Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Refactor metadata object to support custom columns

Browse Source
This commit is contained in:
Kovid Goyal 2010-09-09 09:26:03 -06:00
commit 1e4baacd51
25 changed files with 778 additions and 520 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
resources/content_server/gui.js
View File

@ -59,14 +59,44 @@ function render_book(book) {
title = title.slice(0, title.length-2);
title += ' ({0} MB) '.format(size);
}
title += '<span class="tagdata_short" style="display:all">'
if (tags) {
t = tags.split(':&:', 2);
m = parseInt(t[0]);
t = t[1].split(',', m);
if (t.length == m) t[m] = '...'
title += 'Tags=[{0}] '.format(t.join(','));
}
custcols = book.attr("custcols").split(',')
for ( i = 0; i < custcols.length; i++) {
if (custcols[i].length > 0) {
vals = book.attr(custcols[i]).split(':#:', 2);
if (vals[0].indexOf('#T#') == 0) { //startswith
vals[0] = vals[0].substr(3, vals[0].length)
t = vals[1].split(':&:', 2);
m = parseInt(t[0]);
t = t[1].split(',', m);
if (t.length == m) t[m] = '...';
vals[1] = t.join(',');
}
title += '{0}=[{1}] '.format(vals[0], vals[1]);
}
}
title += '</span>'
title += '<span class="tagdata_long" style="display:none">'
if (tags) title += 'Tags=[{0}] '.format(tags);
custcols = book.attr("custcols").split(',')
for ( i = 0; i < custcols.length; i++) {
if (custcols[i].length > 0) {
vals = book.attr(custcols[i]).split(':#:', 2);
if (vals[0].indexOf('#T#') == 0) { //startswith
vals[0] = vals[0].substr(3, vals[0].length)
vals[1] = (vals[1].split(':&:', 2))[1];
}
title += '{0}=[{1}] '.format(vals[0], vals[1]);
}
}
title += '</span>'
title += '<img style="display:none" alt="" src="get/cover/{0}" /></span>'.format(id);
title += '<div class="comments">{0}</div>'.format(comments)
// Render authors cell
@ -170,11 +200,15 @@ function fetch_library_books(start, num, timeout, sort, order, search) {
var cover = row.find('img').attr('src');
var collapsed = row.find('.comments').css('display') == 'none';
$("#book_list tbody tr * .comments").css('display', 'none');
$("#book_list tbody tr * .tagdata_short").css('display', 'inherit');
$("#book_list tbody tr * .tagdata_long").css('display', 'none');
$('#cover_pane').css('visibility', 'hidden');
if (collapsed) {
row.find('.comments').css('display', 'inherit');
$('#cover_pane img').attr('src', cover);
$('#cover_pane').css('visibility', 'visible');
row.find(".tagdata_short").css('display', 'none');
row.find(".tagdata_long").css('display', 'inherit');
}
});

32
resources/default_tweaks.py
View File

@ -93,6 +93,37 @@ save_template_title_series_sorting = 'library_order'
auto_connect_to_folder = ''
# Specify renaming rules for sony collections. Collections on Sonys are named
# depending upon whether the field is standard or custom. A collection derived
# from a standard field is named for the value in that field. For example, if
# the standard 'series' column contains the name 'Darkover', then the series
# will be named 'Darkover'. A collection derived from a custom field will have
# the name of the field added to the value. For example, if a custom series
# column named 'My Series' contains the name 'Darkover', then the collection
# will be named 'Darkover (My Series)'. If two books have fields that generate
# the same collection name, then both books will be in that collection. This
# tweak lets you specify for a standard or custom field the value to be put
# inside the parentheses. You can use it to add a parenthetical description to a
# standard field, for example 'Foo (Tag)' instead of the 'Foo'. You can also use
# it to force multiple fields to end up in the same collection. For example, you
# could force the values in 'series', '#my_series_1', and '#my_series_2' to
# appear in collections named 'some_value (Series)', thereby merging all of the
# fields into one set of collections. The syntax of this tweak is
# {'field_lookup_name':'name_to_use', 'lookup_name':'name', ...}
# Example 1: I want three series columns to be merged into one set of
# collections. If the column lookup names are 'series', '#series_1' and
# '#series_2', and if I want nothing in the parenthesis, then the value to use
# in the tweak value would be:
# sony_collection_renaming_rules={'series':'', '#series_1':'', '#series_2':''}
# Example 2: I want the word '(Series)' to appear on collections made from
# series, and the word '(Tag)' to appear on collections made from tags. Use:
# sony_collection_renaming_rules={'series':'Series', 'tags':'Tag'}
# Example 3: I want 'series' and '#myseries' to be merged, and for the
# collection name to have '(Series)' appended. The renaming rule is:
# sony_collection_renaming_rules={'series':'Series', '#myseries':'Series'}
sony_collection_renaming_rules={}
# Create search terms to apply a query across several built-in search terms.
# Syntax: {'new term':['existing term 1', 'term 2', ...], 'new':['old'...] ...}
# Example: create the term 'myseries' that when used as myseries:foo would
@ -114,3 +145,4 @@ add_new_book_tags_when_importing_books = False
# Set the maximum number of tags to show per book in the content server
max_content_server_tags_shown=5

4
src/calibre/customize/__init__.py
View File

@ -218,7 +218,7 @@ class MetadataReaderPlugin(Plugin): # {{{
with the input data.
:param type: The type of file. Guaranteed to be one of the entries
in :attr:`file_types`.
:return: A :class:`calibre.ebooks.metadata.MetaInformation` object
:return: A :class:`calibre.ebooks.metadata.book.Metadata` object
'''
return None
# }}}
@ -248,7 +248,7 @@ class MetadataWriterPlugin(Plugin): # {{{
with the input data.
:param type: The type of file. Guaranteed to be one of the entries
in :attr:`file_types`.
:param mi: A :class:`calibre.ebooks.metadata.MetaInformation` object
:param mi: A :class:`calibre.ebooks.metadata.book.Metadata` object
'''
pass

9
src/calibre/devices/apple/driver.py
View File

@ -13,7 +13,8 @@ from calibre.devices.errors import UserFeedback
from calibre.devices.usbms.deviceconfig import DeviceConfig
from calibre.devices.interface import DevicePlugin
from calibre.ebooks.BeautifulSoup import BeautifulSoup
from calibre.ebooks.metadata import MetaInformation, authors_to_string
from calibre.ebooks.metadata import authors_to_string, MetaInformation
from calibre.ebooks.metadata.book.base import Metadata
from calibre.ebooks.metadata.epub import set_metadata
from calibre.library.server.utils import strftime
from calibre.utils.config import config_dir
@ -871,7 +872,7 @@ class ITUNES(DriverBase):
once uploaded to the device. len(names) == len(files)
:return: A list of 3-element tuples. The list is meant to be passed
to L{add_books_to_metadata}.
:metadata: If not None, it is a list of :class:`MetaInformation` objects.
:metadata: If not None, it is a list of :class:`Metadata` objects.
The idea is to use the metadata to determine where on the device to
put the book. len(metadata) == len(files). Apart from the regular
cover (path to cover), there may also be a thumbnail attribute, which should
@ -2999,14 +3000,14 @@ class BookList(list):
'''
return {}
class Book(MetaInformation):
class Book(Metadata):
'''
A simple class describing a book in the iTunes Books Library.
- See ebooks.metadata.__init__ for all fields
'''
def __init__(self,title,author):
MetaInformation.__init__(self, title, authors=[author])
Metadata.__init__(self, title, authors=[author])
@dynamic_property
def title_sorter(self):

4
src/calibre/devices/interface.py
View File

@ -316,7 +316,7 @@ class DevicePlugin(Plugin):
being uploaded to the device.
:param names: A list of file names that the books should have
once uploaded to the device. len(names) == len(files)
:param metadata: If not None, it is a list of :class:`MetaInformation` objects.
:param metadata: If not None, it is a list of :class:`Metadata` objects.
The idea is to use the metadata to determine where on the device to
put the book. len(metadata) == len(files). Apart from the regular
cover (path to cover), there may also be a thumbnail attribute, which should
@ -335,7 +335,7 @@ class DevicePlugin(Plugin):
the device.
:param locations: Result of a call to L{upload_books}
:param metadata: List of :class:`MetaInformation` objects, same as for
:param metadata: List of :class:`Metadata` objects, same as for
:meth:`upload_books`.
:param booklists: A tuple containing the result of calls to
(:meth:`books(oncard=None)`,

83
src/calibre/devices/kobo/books.py
View File

@ -4,37 +4,15 @@ __copyright__ = '2010, Timothy Legge <timlegge at gmail.com>'
'''
import os
import re
import time
from calibre.ebooks.metadata import MetaInformation
from calibre.constants import filesystem_encoding, preferred_encoding
from calibre import isbytestring
from calibre.devices.usbms.books import Book as Book_
class Book(MetaInformation):
class Book(Book_):
BOOK_ATTRS = ['lpath', 'size', 'mime', 'device_collections', '_new_book']
JSON_ATTRS = [
'lpath', 'title', 'authors', 'mime', 'size', 'tags', 'author_sort',
'title_sort', 'comments', 'category', 'publisher', 'series',
'series_index', 'rating', 'isbn', 'language', 'application_id',
'book_producer', 'lccn', 'lcc', 'ddc', 'rights', 'publication_type',
'uuid', 'device_collections',
]
def __init__(self, prefix, lpath, title, authors, mime, date, ContentType, thumbnail_name, size=None, other=None):
MetaInformation.__init__(self, '')
self.device_collections = []
self._new_book = False
self.path = os.path.join(prefix, lpath)
if os.sep == '\\':
self.path = self.path.replace('/', '\\')
self.lpath = lpath.replace('\\', '/')
else:
self.lpath = lpath
def __init__(self, prefix, lpath, title, authors, mime, date, ContentType,
thumbnail_name, size=None, other=None):
Book_.__init__(self, prefix, lpath)
self.title = title
if not authors:
@ -51,63 +29,12 @@ class Book(MetaInformation):
self.datetime = time.gmtime(os.path.getctime(self.path))
except:
self.datetime = time.gmtime()
if thumbnail_name is not None:
self.thumbnail = ImageWrapper(thumbnail_name)
self.tags = []
if other:
self.smart_update(other)
def __eq__(self, other):
return self.path == getattr(other, 'path', None)
@dynamic_property
def db_id(self):
doc = '''The database id in the application database that this file corresponds to'''
def fget(self):
match = re.search(r'_(\d+)$', self.lpath.rpartition('.')[0])
if match:
return int(match.group(1))
return None
return property(fget=fget, doc=doc)
@dynamic_property
def title_sorter(self):
doc = '''String to sort the title. If absent, title is returned'''
def fget(self):
return re.sub('^\s*A\s+|^\s*The\s+|^\s*An\s+', '', self.title).rstrip()
return property(doc=doc, fget=fget)
@dynamic_property
def thumbnail(self):
return None
def smart_update(self, other, replace_metadata=False):
'''
Merge the information in C{other} into self. In case of conflicts, the information
in C{other} takes precedence, unless the information in C{other} is NULL.
'''
MetaInformation.smart_update(self, other)
for attr in self.BOOK_ATTRS:
if hasattr(other, attr):
val = getattr(other, attr, None)
setattr(self, attr, val)
def to_json(self):
json = {}
for attr in self.JSON_ATTRS:
val = getattr(self, attr)
if isbytestring(val):
enc = filesystem_encoding if attr == 'lpath' else preferred_encoding
val = val.decode(enc, 'replace')
elif isinstance(val, (list, tuple)):
val = [x.decode(preferred_encoding, 'replace') if
isbytestring(x) else x for x in val]
json[attr] = val
return json
class ImageWrapper(object):
def __init__(self, image_path):
self.image_path = image_path

129
src/calibre/devices/usbms/books.py
View File

@ -6,29 +6,19 @@ __docformat__ = 'restructuredtext en'
import os, re, time, sys
from calibre.ebooks.metadata import MetaInformation
from calibre.ebooks.metadata.book.base import Metadata
from calibre.devices.mime import mime_type_ext
from calibre.devices.interface import BookList as _BookList
from calibre.constants import filesystem_encoding, preferred_encoding
from calibre.constants import preferred_encoding
from calibre import isbytestring
from calibre.utils.config import prefs
class Book(MetaInformation):
BOOK_ATTRS = ['lpath', 'size', 'mime', 'device_collections', '_new_book']
JSON_ATTRS = [
'lpath', 'title', 'authors', 'mime', 'size', 'tags', 'author_sort',
'title_sort', 'comments', 'category', 'publisher', 'series',
'series_index', 'rating', 'isbn', 'language', 'application_id',
'book_producer', 'lccn', 'lcc', 'ddc', 'rights', 'publication_type',
'uuid',
]
from calibre.utils.config import prefs, tweaks
from calibre.utils.date import format_date
class Book(Metadata):
def __init__(self, prefix, lpath, size=None, other=None):
from calibre.ebooks.metadata.meta import path_to_ext
MetaInformation.__init__(self, '')
Metadata.__init__(self, '')
self._new_book = False
self.device_collections = []
@ -72,32 +62,6 @@ class Book(MetaInformation):
def thumbnail(self):
return None
def smart_update(self, other, replace_metadata=False):
'''
Merge the information in C{other} into self. In case of conflicts, the information
in C{other} takes precedence, unless the information in C{other} is NULL.
'''
MetaInformation.smart_update(self, other, replace_metadata)
for attr in self.BOOK_ATTRS:
if hasattr(other, attr):
val = getattr(other, attr, None)
setattr(self, attr, val)
def to_json(self):
json = {}
for attr in self.JSON_ATTRS:
val = getattr(self, attr)
if isbytestring(val):
enc = filesystem_encoding if attr == 'lpath' else preferred_encoding
val = val.decode(enc, 'replace')
elif isinstance(val, (list, tuple)):
val = [x.decode(preferred_encoding, 'replace') if
isbytestring(x) else x for x in val]
json[attr] = val
return json
class BookList(_BookList):
def __init__(self, oncard, prefix, settings):
@ -131,11 +95,38 @@ class CollectionsBookList(BookList):
def supports_collections(self):
return True
def compute_category_name(self, attr, category, cust_field_meta):
renames = tweaks['sony_collection_renaming_rules']
attr_name = renames.get(attr, None)
if attr_name is None:
if attr in cust_field_meta:
attr_name = '(%s)'%cust_field_meta[attr]['name']
else:
attr_name = ''
elif attr_name != '':
attr_name = '(%s)'%attr_name
if attr not in cust_field_meta:
cat_name = '%s %s'%(category, attr_name)
else:
fm = cust_field_meta[attr]
if fm['datatype'] == 'bool':
if category:
cat_name = '%s %s'%(_('Yes'), attr_name)
else:
cat_name = '%s %s'%(_('No'), attr_name)
elif fm['datatype'] == 'datetime':
cat_name = '%s %s'%(format_date(category,
fm['display'].get('date_format','dd MMM yyyy')), attr_name)
else:
cat_name = '%s %s'%(category, attr_name)
return cat_name.strip()
def get_collections(self, collection_attributes):
from calibre.devices.usbms.driver import debug_print
debug_print('Starting get_collections:', prefs['manage_device_metadata'])
debug_print('Renaming rules:', tweaks['sony_collection_renaming_rules'])
collections = {}
series_categories = set([])
# This map of sets is used to avoid linear searches when testing for
# book equality
collections_lpaths = {}
@ -161,41 +152,55 @@ class CollectionsBookList(BookList):
# For existing books, modify the collections only if the user
# specified 'on_connect'
attrs = collection_attributes
meta_vals = book.get_all_non_none_attributes()
cust_field_meta = book.get_all_user_metadata(make_copy=False)
for attr in attrs:
attr = attr.strip()
val = getattr(book, attr, None)
val = meta_vals.get(attr, None)
if not val: continue
if isbytestring(val):
val = val.decode(preferred_encoding, 'replace')
if isinstance(val, (list, tuple)):
val = list(val)
elif isinstance(val, unicode):
else:
val = [val]
for category in val:
is_series = False
if attr in cust_field_meta: # is a custom field
fm = cust_field_meta[attr]
if fm['datatype'] == 'text' and len(category) > 1 and \
category[0] == '[' and category[-1] == ']':
continue
if fm['datatype'] == 'series':
is_series = True
else: # is a standard field
if attr == 'tags' and len(category) > 1 and \
category[0] == '[' and category[-1] == ']':
continue
if category not in collections:
collections[category] = []
collections_lpaths[category] = set()
if lpath not in collections_lpaths[category]:
collections_lpaths[category].add(lpath)
collections[category].append(book)
if attr == 'series' or \
('series' in collection_attributes and
getattr(book, 'series', None) == category):
series_categories.add(category)
meta_vals.get('series', None) == category):
is_series = True
cat_name = self.compute_category_name(attr, category,
cust_field_meta)
if cat_name not in collections:
collections[cat_name] = []
collections_lpaths[cat_name] = set()
if lpath in collections_lpaths[cat_name]:
continue
collections_lpaths[cat_name].add(lpath)
if is_series:
collections[cat_name].append(
(book, meta_vals.get(attr+'_index', sys.maxint)))
else:
collections[cat_name].append(
(book, meta_vals.get('title_sort', 'zzzz')))
# Sort collections
result = {}
for category, books in collections.items():
def tgetter(x):
return getattr(x, 'title_sort', 'zzzz')
books.sort(cmp=lambda x,y:cmp(tgetter(x), tgetter(y)))
if category in series_categories:
# Ensures books are sub sorted by title
def getter(x):
return getattr(x, 'series_index', sys.maxint)
books.sort(cmp=lambda x,y:cmp(getter(x), getter(y)))
return collections
books.sort(cmp=lambda x,y:cmp(x[1], y[1]))
result[category] = [x[0] for x in books]
return result
def rebuild_collections(self, booklist, oncard):
'''

21
src/calibre/devices/usbms/driver.py
View File

@ -13,7 +13,6 @@ for a particular device.
import os
import re
import time
import json
from itertools import cycle
from calibre import prints, isbytestring
@ -21,6 +20,7 @@ from calibre.constants import filesystem_encoding, DEBUG
from calibre.devices.usbms.cli import CLI
from calibre.devices.usbms.device import Device
from calibre.devices.usbms.books import BookList, Book
from calibre.ebooks.metadata.book.json_codec import JsonCodec
BASE_TIME = None
def debug_print(*args):
@ -288,6 +288,7 @@ class USBMS(CLI, Device):
# at the end just before the return
def sync_booklists(self, booklists, end_session=True):
debug_print('USBMS: starting sync_booklists')
json_codec = JsonCodec()
if not os.path.exists(self.normalize_path(self._main_prefix)):
os.makedirs(self.normalize_path(self._main_prefix))
@ -296,10 +297,8 @@ class USBMS(CLI, Device):
if prefix is not None and isinstance(booklists[listid], self.booklist_class):
if not os.path.exists(prefix):
os.makedirs(self.normalize_path(prefix))
js = [item.to_json() for item in booklists[listid] if
hasattr(item, 'to_json')]
with open(self.normalize_path(os.path.join(prefix, self.METADATA_CACHE)), 'wb') as f:
f.write(json.dumps(js, indent=2, encoding='utf-8'))
json_codec.encode_to_file(f, booklists[listid])
write_prefix(self._main_prefix, 0)
write_prefix(self._card_a_prefix, 1)
write_prefix(self._card_b_prefix, 2)
@ -345,19 +344,13 @@ class USBMS(CLI, Device):
@classmethod
def parse_metadata_cache(cls, bl, prefix, name):
# bl = cls.booklist_class()
js = []
json_codec = JsonCodec()
need_sync = False
cache_file = cls.normalize_path(os.path.join(prefix, name))
if os.access(cache_file, os.R_OK):
try:
with open(cache_file, 'rb') as f:
js = json.load(f, encoding='utf-8')
for item in js:
book = cls.book_class(prefix, item.get('lpath', None))
for key in item.keys():
setattr(book, key, item[key])
bl.append(book)
json_codec.decode_from_file(f, bl, cls.book_class, prefix)
except:
import traceback
traceback.print_exc()
@ -392,7 +385,7 @@ class USBMS(CLI, Device):
@classmethod
def book_from_path(cls, prefix, lpath):
from calibre.ebooks.metadata import MetaInformation
from calibre.ebooks.metadata.book.base import Metadata
if cls.settings().read_metadata or cls.MUST_READ_METADATA:
mi = cls.metadata_from_path(cls.normalize_path(os.path.join(prefix, lpath)))
@ -401,7 +394,7 @@ class USBMS(CLI, Device):
mi = metadata_from_filename(cls.normalize_path(os.path.basename(lpath)),
cls.build_template_regexp())
if mi is None:
mi = MetaInformation(os.path.splitext(os.path.basename(lpath))[0],
mi = Metadata(os.path.splitext(os.path.basename(lpath))[0],
[_('Unknown')])
size = os.stat(cls.normalize_path(os.path.join(prefix, lpath))).st_size
book = cls.book_class(prefix, lpath, other=mi, size=size)

207
src/calibre/ebooks/metadata/__init__.py
View File

@ -10,10 +10,9 @@ import os, mimetypes, sys, re
from urllib import unquote, quote
from urlparse import urlparse
from calibre import relpath, prints
from calibre import relpath
from calibre.utils.config import tweaks
from calibre.utils.date import isoformat
_author_pat = re.compile(',?\s+(and|with)\s+', re.IGNORECASE)
def string_to_authors(raw):
@ -221,214 +220,18 @@ class ResourceCollection(object):
class MetaInformation(object):
'''Convenient encapsulation of book metadata'''
@staticmethod
def copy(mi):
ans = MetaInformation(mi.title, mi.authors)
for attr in ('author_sort', 'title_sort', 'comments', 'category',
'publisher', 'series', 'series_index', 'rating',
'isbn', 'tags', 'cover_data', 'application_id', 'guide',
'manifest', 'spine', 'toc', 'cover', 'language',
'book_producer', 'timestamp', 'lccn', 'lcc', 'ddc',
'author_sort_map',
'pubdate', 'rights', 'publication_type', 'uuid'):
if hasattr(mi, attr):
setattr(ans, attr, getattr(mi, attr))
def __init__(self, title, authors=(_('Unknown'),)):
'''
def MetaInformation(title, authors=(_('Unknown'),)):
''' Convenient encapsulation of book metadata, needed for compatibility
@param title: title or ``_('Unknown')`` or a MetaInformation object
@param authors: List of strings or []
'''
from calibre.ebooks.metadata.book.base import Metadata
mi = None
if hasattr(title, 'title') and hasattr(title, 'authors'):
mi = title
title = mi.title
authors = mi.authors
self.title = title
self.author = list(authors) if authors else []# Needed for backward compatibility
#: List of strings or []
self.authors = list(authors) if authors else []
self.tags = getattr(mi, 'tags', [])
#: mi.cover_data = (ext, data)
self.cover_data = getattr(mi, 'cover_data', (None, None))
self.author_sort_map = getattr(mi, 'author_sort_map', {})
for x in ('author_sort', 'title_sort', 'comments', 'category', 'publisher',
'series', 'series_index', 'rating', 'isbn', 'language',
'application_id', 'manifest', 'toc', 'spine', 'guide', 'cover',
'book_producer', 'timestamp', 'lccn', 'lcc', 'ddc', 'pubdate',
'rights', 'publication_type', 'uuid',
):
setattr(self, x, getattr(mi, x, None))
def print_all_attributes(self):
for x in ('title','author', 'author_sort', 'title_sort', 'comments', 'category', 'publisher',
'series', 'series_index', 'tags', 'rating', 'isbn', 'language',
'application_id', 'manifest', 'toc', 'spine', 'guide', 'cover',
'book_producer', 'timestamp', 'lccn', 'lcc', 'ddc', 'pubdate',
'rights', 'publication_type', 'uuid', 'author_sort_map'
):
prints(x, getattr(self, x, 'None'))
def smart_update(self, mi, replace_metadata=False):
'''
Merge the information in C{mi} into self. In case of conflicts, the
information in C{mi} takes precedence, unless the information in mi is
NULL. If replace_metadata is True, then the information in mi always
takes precedence.
'''
if mi.title and mi.title != _('Unknown'):
self.title = mi.title
if mi.authors and mi.authors[0] != _('Unknown'):
self.authors = mi.authors
for attr in ('author_sort', 'title_sort', 'category',
'publisher', 'series', 'series_index', 'rating',
'isbn', 'application_id', 'manifest', 'spine', 'toc',
'cover', 'guide', 'book_producer',
'timestamp', 'lccn', 'lcc', 'ddc', 'pubdate', 'rights',
'publication_type', 'uuid'):
if replace_metadata:
setattr(self, attr, getattr(mi, attr, 1.0 if \
attr == 'series_index' else None))
elif hasattr(mi, attr):
val = getattr(mi, attr)
if val is not None:
setattr(self, attr, val)
if replace_metadata:
self.tags = mi.tags
elif mi.tags:
self.tags += mi.tags
self.tags = list(set(self.tags))
if mi.author_sort_map:
self.author_sort_map.update(mi.author_sort_map)
if getattr(mi, 'cover_data', False):
other_cover = mi.cover_data[-1]
self_cover = self.cover_data[-1] if self.cover_data else ''
if not self_cover: self_cover = ''
if not other_cover: other_cover = ''
if len(other_cover) > len(self_cover):
self.cover_data = mi.cover_data
if replace_metadata:
self.comments = getattr(mi, 'comments', '')
else:
my_comments = getattr(self, 'comments', '')
other_comments = getattr(mi, 'comments', '')
if not my_comments:
my_comments = ''
if not other_comments:
other_comments = ''
if len(other_comments.strip()) > len(my_comments.strip()):
self.comments = other_comments
other_lang = getattr(mi, 'language', None)
if other_lang and other_lang.lower() != 'und':
self.language = other_lang
def format_series_index(self):
try:
x = float(self.series_index)
except ValueError:
x = 1
return fmt_sidx(x)
def authors_from_string(self, raw):
self.authors = string_to_authors(raw)
def format_authors(self):
return authors_to_string(self.authors)
def format_tags(self):
return u', '.join([unicode(t) for t in self.tags])
def format_rating(self):
return unicode(self.rating)
def __unicode__(self):
ans = []
def fmt(x, y):
ans.append(u'%-20s: %s'%(unicode(x), unicode(y)))
fmt('Title', self.title)
if self.title_sort:
fmt('Title sort', self.title_sort)
if self.authors:
fmt('Author(s)', authors_to_string(self.authors) + \
((' [' + self.author_sort + ']') if self.author_sort else ''))
if self.publisher:
fmt('Publisher', self.publisher)
if getattr(self, 'book_producer', False):
fmt('Book Producer', self.book_producer)
if self.category:
fmt('Category', self.category)
if self.comments:
fmt('Comments', self.comments)
if self.isbn:
fmt('ISBN', self.isbn)
if self.tags:
fmt('Tags', u', '.join([unicode(t) for t in self.tags]))
if self.series:
fmt('Series', self.series + ' #%s'%self.format_series_index())
if self.language:
fmt('Language', self.language)
if self.rating is not None:
fmt('Rating', self.rating)
if self.timestamp is not None:
fmt('Timestamp', isoformat(self.timestamp))
if self.pubdate is not None:
fmt('Published', isoformat(self.pubdate))
if self.rights is not None:
fmt('Rights', unicode(self.rights))
if self.lccn:
fmt('LCCN', unicode(self.lccn))
if self.lcc:
fmt('LCC', unicode(self.lcc))
if self.ddc:
fmt('DDC', unicode(self.ddc))
return u'\n'.join(ans)
def to_html(self):
ans = [(_('Title'), unicode(self.title))]
ans += [(_('Author(s)'), (authors_to_string(self.authors) if self.authors else _('Unknown')))]
ans += [(_('Publisher'), unicode(self.publisher))]
ans += [(_('Producer'), unicode(self.book_producer))]
ans += [(_('Comments'), unicode(self.comments))]
ans += [('ISBN', unicode(self.isbn))]
if self.lccn:
ans += [('LCCN', unicode(self.lccn))]
if self.lcc:
ans += [('LCC', unicode(self.lcc))]
if self.ddc:
ans += [('DDC', unicode(self.ddc))]
ans += [(_('Tags'), u', '.join([unicode(t) for t in self.tags]))]
if self.series:
ans += [(_('Series'), unicode(self.series)+ ' #%s'%self.format_series_index())]
ans += [(_('Language'), unicode(self.language))]
if self.timestamp is not None:
ans += [(_('Timestamp'), unicode(self.timestamp.isoformat(' ')))]
if self.pubdate is not None:
ans += [(_('Published'), unicode(self.pubdate.isoformat(' ')))]
if self.rights is not None:
ans += [(_('Rights'), unicode(self.rights))]
for i, x in enumerate(ans):
ans[i] = u'<tr><td><b>%s</b></td><td>%s</td></tr>'%x
return u'<table>%s</table>'%u'\n'.join(ans)
def __str__(self):
return self.__unicode__().encode('utf-8')
def __nonzero__(self):
return bool(self.title or self.author or self.comments or self.tags)
return Metadata(title, authors, mi)
def check_isbn10(isbn):
try:

90
src/calibre/ebooks/metadata/book/__init__.py
View File

@ -12,47 +12,44 @@ an empty list/dictionary for complex types and (None, None) for cover_data
SOCIAL_METADATA_FIELDS = frozenset([
'tags', # Ordered list
# A floating point number between 0 and 10
'rating',
# A simple HTML enabled string
'comments',
# A simple string
'series',
# A floating point number
'series_index',
'rating', # A floating point number between 0 and 10
'comments', # A simple HTML enabled string
'series', # A simple string
'series_index', # A floating point number
# Of the form { scheme1:value1, scheme2:value2}
# For example: {'isbn':'123456789', 'doi':'xxxx', ... }
'classifiers',
'isbn', # Pseudo field for convenience, should get/set isbn classifier
])
'''
The list of names that convert to classifiers when in get and set.
'''
TOP_LEVEL_CLASSIFIERS = frozenset([
'isbn',
])
PUBLICATION_METADATA_FIELDS = frozenset([
# title must never be None. Should be _('Unknown')
'title',
'title', # title must never be None. Should be _('Unknown')
# Pseudo field that can be set, but if not set is auto generated
# from title and languages
'title_sort',
# Ordered list of authors. Must never be None, can be [_('Unknown')]
'authors',
# Map of sort strings for each author
'author_sort_map',
'authors', # Ordered list. Must never be None, can be [_('Unknown')]
'author_sort_map', # Map of sort strings for each author
# Pseudo field that can be set, but if not set is auto generated
# from authors and languages
'author_sort',
'book_producer',
# Dates and times must be timezone aware
'timestamp',
'timestamp', # Dates and times must be timezone aware
'pubdate',
'rights',
# So far only known publication type is periodical:calibre
# If None, means book
'publication_type',
# A UUID usually of type 4
'uuid',
'uuid', # A UUID usually of type 4
'language', # the primary language of this book
'languages', # ordered list
# Simple string, no special semantics
'publisher',
'publisher', # Simple string, no special semantics
# Absolute path to image file encoded in filesystem_encoding
'cover',
# Of the form (format, data) where format is, for e.g. 'jpeg', 'png', 'gif'...
@ -69,33 +66,56 @@ BOOK_STRUCTURE_FIELDS = frozenset([
])
USER_METADATA_FIELDS = frozenset([
# A dict of a form to be specified
# A dict of dicts similar to field_metadata. Each field description dict
# also contains a value field with the key #value#.
'user_metadata',
])
DEVICE_METADATA_FIELDS = frozenset([
# Ordered list of strings
'device_collections',
'device_collections', # Ordered list of strings
'lpath', # Unicode, / separated
# In bytes
'size',
# Mimetype of the book file being represented
'mime',
'size', # In bytes
'mime', # Mimetype of the book file being represented
])
CALIBRE_METADATA_FIELDS = frozenset([
# An application id
# Semantics to be defined. Is it a db key? a db name + key? A uuid?
'application_id',
'application_id', # An application id, currently set to the db_id.
# the calibre primary key of the item.
'db_id', # the calibre primary key of the item.
# TODO: NEWMETA: May want to remove once Sony's no longer use it
]
)
ALL_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)
# All fields except custom fields
STANDARD_METADATA_FIELDS = SOCIAL_METADATA_FIELDS.union(
PUBLICATION_METADATA_FIELDS).union(
BOOK_STRUCTURE_FIELDS).union(
DEVICE_METADATA_FIELDS).union(
CALIBRE_METADATA_FIELDS)
# Metadata fields that smart update should copy without special handling
COPYABLE_METADATA_FIELDS = SOCIAL_METADATA_FIELDS.union(
PUBLICATION_METADATA_FIELDS).union(
BOOK_STRUCTURE_FIELDS).union(
DEVICE_METADATA_FIELDS).union(
CALIBRE_METADATA_FIELDS) - \
frozenset(['title', 'title_sort', 'authors',
'author_sort', 'author_sort_map' 'comments',
'cover_data', 'tags', 'language', 'lpath',
'size'])
SERIALIZABLE_FIELDS = SOCIAL_METADATA_FIELDS.union(
USER_METADATA_FIELDS).union(
PUBLICATION_METADATA_FIELDS).union(
CALIBRE_METADATA_FIELDS).union(
frozenset(['lpath'])) # I don't think we need device_collections
# Serialization of covers/thumbnails will have to be handled carefully, maybe
# as an option to the serializer class
DEVICE_METADATA_FIELDS) - \
frozenset(['device_collections'])
# device_collections is rebuilt when needed

340
src/calibre/ebooks/metadata/book/base.py
View File

@ -6,8 +6,15 @@ __copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'
import copy
import traceback
from calibre import prints
from calibre.ebooks.metadata.book import COPYABLE_METADATA_FIELDS
from calibre.ebooks.metadata.book import STANDARD_METADATA_FIELDS
from calibre.ebooks.metadata.book import TOP_LEVEL_CLASSIFIERS
from calibre.utils.date import isoformat, format_date
from calibre.ebooks.metadata.book import RESERVED_METADATA_FIELDS
NULL_VALUES = {
'user_metadata': {},
@ -19,103 +26,356 @@ NULL_VALUES = {
'author_sort_map': {},
'authors' : [_('Unknown')],
'title' : _('Unknown'),
'language' : 'und'
}
class Metadata(object):
'''
This class must expose a superset of the API of MetaInformation in terms
of attribute access and methods. Only the __init__ method is different.
MetaInformation will simply become a function that creates and fills in
the attributes of this class.
A class representing all the metadata for a book.
Please keep the method based API of this class to a minimum. Every method
becomes a reserved field name.
'''
def __init__(self):
def __init__(self, title, authors=(_('Unknown'),), other=None):
'''
@param title: title or ``_('Unknown')``
@param authors: List of strings or []
@param other: None or a metadata object
'''
object.__setattr__(self, '_data', copy.deepcopy(NULL_VALUES))
if other is not None:
self.smart_update(other)
else:
if title:
self.title = title
if authors:
#: List of strings or []
self.author = list(authors) if authors else []# Needed for backward compatibility
self.authors = list(authors) if authors else []
def __getattribute__(self, field):
_data = object.__getattribute__(self, '_data')
if field in RESERVED_METADATA_FIELDS:
if field in TOP_LEVEL_CLASSIFIERS:
return _data.get('classifiers').get(field, None)
if field in STANDARD_METADATA_FIELDS:
return _data.get(field, None)
try:
return object.__getattribute__(self, field)
except AttributeError:
pass
if field in _data['user_metadata'].iterkeys():
# TODO: getting user metadata values
pass
return _data['user_metadata'][field]['#value#']
raise AttributeError(
'Metadata object has no attribute named: '+ repr(field))
def __setattr__(self, field, val):
def __setattr__(self, field, val, extra=None):
_data = object.__getattribute__(self, '_data')
if field in RESERVED_METADATA_FIELDS:
if field != 'user_metadata':
if not val:
val = NULL_VALUES[field]
if field in TOP_LEVEL_CLASSIFIERS:
_data['classifiers'].update({field: val})
elif field in STANDARD_METADATA_FIELDS:
if val is None:
val = NULL_VALUES.get(field, None)
_data[field] = val
else:
raise AttributeError('You cannot set user_metadata directly.')
elif field in _data['user_metadata'].iterkeys():
# TODO: Setting custom column values
pass
_data['user_metadata'][field]['#value#'] = val
_data['user_metadata'][field]['#extra#'] = extra
else:
# You are allowed to stick arbitrary attributes onto this object as
# long as they dont conflict with global or user metadata names
# long as they don't conflict with global or user metadata names
# Don't abuse this privilege
self.__dict__[field] = val
def get(self, field, default=None):
if default is not None:
try:
return self.__getattribute__(field)
except AttributeError:
return default
return self.__getattribute__(field)
def get_extra(self, field):
_data = object.__getattribute__(self, '_data')
if field in _data['user_metadata'].iterkeys():
return _data['user_metadata'][field]['#extra#']
raise AttributeError(
'Metadata object has no attribute named: '+ repr(field))
def set(self, field, val, extra=None):
self.__setattr__(field, val, extra)
@property
def user_metadata_names(self):
def user_metadata_keys(self):
'The set of user metadata names this object knows about'
_data = object.__getattribute__(self, '_data')
return frozenset(_data['user_metadata'].iterkeys())
# Old MetaInformation API {{{
def copy(self):
pass
def get_all_user_metadata(self, make_copy):
'''
return a dict containing all the custom field metadata associated with
the book.
'''
_data = object.__getattribute__(self, '_data')
user_metadata = _data['user_metadata']
if not make_copy:
return user_metadata
res = {}
for k in user_metadata:
res[k] = copy.deepcopy(user_metadata[k])
return res
def get_user_metadata(self, field, make_copy):
'''
return field metadata from the object if it is there. Otherwise return
None. field is the key name, not the label. Return a copy if requested,
just in case the user wants to change values in the dict.
'''
_data = object.__getattribute__(self, '_data')
_data = _data['user_metadata']
if field in _data:
if make_copy:
return copy.deepcopy(_data[field])
return _data[field]
return None
def set_all_user_metadata(self, metadata):
'''
store custom field metadata into the object. Field is the key name
not the label
'''
if metadata is None:
traceback.print_stack()
else:
for key in metadata:
self.set_user_metadata(key, metadata[key])
def set_user_metadata(self, field, metadata):
'''
store custom field metadata for one column into the object. Field is
the key name not the label
'''
if field is not None:
if metadata is None:
traceback.print_stack()
metadata = copy.deepcopy(metadata)
if '#value#' not in metadata:
if metadata['datatype'] == 'text' and metadata['is_multiple']:
metadata['#value#'] = []
else:
metadata['#value#'] = None
_data = object.__getattribute__(self, '_data')
_data['user_metadata'][field] = metadata
def get_all_non_none_attributes(self):
'''
Return a dictionary containing all non-None metadata fields, including
the custom ones.
'''
result = {}
_data = object.__getattribute__(self, '_data')
for attr in STANDARD_METADATA_FIELDS:
v = _data.get(attr, None)
if v is not None:
result[attr] = v
for attr in _data['user_metadata'].iterkeys():
v = _data['user_metadata'][attr]['#value#']
if v is not None:
result[attr] = v
if _data['user_metadata'][attr]['datatype'] == 'series':
result[attr+'_index'] = _data['user_metadata'][attr]['#extra#']
return result
# Old Metadata API {{{
def print_all_attributes(self):
pass
for x in STANDARD_METADATA_FIELDS:
prints('%s:'%x, getattr(self, x, 'None'))
for x in self.user_metadata_keys:
meta = self.get_user_metadata(x, make_copy=False)
if meta is not None:
prints(x, meta)
prints('--------------')
def smart_update(self, other, replace_metadata=False):
pass
'''
Merge the information in C{other} into self. In case of conflicts, the information
in C{other} takes precedence, unless the information in other is NULL.
'''
def copy_not_none(dest, src, attr):
v = getattr(src, attr, None)
if v is not None:
setattr(dest, attr, copy.deepcopy(v))
def format_series_index(self):
pass
if other.title and other.title != _('Unknown'):
self.title = other.title
if hasattr(other, 'title_sort'):
self.title_sort = other.title_sort
if other.authors and other.authors[0] != _('Unknown'):
self.authors = other.authors
if hasattr(other, 'author_sort_map'):
self.author_sort_map = other.author_sort_map
if hasattr(other, 'author_sort'):
self.author_sort = other.author_sort
if replace_metadata:
for attr in COPYABLE_METADATA_FIELDS:
setattr(self, attr, getattr(other, attr, 1.0 if \
attr == 'series_index' else None))
self.tags = other.tags
self.cover_data = getattr(other, 'cover_data', '')
self.set_all_user_metadata(other.get_all_user_metadata(make_copy=True))
copy_not_none(self, other, 'lpath')
copy_not_none(self, other, 'size')
copy_not_none(self, other, 'comments')
# language is handled below
else:
for attr in COPYABLE_METADATA_FIELDS:
if hasattr(other, attr):
copy_not_none(self, other, attr)
val = getattr(other, attr)
if val is not None:
setattr(self, attr, copy.deepcopy(val))
if other.tags:
self.tags += list(set(self.tags + other.tags))
if getattr(other, 'cover_data', False):
other_cover = other.cover_data[-1]
self_cover = self.cover_data[-1] if self.cover_data else ''
if not self_cover: self_cover = ''
if not other_cover: other_cover = ''
if len(other_cover) > len(self_cover):
self.cover_data = other.cover_data
if getattr(other, 'user_metadata_keys', None):
for x in other.user_metadata_keys:
meta = other.get_user_metadata(x, make_copy=True)
if meta is not None:
self.set_user_metadata(x, meta) # get... did the deepcopy
my_comments = getattr(self, 'comments', '')
other_comments = getattr(other, 'comments', '')
if not my_comments:
my_comments = ''
if not other_comments:
other_comments = ''
if len(other_comments.strip()) > len(my_comments.strip()):
self.comments = other_comments
other_lang = getattr(other, 'language', None)
if other_lang and other_lang.lower() != 'und':
self.language = other_lang
def format_series_index(self, val=None):
from calibre.ebooks.metadata import fmt_sidx
v = self.series_index if val is None else val
try:
x = float(v)
except ValueError:
x = 1
return fmt_sidx(x)
def authors_from_string(self, raw):
pass
from calibre.ebooks.metadata import string_to_authors
self.authors = string_to_authors(raw)
def format_authors(self):
pass
from calibre.ebooks.metadata import authors_to_string
return authors_to_string(self.authors)
def format_tags(self):
pass
return u', '.join([unicode(t) for t in self.tags])
def format_rating(self):
return unicode(self.rating)
def format_custom_field(self, key):
'''
returns the tuple (field_name, formatted_value)
'''
cmeta = self.get_user_metadata(key, make_copy=False)
name = unicode(cmeta['name'])
res = self.get(key, None)
if res is not None:
datatype = cmeta['datatype']
if datatype == 'text' and cmeta['is_multiple']:
res = u', '.join(res)
elif datatype == 'series':
res = res + ' [%s]'%self.format_series_index(val=self.get_extra(key))
elif datatype == 'datetime':
res = format_date(res, cmeta['display'].get('date_format','dd MMM yyyy'))
elif datatype == 'bool':
res = _('Yes') if res else _('No')
return (name, unicode(res))
def __unicode__(self):
pass
from calibre.ebooks.metadata import authors_to_string
ans = []
def fmt(x, y):
ans.append(u'%-20s: %s'%(unicode(x), unicode(y)))
fmt('Title', self.title)
if self.title_sort:
fmt('Title sort', self.title_sort)
if self.authors:
fmt('Author(s)', authors_to_string(self.authors) + \
((' [' + self.author_sort + ']') if self.author_sort else ''))
if self.publisher:
fmt('Publisher', self.publisher)
if getattr(self, 'book_producer', False):
fmt('Book Producer', self.book_producer)
if self.comments:
fmt('Comments', self.comments)
if self.isbn:
fmt('ISBN', self.isbn)
if self.tags:
fmt('Tags', u', '.join([unicode(t) for t in self.tags]))
if self.series:
fmt('Series', self.series + ' #%s'%self.format_series_index())
if self.language:
fmt('Language', self.language)
if self.rating is not None:
fmt('Rating', self.rating)
if self.timestamp is not None:
fmt('Timestamp', isoformat(self.timestamp))
if self.pubdate is not None:
fmt('Published', isoformat(self.pubdate))
if self.rights is not None:
fmt('Rights', unicode(self.rights))
for key in self.user_metadata_keys:
val = self.get(key, None)
if val is not None:
(name, val) = self.format_custom_field(key)
fmt(name, unicode(val))
return u'\n'.join(ans)
def to_html(self):
pass
from calibre.ebooks.metadata import authors_to_string
ans = [(_('Title'), unicode(self.title))]
ans += [(_('Author(s)'), (authors_to_string(self.authors) if self.authors else _('Unknown')))]
ans += [(_('Publisher'), unicode(self.publisher))]
ans += [(_('Producer'), unicode(self.book_producer))]
ans += [(_('Comments'), unicode(self.comments))]
ans += [('ISBN', unicode(self.isbn))]
ans += [(_('Tags'), u', '.join([unicode(t) for t in self.tags]))]
if self.series:
ans += [(_('Series'), unicode(self.series)+ ' #%s'%self.format_series_index())]
ans += [(_('Language'), unicode(self.language))]
if self.timestamp is not None:
ans += [(_('Timestamp'), unicode(self.timestamp.isoformat(' ')))]
if self.pubdate is not None:
ans += [(_('Published'), unicode(self.pubdate.isoformat(' ')))]
if self.rights is not None:
ans += [(_('Rights'), unicode(self.rights))]
for key in self.user_metadata_keys:
val = self.get(key, None)
if val is not None:
(name, val) = self.format_custom_field(key)
ans += [(name, val)]
for i, x in enumerate(ans):
ans[i] = u'<tr><td><b>%s</b></td><td>%s</td></tr>'%x
return u'<table>%s</table>'%u'\n'.join(ans)
def __str__(self):
return self.__unicode__().encode('utf-8')
def __nonzero__(self):
return True
return bool(self.title or self.author or self.comments or self.tags)
# }}}
# We don't need reserved field names for this object any more. Lets just use a
# protocol like the last char of a user field label should be _ when using this
# object
# So mi.tags returns the builtin tags and mi.tags_ returns the user tags

124
src/calibre/ebooks/metadata/book/json_codec.py Normal file
View File

@ -0,0 +1,124 @@
'''
Created on 4 Jun 2010
@author: charles
'''
from base64 import b64encode, b64decode
import json
import traceback
from calibre.ebooks.metadata.book import SERIALIZABLE_FIELDS
from calibre.constants import filesystem_encoding, preferred_encoding
from calibre.library.field_metadata import FieldMetadata
from calibre.utils.date import parse_date, isoformat, UNDEFINED_DATE
# Translate datetimes to and from strings. The string form is the datetime in
# UTC. The returned date is also UTC
def string_to_datetime(src):
if src == "None":
return None
# dt = strptime(src, '%d %m %Y %H:%M:%S', assume_utc=True, as_utc=True)
# if dt == UNDEFINED_DATE:
# return None
return parse_date(src)
def datetime_to_string(dateval):
if dateval is None or dateval == UNDEFINED_DATE:
return "None"
# tt = date_to_utc(dateval).timetuple()
# res = "%02d %02d %04d %02d:%02d:%02d"%(tt.tm_mday, tt.tm_mon, tt.tm_year,
# tt.tm_hour, tt.tm_min, tt.tm_sec)
return isoformat(dateval)
def encode_thumbnail(thumbnail):
'''
Encode the image part of a thumbnail, then return the 3 part tuple
'''
if thumbnail is None:
return None
return (thumbnail[0], thumbnail[1], b64encode(str(thumbnail[2])))
def decode_thumbnail(tup):
'''
Decode an encoded thumbnail into its 3 component parts
'''
if tup is None:
return None
return (tup[0], tup[1], b64decode(tup[2]))
class JsonCodec(object):
def __init__(self):
self.field_metadata = FieldMetadata()
def encode_to_file(self, file, booklist):
json.dump(self.encode_booklist_metadata(booklist), file, indent=2, encoding='utf-8')
def encode_booklist_metadata(self, booklist):
result = []
for book in booklist:
result.append(self.encode_book_metadata(book))
return result
def encode_book_metadata(self, book):
result = {}
for key in SERIALIZABLE_FIELDS:
result[key] = self.encode_metadata_attr(book, key)
return result
def encode_metadata_attr(self, book, key):
if key == 'user_metadata':
meta = book.get_all_user_metadata(make_copy=True)
for k in meta:
if meta[k]['datatype'] == 'datetime':
meta[k]['#value#'] = datetime_to_string(meta[k]['#value#'])
return meta
if key in self.field_metadata:
datatype = self.field_metadata[key]['datatype']
else:
datatype = None
value = book.get(key)
if key == 'thumbnail':
return encode_thumbnail(value)
elif isinstance(value, str): # str includes bytes
enc = filesystem_encoding if key == 'lpath' else preferred_encoding
return value.decode(enc, 'replace')
elif isinstance(value, (list, tuple)):
return [x.decode(preferred_encoding, 'replace') if
isinstance(x, str) else x for x in value]
elif datatype == 'datetime':
return datetime_to_string(value)
else:
return value
def decode_from_file(self, file, booklist, book_class, prefix):
js = []
try:
js = json.load(file, encoding='utf-8')
for item in js:
book = book_class(prefix, item.get('lpath', None))
for key in item.keys():
meta = self.decode_metadata(key, item[key])
if key == 'user_metadata':
book.set_all_user_metadata(meta)
else:
setattr(book, key, meta)
booklist.append(book)
except:
print 'exception during JSON decoding'
traceback.print_exc()
booklist = []
def decode_metadata(self, key, value):
if key == 'user_metadata':
for k in value:
if value[k]['datatype'] == 'datetime':
value[k]['#value#'] = string_to_datetime(value[k]['#value#'])
return value
elif key in self.field_metadata:
if self.field_metadata[key]['datatype'] == 'datetime':
return string_to_datetime(value)
if key == 'thumbnail':
return decode_thumbnail(value)
return value

2
src/calibre/ebooks/metadata/epub.py
View File

@ -164,7 +164,7 @@ def get_cover(opf, opf_path, stream, reader=None):
return render_html_svg_workaround(cpage, default_log)
def get_metadata(stream, extract_cover=True):
""" Return metadata as a :class:`MetaInformation` object """
""" Return metadata as a :class:`Metadata` object """
stream.seek(0)
reader = OCFZipReader(stream)
mi = MetaInformation(reader.opf)

2
src/calibre/ebooks/metadata/fetch.py
View File

@ -29,7 +29,7 @@ class MetadataSource(Plugin): # {{{
future use.
The fetch method must store the results in `self.results` as a list of
:class:`MetaInformation` objects. If there is an error, it should be stored
:class:`Metadata` objects. If there is an error, it should be stored
in `self.exception` and `self.tb` (for the traceback).
'''

6
src/calibre/ebooks/metadata/isbndb.py
View File

@ -8,7 +8,7 @@ import sys, re
from urllib import quote
from calibre.utils.config import OptionParser
from calibre.ebooks.metadata import MetaInformation
from calibre.ebooks.metadata.book.base import Metadata
from calibre.ebooks.BeautifulSoup import BeautifulStoneSoup
from calibre import browser
@ -42,10 +42,10 @@ def fetch_metadata(url, max=100, timeout=5.):
return books
class ISBNDBMetadata(MetaInformation):
class ISBNDBMetadata(Metadata):
def __init__(self, book):
MetaInformation.__init__(self, None, [])
Metadata.__init__(self, None, [])
self.isbn = book.get('isbn13', book.get('isbn'))
self.title = book.find('titlelong').string

11
src/calibre/ebooks/metadata/opf2.py
View File

@ -16,7 +16,8 @@ from lxml import etree
from calibre.ebooks.chardet import xml_to_unicode
from calibre.constants import __appname__, __version__, filesystem_encoding
from calibre.ebooks.metadata.toc import TOC
from calibre.ebooks.metadata import MetaInformation, string_to_authors
from calibre.ebooks.metadata import string_to_authors, MetaInformation
from calibre.ebooks.metadata.book.base import Metadata
from calibre.utils.date import parse_date, isoformat
from calibre.utils.localization import get_lang
@ -926,16 +927,16 @@ class OPF(object):
setattr(self, attr, val)
class OPFCreator(MetaInformation):
class OPFCreator(Metadata):
def __init__(self, base_path, *args, **kwargs):
def __init__(self, base_path, other):
'''
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)
Metadata.__init__(self, title='', other=other)
self.base_path = os.path.abspath(base_path)
if self.application_id is None:
self.application_id = str(uuid.uuid4())
@ -1187,7 +1188,7 @@ def metadata_to_opf(mi, as_string=True):
factory(DC('contributor'), mi.book_producer, __appname__, 'bkp')
if hasattr(mi.pubdate, 'isoformat'):
factory(DC('date'), isoformat(mi.pubdate))
if mi.category:
if hasattr(mi, 'category') and mi.category:
factory(DC('type'), mi.category)
if mi.comments:
factory(DC('description'), mi.comments)

2
src/calibre/gui2/book_details.py
View File

@ -28,6 +28,8 @@ WEIGHTS[_('Tags')] = 4
def render_rows(data):
keys = data.keys()
# First sort by name. The WEIGHTS sort will preserve this sub-order
keys.sort(cmp=lambda x, y: cmp(x.lower(), y.lower()))
keys.sort(cmp=lambda x, y: cmp(WEIGHTS[x], WEIGHTS[y]))
rows = []
for key in keys:

9
src/calibre/gui2/library/models.py
View File

@ -323,7 +323,11 @@ class BooksModel(QAbstractTableModel): # {{{
data[_('Series')] = \
_('Book <font face="serif">%s</font> of %s.')%\
(sidx, prepare_string_for_xml(series))
mi = self.db.get_metadata(idx)
for key in mi.user_metadata_keys:
name, val = mi.format_custom_field(key)
if val is not None:
data[name] = val
return data
def set_cache(self, idx):
@ -372,7 +376,6 @@ class BooksModel(QAbstractTableModel): # {{{
return ans
def get_metadata(self, rows, rows_are_ids=False, full_metadata=False):
# Should this add the custom columns? It doesn't at the moment
metadata, _full_metadata = [], []
if not rows_are_ids:
rows = [self.db.id(row.row()) for row in rows]
@ -1053,7 +1056,7 @@ class DeviceBooksModel(BooksModel): # {{{
if hasattr(cdata, 'image_path'):
img.load(cdata.image_path)
else:
img.loadFromData(cdata)
img.loadFromData(cdata[2])
if img.isNull():
img = self.default_image
data['cover'] = img

29
src/calibre/gui2/preferences/save_template.py
View File

@ -8,10 +8,8 @@ __docformat__ = 'restructuredtext en'
from PyQt4.Qt import QWidget, pyqtSignal
from calibre.gui2 import error_dialog
from calibre.gui2.preferences.save_template_ui import Ui_Form
from calibre.library.save_to_disk import FORMAT_ARG_DESCS, \
preprocess_template
from calibre.library.save_to_disk import FORMAT_ARG_DESCS
class SaveTemplate(QWidget, Ui_Form):
@ -41,18 +39,21 @@ class SaveTemplate(QWidget, Ui_Form):
self.changed_signal.emit()
def validate(self):
tmpl = preprocess_template(self.opt_template.text())
fa = {}
for x in FORMAT_ARG_DESCS.keys():
fa[x]='random long string'
try:
tmpl.format(**fa)
except Exception, err:
error_dialog(self, _('Invalid template'),
'<p>'+_('The template %s is invalid:')%tmpl + \
'<br>'+str(err), show=True)
return False
# TODO: NEWMETA: I haven't figured out how to get the custom columns
# into here, so for the moment make all templates valid.
return True
# tmpl = preprocess_template(self.opt_template.text())
# fa = {}
# for x in FORMAT_ARG_DESCS.keys():
# fa[x]='random long string'
# try:
# tmpl.format(**fa)
# except Exception, err:
# error_dialog(self, _('Invalid template'),
# '<p>'+_('The template %s is invalid:')%tmpl + \
# '<br>'+str(err), show=True)
# return False
# return True
def set_value(self, val):
self.opt_template.set_value(val)

43
src/calibre/library/database2.py
View File

@ -521,15 +521,15 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
'''
Convenience method to return metadata as a L{MetaInformation} object.
'''
aum = self.authors(idx, index_is_id=index_is_id)
if aum: aum = [a.strip().replace('|', ',') for a in aum.split(',')]
aut_list = self.authors_with_sort_strings(idx, index_is_id=index_is_id)
aum = []
aus = {}
for (author, author_sort) in aut_list:
aum.append(author)
aus[author] = author_sort
mi = MetaInformation(self.title(idx, index_is_id=index_is_id), aum)
mi.author_sort = self.author_sort(idx, index_is_id=index_is_id)
if mi.authors:
mi.author_sort_map = {}
for name, sort in zip(mi.authors, self.authors_sort_strings(idx,
index_is_id)):
mi.author_sort_map[name] = sort
mi.author_sort_map = aus
mi.comments = self.comments(idx, index_is_id=index_is_id)
mi.publisher = self.publisher(idx, index_is_id=index_is_id)
mi.timestamp = self.timestamp(idx, index_is_id=index_is_id)
@ -546,6 +546,13 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
mi.isbn = self.isbn(idx, index_is_id=index_is_id)
id = idx if index_is_id else self.id(idx)
mi.application_id = id
for key,meta in self.field_metadata.iteritems():
if meta['is_custom']:
mi.set_user_metadata(key, meta)
mi.set(key, val=self.get_custom(idx, label=meta['label'],
index_is_id=index_is_id),
extra=self.get_custom_extra(idx, label=meta['label'],
index_is_id=index_is_id))
if get_cover:
mi.cover = self.cover(id, index_is_id=True, as_path=True)
return mi
@ -1084,6 +1091,15 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
if getattr(mi, 'timestamp', None) is not None:
doit(self.set_timestamp, id, mi.timestamp, notify=False)
self.set_path(id, True)
user_mi = mi.get_all_user_metadata(make_copy=False)
for key in user_mi.iterkeys():
if key in self.field_metadata and \
user_mi[key]['datatype'] == self.field_metadata[key]['datatype']:
doit(self.set_custom, id,
val=mi.get(key),
extra=mi.get_extra(key),
label=user_mi[key]['label'])
self.notify('metadata', [id])
# Given a book, return the list of author sort strings for the book's authors
@ -1099,6 +1115,19 @@ class LibraryDatabase2(LibraryDatabase, SchemaUpgrade, CustomColumns):
result.append(sort)
return result
# Given a book, return the map of author sort strings for the book's authors
def authors_with_sort_strings(self, id, index_is_id=False):
id = id if index_is_id else self.id(id)
aut_strings = self.conn.get('''
SELECT authors.name, authors.sort
FROM authors, books_authors_link as bl
WHERE bl.book=? and authors.id=bl.author
ORDER BY bl.id''', (id,))
result = []
for (author, sort,) in aut_strings:
result.append((author.replace('|', ','), sort))
return result
# Given a book, return the author_sort string for authors of the book
def author_sort_from_book(self, id, index_is_id=False):
auts = self.authors_sort_strings(id, index_is_id)

45
src/calibre/library/save_to_disk.py
View File

@ -6,7 +6,7 @@ __license__ = 'GPL v3'
__copyright__ = '2009, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'
import os, traceback, cStringIO, re
import os, traceback, cStringIO, re, string
from calibre.utils.config import Config, StringConfig, tweaks
from calibre.utils.filenames import shorten_components_to, supports_long_names, \
@ -14,6 +14,7 @@ from calibre.utils.filenames import shorten_components_to, supports_long_names,
from calibre.ebooks.metadata.opf2 import metadata_to_opf
from calibre.ebooks.metadata.meta import set_metadata
from calibre.constants import preferred_encoding, filesystem_encoding
from calibre.ebooks.metadata import fmt_sidx
from calibre.ebooks.metadata import title_sort
from calibre import strftime
@ -97,29 +98,33 @@ def preprocess_template(template):
template = template.decode(preferred_encoding, 'replace')
return template
def safe_format(x, format_args):
class SafeFormat(string.Formatter):
'''
Provides a format function that substitutes '' for any missing value
'''
def get_value(self, key, args, kwargs):
try:
ans = x.format(**format_args).strip()
return re.sub(r'\s+', ' ', ans)
except IndexError: # Thrown if user used [] and index is out of bounds
pass
except AttributeError: # Thrown if user used a non existing attribute
pass
return kwargs[key]
except:
return ''
safe_formatter = SafeFormat()
def safe_format(x, format_args):
ans = safe_formatter.vformat(x, [], format_args).strip()
return re.sub(r'\s+', ' ', ans)
def get_components(template, mi, id, timefmt='%b %Y', length=250,
sanitize_func=ascii_filename, replace_whitespace=False,
to_lowercase=False):
library_order = tweaks['save_template_title_series_sorting'] == 'library_order'
tsfmt = title_sort if library_order else lambda x: x
format_args = dict(**FORMAT_ARGS)
format_args = FORMAT_ARGS.copy()
format_args.update(mi.get_all_non_none_attributes())
if mi.title:
format_args['title'] = tsfmt(mi.title)
if mi.authors:
format_args['authors'] = mi.format_authors()
format_args['author'] = format_args['authors']
if mi.author_sort:
format_args['author_sort'] = mi.author_sort
if mi.tags:
format_args['tags'] = mi.format_tags()
if format_args['tags'].startswith('/'):
@ -132,15 +137,25 @@ def get_components(template, mi, id, timefmt='%b %Y', length=250,
template = re.sub(r'\{series_index[^}]*?\}', '', template)
if mi.rating is not None:
format_args['rating'] = mi.format_rating()
if mi.isbn:
format_args['isbn'] = mi.isbn
if mi.publisher:
format_args['publisher'] = mi.publisher
if hasattr(mi.timestamp, 'timetuple'):
format_args['timestamp'] = strftime(timefmt, mi.timestamp.timetuple())
if hasattr(mi.pubdate, 'timetuple'):
format_args['pubdate'] = strftime(timefmt, mi.pubdate.timetuple())
format_args['id'] = str(id)
# Now format the custom fields
custom_metadata = mi.get_all_user_metadata(make_copy=False)
for key in custom_metadata:
if key in format_args:
## TODO: NEWMETA: should ratings be divided by 2? The standard rating isn't...
if custom_metadata[key]['datatype'] == 'series':
format_args[key] = tsfmt(format_args[key])
if key+'_index' in format_args:
format_args[key+'_index'] = fmt_sidx(format_args[key+'_index'])
elif custom_metadata[key]['datatype'] == 'datetime':
format_args[key] = strftime(timefmt, format_args[key].timetuple())
elif custom_metadata[key]['datatype'] == 'bool':
format_args[key] = _('yes') if format_args[key] else _('no')
components = [x.strip() for x in template.split('/') if x.strip()]
components = [safe_format(x, format_args) for x in components]
components = [sanitize_func(x) for x in components if x]

4
src/calibre/library/server/mobile.py
View File

@ -199,6 +199,10 @@ class MobileServer(object):
CKEYS = [key for key in sorted(CFM.get_custom_fields(),
cmp=lambda x,y: cmp(CFM[x]['name'].lower(),
CFM[y]['name'].lower()))]
# This method uses its own book dict, not the Metadata dict. The loop
# below could be changed to use db.get_metadata instead of reading
# info directly from the record made by the view, but it doesn't seem
# worth it at the moment.
books = []
for record in items[(start-1):(start-1)+num]:
book = {'formats':record[FM['formats']], 'size':record[FM['size']]}

9
src/calibre/library/server/utils.py
View File

@ -5,7 +5,7 @@ __license__ = 'GPL v3'
__copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
__docformat__ = 'restructuredtext en'
import time
import time, sys
import cherrypy
@ -44,8 +44,8 @@ def strftime(fmt='%Y/%m/%d %H:%M:%S', dt=None):
except:
return _strftime(fmt, nowf().timetuple())
def format_tag_string(tags, sep):
MAX = tweaks['max_content_server_tags_shown']
def format_tag_string(tags, sep, ignore_max=False):
MAX = sys.maxint if ignore_max else tweaks['max_content_server_tags_shown']
if tags:
tlist = [t.strip() for t in tags.split(sep)]
else:
@ -53,5 +53,6 @@ def format_tag_string(tags, sep):
tlist.sort(cmp=lambda x,y:cmp(x.lower(), y.lower()))
if len(tlist) > MAX:
tlist = tlist[:MAX]+['...']
return u'%s'%(', '.join(tlist)) if tlist else ''
return u'%s:&:%s'%(tweaks['max_content_server_tags_shown'],
', '.join(tlist)) if tlist else ''

13
src/calibre/library/server/xml.py
View File

@ -66,6 +66,10 @@ class XMLServer(object):
return x.decode(preferred_encoding, 'replace')
return unicode(x)
# This method uses its own book dict, not the Metadata dict. The loop
# below could be changed to use db.get_metadata instead of reading
# info directly from the record made by the view, but it doesn't seem
# worth it at the moment.
for record in items[start:start+num]:
kwargs = {}
aus = record[FM['authors']] if record[FM['authors']] else __builtin__._('Unknown')
@ -85,7 +89,7 @@ class XMLServer(object):
'comments'):
y = record[FM[x]]
if x == 'tags':
y = format_tag_string(y, ',')
y = format_tag_string(y, ',', ignore_max=True)
kwargs[x] = serialize(y) if y else ''
c = kwargs.pop('comments')
@ -107,7 +111,9 @@ class XMLServer(object):
name = CFM[key]['name']
custcols.append(k)
if datatype == 'text' and CFM[key]['is_multiple']:
kwargs[k] = concat(name, format_tag_string(val,'|'))
kwargs[k] = concat('#T#'+name,
format_tag_string(val,'|',
ignore_max=True))
elif datatype == 'series':
kwargs[k] = concat(name, '%s [%s]'%(val,
fmt_sidx(record[CFM.cc_series_index_column_for(key)])))
@ -138,6 +144,3 @@ class XMLServer(object):
return etree.tostring(ans, encoding='utf-8', pretty_print=True,
xml_declaration=True)