formats API

src/calibre/db/backend.py

@@ -8,7 +8,7 @@ __copyright__ = '2011, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
 # Imports {{{
-import os, shutil, uuid, json, glob, time, cPickle
+import os, shutil, uuid, json, glob, time, cPickle, hashlib
 from functools import partial
 
 import apsw
@@ -17,7 +17,9 @@ from calibre import isbytestring, force_unicode, prints
 from calibre.constants import (iswindows, filesystem_encoding,
         preferred_encoding)
 from calibre.ptempfile import PersistentTemporaryFile
+from calibre.db import SPOOL_SIZE
 from calibre.db.schema_upgrades import SchemaUpgrade
+from calibre.db.errors import NoSuchFormat
 from calibre.library.field_metadata import FieldMetadata
 from calibre.ebooks.metadata import title_sort, author_to_author_sort
 from calibre.utils.icu import sort_key
@@ -926,6 +928,19 @@ class DB(object):
             shutil.copyfile(candidates[0], fmt_path)
             return fmt_path
 
+    def format_hash(self, book_id, fmt, fname, path):
+        path = self.format_abspath(book_id, fmt, fname, path)
+        if path is None:
+            raise NoSuchFormat('Record %d has no fmt: %s'%(book_id, fmt))
+        sha = hashlib.sha256()
+        with lopen(path, 'rb') as f:
+            while True:
+                raw = f.read(SPOOL_SIZE)
+                sha.update(raw)
+                if len(raw) < SPOOL_SIZE:
+                    break
+        return sha.hexdigest()
+
     def format_metadata(self, book_id, fmt, fname, path):
         path = self.format_abspath(book_id, fmt, fname, path)
         ans = {}

src/calibre/db/cache.py

@@ -408,7 +408,16 @@ class Cache(object):
         return {aid:af.author_data(aid) for aid in author_ids if aid in af.table.id_map}
 
     @read_api
-    def format_metadata(self, book_id, fmt, allow_cache=True):
+    def format_hash(self, book_id, fmt):
+        try:
+            name = self.fields['formats'].format_fname(book_id, fmt)
+            path = self._field_for('path', book_id).replace('/', os.sep)
+        except:
+            raise NoSuchFormat('Record %d has no fmt: %s'%(book_id, fmt))
+        return self.backend.format_hash(book_id, fmt, name, path)
+
+    @api
+    def format_metadata(self, book_id, fmt, allow_cache=True, update_db=False):
         if not fmt:
             return {}
         fmt = fmt.upper()
@@ -416,6 +425,7 @@ class Cache(object):
             x = self.format_metadata_cache[book_id].get(fmt, None)
             if x is not None:
                 return x
+        with self.read_lock:
             try:
                 name = self.fields['formats'].format_fname(book_id, fmt)
                 path = self._field_for('path', book_id).replace('/', os.sep)
@@ -426,8 +436,19 @@ class Cache(object):
             if path and name:
                 ans = self.backend.format_metadata(book_id, fmt, name, path)
                 self.format_metadata_cache[book_id][fmt] = ans
+        if update_db and 'size' in ans:
+            with self.write_lock:
+                max_size = self.fields['formats'].table.update_fmt(book_id, fmt, name, ans['size'], self.backend)
+                self.fields['size'].table.update_sizes({book_id: max_size})
+
         return ans
 
+    @read_api
+    def format_files(self, book_id):
+        field = self.fields['formats']
+        fmts = field.table.book_col_map.get(book_id, ())
+        return {fmt:field.format_fname(book_id, fmt) for fmt in fmts}
+
     @read_api
     def pref(self, name, default=None):
         return self.backend.prefs.get(name, default)
@@ -524,6 +545,7 @@ class Cache(object):
         the path is different from the current path (taking case sensitivity
         into account).
         '''
+        fmt = (fmt or '').upper()
         try:
             name = self.fields['formats'].format_fname(book_id, fmt)
             path = self._field_for('path', book_id).replace('/', os.sep)
@@ -544,6 +566,7 @@ class Cache(object):
         Apart from the viewer, I don't believe any of the others do any file
         I/O with the results of this call.
         '''
+        fmt = (fmt or '').upper()
         try:
             name = self.fields['formats'].format_fname(book_id, fmt)
             path = self._field_for('path', book_id).replace('/', os.sep)
@@ -555,6 +578,7 @@ class Cache(object):
     @read_api
     def has_format(self, book_id, fmt):
         'Return True iff the format exists on disk'
+        fmt = (fmt or '').upper()
         try:
             name = self.fields['formats'].format_fname(book_id, fmt)
             path = self._field_for('path', book_id).replace('/', os.sep)
@@ -601,6 +625,7 @@ class Cache(object):
                                   this means that repeated calls yield the same
                                   temp file (which is re-created each time)
         '''
+        fmt = (fmt or '').upper()
         ext = ('.'+fmt.lower()) if fmt else ''
         if as_path:
             if preserve_filename:

src/calibre/db/legacy.py

@@ -17,6 +17,7 @@ from calibre.db.adding import (
     import_book_directory, recursive_import, add_catalog, add_news)
 from calibre.db.backend import DB
 from calibre.db.cache import Cache
+from calibre.db.errors import NoSuchFormat
 from calibre.db.categories import CATEGORY_SORTS
 from calibre.db.view import View
 from calibre.db.write import clean_identifier
@@ -436,6 +437,43 @@ class LibraryDatabase(object):
     def has_id(self, book_id):
         return book_id in self.new_api.all_book_ids()
 
+    def format(self, index, fmt, index_is_id=False, as_file=False, mode='r+b', as_path=False, preserve_filename=False):
+        book_id = index if index_is_id else self.id(index)
+        return self.new_api.format(book_id, fmt, as_file=as_file, as_path=as_path, preserve_filename=preserve_filename)
+
+    def format_abspath(self, index, fmt, index_is_id=False):
+        book_id = index if index_is_id else self.id(index)
+        return self.new_api.format_abspath(book_id, fmt)
+
+    def format_path(self, index, fmt, index_is_id=False):
+        book_id = index if index_is_id else self.id(index)
+        ans = self.new_api.format_abspath(book_id, fmt)
+        if ans is None:
+            raise NoSuchFormat('Record %d has no format: %s'%(book_id, fmt))
+        return ans
+
+    def format_files(self, index, index_is_id=False):
+        book_id = index if index_is_id else self.id(index)
+        return [(v, k) for k, v in self.new_api.format_files(book_id).iteritems()]
+
+    def format_metadata(self, book_id, fmt, allow_cache=True, update_db=False, commit=False):
+        return self.new_api.format_metadata(book_id, fmt, allow_cache=allow_cache, update_db=update_db)
+
+    def format_last_modified(self, book_id, fmt):
+        m = self.format_metadata(book_id, fmt)
+        if m:
+            return m['mtime']
+
+    def formats(self, index, index_is_id=False, verify_formats=True):
+        book_id = index if index_is_id else self.id(index)
+        ans = self.new_api.formats(book_id, verify_formats=verify_formats)
+        if ans:
+            return ','.join(ans)
+
+    def has_format(self, index, fmt, index_is_id=False):
+        book_id = index if index_is_id else self.id(index)
+        return self.new_api.has_format(book_id, fmt)
+
     # Private interface {{{
     def __iter__(self):
         for row in self.data.iterall():
@@ -463,6 +501,7 @@ for prop in ('author_sort', 'authors', 'comment', 'comments', 'publisher',
         return func
     setattr(LibraryDatabase, prop, MT(getter(prop)))
 
+LibraryDatabase.format_hash = MT(lambda self, book_id, fmt:self.new_api.format_hash(book_id, fmt))
 LibraryDatabase.index = MT(lambda self, book_id, cache=False:self.data.id_to_index(book_id))
 LibraryDatabase.has_cover = MT(lambda self, book_id:self.new_api.field_for('cover', book_id))
 LibraryDatabase.get_tags = MT(lambda self, book_id:set(self.new_api.field_for('tags', book_id)))

src/calibre/db/tests/legacy.py

@@ -11,6 +11,7 @@ from io import BytesIO
 from repr import repr
 from functools import partial
 from tempfile import NamedTemporaryFile
+from operator import itemgetter
 
 from calibre.db.tests.base import BaseTest
 
@@ -159,6 +160,11 @@ class LegacyTest(BaseTest):
 
         for meth, args in {
             'get_next_series_num_for': [('A Series One',)],
+            'format':[(1, 'FMT1', True), (2, 'FMT1', True), (0, 'xxxxxx')],
+            'has_format':[(1, 'FMT1', True), (2, 'FMT1', True), (0, 'xxxxxx')],
+            '@format_files':[(0,),(1,),(2,)],
+            'formats':[(0,),(1,),(2,)],
+            'format_hash':[(1, 'FMT1'),(1, 'FMT2'), (2, 'FMT1')],
             'author_sort_from_authors': [(['Author One', 'Author Two', 'Unknown'],)],
             'has_book':[(Metadata('title one'),), (Metadata('xxxx1111'),)],
             'has_id':[(1,), (2,), (3,), (9999,)],
@@ -330,6 +336,7 @@ class LegacyTest(BaseTest):
             'author_id',  # replaced by get_author_id
             'books_for_author',  # broken
             'books_in_old_database',  # unused
+            'migrate_old',  # no longer supported
 
             # Internal API
             'clean_user_categories',  'cleanup_tags',  'books_list_filter', 'conn', 'connect', 'construct_file_name',
@@ -337,6 +344,7 @@ class LegacyTest(BaseTest):
             'run_import_plugins', 'vacuum', 'set_path', 'row', 'row_factory', 'rows', 'rmtree', 'series_index_pat',
             'import_old_database', 'dirtied_lock', 'dirtied_cache', 'dirty_queue_length', 'dirty_books_referencing',
             'windows_check_if_files_in_use', 'get_metadata_for_dump', 'get_a_dirtied_book', 'dirtied_sequence',
+            'format_filename_cache', 'format_metadata_cache', 'filter', 'create_version1',
         }
         SKIP_ARGSPEC = {
             '__init__',
@@ -411,6 +419,9 @@ class LegacyTest(BaseTest):
         ndb = self.init_legacy(self.cloned_library)
         db = self.init_old(self.cloned_library)
         run_funcs(self, db, ndb, (
+            ('+format_metadata', 1, 'FMT1', itemgetter('size')),
+            ('+format_metadata', 1, 'FMT2', itemgetter('size')),
+            ('+format_metadata', 2, 'FMT1', itemgetter('size')),
             ('get_tags', 0), ('get_tags', 1), ('get_tags', 2),
             ('is_tag_used', 'News'), ('is_tag_used', 'xchkjgfh'),
             ('bulk_modify_tags', (1,), ['t1'], ['News']),