer(object):
             cherrypy.response.headers['Last-Modified'] = self.last_modified(lm)
         return open(path, 'rb').read()
 
-    @expose
     def index(self, **kwargs):
         'The / URL'
         ua = cherrypy.request.headers.get('User-Agent', '').strip()

src/calibre/library/server/mobile.py

@@ -5,34 +5,143 @@ __license__   = 'GPL v3'
 __copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-import re, copy
+import re
 import __builtin__
 
 import cherrypy
+from lxml import html
+from lxml.html.builder import HTML, HEAD, TITLE, STYLE, LINK, DIV, IMG, BODY, \
+        OPTION, SELECT, INPUT, FORM, SPAN, TABLE, TR, TD, A, HR
 
-from calibre.utils.genshi.template import MarkupTemplate
-from calibre.library.server.utils import strftime, expose
+from calibre.library.server.utils import strftime
 from calibre.ebooks.metadata import fmt_sidx
+from calibre.constants import __appname__
+from calibre import human_readable
 
-# Templates {{{
-MOBILE_BOOK = '''\
-<tr xmlns:py="http://genshi.edgewall.org/">
-<td class="thumbnail">
-    <img type="image/jpeg" src="/get/thumb/${r[FM['id']]}" border="0"/>
-</td>
-<td>
-    <py:for each="format in r[FM['formats']].split(',')">
-        <span class="button"><a href="/get/${format}/${authors}-${r[FM['title']]}_${r[FM['id']]}.${format}">${format.lower()}</a></span>&nbsp;
-    </py:for>
-    ${r[FM['title']]}${(' ['+r[FM['series']]+'-'+r[FM['series_index']]+']') if r[FM['series']] else ''} by ${authors} - ${r[FM['size']]/1024}k - ${r[FM['publisher']] if r[FM['publisher']] else ''} ${pubdate} ${'['+r[FM['tags']]+']' if r[FM['tags']] else ''}
-</td>
-</tr>
-'''
+def CLASS(*args, **kwargs): # class is a reserved word in Python
+    kwargs['class'] = ' '.join(args)
+    return kwargs
 
-MOBILE = MarkupTemplate('''\
-<html xmlns:py="http://genshi.edgewall.org/">
-<head>
-<style>
+
+def build_search_box(num, search, sort, order): # {{{
+    div = DIV(id='search_box')
+    form = FORM('Show ', method='get', action='mobile')
+    div.append(form)
+
+    num_select = SELECT(name='num')
+    for option in (5, 10, 25, 100):
+        kwargs = {'value':str(option)}
+        if option == num:
+            kwargs['SELECTED'] = 'SELECTED'
+        num_select.append(OPTION(str(option), **kwargs))
+    num_select.tail = ' books matching '
+    form.append(num_select)
+
+    searchf = INPUT(name='search', id='s', value=search if search else '')
+    searchf.tail = ' sorted by '
+    form.append(searchf)
+
+    sort_select = SELECT(name='sort')
+    for option in ('date','author','title','rating','size','tags','series'):
+        kwargs = {'value':option}
+        if option == sort:
+            kwargs['SELECTED'] = 'SELECTED'
+        sort_select.append(OPTION(option, **kwargs))
+    form.append(sort_select)
+
+    order_select = SELECT(name='order')
+    for option in ('ascending','descending'):
+        kwargs = {'value':option}
+        if option == order:
+            kwargs['SELECTED'] = 'SELECTED'
+        order_select.append(OPTION(option, **kwargs))
+    form.append(order_select)
+
+    form.append(INPUT(id='go', type='submit', value='Search'))
+
+    return div
+    # }}}
+
+def build_navigation(start, num, total, url_base): # {{{
+    end = min((start+num-1), total)
+    tagline = SPAN('Books %d to %d of %d'%(start, end, total),
+            style='display: block; text-align: center;')
+    left_buttons = TD(CLASS('button', style='text-align:left'))
+    right_buttons = TD(CLASS('button', style='text-align:right'))
+
+    if start > 1:
+        for t,s in [('First', 1), ('Previous', max(start-(num+1),1))]:
+            left_buttons.append(A(t, href='%s;start=%d'%(url_base, s)))
+
+    if total > start + num:
+        for t,s in [('Next', start+num), ('Last', total-num+1)]:
+            right_buttons.append(A(t, href='%s;start=%d'%(url_base, s)))
+
+    buttons = TABLE(
+            TR(left_buttons, right_buttons),
+            CLASS('buttons'))
+    return DIV(tagline, buttons, CLASS('navigation'))
+
+    # }}}
+
+def build_index(books, num, search, sort, order, start, total, url_base):
+    logo = DIV(IMG(src='/static/calibre.png', alt=__appname__), id='logo')
+
+    search_box = build_search_box(num, search, sort, order)
+    navigation = build_navigation(start, num, total, url_base)
+    bookt = TABLE(id='listing')
+
+    body = BODY(
+        logo,
+        search_box,
+        navigation,
+        HR(CLASS('spacer')),
+        bookt
+    )
+
+    # Book list {{{
+    for book in books:
+        thumbnail = TD(
+                IMG(type='image/jpeg', border='0', src='/get/thumb/%s' %
+                            book['id']),
+                CLASS('thumbnail'))
+
+        data = TD()
+        last = None
+        for fmt in book['formats'].split(','):
+            s = SPAN(
+                A(
+                    fmt.lower(),
+                    href='/get/%s/%s-%s_%d.%s' % (fmt, book['authors'],
+                        book['title'], book['id'], fmt)
+                ),
+                CLASS('button'))
+            s.tail = u'\u202f' # &nbsp;
+            last = s
+            data.append(s)
+
+        series = u'[%s - %s]'%(book['series'], book['series_index']) \
+                if book['series'] else ''
+        tags = u'[%s]'%book['tags'] if book['tags'] else ''
+
+        text = u'\u202f%s %s by %s - %s - %s %s' % (book['title'], series,
+                book['authors'], book['size'], book['timestamp'], tags)
+
+        if last is None:
+            data.text = text
+        else:
+            last.tail += text
+
+        bookt.append(TR(thumbnail, data))
+    # }}}
+
+    return HTML(
+        HEAD(
+            TITLE(__appname__ + ' Library'),
+            LINK(rel='icon', href='http://calibre-ebook.com/favicon.ico',
+                type='image/x-icon'),
+            STYLE( # {{{
+                '''
 .navigation table.buttons {
     width: 100%;
 }
@@ -109,71 +218,20 @@ div.navigation {
     clear: both;
 }
 
-</style>
-<link rel="icon" href="http://calibre-ebook.com/favicon.ico" type="image/x-icon" />
-</head>
-<body>
-    <div id="logo">
-    <img src="/static/calibre.png" alt="Calibre" />
-    </div>
-    <div id="search_box">
-    <form method="get" action="/mobile">
-    Show <select name="num">
-        <py:for each="option in [5,10,25,100]">
-                <option py:if="option == num" value="${option}" SELECTED="SELECTED">${option}</option>
-                <option py:if="option != num" value="${option}">${option}</option>
-        </py:for>
-        </select>
-    books matching <input name="search" id="s" value="${search}" /> sorted by
+                ''', type='text/css') # }}}
+        ), # End head
+        body
+    ) # End html
 
-    <select name="sort">
-        <py:for each="option in ['date','author','title','rating','size','tags','series']">
-                <option py:if="option == sort" value="${option}" SELECTED="SELECTED">${option}</option>
-                <option py:if="option != sort" value="${option}">${option}</option>
-        </py:for>
-    </select>
-    <select name="order">
-        <py:for each="option in ['ascending','descending']">
-                <option py:if="option == order" value="${option}" SELECTED="SELECTED">${option}</option>
-                <option py:if="option != order" value="${option}">${option}</option>
-        </py:for>
-    </select>
-    <input id="go" type="submit" value="Search"/>
-    </form>
-    </div>
-    <div class="navigation">
-    <span style="display: block; text-align: center;">Books ${start} to ${ min((start+num-1) , total) } of ${total}</span>
-    <table class="buttons">
-    <tr>
-    <td class="button" style="text-align:left;">
-        <a py:if="start > 1" href="${url_base};start=1">First</a>
-        <a py:if="start > 1" href="${url_base};start=${max(start-(num+1),1)}">Previous</a>
-    </td>
-    <td class="button" style="text-align: right;">
-        <a py:if=" total > (start + num) " href="${url_base};start=${start+num}">Next</a>
-        <a py:if=" total > (start + num) " href="${url_base};start=${total-num+1}">Last</a>
-    </td>
-    </tr>
-    </table>
-    </div>
-    <hr class="spacer" />
-    <table id="listing">
-        <py:for each="book in books">
-            ${Markup(book)}
-        </py:for>
-    </table>
-</body>
-</html>
-''')
-
-# }}}
 
 class MobileServer(object):
     'A view optimized for browsers in mobile devices'
 
     MOBILE_UA = re.compile('(?i)(?:iPhone|Opera Mini|NetFront|webOS|Mobile|Android|imode|DoCoMo|Minimo|Blackberry|MIDP|Symbian|HD2)')
 
-    @expose
+    def add_routes(self, connect):
+        connect('mobile', '/mobile', self.mobile)
+
     def mobile(self, start='1', num='25', sort='date', search='',
                 _=None, order='descending'):
         '''
@@ -193,26 +251,31 @@ class MobileServer(object):
         except ValueError:
             raise cherrypy.HTTPError(400, 'num: %s is not an integer'%num)
         ids = self.db.data.parse(search) if search and search.strip() else self.db.data.universal_set()
-        ids = sorted(ids)
         FM = self.db.FIELD_MAP
-        items = copy.deepcopy([r for r in iter(self.db) if r[FM['id']] in ids])
+        items = [r for r in iter(self.db) if r[FM['id']] in ids]
         if sort is not None:
             self.sort(items, sort, (order.lower().strip() == 'ascending'))
 
-        book, books = MarkupTemplate(MOBILE_BOOK), []
+        books = []
         for record in items[(start-1):(start-1)+num]:
-            if record[FM['formats']] is None:
-                record[FM['formats']] = ''
-            if record[FM['size']] is None:
-                record[FM['size']] = 0
+            book = {'formats':record[FM['formats']], 'size':record[FM['size']]}
+            if not book['formats']:
+                book['formats'] = ''
+            if not book['size']:
+                book['size'] = 0
+            book['size'] = human_readable(book['size'])
+
             aus = record[FM['authors']] if record[FM['authors']] else __builtin__._('Unknown')
             authors = '|'.join([i.replace('|', ',') for i in aus.split(',')])
-            record[FM['series_index']] = \
-                    fmt_sidx(float(record[FM['series_index']]))
-            ts, pd = strftime('%Y/%m/%d %H:%M:%S', record[FM['timestamp']]), \
-                strftime('%Y/%m/%d %H:%M:%S', record[FM['pubdate']])
-            books.append(book.generate(r=record, authors=authors, timestamp=ts,
-                pubdate=pd, FM=FM).render('xml').decode('utf-8'))
+            book['authors'] = authors
+            book['series_index'] = fmt_sidx(float(record[FM['series_index']]))
+            book['series'] = record[FM['series']]
+            book['tags'] = record[FM['tags']]
+            book['title'] = record[FM['title']]
+            for x in ('timestamp', 'pubdate'):
+                book[x] = strftime('%Y/%m/%d %H:%M:%S', record[FM[x]])
+            book['id'] = record[FM['id']]
+            books.append(book)
         updated = self.db.last_modified()
 
         cherrypy.response.headers['Content-Type'] = 'text/html; charset=utf-8'
@@ -221,8 +284,8 @@ class MobileServer(object):
 
         url_base = "/mobile?search=" + search+";order="+order+";sort="+sort+";num="+str(num)
 
-        return MOBILE.generate(books=books, start=start, updated=updated,
-                search=search, sort=sort, order=order, num=num, FM=FM,
-                total=len(ids), url_base=url_base).render('html')
-
+        return html.tostring(build_index(books, num, search, sort, order,
+                             start, len(ids), url_base),
+                             encoding='utf-8', include_meta_content_type=True,
+                             pretty_print=True)
 

src/calibre/library/server/opds.py

@@ -5,296 +5,163 @@ __license__   = 'GPL v3'
 __copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-import re
-from itertools import repeat
+import hashlib, binascii
+from functools import partial
 
+from lxml import etree
+from lxml.builder import ElementMaker
 import cherrypy
 
-from calibre.utils.genshi.template import MarkupTemplate
-from calibre.library.server.utils import strftime, expose
-from calibre.ebooks.metadata import fmt_sidx, title_sort
-from calibre import guess_type, prepare_string_for_xml
+from calibre.constants import __appname__
 
-# Templates {{{
+BASE_HREFS = {
+        0 : '/stanza',
+        1 : '/opds',
+}
 
-STANZA_ENTRY=MarkupTemplate('''\
-<entry xmlns:py="http://genshi.edgewall.org/">
-    <title>${record[FM['title']]}</title>
-    <id>urn:calibre:${urn}</id>
-    <author><name>${authors}</name></author>
-    <updated>${timestamp}</updated>
-    <link type="${mimetype}" href="/get/${fmt}/${record[FM['id']]}" />
-    <link rel="x-stanza-cover-image" type="image/jpeg" href="/get/cover/${record[FM['id']]}" />
-    <link rel="x-stanza-cover-image-thumbnail" type="image/jpeg" href="/get/thumb/${record[FM['id']]}" />
-    <content type="xhtml">
-        <div xmlns="http://www.w3.org/1999/xhtml" style="text-align: center">${Markup(extra)}${record[FM['comments']]}</div>
-    </content>
-</entry>
-''')
+# Vocabulary for building OPDS feeds {{{
+E = ElementMaker(namespace='http://www.w3.org/2005/Atom',
+                 nsmap={
+                     None   : 'http://www.w3.org/2005/Atom',
+                     'dc'   : 'http://purl.org/dc/terms/',
+                     'opds' : 'http://opds-spec.org/2010/catalog',
+                     })
 
-STANZA_SUBCATALOG_ENTRY=MarkupTemplate('''\
-<entry xmlns:py="http://genshi.edgewall.org/">
-    <title>${title}</title>
-    <id>urn:calibre:${id}</id>
-    <updated>${updated.strftime('%Y-%m-%dT%H:%M:%S+00:00')}</updated>
-    <link type="application/atom+xml" href="/stanza/?${what}id=${id}" />
-    <content type="text">${count} books</content>
-</entry>
-''')
 
-STANZA = MarkupTemplate('''\
-<?xml version="1.0" encoding="utf-8"?>
-<feed xmlns="http://www.w3.org/2005/Atom" xmlns:py="http://genshi.edgewall.org/">
-    <title>calibre Library</title>
-    <id>$id</id>
-    <updated>${updated.strftime('%Y-%m-%dT%H:%M:%S+00:00')}</updated>
-    <link rel="search" title="Search" type="application/atom+xml" href="/stanza/?search={searchTerms}"/>
-    ${Markup(next_link)}
-    <author>
-    <name>calibre</name>
-    <uri>http://calibre-ebook.com</uri>
-    </author>
-    <subtitle>
-        ${subtitle}
-    </subtitle>
-    <py:for each="entry in data">
-    ${Markup(entry)}
-    </py:for>
-</feed>
-''')
+FEED    = E.feed
+TITLE   = E.title
+ID      = E.id
 
-STANZA_MAIN = MarkupTemplate('''\
-<?xml version="1.0" encoding="utf-8"?>
-<feed xmlns="http://www.w3.org/2005/Atom" xmlns:py="http://genshi.edgewall.org/">
-    <title>calibre Library</title>
-    <id>$id</id>
-    <updated>${updated.strftime('%Y-%m-%dT%H:%M:%S+00:00')}</updated>
-    <link rel="search" title="Search" type="application/atom+xml" href="/stanza/?search={searchTerms}"/>
-    <author>
-    <name>calibre</name>
-    <uri>http://calibre-ebook.com</uri>
-    </author>
-    <subtitle>
-        ${subtitle}
-    </subtitle>
-    <entry>
-    <title>By Author</title>
-    <id>urn:uuid:fc000fa0-8c23-11de-a31d-0002a5d5c51b</id>
-    <updated>${updated.strftime('%Y-%m-%dT%H:%M:%S+00:00')}</updated>
-    <link type="application/atom+xml" href="/stanza/?sortby=byauthor" />
-    <content type="text">Books sorted by Author</content>
-    </entry>
-    <entry>
-    <title>By Title</title>
-    <id>urn:uuid:1df4fe40-8c24-11de-b4c6-0002a5d5c51b</id>
-    <updated>${updated.strftime('%Y-%m-%dT%H:%M:%S+00:00')}</updated>
-    <link type="application/atom+xml" href="/stanza/?sortby=bytitle" />
-    <content type="text">Books sorted by Title</content>
-    </entry>
-    <entry>
-    <title>By Newest</title>
-    <id>urn:uuid:3c6d4940-8c24-11de-a4d7-0002a5d5c51b</id>
-    <updated>${updated.strftime('%Y-%m-%dT%H:%M:%S+00:00')}</updated>
-    <link type="application/atom+xml" href="/stanza/?sortby=bynewest" />
-    <content type="text">Books sorted by Date</content>
-    </entry>
-    <entry>
-    <title>By Tag</title>
-    <id>urn:uuid:824921e8-db8a-4e61-7d38-f1ce41502853</id>
-    <updated>${updated.strftime('%Y-%m-%dT%H:%M:%S+00:00')}</updated>
-    <link type="application/atom+xml" href="/stanza/?sortby=bytag" />
-    <content type="text">Books sorted by Tags</content>
-    </entry>
-    <entry>
-    <title>By Series</title>
-    <id>urn:uuid:512a5e50-a88f-f6b8-82aa-8f129c719f61</id>
-    <updated>${updated.strftime('%Y-%m-%dT%H:%M:%S+00:00')}</updated>
-    <link type="application/atom+xml" href="/stanza/?sortby=byseries" />
-    <content type="text">Books sorted by Series</content>
-    </entry>
-</feed>
-''')
+def UPDATED(dt, *args, **kwargs):
+    return E.updated(dt.strftime('%Y-%m-%dT%H:%M:%S+00:00'), *args, **kwargs)
+
+LINK = partial(E.link, type='application/atom+xml')
+NAVLINK = partial(E.link,
+        type='application/atom+xml;type=feed;profile=opds-catalog')
+
+def SEARCH(base_href, *args, **kwargs):
+    kwargs['rel'] = 'search'
+    kwargs['title'] = 'Search'
+    kwargs['href'] = base_href+'/search/{searchTerms}'
+    return LINK(*args, **kwargs)
+
+def AUTHOR(name, uri=None):
+    args = [E.name(name)]
+    if uri is not None:
+        args.append(E.uri(uri))
+    return E.author(*args)
+
+SUBTITLE = E.subtitle
+
+def NAVCATALOG_ENTRY(base_href, updated, title, description, query):
+    href = base_href+'/navcatalog/'+binascii.hexlify(query)
+    id_ = 'calibre-navcatalog:'+str(hashlib.sha1(href).hexdigest())
+    return E.entry(
+        TITLE(title),
+        ID(id_),
+        UPDATED(updated),
+        E.content(description, type='text'),
+        NAVLINK(href=href)
+    )
 
 # }}}
 
+class Feed(object):
+
+    def __str__(self):
+        return etree.tostring(self.root, pretty_print=True, encoding='utf-8',
+                xml_declaration=True)
+
+class TopLevel(Feed):
+
+    def __init__(self,
+            updated,  # datetime object in UTC
+            categories,
+            version,
+            id_       = 'urn:calibre:main',
+            ):
+        base_href = BASE_HREFS[version]
+        self.base_href = base_href
+        subc = partial(NAVCATALOG_ENTRY, base_href, updated)
+
+        subcatalogs = [subc(_('By ')+title,
+            _('Books sorted by ') + desc, q) for title, desc, q in
+            categories]
+
+        self.root = \
+            FEED(
+                    TITLE(__appname__ + ' ' + _('Library')),
+                    ID(id_),
+                    UPDATED(updated),
+                    SEARCH(base_href),
+                    AUTHOR(__appname__, uri='http://calibre-ebook.com'),
+                    SUBTITLE(_('Books in your library')),
+                    *subcatalogs
+                )
+
+STANZA_FORMATS = frozenset(['epub', 'pdb'])
+
 class OPDSServer(object):
 
-    def get_matches(self, location, query):
-        base = self.db.data.get_matches(location, query)
-        epub = self.db.data.get_matches('format', '=epub')
-        pdb = self.db.data.get_matches('format', '=pdb')
-        return base.intersection(epub.union(pdb))
+    def add_routes(self, connect):
+        for base in ('stanza', 'opds'):
+            version = 0 if base == 'stanza' else 1
+            base_href = BASE_HREFS[version]
+            connect(base, base_href, self.opds, version=version)
+            connect('opdsnavcatalog_'+base, base_href+'/navcatalog/{which}',
+                    self.opds_navcatalog, version=version)
+            connect('opdssearch_'+base, base_href+'/search/{terms}',
+                    self.opds_search, version=version)
 
-    def stanza_sortby_subcategory(self, updated, sortby, offset):
-        pat = re.compile(r'\(.*\)')
+    def get_opds_allowed_ids_for_version(self, version):
+        search = '' if version > 0 else ' '.join(['format:='+x for x in
+            STANZA_FORMATS])
+        self.seach_cache(search)
 
-        def clean_author(x):
-            return pat.sub('', x).strip()
+    def opds_search(self, terms=None, version=0):
+        version = int(version)
+        if not terms or version not in BASE_HREFS:
+            raise cherrypy.HTTPError(404, 'Not found')
 
-        def author_cmp(x, y):
-            x = x if ',' in x else clean_author(x).rpartition(' ')[-1]
-            y = y if ',' in y else clean_author(y).rpartition(' ')[-1]
-            return cmp(x.lower(), y.lower())
+    def opds_navcatalog(self, which=None, version=0):
+        version = int(version)
+        if not which or version not in BASE_HREFS:
+            raise cherrypy.HTTPError(404, 'Not found')
+        which = binascii.unhexlify(which)
+        type_ = which[0]
+        which = which[1:]
+        if type_ == 'O':
+            return self.get_opds_all_books(which)
+        elif type_ == 'N':
+            return self.get_opds_navcatalog(which)
+        raise cherrypy.HTTPError(404, 'Not found')
 
-        def get_author(x):
-            pref, ___, suff = clean_author(x).rpartition(' ')
-            return suff + (', '+pref) if pref else suff
-
-
-        what, subtitle = sortby[2:], ''
-        if sortby == 'byseries':
-            data = self.db.all_series()
-            data = [(x[0], x[1], len(self.get_matches('series', '='+x[1]))) for x in data]
-            subtitle = 'Books by series'
-        elif sortby == 'byauthor':
-            data = self.db.all_authors()
-            data = [(x[0], x[1], len(self.get_matches('authors', '='+x[1]))) for x in data]
-            subtitle = 'Books by author'
-        elif sortby == 'bytag':
-            data = self.db.all_tags2()
-            data = [(x[0], x[1], len(self.get_matches('tags', '='+x[1]))) for x in data]
-            subtitle = 'Books by tag'
-        fcmp = author_cmp if sortby == 'byauthor' else cmp
-        data = [x for x in data if x[2] > 0]
-        data.sort(cmp=lambda x, y: fcmp(x[1], y[1]))
-        next_offset = offset + self.max_stanza_items
-        rdata = data[offset:next_offset]
-        if next_offset >= len(data):
-            next_offset = -1
-        gt = get_author if sortby == 'byauthor' else lambda x: x
-        entries = [STANZA_SUBCATALOG_ENTRY.generate(title=gt(title), id=id,
-            what=what, updated=updated, count=c).render('xml').decode('utf-8') for id,
-            title, c in rdata]
-        next_link = ''
-        if next_offset > -1:
-            next_link = ('<link rel="next" title="Next" '
-            'type="application/atom+xml" href="/stanza/?sortby=%s&amp;offset=%d"/>\n'
-            ) % (sortby, next_offset)
-        return STANZA.generate(subtitle=subtitle, data=entries, FM=self.db.FIELD_MAP,
-                    updated=updated, id='urn:calibre:main', next_link=next_link).render('xml')
-
-    def stanza_main(self, updated):
-        return STANZA_MAIN.generate(subtitle='', data=[], FM=self.db.FIELD_MAP,
-                    updated=updated, id='urn:calibre:main').render('xml')
-
-    @expose
-    def stanza(self, search=None, sortby=None, authorid=None, tagid=None,
-            seriesid=None, offset=0):
-        'Feeds to read calibre books on a ipod with stanza.'
-        books = []
+    def opds(self, version=0):
+        version = int(version)
+        if version not in BASE_HREFS:
+            raise cherrypy.HTTPError(404, 'Not found')
+        categories = self.categories_cache(
+                self.get_opds_allowed_ids_for_version(version))
+        category_meta = self.db.get_tag_browser_categories()
+        cats = [
+                (_('Newest'), _('Date'), 'Onewest'),
+                (_('Title'), _('Title'), 'Otitle'),
+                ]
+        for category in categories:
+            if category == 'formats':
+                continue
+            meta = category_meta.get(category, None)
+            if meta is None:
+                continue
+            cats.append((meta['name'], meta['name'], 'N'+category))
         updated = self.db.last_modified()
-        offset = int(offset)
+
         cherrypy.response.headers['Last-Modified'] = self.last_modified(updated)
         cherrypy.response.headers['Content-Type'] = 'text/xml'
-        # Main feed
-        if not sortby and not search and not authorid and not tagid and not seriesid:
-            return self.stanza_main(updated)
-        if sortby in ('byseries', 'byauthor', 'bytag'):
-            return self.stanza_sortby_subcategory(updated, sortby, offset)
 
-        # Get matching ids
-        if authorid:
-            authorid=int(authorid)
-            au = self.db.author_name(authorid)
-            ids = self.get_matches('authors', au)
-        elif tagid:
-            tagid=int(tagid)
-            ta = self.db.tag_name(tagid)
-            ids = self.get_matches('tags', ta)
-        elif seriesid:
-            seriesid=int(seriesid)
-            se = self.db.series_name(seriesid)
-            ids = self.get_matches('series', se)
-        else:
-            ids = self.db.data.parse(search) if search and search.strip() else self.db.data.universal_set()
-        record_list = list(iter(self.db))
-
-        FM = self.db.FIELD_MAP
-        # Sort the record list
-        if sortby == "bytitle" or authorid or tagid:
-            record_list.sort(lambda x, y:
-                    cmp(title_sort(x[FM['title']]),
-                        title_sort(y[FM['title']])))
-        elif seriesid:
-            record_list.sort(lambda x, y:
-                    cmp(x[FM['series_index']],
-                        y[FM['series_index']]))
-        else: # Sort by date
-            record_list = reversed(record_list)
-
-
-        fmts = FM['formats']
-        pat = re.compile(r'EPUB|PDB', re.IGNORECASE)
-        record_list = [x for x in record_list if x[FM['id']] in ids and
-                pat.search(x[fmts] if x[fmts] else '') is not None]
-        next_offset = offset + self.max_stanza_items
-        nrecord_list = record_list[offset:next_offset]
-        if next_offset >= len(record_list):
-            next_offset = -1
-
-        next_link = ''
-        if next_offset > -1:
-            q = ['offset=%d'%next_offset]
-            for x in ('search', 'sortby', 'authorid', 'tagid', 'seriesid'):
-                val = locals()[x]
-                if val is not None:
-                    val = prepare_string_for_xml(unicode(val), True)
-                    q.append('%s=%s'%(x, val))
-            next_link = ('<link rel="next" title="Next" '
-            'type="application/atom+xml" href="/stanza/?%s"/>\n'
-            ) % '&amp;'.join(q)
-
-        for record in nrecord_list:
-            r = record[FM['formats']]
-            r = r.upper() if r else ''
-
-            z = record[FM['authors']]
-            if not z:
-                z = _('Unknown')
-            authors = ' & '.join([i.replace('|', ',') for i in
-                                    z.split(',')])
-
-            # Setup extra description
-            extra = []
-            rating = record[FM['rating']]
-            if rating > 0:
-                rating = ''.join(repeat('&#9733;', rating))
-                extra.append('RATING: %s<br />'%rating)
-            tags = record[FM['tags']]
-            if tags:
-                extra.append('TAGS: %s<br />'%\
-                        prepare_string_for_xml(', '.join(tags.split(','))))
-            series = record[FM['series']]
-            if series:
-                extra.append('SERIES: %s [%s]<br />'%\
-                        (prepare_string_for_xml(series),
-                        fmt_sidx(float(record[FM['series_index']]))))
-
-            fmt = 'epub' if 'EPUB' in r else 'pdb'
-            mimetype = guess_type('dummy.'+fmt)[0]
-
-            # Create the sub-catalog, which is either a list of
-            # authors/tags/series or a list of books
-            data = dict(
-                    record=record,
-                    updated=updated,
-                    authors=authors,
-                    tags=tags,
-                    series=series,
-                    FM=FM,
-                    extra='\n'.join(extra),
-                    mimetype=mimetype,
-                    fmt=fmt,
-                    urn=record[FM['uuid']],
-                    timestamp=strftime('%Y-%m-%dT%H:%M:%S+00:00',
-                        record[FM['timestamp']])
-                    )
-            books.append(STANZA_ENTRY.generate(**data)\
-                                        .render('xml').decode('utf8'))
-
-        return STANZA.generate(subtitle='', data=books, FM=FM,
-                next_link=next_link, updated=updated, id='urn:calibre:main').render('xml')
+        feed = TopLevel(updated, cats, version)
 
+        return str(feed)
 
 
 

src/calibre/library/server/utils.py

@@ -5,20 +5,34 @@ __license__   = 'GPL v3'
 __copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-from calibre import strftime as _strftime
+import time
+
+import cherrypy
+
+from calibre import strftime as _strftime, prints
 from calibre.utils.date import now as nowf
 
 
 def expose(func):
-    import cherrypy
 
-    def do(self, *args, **kwargs):
+    def do(*args, **kwargs):
+        self = func.im_self
+        if self.opts.develop:
+            start = time.time()
+
         dict.update(cherrypy.response.headers, {'Server':self.server_name})
         if not self.embedded:
             self.db.check_if_modified()
-        return func(self, *args, **kwargs)
+        ans = func(*args, **kwargs)
+        if self.opts.develop:
+            prints('Function', func.__name__, 'called with args:', args, kwargs)
+            prints('\tTime:', func.__name__, time.time()-start)
+        return ans
+
+    do.__name__ = func.__name__
+
+    return do
 
-    return cherrypy.expose(do)
 
 def strftime(fmt='%Y/%m/%d %H:%M:%S', dt=None):
     if not hasattr(dt, 'timetuple'):

src/calibre/library/server/xml.py

@@ -5,52 +5,26 @@ __license__   = 'GPL v3'
 __copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-import copy, __builtin__
+import __builtin__
 
 import cherrypy
+from lxml.builder import ElementMaker
+from lxml import etree
 
-from calibre.utils.genshi.template import MarkupTemplate
-from calibre.library.server.utils import strftime, expose
+from calibre.library.server.utils import strftime
 from calibre.ebooks.metadata import fmt_sidx
+from calibre.constants import preferred_encoding
+from calibre import isbytestring
 
-# Templates {{{
-BOOK = '''\
-<book xmlns:py="http://genshi.edgewall.org/"
-    id="${r[FM['id']]}"
-    title="${r[FM['title']]}"
-    sort="${r[FM['sort']]}"
-    author_sort="${r[FM['author_sort']]}"
-    authors="${authors}"
-    rating="${r[FM['rating']]}"
-    timestamp="${timestamp}"
-    pubdate="${pubdate}"
-    size="${r[FM['size']]}"
-    isbn="${r[FM['isbn']] if r[FM['isbn']] else ''}"
-    formats="${r[FM['formats']] if r[FM['formats']] else ''}"
-    series = "${r[FM['series']] if r[FM['series']] else ''}"
-    series_index="${r[FM['series_index']]}"
-    tags="${r[FM['tags']] if r[FM['tags']] else ''}"
-    publisher="${r[FM['publisher']] if r[FM['publisher']] else ''}">${r[FM['comments']] if r[FM['comments']] else ''}
-</book>
-'''
-
-
-LIBRARY = MarkupTemplate('''\
-<?xml version="1.0" encoding="utf-8"?>
-<library xmlns:py="http://genshi.edgewall.org/" start="$start" num="${len(books)}" total="$total" updated="${updated.strftime('%Y-%m-%dT%H:%M:%S+00:00')}">
-<py:for each="book in books">
-    ${Markup(book)}
-</py:for>
-</library>
-''')
-
-# }}}
+E = ElementMaker()
 
 class XMLServer(object):
     'Serves XML and the Ajax based HTML frontend'
 
-    @expose
-    def library(self, start='0', num='50', sort=None, search=None,
+    def add_routes(self, connect):
+        connect('xml', '/xml', self.xml)
+
+    def xml(self, start='0', num='50', sort=None, search=None,
                 _=None, order='ascending'):
         '''
         Serves metadata from the calibre database as XML.
@@ -68,30 +42,63 @@ class XMLServer(object):
             num = int(num)
         except ValueError:
             raise cherrypy.HTTPError(400, 'num: %s is not an integer'%num)
+
         order = order.lower().strip() == 'ascending'
+
         ids = self.db.data.parse(search) if search and search.strip() else self.db.data.universal_set()
-        ids = sorted(ids)
+
         FM = self.db.FIELD_MAP
-        items = copy.deepcopy([r for r in iter(self.db) if r[FM['id']] in ids])
+
+        items = [r for r in iter(self.db) if r[FM['id']] in ids]
         if sort is not None:
             self.sort(items, sort, order)
 
-        book, books = MarkupTemplate(BOOK), []
+
+        books = []
+
+        def serialize(x):
+            if isinstance(x, unicode):
+                return x
+            if isbytestring(x):
+                return x.decode(preferred_encoding, 'replace')
+            return unicode(x)
+
         for record in items[start:start+num]:
+            kwargs = {}
             aus = record[FM['authors']] if record[FM['authors']] else __builtin__._('Unknown')
             authors = '|'.join([i.replace('|', ',') for i in aus.split(',')])
-            record[FM['series_index']] = \
+            kwargs['authors'] = authors
+
+            kwargs['series_index'] = \
                 fmt_sidx(float(record[FM['series_index']]))
-            ts, pd = strftime('%Y/%m/%d %H:%M:%S', record[FM['timestamp']]), \
-                strftime('%Y/%m/%d %H:%M:%S', record[FM['pubdate']])
-            books.append(book.generate(r=record, authors=authors, timestamp=ts,
-                pubdate=pd, FM=FM).render('xml').decode('utf-8'))
+
+            for x in ('timestamp', 'pubdate'):
+                kwargs[x] = strftime('%Y/%m/%d %H:%M:%S', record[FM[x]])
+
+            for x in ('id', 'title', 'sort', 'author_sort', 'rating', 'size'):
+                kwargs[x] = serialize(record[FM[x]])
+
+            for x in ('isbn', 'formats', 'series', 'tags', 'publisher',
+                    'comments'):
+                y = record[FM[x]]
+                kwargs[x] = serialize(y) if y else ''
+
+            c = kwargs.pop('comments')
+            books.append(E.book(c, **kwargs))
+
         updated = self.db.last_modified()
+        kwargs = dict(
+                start = str(start),
+                updated=updated.strftime('%Y-%m-%dT%H:%M:%S+00:00'),
+                total=str(len(ids)),
+                num=str(len(books)))
+        ans = E.library(*books, **kwargs)
 
         cherrypy.response.headers['Content-Type'] = 'text/xml'
         cherrypy.response.headers['Last-Modified'] = self.last_modified(updated)
-        return LIBRARY.generate(books=books, start=start, updated=updated,
-                                     total=len(ids), FM=FM).render('xml')
+
+        return etree.tostring(ans, encoding='utf-8', pretty_print=True,
+                xml_declaration=True)
 
 
 

src/calibre/linux.py

@@ -132,7 +132,6 @@ class PostInstall:
         self.mime_resources = []
         if islinux:
             self.setup_completion()
-            self.setup_udev_rules()
         self.install_man_pages()
         if islinux:
             self.setup_desktop_integration()
@@ -286,40 +285,6 @@ class PostInstall:
                 raise
             self.task_failed('Setting up completion failed')
 
-    def setup_udev_rules(self):
-        self.info('Trying to setup udev rules...')
-        try:
-            group_file = os.path.join(self.opts.staging_etc, 'group')
-            if not os.path.exists(group_file):
-                group_file = '/etc/group'
-            groups = open(group_file, 'rb').read()
-            group = 'plugdev' if 'plugdev' in groups else 'usb'
-            old_udev = '/etc/udev/rules.d/95-calibre.rules'
-            if not os.path.exists(old_udev):
-                old_udev = os.path.join(self.opts.staging_etc, 'udev/rules.d/95-calibre.rules')
-            if os.path.exists(old_udev):
-                try:
-                    os.remove(old_udev)
-                except:
-                    self.warn('Old udev rules found, please delete manually:',
-                            old_udev)
-            if self.opts.staging_root == '/usr':
-                base = '/lib'
-            else:
-                base = os.path.join(self.opts.staging_root, 'lib')
-            base = os.path.join(base, 'udev', 'rules.d')
-            if not os.path.exists(base):
-                os.makedirs(base)
-            with open(os.path.join(base, '95-calibre.rules'), 'wb') as udev:
-                self.manifest.append(udev.name)
-                udev.write('''# Sony Reader PRS-500\n'''
-                        '''SUBSYSTEMS=="usb", SYSFS{idProduct}=="029b", SYSFS{idVendor}=="054c", MODE="660", GROUP="%s"\n'''%(group,)
-                        )
-        except:
-            if self.opts.fatal_errors:
-                raise
-            self.task_failed('Setting up udev rules failed')
-
     def install_man_pages(self):
         try:
             from calibre.utils.help2man import create_man_page

src/calibre/utils/config.py

@@ -694,8 +694,10 @@ def _prefs():
             help=_('Add new formats to existing book records'))
     c.add_opt('installation_uuid', default=None, help='Installation UUID')
 
-    # this is here instead of the gui preferences because calibredb can execute searches
+    # these are here instead of the gui preferences because calibredb and
+    # calibre server can execute searches
     c.add_opt('saved_searches', default={}, help=_('List of named saved searches'))
+    c.add_opt('user_categories', default={}, help=_('User-created tag browser categories'))
 
     c.add_opt('migrated', default=False, help='For Internal use. Don\'t modify.')
     return c

src/calibre/utils/ordered_dict.py

@@ -0,0 +1,114 @@
+#!/usr/bin/env  python
+
+__license__   = 'GPL v3'
+__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
+
+'''
+A ordered dictionary. Use the builtin type on python >= 2.7
+'''
+
+
+try:
+    from collections import OrderedDict
+    OrderedDict
+except ImportError:
+    from UserDict import DictMixin
+
+    class OrderedDict(dict, DictMixin):
+
+        def __init__(self, *args, **kwds):
+            if len(args) > 1:
+                raise TypeError('expected at most 1 arguments, got %d' % len(args))
+            try:
+                self.__end
+            except AttributeError:
+                self.clear()
+            self.update(*args, **kwds)
+
+        def clear(self):
+            self.__end = end = []
+            end += [None, end, end]         # sentinel node for doubly linked list
+            self.__map = {}                 # key --> [key, prev, next]
+            dict.clear(self)
+
+        def __setitem__(self, key, value):
+            if key not in self:
+                end = self.__end
+                curr = end[1]
+                curr[2] = end[1] = self.__map[key] = [key, curr, end]
+            dict.__setitem__(self, key, value)
+
+        def __delitem__(self, key):
+            dict.__delitem__(self, key)
+            key, prev, next = self.__map.pop(key)
+            prev[2] = next
+            next[1] = prev
+
+        def __iter__(self):
+            end = self.__end
+            curr = end[2]
+            while curr is not end:
+                yield curr[0]
+                curr = curr[2]
+
+        def __reversed__(self):
+            end = self.__end
+            curr = end[1]
+            while curr is not end:
+                yield curr[0]
+                curr = curr[1]
+
+        def popitem(self, last=True):
+            if not self:
+                raise KeyError('dictionary is empty')
+            if last:
+                key = reversed(self).next()
+            else:
+                key = iter(self).next()
+            value = self.pop(key)
+            return key, value
+
+        def __reduce__(self):
+            items = [[k, self[k]] for k in self]
+            tmp = self.__map, self.__end
+            del self.__map, self.__end
+            inst_dict = vars(self).copy()
+            self.__map, self.__end = tmp
+            if inst_dict:
+                return (self.__class__, (items,), inst_dict)
+            return self.__class__, (items,)
+
+        def keys(self):
+            return list(self)
+
+        setdefault = DictMixin.setdefault
+        update = DictMixin.update
+        pop = DictMixin.pop
+        values = DictMixin.values
+        items = DictMixin.items
+        iterkeys = DictMixin.iterkeys
+        itervalues = DictMixin.itervalues
+        iteritems = DictMixin.iteritems
+
+        def __repr__(self):
+            if not self:
+                return '%s()' % (self.__class__.__name__,)
+            return '%s(%r)' % (self.__class__.__name__, self.items())
+
+        def copy(self):
+            return self.__class__(self)
+
+        @classmethod
+        def fromkeys(cls, iterable, value=None):
+            d = cls()
+            for key in iterable:
+                d[key] = value
+            return d
+
+        def __eq__(self, other):
+            if isinstance(other, OrderedDict):
+                return len(self)==len(other) and self.items() == other.items()
+            return dict.__eq__(self, other)
+
+        def __ne__(self, other):
+            return not self == other

src/calibre/web/feeds/templates.py

@@ -2,207 +2,193 @@
 
 __license__   = 'GPL v3'
 __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
-from calibre.utils.genshi.template import MarkupTemplate
-from calibre import preferred_encoding, strftime
 
+from lxml import html, etree
+from lxml.html.builder import HTML, HEAD, TITLE, STYLE, DIV, BODY, \
+        STRONG, BR, H1, SPAN, A, HR, UL, LI, H2, IMG, P as PT
 
-class Template(MarkupTemplate):
+from calibre import preferred_encoding, strftime, isbytestring
+
+def CLASS(*args, **kwargs): # class is a reserved word in Python
+    kwargs['class'] = ' '.join(args)
+    return kwargs
+
+class Template(object):
+
+    IS_HTML = True
 
     def generate(self, *args, **kwargs):
         if not kwargs.has_key('style'):
             kwargs['style'] = ''
         for key in kwargs.keys():
-            if isinstance(kwargs[key], basestring) and not isinstance(kwargs[key], unicode):
-                kwargs[key] = unicode(kwargs[key], 'utf-8', 'replace')
-        for arg in args:
-            if isinstance(arg, basestring) and not isinstance(arg, unicode):
-                arg = unicode(arg, 'utf-8', 'replace')
+            if isbytestring(kwargs[key]):
+                kwargs[key] = kwargs[key].decode('utf-8', 'replace')
+            if kwargs[key] is None:
+                kwargs[key] = u''
+        args = list(args)
+        for i in range(len(args)):
+            if isbytestring(args[i]):
+                args[i] = args[i].decode('utf-8', 'replace')
+            if args[i] is None:
+                args[i] = u''
 
-        return MarkupTemplate.generate(self, *args, **kwargs)
+        self._generate(*args, **kwargs)
+
+        return self
+
+    def render(self, *args, **kwargs):
+        if self.IS_HTML:
+            return html.tostring(self.root, encoding='utf-8',
+                    include_meta_content_type=True, pretty_print=True)
+        return etree.tostring(self.root, encoding='utf-8', xml_declaration=True,
+                pretty_print=True)
 
 class NavBarTemplate(Template):
 
-    def __init__(self):
-        Template.__init__(self, u'''\
-<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
-    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
-<html xmlns="http://www.w3.org/1999/xhtml"
-      xml:lang="en"
-      xmlns:xi="http://www.w3.org/2001/XInclude"
-      xmlns:py="http://genshi.edgewall.org/"
-
->
-    <head>
-        <style py:if="extra_css" type="text/css">
-        ${extra_css}
-        </style>
-    </head>
-    <body>
-        <div class="calibre_navbar calibre_rescale_70" style="text-align:${'center' if center else 'left'};">
-            <hr py:if="bottom" />
-            <p py:if="bottom" style="text-align:left">
-                This article was downloaded by <b>${__appname__}</b> from <a href="${url}">${url}</a>
-            </p>
-            <br py:if="bottom" /><br py:if="bottom" />
-            <py:if test="art != num - 1 and not bottom">
-            | <a href="${prefix}../article_${str(art+1)}/index.html">Next</a>
-            </py:if>
-            <py:if test="art == num - 1 and not bottom">
-            | <a href="${prefix}../../feed_${str(feed+1)}/index.html">Next</a>
-            </py:if>
-            | <a href="${prefix}../index.html#article_${str(art)}">Section menu</a>
-            <py:if test="two_levels">
-            | <a href="${prefix}../../index.html#feed_${str(feed)}">Main menu</a>
-            </py:if>
-            <py:if test="art != 0 and not bottom">
-            | <a href="${prefix}../article_${str(art-1)}/index.html">Previous</a>
-            </py:if>
-            |
-            <hr py:if="not bottom" />
-        </div>
-    </body>
-</html>
-''')
-
-    def generate(self, bottom, feed, art, number_of_articles_in_feed,
+    def _generate(self, bottom, feed, art, number_of_articles_in_feed,
                  two_levels, url, __appname__, prefix='', center=True,
-                 extra_css=None):
+                 extra_css=None, style=None):
+        head = HEAD(TITLE('navbar'))
+        if style:
+            head.append(STYLE(style, type='text/css'))
+        if extra_css:
+            head.append(STYLE(extra_css, type='text/css'))
+
         if prefix and not prefix.endswith('/'):
             prefix += '/'
-        return Template.generate(self, bottom=bottom, art=art, feed=feed,
-                                 num=number_of_articles_in_feed,
-                                 two_levels=two_levels, url=url,
-                                 __appname__=__appname__, prefix=prefix,
-                                 center=center, extra_css=extra_css)
+        align = 'center' if center else 'left'
+        navbar = DIV(CLASS('calibre_navbar', 'calibre_rescale_70',
+            style='text-align:'+align))
+        if bottom:
+            navbar.append(HR())
+            text = 'This article was downloaded by '
+            p = PT(text, STRONG(__appname__), A(url, href=url), style='text-align:left')
+            p[0].tail = ' from '
+            navbar.append(BR())
+            navbar.append(BR())
+        else:
+            next = 'feed_%d'%(feed+1) if art == number_of_articles_in_feed - 1 \
+                    else 'article_%d'%(art+1)
+            up = '../..' if art == number_of_articles_in_feed - 1 else '..'
+            href = '%s%s/%s/index.html'%(prefix, up, next)
+            navbar.text = '| '
+            navbar.append(A('Next', href=href))
+        href = '%s../index.html#article_%d'%(prefix, art)
+        navbar.iterchildren(reversed=True).next().tail = ' | '
+        navbar.append(A('Section Menu', href=href))
+        href = '%s../../index.html#feed_%d'%(prefix, feed)
+        navbar.iterchildren(reversed=True).next().tail = ' | '
+        navbar.append(A('Main Menu', href=href))
+        if art > 0 and not bottom:
+            href = '%s../article_%d/index.html'%(prefix, art-1)
+            navbar.iterchildren(reversed=True).next().tail = ' | '
+            navbar.append(A('Previous', href=href))
+        navbar.iterchildren(reversed=True).next().tail = ' | '
+        if not bottom:
+            navbar.append(HR())
+
+        self.root = HTML(head, BODY(navbar))
+
+
 
 
 class IndexTemplate(Template):
 
-    def __init__(self):
-        Template.__init__(self, u'''\
-<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
-    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
-<html xmlns="http://www.w3.org/1999/xhtml"
-      xml:lang="en"
-      xmlns:xi="http://www.w3.org/2001/XInclude"
-      xmlns:py="http://genshi.edgewall.org/"
-
->
-    <head>
-        <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
-        <title>${title}</title>
-        <style py:if="style" type="text/css">
-            ${style}
-        </style>
-        <style py:if="extra_css" type="text/css">
-            ${extra_css}
-        </style>
-    </head>
-    <body>
-        <div class="calibre_rescale_100">
-            <h1 class="calibre_recipe_title calibre_rescale_180">${title}</h1>
-            <p style="text-align:right">${date}</p>
-            <ul class="calibre_feed_list">
-                <py:for each="i, feed in enumerate(feeds)">
-                <li py:if="feed" id="feed_${str(i)}">
-                    <a class="feed calibre_rescale_120" href="${'feed_%d/index.html'%i}">${feed.title}</a>
-                </li>
-                </py:for>
-            </ul>
-        </div>
-    </body>
-</html>
-''')
-
-    def generate(self, title, datefmt, feeds, extra_css=None):
+    def _generate(self, title, datefmt, feeds, extra_css=None, style=None):
         if isinstance(datefmt, unicode):
             datefmt = datefmt.encode(preferred_encoding)
         date = strftime(datefmt)
-        return Template.generate(self, title=title, date=date, feeds=feeds,
-                                 extra_css=extra_css)
-
+        head = HEAD(TITLE(title))
+        if style:
+            head.append(STYLE(style, type='text/css'))
+        if extra_css:
+            head.append(STYLE(extra_css, type='text/css'))
+        ul = UL(CLASS('calibre_feed_list'))
+        for i, feed in enumerate(feeds):
+            if feed:
+                li = LI(A(feed.title, CLASS('feed', 'calibre_rescale_120',
+                    href='feed_%d/index.html'%i)), id='feed_%d'%i)
+                ul.append(li)
+        div = DIV(
+                H1(title, CLASS('calibre_recipe_title', 'calibre_rescale_180')),
+                PT(date, style='text-align:right'),
+                ul,
+                CLASS('calibre_rescale_100'))
+        self.root = HTML(head, BODY(div))
 
 class FeedTemplate(Template):
 
-    def __init__(self):
-        Template.__init__(self, u'''\
-<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
-    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
-<html xmlns="http://www.w3.org/1999/xhtml"
-      xml:lang="en"
-      xmlns:xi="http://www.w3.org/2001/XInclude"
-      xmlns:py="http://genshi.edgewall.org/"
+    def _generate(self, feed, cutoff, extra_css=None, style=None):
+        head = HEAD(TITLE(feed.title))
+        if style:
+            head.append(STYLE(style, type='text/css'))
+        if extra_css:
+            head.append(STYLE(extra_css, type='text/css'))
+        body = BODY(style='page-break-before:always')
+        div = DIV(
+                H2(feed.title,
+                    CLASS('calibre_feed_title', 'calibre_rescale_160')),
+                CLASS('calibre_rescale_100')
+              )
+        body.append(div)
+        if getattr(feed, 'image', None):
+            div.append(DIV(IMG(
+                alt = feed.image_alt if feed.image_alt else '',
+                src = feed.image_url
+                ),
+                CLASS('calibre_feed_image')))
+        if getattr(feed, 'description', None):
+            d = DIV(feed.description, CLASS('calibre_feed_description',
+                'calibre_rescale_80'))
+            d.append(BR())
+            div.append(d)
+        ul = UL(CLASS('calibre_article_list'))
+        for i, article in enumerate(feed.articles):
+            if not getattr(article, 'downloaded', False):
+                continue
+            li = LI(
+                    A(article.title, CLASS('article calibre_rescale_120',
+                                    href=article.url)),
+                    SPAN(article.formatted_date, CLASS('article_date')),
+                    CLASS('calibre_rescale_100', id='article_%d'%i,
+                            style='padding-bottom:0.5em')
+                    )
+            if article.summary:
+                li.append(DIV(cutoff(article.text_summary),
+                    CLASS('article_description', 'calibre_rescale_70')))
+            ul.append(li)
+        div.append(ul)
+        navbar = DIV('| ', CLASS('calibre_navbar', 'calibre_rescale_70'))
+        link = A('Up one level', href="../index.html")
+        link.tail = ' |'
+        navbar.append(link)
+        div.append(navbar)
 
->
-    <head>
-        <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
-        <title>${feed.title}</title>
-        <style py:if="style" type="text/css">
-            ${style}
-        </style>
-        <style py:if="extra_css" type="text/css">
-            ${extra_css}
-        </style>
-    </head>
-    <body style="page-break-before:always">
-    <div class="calibre_rescale_100">
-        <h2 class="calibre_feed_title calibre_rescale_160">${feed.title}</h2>
-        <py:if test="getattr(feed, 'image', None)">
-        <div class="calibre_feed_image">
-            <img alt="${feed.image_alt}" src="${feed.image_url}" />
-        </div>
-        </py:if>
-        <div class="calibre_feed_description calibre_rescale_80" py:if="getattr(feed, 'description', None)">
-            ${feed.description}<br />
-        </div>
-        <ul class="calibre_article_list">
-            <py:for each="i, article in enumerate(feed.articles)">
-            <li id="${'article_%d'%i}" py:if="getattr(article, 'downloaded',
-            False)" style="padding-bottom:0.5em" class="calibre_rescale_100">
-                <a class="article calibre_rescale_120" href="${article.url}">${article.title}</a>
-                <span class="article_date">${article.formatted_date}</span>
-                <div class="article_description calibre_rescale_70" py:if="article.summary">
-                    ${Markup(cutoff(article.text_summary))}
-                </div>
-            </li>
-            </py:for>
-        </ul>
-        <div class="calibre_navbar calibre_rescale_70">
-            | <a href="../index.html">Up one level</a> |
-        </div>
-        </div>
-    </body>
-</html>
-''')
+        self.root = HTML(head, body)
 
-    def generate(self, feed, cutoff, extra_css=None):
-        return Template.generate(self, feed=feed, cutoff=cutoff,
-                                 extra_css=extra_css)
 
 class EmbeddedContent(Template):
 
-    def __init__(self):
-        Template.__init__(self, u'''\
-<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
-    "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
-<html xmlns="http://www.w3.org/1999/xhtml"
-      xml:lang="en"
-      xmlns:xi="http://www.w3.org/2001/XInclude"
-      xmlns:py="http://genshi.edgewall.org/"
+    def _generate(self, article, style=None, extra_css=None):
+        content = article.content if article.content else ''
+        summary = article.summary if article.summary else ''
+        text = content if len(content) > len(summary) else summary
+        head = HEAD(TITLE(article.title))
+        if style:
+            head.append(STYLE(style, type='text/css'))
+        if extra_css:
+            head.append(STYLE(extra_css, type='text/css'))
 
->
-    <head>
-        <title>${article.title}</title>
-    </head>
+        if isbytestring(text):
+            text = text.decode('utf-8', 'replace')
+        elements = html.fragments_fromstring(text)
+        self.root = HTML(head,
+                BODY(H2(article.title), DIV()))
+        div = self.root.find('body').find('div')
+        if elements and isinstance(elements[0], unicode):
+            div.text = elements[0]
+            elements = list(elements)[1:]
+        for elem in elements:
+            elem.getparent().remove(elem)
+            div.append(elem)
 
-    <body>
-        <h2>${article.title}</h2>
-        <div>
-            ${Markup(article.content if len(article.content if article.content else '') > len(article.summary if article.summary else '') else article.summary)}
-        </div>
-    </body>
-</html>
-''')
-
-    def generate(self, article):
-        return Template.generate(self, article=article)

src/routes/__init__.py

@@ -0,0 +1,142 @@
+"""Provides common classes and functions most users will want access to."""
+import threading, sys
+
+class _RequestConfig(object):
+    """
+    RequestConfig thread-local singleton
+    
+    The Routes RequestConfig object is a thread-local singleton that should 
+    be initialized by the web framework that is utilizing Routes.
+    """
+    __shared_state = threading.local()
+    
+    def __getattr__(self, name):
+        return getattr(self.__shared_state, name)
+
+    def __setattr__(self, name, value):
+        """
+        If the name is environ, load the wsgi envion with load_wsgi_environ
+        and set the environ
+        """
+        if name == 'environ':
+            self.load_wsgi_environ(value)
+            return self.__shared_state.__setattr__(name, value)
+        return self.__shared_state.__setattr__(name, value)
+    
+    def __delattr__(self, name):
+        delattr(self.__shared_state, name)
+    
+    def load_wsgi_environ(self, environ):
+        """
+        Load the protocol/server info from the environ and store it.
+        Also, match the incoming URL if there's already a mapper, and
+        store the resulting match dict in mapper_dict.
+        """
+        if 'HTTPS' in environ or environ.get('wsgi.url_scheme') == 'https' \
+           or environ.get('HTTP_X_FORWARDED_PROTO') == 'https':
+            self.__shared_state.protocol = 'https'
+        else:
+            self.__shared_state.protocol = 'http'
+        try:
+            self.mapper.environ = environ
+        except AttributeError:
+            pass
+        
+        # Wrap in try/except as common case is that there is a mapper
+        # attached to self
+        try:
+            if 'PATH_INFO' in environ:
+                mapper = self.mapper
+                path = environ['PATH_INFO']
+                result = mapper.routematch(path)
+                if result is not None:
+                    self.__shared_state.mapper_dict = result[0]
+                    self.__shared_state.route = result[1]
+                else:
+                    self.__shared_state.mapper_dict = None
+                    self.__shared_state.route = None
+        except AttributeError:
+            pass
+        
+        if 'HTTP_X_FORWARDED_HOST' in environ:
+            self.__shared_state.host = environ['HTTP_X_FORWARDED_HOST']
+        elif 'HTTP_HOST' in environ:
+            self.__shared_state.host = environ['HTTP_HOST']
+        else:
+            self.__shared_state.host = environ['SERVER_NAME']
+            if environ['wsgi.url_scheme'] == 'https':
+                if environ['SERVER_PORT'] != '443':
+                    self.__shared_state.host += ':' + environ['SERVER_PORT']
+            else:
+                if environ['SERVER_PORT'] != '80':
+                    self.__shared_state.host += ':' + environ['SERVER_PORT']
+
+def request_config(original=False):
+    """
+    Returns the Routes RequestConfig object.
+    
+    To get the Routes RequestConfig:
+    
+    >>> from routes import *
+    >>> config = request_config()
+    
+    The following attributes must be set on the config object every request:
+    
+    mapper
+        mapper should be a Mapper instance thats ready for use
+    host
+        host is the hostname of the webapp
+    protocol
+        protocol is the protocol of the current request
+    mapper_dict
+        mapper_dict should be the dict returned by mapper.match()
+    redirect
+        redirect should be a function that issues a redirect, 
+        and takes a url as the sole argument
+    prefix (optional)
+        Set if the application is moved under a URL prefix. Prefix
+        will be stripped before matching, and prepended on generation
+    environ (optional)
+        Set to the WSGI environ for automatic prefix support if the
+        webapp is underneath a 'SCRIPT_NAME'
+        
+        Setting the environ will use information in environ to try and
+        populate the host/protocol/mapper_dict options if you've already
+        set a mapper.
+    
+    **Using your own requst local**
+    
+    If you have your own request local object that you'd like to use instead 
+    of the default thread local provided by Routes, you can configure Routes 
+    to use it::
+        
+        from routes import request_config()
+        config = request_config()
+        if hasattr(config, 'using_request_local'):
+            config.request_local = YourLocalCallable
+            config = request_config()
+    
+    Once you have configured request_config, its advisable you retrieve it 
+    again to get the object you wanted. The variable you assign to 
+    request_local is assumed to be a callable that will get the local config 
+    object you wish.
+    
+    This example tests for the presence of the 'using_request_local' attribute
+    which will be present if you haven't assigned it yet. This way you can 
+    avoid repeat assignments of the request specific callable.
+    
+    Should you want the original object, perhaps to change the callable its 
+    using or stop this behavior, call request_config(original=True).
+    """
+    obj = _RequestConfig()
+    try:
+        if obj.request_local and original is False:
+            return getattr(obj, 'request_local')()
+    except AttributeError:
+        obj.request_local = False
+        obj.using_request_local = False
+    return _RequestConfig()
+    
+from routes.mapper import Mapper
+from routes.util import redirect_to, url_for, URLGenerator
+__all__=['Mapper', 'url_for', 'URLGenerator', 'redirect_to', 'request_config']

src/routes/base.py

@@ -0,0 +1,4 @@
+"""Route and Mapper core classes"""
+from routes import request_config
+from routes.mapper import Mapper
+from routes.route import Route

src/routes/lru.py

@@ -0,0 +1,70 @@
+"""LRU caching class and decorator"""
+import threading
+
+_marker = object()
+
+class LRUCache(object):
+    def __init__(self, size):
+        """ Implements a psueudo-LRU algorithm (CLOCK) """
+        if size < 1:
+            raise ValueError('size must be >1')
+        self.clock = []
+        for i in xrange(0, size):
+            self.clock.append({'key':_marker, 'ref':False})
+        self.size = size
+        self.maxpos = size - 1
+        self.hand = 0
+        self.data = {}
+        self.lock = threading.Lock()
+    
+    def __contains__(self, key):
+        return key in self.data
+
+    def __getitem__(self, key, default=None):
+        try:
+            datum = self.data[key]
+        except KeyError:
+            return default
+        pos, val = datum
+        self.clock[pos]['ref'] = True
+        hand = pos + 1
+        if hand > self.maxpos:
+            hand = 0
+        self.hand = hand
+        return val
+
+    def __setitem__(self, key, val, _marker=_marker):
+        hand = self.hand
+        maxpos = self.maxpos
+        clock = self.clock
+        data = self.data
+        lock = self.lock
+
+        end = hand - 1
+        if end < 0:
+            end = maxpos
+
+        while 1:
+            current = clock[hand]
+            ref = current['ref']
+            if ref is True:
+                current['ref'] = False
+                hand = hand + 1
+                if hand > maxpos:
+                    hand = 0
+            elif ref is False or hand == end:
+                lock.acquire()
+                try:
+                    oldkey = current['key']
+                    if oldkey in data:
+                        del data[oldkey]
+                    current['key'] = key
+                    current['ref'] = True
+                    data[key] = (hand, val)
+                    hand += 1
+                    if hand > maxpos:
+                        hand = 0
+                    self.hand = hand
+                finally:
+                    lock.release()
+                break

src/routes/middleware.py

@@ -0,0 +1,146 @@
+"""Routes WSGI Middleware"""
+import re
+import logging
+
+from webob import Request
+
+from routes.base import request_config
+from routes.util import URLGenerator, url_for
+
+log = logging.getLogger('routes.middleware')
+
+class RoutesMiddleware(object):
+    """Routing middleware that handles resolving the PATH_INFO in
+    addition to optionally recognizing method overriding."""
+    def __init__(self, wsgi_app, mapper, use_method_override=True, 
+                 path_info=True, singleton=True):
+        """Create a Route middleware object
+        
+        Using the use_method_override keyword will require Paste to be
+        installed, and your application should use Paste's WSGIRequest
+        object as it will properly handle POST issues with wsgi.input
+        should Routes check it.
+        
+        If path_info is True, then should a route var contain
+        path_info, the SCRIPT_NAME and PATH_INFO will be altered
+        accordingly. This should be used with routes like:
+        
+        .. code-block:: python
+        
+            map.connect('blog/*path_info', controller='blog', path_info='')
+        
+        """
+        self.app = wsgi_app
+        self.mapper = mapper
+        self.singleton = singleton
+        self.use_method_override = use_method_override
+        self.path_info = path_info
+        log_debug = self.log_debug = logging.DEBUG >= log.getEffectiveLevel()
+        if self.log_debug:
+            log.debug("Initialized with method overriding = %s, and path "
+                  "info altering = %s", use_method_override, path_info)
+    
+    def __call__(self, environ, start_response):
+        """Resolves the URL in PATH_INFO, and uses wsgi.routing_args
+        to pass on URL resolver results."""
+        old_method = None
+        if self.use_method_override:
+            req = None
+            
+            # In some odd cases, there's no query string
+            try:
+                qs = environ['QUERY_STRING']
+            except KeyError:
+                qs = ''
+            if '_method' in qs:
+                req = Request(environ)
+                req.errors = 'ignore'
+                if '_method' in req.GET:
+                    old_method = environ['REQUEST_METHOD']
+                    environ['REQUEST_METHOD'] = req.GET['_method'].upper()
+                    if self.log_debug:
+                        log.debug("_method found in QUERY_STRING, altering request"
+                                " method to %s", environ['REQUEST_METHOD'])
+            elif environ['REQUEST_METHOD'] == 'POST' and is_form_post(environ):
+                if req is None:
+                    req = Request(environ)
+                    req.errors = 'ignore'
+                if '_method' in req.POST:
+                    old_method = environ['REQUEST_METHOD']
+                    environ['REQUEST_METHOD'] = req.POST['_method'].upper()
+                    if self.log_debug:
+                        log.debug("_method found in POST data, altering request "
+                                  "method to %s", environ['REQUEST_METHOD'])
+        
+        # Run the actual route matching
+        # -- Assignment of environ to config triggers route matching
+        if self.singleton:
+            config = request_config()
+            config.mapper = self.mapper
+            config.environ = environ
+            match = config.mapper_dict
+            route = config.route
+        else:
+            results = self.mapper.routematch(environ=environ)
+            if results:
+                match, route = results[0], results[1]
+            else:
+                match = route = None
+                
+        if old_method:
+            environ['REQUEST_METHOD'] = old_method
+        
+        if not match:
+            match = {}
+            if self.log_debug:
+                urlinfo = "%s %s" % (environ['REQUEST_METHOD'], environ['PATH_INFO'])
+                log.debug("No route matched for %s", urlinfo)
+        elif self.log_debug:
+            urlinfo = "%s %s" % (environ['REQUEST_METHOD'], environ['PATH_INFO'])
+            log.debug("Matched %s", urlinfo)
+            log.debug("Route path: '%s', defaults: %s", route.routepath, 
+                      route.defaults)
+            log.debug("Match dict: %s", match)
+                
+        url = URLGenerator(self.mapper, environ)
+        environ['wsgiorg.routing_args'] = ((url), match)
+        environ['routes.route'] = route
+        environ['routes.url'] = url
+
+        if route and route.redirect:
+            route_name = '_redirect_%s' % id(route)
+            location = url(route_name, **match)
+            log.debug("Using redirect route, redirect to '%s' with status"
+                      "code: %s", location, route.redirect_status)
+            start_response(route.redirect_status, 
+                           [('Content-Type', 'text/plain; charset=utf8'), 
+                            ('Location', location)])
+            return []
+
+        # If the route included a path_info attribute and it should be used to
+        # alter the environ, we'll pull it out
+        if self.path_info and 'path_info' in match:
+            oldpath = environ['PATH_INFO']
+            newpath = match.get('path_info') or ''
+            environ['PATH_INFO'] = newpath
+            if not environ['PATH_INFO'].startswith('/'):
+                environ['PATH_INFO'] = '/' + environ['PATH_INFO']
+            environ['SCRIPT_NAME'] += re.sub(r'^(.*?)/' + re.escape(newpath) + '$', 
+                                             r'\1', oldpath)
+        
+        response = self.app(environ, start_response)
+        
+        # Wrapped in try as in rare cases the attribute will be gone already
+        try:
+            del self.mapper.environ
+        except AttributeError:
+            pass
+        return response
+
+def is_form_post(environ):
+    """Determine whether the request is a POSTed html form"""
+    content_type = environ.get('CONTENT_TYPE', '').lower()
+    if ';' in content_type:
+        content_type = content_type.split(';', 1)[0]
+    return content_type in ('application/x-www-form-urlencoded',
+                            'multipart/form-data')

src/routes/route.py

@@ -0,0 +1,742 @@
+import re
+import sys
+import urllib
+
+if sys.version < '2.4':
+    from sets import ImmutableSet as frozenset
+
+from routes.util import _url_quote as url_quote, _str_encode
+
+
+class Route(object):
+    """The Route object holds a route recognition and generation
+    routine.
+    
+    See Route.__init__ docs for usage.
+    
+    """
+    # reserved keys that don't count
+    reserved_keys = ['requirements']
+    
+    # special chars to indicate a natural split in the URL
+    done_chars = ('/', ',', ';', '.', '#')
+    
+    def __init__(self, name, routepath, **kargs):
+        """Initialize a route, with a given routepath for
+        matching/generation
+        
+        The set of keyword args will be used as defaults.
+        
+        Usage::
+        
+            >>> from routes.base import Route
+            >>> newroute = Route(None, ':controller/:action/:id')
+            >>> sorted(newroute.defaults.items())
+            [('action', 'index'), ('id', None)]
+            >>> newroute = Route(None, 'date/:year/:month/:day',  
+            ...     controller="blog", action="view")
+            >>> newroute = Route(None, 'archives/:page', controller="blog", 
+            ...     action="by_page", requirements = { 'page':'\d{1,2}' })
+            >>> newroute.reqs
+            {'page': '\\\d{1,2}'}
+        
+        .. Note:: 
+            Route is generally not called directly, a Mapper instance
+            connect method should be used to add routes.
+        
+        """
+        self.routepath = routepath
+        self.sub_domains = False
+        self.prior = None
+        self.redirect = False
+        self.name = name
+        self._kargs = kargs
+        self.minimization = kargs.pop('_minimize', False)
+        self.encoding = kargs.pop('_encoding', 'utf-8')
+        self.reqs = kargs.get('requirements', {})
+        self.decode_errors = 'replace'
+        
+        # Don't bother forming stuff we don't need if its a static route
+        self.static = kargs.pop('_static', False)
+        self.filter = kargs.pop('_filter', None)
+        self.absolute = kargs.pop('_absolute', False)
+        
+        # Pull out the member/collection name if present, this applies only to
+        # map.resource
+        self.member_name = kargs.pop('_member_name', None)
+        self.collection_name = kargs.pop('_collection_name', None)
+        self.parent_resource = kargs.pop('_parent_resource', None)
+        
+        # Pull out route conditions
+        self.conditions = kargs.pop('conditions', None)
+        
+        # Determine if explicit behavior should be used
+        self.explicit = kargs.pop('_explicit', False)
+                
+        # Since static need to be generated exactly, treat them as
+        # non-minimized
+        if self.static:
+            self.external = '://' in self.routepath
+            self.minimization = False
+        
+        # Strip preceding '/' if present, and not minimizing
+        if routepath.startswith('/') and self.minimization:
+            self.routepath = routepath[1:]
+        self._setup_route()
+        
+    def _setup_route(self):
+        # Build our routelist, and the keys used in the route
+        self.routelist = routelist = self._pathkeys(self.routepath)
+        routekeys = frozenset([key['name'] for key in routelist
+                               if isinstance(key, dict)])
+        self.dotkeys = frozenset([key['name'] for key in routelist
+                                  if isinstance(key, dict) and 
+                                     key['type'] == '.'])
+
+        if not self.minimization:
+            self.make_full_route()
+        
+        # Build a req list with all the regexp requirements for our args
+        self.req_regs = {}
+        for key, val in self.reqs.iteritems():
+            self.req_regs[key] = re.compile('^' + val + '$')
+        # Update our defaults and set new default keys if needed. defaults
+        # needs to be saved
+        (self.defaults, defaultkeys) = self._defaults(routekeys, 
+                                                      self.reserved_keys, 
+                                                      self._kargs.copy())
+        # Save the maximum keys we could utilize
+        self.maxkeys = defaultkeys | routekeys
+        
+        # Populate our minimum keys, and save a copy of our backward keys for
+        # quicker generation later
+        (self.minkeys, self.routebackwards) = self._minkeys(routelist[:])
+        
+        # Populate our hardcoded keys, these are ones that are set and don't 
+        # exist in the route
+        self.hardcoded = frozenset([key for key in self.maxkeys \
+            if key not in routekeys and self.defaults[key] is not None])
+        
+        # Cache our default keys
+        self._default_keys = frozenset(self.defaults.keys())
+    
+    def make_full_route(self):
+        """Make a full routelist string for use with non-minimized
+        generation"""
+        regpath = ''
+        for part in self.routelist:
+            if isinstance(part, dict):
+                regpath += '%(' + part['name'] + ')s'
+            else:
+                regpath += part
+        self.regpath = regpath
+    
+    def make_unicode(self, s):
+        """Transform the given argument into a unicode string."""
+        if isinstance(s, unicode):
+            return s
+        elif isinstance(s, str):
+            return s.decode(self.encoding)
+        elif callable(s):
+            return s
+        else:
+            return unicode(s)
+    
+    def _pathkeys(self, routepath):
+        """Utility function to walk the route, and pull out the valid
+        dynamic/wildcard keys."""
+        collecting = False
+        current = ''
+        done_on = ''
+        var_type = ''
+        just_started = False
+        routelist = []
+        for char in routepath:
+            if char in [':', '*', '{'] and not collecting and not self.static \
+               or char in ['{'] and not collecting:
+                just_started = True
+                collecting = True
+                var_type = char
+                if char == '{':
+                    done_on = '}'
+                    just_started = False
+                if len(current) > 0:
+                    routelist.append(current)
+                    current = ''
+            elif collecting and just_started:
+                just_started = False
+                if char == '(':
+                    done_on = ')'
+                else:
+                    current = char
+                    done_on = self.done_chars + ('-',)
+            elif collecting and char not in done_on:
+                current += char
+            elif collecting:
+                collecting = False
+                if var_type == '{':
+                    if current[0] == '.':
+                        var_type = '.'
+                        current = current[1:]
+                    else:
+                        var_type = ':'
+                    opts = current.split(':')
+                    if len(opts) > 1:
+                        current = opts[0]
+                        self.reqs[current] = opts[1]
+                routelist.append(dict(type=var_type, name=current))
+                if char in self.done_chars:
+                    routelist.append(char)
+                done_on = var_type = current = ''
+            else:
+                current += char
+        if collecting:
+            routelist.append(dict(type=var_type, name=current))
+        elif current:
+            routelist.append(current)
+        return routelist
+
+    def _minkeys(self, routelist):
+        """Utility function to walk the route backwards
+        
+        Will also determine the minimum keys we can handle to generate
+        a working route.
+        
+        routelist is a list of the '/' split route path
+        defaults is a dict of all the defaults provided for the route
+        
+        """
+        minkeys = []
+        backcheck = routelist[:]
+        
+        # If we don't honor minimization, we need all the keys in the
+        # route path
+        if not self.minimization:
+            for part in backcheck:
+                if isinstance(part, dict):
+                    minkeys.append(part['name'])
+            return (frozenset(minkeys), backcheck)
+        
+        gaps = False
+        backcheck.reverse()
+        for part in backcheck:
+            if not isinstance(part, dict) and part not in self.done_chars:
+                gaps = True
+                continue
+            elif not isinstance(part, dict):
+                continue
+            key = part['name']
+            if self.defaults.has_key(key) and not gaps:
+                continue
+            minkeys.append(key)
+            gaps = True
+        return  (frozenset(minkeys), backcheck)
+    
+    def _defaults(self, routekeys, reserved_keys, kargs):
+        """Creates default set with values stringified
+        
+        Put together our list of defaults, stringify non-None values
+        and add in our action/id default if they use it and didn't
+        specify it.
+        
+        defaultkeys is a list of the currently assumed default keys
+        routekeys is a list of the keys found in the route path
+        reserved_keys is a list of keys that are not
+        
+        """
+        defaults = {}
+        # Add in a controller/action default if they don't exist
+        if 'controller' not in routekeys and 'controller' not in kargs \
+           and not self.explicit:
+            kargs['controller'] = 'content'
+        if 'action' not in routekeys and 'action' not in kargs \
+           and not self.explicit:
+            kargs['action'] = 'index'
+        defaultkeys = frozenset([key for key in kargs.keys() \
+                                 if key not in reserved_keys])
+        for key in defaultkeys:
+            if kargs[key] is not None:
+                defaults[key] = self.make_unicode(kargs[key])
+            else:
+                defaults[key] = None
+        if 'action' in routekeys and not defaults.has_key('action') \
+           and not self.explicit:
+            defaults['action'] = 'index'
+        if 'id' in routekeys and not defaults.has_key('id') \
+           and not self.explicit:
+            defaults['id'] = None
+        newdefaultkeys = frozenset([key for key in defaults.keys() \
+                                    if key not in reserved_keys])
+        
+        return (defaults, newdefaultkeys)
+        
+    def makeregexp(self, clist, include_names=True):
+        """Create a regular expression for matching purposes
+        
+        Note: This MUST be called before match can function properly.
+        
+        clist should be a list of valid controller strings that can be 
+        matched, for this reason makeregexp should be called by the web
+        framework after it knows all available controllers that can be
+        utilized.
+        
+        include_names indicates whether this should be a match regexp
+        assigned to itself using regexp grouping names, or if names
+        should be excluded for use in a single larger regexp to
+        determine if any routes match
+        
+        """
+        if self.minimization:
+            reg = self.buildnextreg(self.routelist, clist, include_names)[0]
+            if not reg:
+                reg = '/'
+            reg = reg + '/?' + '$'
+        
+            if not reg.startswith('/'):
+                reg = '/' + reg
+        else:
+            reg = self.buildfullreg(clist, include_names)
+        
+        reg = '^' + reg
+        
+        if not include_names:
+            return reg
+        
+        self.regexp = reg
+        self.regmatch = re.compile(reg)
+    
+    def buildfullreg(self, clist, include_names=True):
+        """Build the regexp by iterating through the routelist and
+        replacing dicts with the appropriate regexp match"""
+        regparts = []
+        for part in self.routelist:
+            if isinstance(part, dict):
+                var = part['name']
+                if var == 'controller':
+                    partmatch = '|'.join(map(re.escape, clist))
+                elif part['type'] == ':':
+                    partmatch = self.reqs.get(var) or '[^/]+?'
+                elif part['type'] == '.':
+                    partmatch = self.reqs.get(var) or '[^/.]+?'
+                else:
+                    partmatch = self.reqs.get(var) or '.+?'
+                if include_names:
+                    regpart = '(?P<%s>%s)' % (var, partmatch)
+                else:
+                    regpart = '(?:%s)' % partmatch
+                if part['type'] == '.':
+                    regparts.append('(?:\.%s)??' % regpart)
+                else:
+                    regparts.append(regpart)
+            else:
+                regparts.append(re.escape(part))
+        regexp = ''.join(regparts) + '$'
+        return regexp
+    
+    def buildnextreg(self, path, clist, include_names=True):
+        """Recursively build our regexp given a path, and a controller
+        list.
+        
+        Returns the regular expression string, and two booleans that
+        can be ignored as they're only used internally by buildnextreg.
+        
+        """
+        if path:
+            part = path[0]
+        else:
+            part = ''
+        reg = ''
+        
+        # noreqs will remember whether the remainder has either a string 
+        # match, or a non-defaulted regexp match on a key, allblank remembers
+        # if the rest could possible be completely empty
+        (rest, noreqs, allblank) = ('', True, True)
+        if len(path[1:]) > 0:
+            self.prior = part
+            (rest, noreqs, allblank) = self.buildnextreg(path[1:], clist, include_names)
+        
+        if isinstance(part, dict) and part['type'] in (':', '.'):
+            var = part['name']
+            typ = part['type']
+            partreg = ''
+            
+            # First we plug in the proper part matcher
+            if self.reqs.has_key(var):
+                if include_names:
+                    partreg = '(?P<%s>%s)' % (var, self.reqs[var])
+                else:
+                    partreg = '(?:%s)' % self.reqs[var]
+                if typ == '.':
+                    partreg = '(?:\.%s)??' % partreg
+            elif var == 'controller':
+                if include_names:
+                    partreg = '(?P<%s>%s)' % (var, '|'.join(map(re.escape, clist)))
+                else:
+                    partreg = '(?:%s)' % '|'.join(map(re.escape, clist))
+            elif self.prior in ['/', '#']:
+                if include_names:
+                    partreg = '(?P<' + var + '>[^' + self.prior + ']+?)'
+                else:
+                    partreg = '(?:[^' + self.prior + ']+?)'
+            else:
+                if not rest:
+                    if typ == '.':
+                        exclude_chars = '/.'
+                    else:
+                        exclude_chars = '/'
+                    if include_names:
+                        partreg = '(?P<%s>[^%s]+?)' % (var, exclude_chars)
+                    else:
+                        partreg = '(?:[^%s]+?)' % exclude_chars
+                    if typ == '.':
+                        partreg = '(?:\.%s)??' % partreg
+                else:
+                    end = ''.join(self.done_chars)
+                    rem = rest
+                    if rem[0] == '\\' and len(rem) > 1:
+                        rem = rem[1]
+                    elif rem.startswith('(\\') and len(rem) > 2:
+                        rem = rem[2]
+                    else:
+                        rem = end
+                    rem = frozenset(rem) | frozenset(['/'])
+                    if include_names:
+                        partreg = '(?P<%s>[^%s]+?)' % (var, ''.join(rem))
+                    else:
+                        partreg = '(?:[^%s]+?)' % ''.join(rem)
+            
+            if self.reqs.has_key(var):
+                noreqs = False
+            if not self.defaults.has_key(var): 
+                allblank = False
+                noreqs = False
+            
+            # Now we determine if its optional, or required. This changes 
+            # depending on what is in the rest of the match. If noreqs is 
+            # true, then its possible the entire thing is optional as there's
+            # no reqs or string matches.
+            if noreqs:
+                # The rest is optional, but now we have an optional with a 
+                # regexp. Wrap to ensure that if we match anything, we match
+                # our regexp first. It's still possible we could be completely
+                # blank as we have a default
+                if self.reqs.has_key(var) and self.defaults.has_key(var):
+                    reg = '(' + partreg + rest + ')?'
+                
+                # Or we have a regexp match with no default, so now being 
+                # completely blank form here on out isn't possible
+                elif self.reqs.has_key(var):
+                    allblank = False
+                    reg = partreg + rest
+                
+                # If the character before this is a special char, it has to be
+                # followed by this
+                elif self.defaults.has_key(var) and \
+                     self.prior in (',', ';', '.'):
+                    reg = partreg + rest
+                
+                # Or we have a default with no regexp, don't touch the allblank
+                elif self.defaults.has_key(var):
+                    reg = partreg + '?' + rest
+                
+                # Or we have a key with no default, and no reqs. Not possible
+                # to be all blank from here
+                else:
+                    allblank = False
+                    reg = partreg + rest
+            # In this case, we have something dangling that might need to be
+            # matched
+            else:
+                # If they can all be blank, and we have a default here, we know
+                # its safe to make everything from here optional. Since 
+                # something else in the chain does have req's though, we have
+                # to make the partreg here required to continue matching
+                if allblank and self.defaults.has_key(var):
+                    reg = '(' + partreg + rest + ')?'
+                    
+                # Same as before, but they can't all be blank, so we have to 
+                # require it all to ensure our matches line up right
+                else:
+                    reg = partreg + rest
+        elif isinstance(part, dict) and part['type'] == '*':
+            var = part['name']
+            if noreqs:
+                if include_names:
+                    reg = '(?P<%s>.*)' % var + rest
+                else:
+                    reg = '(?:.*)' + rest
+                if not self.defaults.has_key(var):
+                    allblank = False
+                    noreqs = False
+            else:
+                if allblank and self.defaults.has_key(var):
+                    if include_names:
+                        reg = '(?P<%s>.*)' % var + rest
+                    else:
+                        reg = '(?:.*)' + rest
+                elif self.defaults.has_key(var):
+                    if include_names:
+                        reg = '(?P<%s>.*)' % var + rest
+                    else:
+                        reg = '(?:.*)' + rest
+                else:
+                    if include_names:
+                        reg = '(?P<%s>.*)' % var + rest
+                    else:
+                        reg = '(?:.*)' + rest
+                    allblank = False
+                    noreqs = False
+        elif part and part[-1] in self.done_chars:
+            if allblank:
+                reg = re.escape(part[:-1]) + '(' + re.escape(part[-1]) + rest
+                reg += ')?'
+            else:
+                allblank = False
+                reg = re.escape(part) + rest
+        
+        # We have a normal string here, this is a req, and it prevents us from 
+        # being all blank
+        else:
+            noreqs = False
+            allblank = False
+            reg = re.escape(part) + rest
+        
+        return (reg, noreqs, allblank)
+    
+    def match(self, url, environ=None, sub_domains=False, 
+              sub_domains_ignore=None, domain_match=''):
+        """Match a url to our regexp. 
+        
+        While the regexp might match, this operation isn't
+        guaranteed as there's other factors that can cause a match to
+        fail even though the regexp succeeds (Default that was relied
+        on wasn't given, requirement regexp doesn't pass, etc.).
+        
+        Therefore the calling function shouldn't assume this will
+        return a valid dict, the other possible return is False if a
+        match doesn't work out.
+        
+        """
+        # Static routes don't match, they generate only
+        if self.static:
+            return False
+        
+        match = self.regmatch.match(url)
+        
+        if not match:
+            return False
+            
+        sub_domain = None
+        
+        if sub_domains and environ and 'HTTP_HOST' in environ:
+            host = environ['HTTP_HOST'].split(':')[0]
+            sub_match = re.compile('^(.+?)\.%s$' % domain_match)
+            subdomain = re.sub(sub_match, r'\1', host)
+            if subdomain not in sub_domains_ignore and host != subdomain:
+                sub_domain = subdomain
+        
+        if self.conditions:
+            if 'method' in self.conditions and environ and \
+                environ['REQUEST_METHOD'] not in self.conditions['method']:
+                return False
+            
+            # Check sub-domains?
+            use_sd = self.conditions.get('sub_domain')
+            if use_sd and not sub_domain:
+                return False
+            elif not use_sd and 'sub_domain' in self.conditions and sub_domain:
+                return False
+            if isinstance(use_sd, list) and sub_domain not in use_sd:
+                return False
+        
+        matchdict = match.groupdict()
+        result = {}
+        extras = self._default_keys - frozenset(matchdict.keys())
+        for key, val in matchdict.iteritems():
+            if key != 'path_info' and self.encoding:
+                # change back into python unicode objects from the URL 
+                # representation
+                try:
+                    val = val and val.decode(self.encoding, self.decode_errors)
+                except UnicodeDecodeError:
+                    return False
+            
+            if not val and key in self.defaults and self.defaults[key]:
+                result[key] = self.defaults[key]
+            else:
+                result[key] = val
+        for key in extras:
+            result[key] = self.defaults[key]
+        
+        # Add the sub-domain if there is one
+        if sub_domains:
+            result['sub_domain'] = sub_domain
+        
+        # If there's a function, call it with environ and expire if it
+        # returns False
+        if self.conditions and 'function' in self.conditions and \
+            not self.conditions['function'](environ, result):
+            return False
+        
+        return result
+    
+    def generate_non_minimized(self, kargs):
+        """Generate a non-minimal version of the URL"""
+        # Iterate through the keys that are defaults, and NOT in the route
+        # path. If its not in kargs, or doesn't match, or is None, this
+        # route won't work
+        for k in self.maxkeys - self.minkeys:
+            if k not in kargs:
+                return False
+            elif self.make_unicode(kargs[k]) != \
+                self.make_unicode(self.defaults[k]):
+                return False
+                
+        # Ensure that all the args in the route path are present and not None
+        for arg in self.minkeys:
+            if arg not in kargs or kargs[arg] is None:
+                if arg in self.dotkeys:
+                    kargs[arg] = ''
+                else:
+                    return False
+
+        # Encode all the argument that the regpath can use
+        for k in kargs:
+            if k in self.maxkeys:
+                if k in self.dotkeys:
+                    if kargs[k]:
+                        kargs[k] = url_quote('.' + kargs[k], self.encoding)
+                else:
+                    kargs[k] = url_quote(kargs[k], self.encoding)
+
+        return self.regpath % kargs
+    
+    def generate_minimized(self, kargs):
+        """Generate a minimized version of the URL"""
+        routelist = self.routebackwards
+        urllist = []
+        gaps = False
+        for part in routelist:
+            if isinstance(part, dict) and part['type'] in (':', '.'):
+                arg = part['name']
+                
+                # For efficiency, check these just once
+                has_arg = kargs.has_key(arg)
+                has_default = self.defaults.has_key(arg)
+                
+                # Determine if we can leave this part off
+                # First check if the default exists and wasn't provided in the 
+                # call (also no gaps)
+                if has_default and not has_arg and not gaps:
+                    continue
+                    
+                # Now check to see if there's a default and it matches the 
+                # incoming call arg
+                if (has_default and has_arg) and self.make_unicode(kargs[arg]) == \
+                    self.make_unicode(self.defaults[arg]) and not gaps: 
+                    continue
+                
+                # We need to pull the value to append, if the arg is None and 
+                # we have a default, use that
+                if has_arg and kargs[arg] is None and has_default and not gaps:
+                    continue
+                
+                # Otherwise if we do have an arg, use that
+                elif has_arg:
+                    val = kargs[arg]
+                
+                elif has_default and self.defaults[arg] is not None:
+                    val = self.defaults[arg]
+                # Optional format parameter?
+                elif part['type'] == '.':
+                    continue
+                # No arg at all? This won't work
+                else:
+                    return False
+                    
+                urllist.append(url_quote(val, self.encoding))
+                if part['type'] == '.':
+                    urllist.append('.')
+
+                if has_arg:
+                    del kargs[arg]
+                gaps = True
+            elif isinstance(part, dict) and part['type'] == '*':
+                arg = part['name']
+                kar = kargs.get(arg)
+                if kar is not None:
+                    urllist.append(url_quote(kar, self.encoding))
+                    gaps = True
+            elif part and part[-1] in self.done_chars:
+                if not gaps and part in self.done_chars:
+                    continue
+                elif not gaps:
+                    urllist.append(part[:-1])
+                    gaps = True
+                else:
+                    gaps = True
+                    urllist.append(part)
+            else:
+                gaps = True
+                urllist.append(part)
+        urllist.reverse()
+        url = ''.join(urllist)
+        return url
+    
+    def generate(self, _ignore_req_list=False, _append_slash=False, **kargs):
+        """Generate a URL from ourself given a set of keyword arguments
+        
+        Toss an exception if this
+        set of keywords would cause a gap in the url.
+        
+        """
+        # Verify that our args pass any regexp requirements
+        if not _ignore_req_list:
+            for key in self.reqs.keys():
+                val = kargs.get(key)
+                if val and not self.req_regs[key].match(self.make_unicode(val)):
+                    return False
+        
+        # Verify that if we have a method arg, its in the method accept list. 
+        # Also, method will be changed to _method for route generation
+        meth = kargs.get('method')
+        if meth:
+            if self.conditions and 'method' in self.conditions \
+                and meth.upper() not in self.conditions['method']:
+                return False
+            kargs.pop('method')
+        
+        if self.minimization:
+            url = self.generate_minimized(kargs)
+        else:
+            url = self.generate_non_minimized(kargs)
+        
+        if url is False:
+            return url
+        
+        if not url.startswith('/') and not self.static:
+            url = '/' + url
+        extras = frozenset(kargs.keys()) - self.maxkeys
+        if extras:
+            if _append_slash and not url.endswith('/'):
+                url += '/'
+            fragments = []
+            # don't assume the 'extras' set preserves order: iterate
+            # through the ordered kargs instead
+            for key in kargs:
+                if key not in extras:
+                    continue
+                if key == 'action' or key == 'controller':
+                    continue
+                val = kargs[key]
+                if isinstance(val, (tuple, list)):
+                    for value in val:
+                        fragments.append((key, _str_encode(value, self.encoding)))
+                else:
+                    fragments.append((key, _str_encode(val, self.encoding)))
+            if fragments:
+                url += '?'
+                url += urllib.urlencode(fragments)
+        elif _append_slash and not url.endswith('/'):
+            url += '/'
+        return url

src/routes/util.py

@@ -0,0 +1,503 @@
+"""Utility functions for use in templates / controllers
+
+*PLEASE NOTE*: Many of these functions expect an initialized RequestConfig
+object. This is expected to have been initialized for EACH REQUEST by the web
+framework.
+
+"""
+import os
+import re
+import urllib
+from routes import request_config
+
+
+class RoutesException(Exception):
+    """Tossed during Route exceptions"""
+
+
+class MatchException(RoutesException):
+    """Tossed during URL matching exceptions"""
+
+
+class GenerationException(RoutesException):
+    """Tossed during URL generation exceptions"""
+
+
+def _screenargs(kargs, mapper, environ, force_explicit=False):
+    """
+    Private function that takes a dict, and screens it against the current 
+    request dict to determine what the dict should look like that is used. 
+    This is responsible for the requests "memory" of the current.
+    """
+    # Coerce any unicode args with the encoding
+    encoding = mapper.encoding
+    for key, val in kargs.iteritems():
+        if isinstance(val, unicode):
+            kargs[key] = val.encode(encoding)
+    
+    if mapper.explicit and mapper.sub_domains and not force_explicit:
+        return _subdomain_check(kargs, mapper, environ)
+    elif mapper.explicit and not force_explicit:
+        return kargs
+    
+    controller_name = kargs.get('controller')
+    
+    if controller_name and controller_name.startswith('/'):
+        # If the controller name starts with '/', ignore route memory
+        kargs['controller'] = kargs['controller'][1:]
+        return kargs
+    elif controller_name and not kargs.has_key('action'):
+        # Fill in an action if we don't have one, but have a controller
+        kargs['action'] = 'index'
+    
+    route_args = environ.get('wsgiorg.routing_args')
+    if route_args:
+        memory_kargs = route_args[1].copy()
+    else:
+        memory_kargs = {}
+     
+    # Remove keys from memory and kargs if kargs has them as None
+    for key in [key for key in kargs.keys() if kargs[key] is None]:
+        del kargs[key]
+        if memory_kargs.has_key(key):
+            del memory_kargs[key]
+    
+    # Merge the new args on top of the memory args
+    memory_kargs.update(kargs)
+    
+    # Setup a sub-domain if applicable
+    if mapper.sub_domains:
+        memory_kargs = _subdomain_check(memory_kargs, mapper, environ)
+    return memory_kargs
+
+
+def _subdomain_check(kargs, mapper, environ):
+    """Screen the kargs for a subdomain and alter it appropriately depending
+    on the current subdomain or lack therof."""
+    if mapper.sub_domains:
+        subdomain = kargs.pop('sub_domain', None)
+        if isinstance(subdomain, unicode):
+            subdomain = str(subdomain)
+        
+        fullhost = environ.get('HTTP_HOST') or environ.get('SERVER_NAME')
+        
+        # In case environ defaulted to {}
+        if not fullhost:
+            return kargs
+        
+        hostmatch = fullhost.split(':')
+        host = hostmatch[0]
+        port = ''
+        if len(hostmatch) > 1:
+            port += ':' + hostmatch[1]
+        sub_match = re.compile('^.+?\.(%s)$' % mapper.domain_match)
+        domain = re.sub(sub_match, r'\1', host)
+        if subdomain and not host.startswith(subdomain) and \
+            subdomain not in mapper.sub_domains_ignore:
+            kargs['_host'] = subdomain + '.' + domain + port
+        elif (subdomain in mapper.sub_domains_ignore or \
+            subdomain is None) and domain != host:
+            kargs['_host'] = domain + port
+        return kargs
+    else:
+        return kargs
+
+
+def _url_quote(string, encoding):
+    """A Unicode handling version of urllib.quote."""
+    if encoding:
+        if isinstance(string, unicode):
+            s = string.encode(encoding)
+        elif isinstance(string, str):
+            # assume the encoding is already correct
+            s = string
+        else:
+            s = unicode(string).encode(encoding)
+    else:
+        s = str(string)
+    return urllib.quote(s, '/')
+
+
+def _str_encode(string, encoding):
+    if encoding:
+        if isinstance(string, unicode):
+            s = string.encode(encoding)
+        elif isinstance(string, str):
+            # assume the encoding is already correct
+            s = string
+        else:
+            s = unicode(string).encode(encoding)
+    return s
+
+
+def url_for(*args, **kargs):
+    """Generates a URL 
+    
+    All keys given to url_for are sent to the Routes Mapper instance for 
+    generation except for::
+        
+        anchor          specified the anchor name to be appened to the path
+        host            overrides the default (current) host if provided
+        protocol        overrides the default (current) protocol if provided
+        qualified       creates the URL with the host/port information as 
+                        needed
+        
+    The URL is generated based on the rest of the keys. When generating a new 
+    URL, values will be used from the current request's parameters (if 
+    present). The following rules are used to determine when and how to keep 
+    the current requests parameters:
+    
+    * If the controller is present and begins with '/', no defaults are used
+    * If the controller is changed, action is set to 'index' unless otherwise 
+      specified
+    
+    For example, if the current request yielded a dict of
+    {'controller': 'blog', 'action': 'view', 'id': 2}, with the standard 
+    ':controller/:action/:id' route, you'd get the following results::
+    
+        url_for(id=4)                    =>  '/blog/view/4',
+        url_for(controller='/admin')     =>  '/admin',
+        url_for(controller='admin')      =>  '/admin/view/2'
+        url_for(action='edit')           =>  '/blog/edit/2',
+        url_for(action='list', id=None)  =>  '/blog/list'
+    
+    **Static and Named Routes**
+    
+    If there is a string present as the first argument, a lookup is done 
+    against the named routes table to see if there's any matching routes. The
+    keyword defaults used with static routes will be sent in as GET query 
+    arg's if a route matches.
+    
+    If no route by that name is found, the string is assumed to be a raw URL. 
+    Should the raw URL begin with ``/`` then appropriate SCRIPT_NAME data will
+    be added if present, otherwise the string will be used as the url with 
+    keyword args becoming GET query args.
+    
+    """
+    anchor = kargs.get('anchor')
+    host = kargs.get('host')
+    protocol = kargs.get('protocol')
+    qualified = kargs.pop('qualified', None)
+    
+    # Remove special words from kargs, convert placeholders
+    for key in ['anchor', 'host', 'protocol']:
+        if kargs.get(key):
+            del kargs[key]
+    config = request_config()
+    route = None
+    static = False
+    encoding = config.mapper.encoding
+    url = ''
+    if len(args) > 0:
+        route = config.mapper._routenames.get(args[0])
+        
+        # No named route found, assume the argument is a relative path
+        if not route:
+            static = True
+            url = args[0]
+        
+        if url.startswith('/') and hasattr(config, 'environ') \
+                and config.environ.get('SCRIPT_NAME'):
+            url = config.environ.get('SCRIPT_NAME') + url
+        
+        if static:
+            if kargs:
+                url += '?'
+                query_args = []
+                for key, val in kargs.iteritems():
+                    if isinstance(val, (list, tuple)):
+                        for value in val:
+                            query_args.append("%s=%s" % (
+                                urllib.quote(unicode(key).encode(encoding)),
+                                urllib.quote(unicode(value).encode(encoding))))
+                    else:
+                        query_args.append("%s=%s" % (
+                            urllib.quote(unicode(key).encode(encoding)),
+                            urllib.quote(unicode(val).encode(encoding))))
+                url += '&'.join(query_args)
+    environ = getattr(config, 'environ', {})
+    if 'wsgiorg.routing_args' not in environ:
+        environ = environ.copy()
+        mapper_dict = getattr(config, 'mapper_dict', None)
+        if mapper_dict is not None:
+            match_dict = mapper_dict.copy()
+        else:
+            match_dict = {}
+        environ['wsgiorg.routing_args'] = ((), match_dict)
+    
+    if not static:
+        route_args = []
+        if route:
+            if config.mapper.hardcode_names:
+                route_args.append(route)
+            newargs = route.defaults.copy()
+            newargs.update(kargs)
+            
+            # If this route has a filter, apply it
+            if route.filter:
+                newargs = route.filter(newargs)
+            
+            if not route.static:
+                # Handle sub-domains
+                newargs = _subdomain_check(newargs, config.mapper, environ)
+        else:
+            newargs = _screenargs(kargs, config.mapper, environ)
+        anchor = newargs.pop('_anchor', None) or anchor
+        host = newargs.pop('_host', None) or host
+        protocol = newargs.pop('_protocol', None) or protocol
+        url = config.mapper.generate(*route_args, **newargs)
+    if anchor is not None:
+        url += '#' + _url_quote(anchor, encoding)
+    if host or protocol or qualified:
+        if not host and not qualified:
+            # Ensure we don't use a specific port, as changing the protocol
+            # means that we most likely need a new port
+            host = config.host.split(':')[0]
+        elif not host:
+            host = config.host
+        if not protocol:
+            protocol = config.protocol
+        if url is not None:
+            url = protocol + '://' + host + url
+    
+    if not isinstance(url, str) and url is not None:
+        raise GenerationException("url_for can only return a string, got "
+                        "unicode instead: %s" % url)
+    if url is None:
+        raise GenerationException(
+            "url_for could not generate URL. Called with args: %s %s" % \
+            (args, kargs))
+    return url
+
+
+class URLGenerator(object):
+    """The URL Generator generates URL's
+    
+    It is automatically instantiated by the RoutesMiddleware and put
+    into the ``wsgiorg.routing_args`` tuple accessible as::
+    
+        url = environ['wsgiorg.routing_args'][0][0]
+    
+    Or via the ``routes.url`` key::
+    
+        url = environ['routes.url']
+    
+    The url object may be instantiated outside of a web context for use
+    in testing, however sub_domain support and fully qualified URL's
+    cannot be generated without supplying a dict that must contain the
+    key ``HTTP_HOST``.
+    
+    """
+    def __init__(self, mapper, environ):
+        """Instantiate the URLGenerator
+        
+        ``mapper``
+            The mapper object to use when generating routes.
+        ``environ``
+            The environment dict used in WSGI, alternately, any dict
+            that contains at least an ``HTTP_HOST`` value.
+        
+        """
+        self.mapper = mapper
+        if 'SCRIPT_NAME' not in environ:
+            environ['SCRIPT_NAME'] = ''
+        self.environ = environ
+    
+    def __call__(self, *args, **kargs):
+        """Generates a URL 
+
+        All keys given to url_for are sent to the Routes Mapper instance for 
+        generation except for::
+
+            anchor          specified the anchor name to be appened to the path
+            host            overrides the default (current) host if provided
+            protocol        overrides the default (current) protocol if provided
+            qualified       creates the URL with the host/port information as 
+                            needed
+
+        """
+        anchor = kargs.get('anchor')
+        host = kargs.get('host')
+        protocol = kargs.get('protocol')
+        qualified = kargs.pop('qualified', None)
+
+        # Remove special words from kargs, convert placeholders
+        for key in ['anchor', 'host', 'protocol']:
+            if kargs.get(key):
+                del kargs[key]
+        
+        route = None
+        use_current = '_use_current' in kargs and kargs.pop('_use_current')
+        
+        static = False
+        encoding = self.mapper.encoding
+        url = ''
+                
+        more_args = len(args) > 0
+        if more_args:
+            route = self.mapper._routenames.get(args[0])
+        
+        if not route and more_args:
+            static = True
+            url = args[0]
+            if url.startswith('/') and self.environ.get('SCRIPT_NAME'):
+                url = self.environ.get('SCRIPT_NAME') + url
+
+            if static:
+                if kargs:
+                    url += '?'
+                    query_args = []
+                    for key, val in kargs.iteritems():
+                        if isinstance(val, (list, tuple)):
+                            for value in val:
+                                query_args.append("%s=%s" % (
+                                    urllib.quote(unicode(key).encode(encoding)),
+                                    urllib.quote(unicode(value).encode(encoding))))
+                        else:
+                            query_args.append("%s=%s" % (
+                                urllib.quote(unicode(key).encode(encoding)),
+                                urllib.quote(unicode(val).encode(encoding))))
+                    url += '&'.join(query_args)
+        if not static:
+            route_args = []
+            if route:
+                if self.mapper.hardcode_names:
+                    route_args.append(route)
+                newargs = route.defaults.copy()
+                newargs.update(kargs)
+                
+                # If this route has a filter, apply it
+                if route.filter:
+                    newargs = route.filter(newargs)
+                if not route.static or (route.static and not route.external):
+                    # Handle sub-domains, retain sub_domain if there is one
+                    sub = newargs.get('sub_domain', None)
+                    newargs = _subdomain_check(newargs, self.mapper,
+                                               self.environ)
+                    # If the route requires a sub-domain, and we have it, restore
+                    # it
+                    if 'sub_domain' in route.defaults:
+                        newargs['sub_domain'] = sub
+                    
+            elif use_current:
+                newargs = _screenargs(kargs, self.mapper, self.environ, force_explicit=True)
+            elif 'sub_domain' in kargs:
+                newargs = _subdomain_check(kargs, self.mapper, self.environ)
+            else:
+                newargs = kargs
+            
+            anchor = anchor or newargs.pop('_anchor', None)
+            host = host or newargs.pop('_host', None)
+            protocol = protocol or newargs.pop('_protocol', None)
+            url = self.mapper.generate(*route_args, **newargs)
+        if anchor is not None:
+            url += '#' + _url_quote(anchor, encoding)
+        if host or protocol or qualified:
+            if 'routes.cached_hostinfo' not in self.environ:
+                cache_hostinfo(self.environ)
+            hostinfo = self.environ['routes.cached_hostinfo']
+            
+            if not host and not qualified:
+                # Ensure we don't use a specific port, as changing the protocol
+                # means that we most likely need a new port
+                host = hostinfo['host'].split(':')[0]
+            elif not host:
+                host = hostinfo['host']
+            if not protocol:
+                protocol = hostinfo['protocol']
+            if url is not None:
+                if host[-1] != '/':
+                    host += '/'
+                url = protocol + '://' + host + url.lstrip('/')
+
+        if not isinstance(url, str) and url is not None:
+            raise GenerationException("Can only return a string, got "
+                            "unicode instead: %s" % url)
+        if url is None:
+            raise GenerationException(
+                "Could not generate URL. Called with args: %s %s" % \
+                (args, kargs))
+        return url
+    
+    def current(self, *args, **kwargs):
+        """Generate a route that includes params used on the current
+        request
+        
+        The arguments for this method are identical to ``__call__``
+        except that arguments set to None will remove existing route
+        matches of the same name from the set of arguments used to
+        construct a URL.
+        """
+        return self(_use_current=True, *args, **kwargs)
+
+
+def redirect_to(*args, **kargs):
+    """Issues a redirect based on the arguments. 
+    
+    Redirect's *should* occur as a "302 Moved" header, however the web 
+    framework may utilize a different method.
+    
+    All arguments are passed to url_for to retrieve the appropriate URL, then
+    the resulting URL it sent to the redirect function as the URL.
+    """
+    target = url_for(*args, **kargs)
+    config = request_config()
+    return config.redirect(target)
+
+
+def cache_hostinfo(environ):
+    """Processes the host information and stores a copy
+    
+    This work was previously done but wasn't stored in environ, nor is
+    it guaranteed to be setup in the future (Routes 2 and beyond).
+    
+    cache_hostinfo processes environ keys that may be present to
+    determine the proper host, protocol, and port information to use
+    when generating routes.
+    
+    """
+    hostinfo = {}
+    if environ.get('HTTPS') or environ.get('wsgi.url_scheme') == 'https' \
+       or environ.get('HTTP_X_FORWARDED_PROTO') == 'https':
+        hostinfo['protocol'] = 'https'
+    else:
+        hostinfo['protocol'] = 'http'
+    if environ.get('HTTP_X_FORWARDED_HOST'):
+        hostinfo['host'] = environ['HTTP_X_FORWARDED_HOST']
+    elif environ.get('HTTP_HOST'):
+        hostinfo['host'] = environ['HTTP_HOST']
+    else:
+        hostinfo['host'] = environ['SERVER_NAME']
+        if environ.get('wsgi.url_scheme') == 'https':
+            if environ['SERVER_PORT'] != '443':
+                hostinfo['host'] += ':' + environ['SERVER_PORT']
+        else:
+            if environ['SERVER_PORT'] != '80':
+                hostinfo['host'] += ':' + environ['SERVER_PORT']
+    environ['routes.cached_hostinfo'] = hostinfo
+    return hostinfo
+
+
+def controller_scan(directory=None):
+    """Scan a directory for python files and use them as controllers"""
+    if directory is None:
+        return []
+    
+    def find_controllers(dirname, prefix=''):
+        """Locate controllers in a directory"""
+        controllers = []
+        for fname in os.listdir(dirname):
+            filename = os.path.join(dirname, fname)
+            if os.path.isfile(filename) and \
+                re.match('^[^_]{1,1}.*\.py$', fname):
+                controllers.append(prefix + fname[:-3])
+            elif os.path.isdir(filename):
+                controllers.extend(find_controllers(filename, 
+                                                    prefix=prefix+fname+'/'))
+        return controllers
+    def longest_first(fst, lst):
+        """Compare the length of one string to another, shortest goes first"""
+        return cmp(len(lst), len(fst))
+    controllers = find_controllers(directory)
+    controllers.sort(longest_first)
+    return controllers