def test_header(self): '''Make sure magic plus table address are present in files.''' open('a', 'w').write('as;dlfkja;ldskfjasd;lfkj') writer = IndexWriter('a') writer.init() writer.terminate() header = open('a').read(8) self.assert_equals('pdko\x00\x00\x00\x01', header)
def build(self, sections_iterator, index_file): '''Build up IndexedWorldData from the data in the given sections. ''' indexed_field_names = ('name', 'sp-name', 'source-rpm', 'filename') indexed_fields = [ ('pdk', f) for f in indexed_field_names ] index_writer = IndexWriter(index_file) index_writer.init() try: for section_name, section in sections_iterator: section_iterator = section.iter_package_info() for ghost, header, blob_id, locator in section_iterator: if ghost: type_string = ghost.type else: type_string = None addresses = index_writer.add(type_string, header, blob_id, locator) if ghost: index_keys = [] for field in indexed_fields: try: value = ghost[field] except KeyError: continue key = (section_name, field, value) index_keys.append(key) index_writer.index(index_keys, addresses) ent_id_key = ('ent-id', blob_id) channel_key = section_name index_writer.index([ ent_id_key, channel_key ], addresses) index_writer.terminate() del index_writer # Next time index_file is accessed, it will be reloaded, # therefore actually reading the new file we just wrote! del self.index_file except MissingChannelDataError: message = 'Missing cached data. ' + \ 'Consider running pdk channel update. ' + \ '(%s)' % section_name raise SemanticError(message)
def build(self, sections_iterator, index_file): '''Build up IndexedWorldData from the data in the given sections. ''' indexed_field_names = ('name', 'sp-name', 'source-rpm', 'filename') indexed_fields = [('pdk', f) for f in indexed_field_names] index_writer = IndexWriter(index_file) index_writer.init() try: for section_name, section in sections_iterator: section_iterator = section.iter_package_info() for ghost, header, blob_id, locator in section_iterator: if ghost: type_string = ghost.type else: type_string = None addresses = index_writer.add(type_string, header, blob_id, locator) if ghost: index_keys = [] for field in indexed_fields: try: value = ghost[field] except KeyError: continue key = (section_name, field, value) index_keys.append(key) index_writer.index(index_keys, addresses) ent_id_key = ('ent-id', blob_id) channel_key = section_name index_writer.index([ent_id_key, channel_key], addresses) index_writer.terminate() del index_writer # Next time index_file is accessed, it will be reloaded, # therefore actually reading the new file we just wrote! del self.index_file except MissingChannelDataError: message = 'Missing cached data. ' + \ 'Consider running pdk channel update. ' + \ '(%s)' % section_name raise SemanticError(message)
def test_write_then_read(self): writer = IndexWriter('a') writer.init() address = writer.add('full', 'time', 3) writer.index([ ('a', 'b') ], address) address = writer.add(2, 7, 5) writer.index([ ('a', 'b'), ('z', 'c') ], address) writer.terminate() reader = IndexFile('a') self.assert_equal(['time', 7], list(reader.get(('a', 'b'), 1))) self.assert_equal([7], list(reader.get(('z', 'c'), 1))) self.assert_equal([], list(reader.get(('l', 'm'), 1))) self.assert_equal([ ('full', 'time', 3), (2, 7, 5) ], list(reader.get_all(('a', 'b')))) self.assert_equal([ (2, 7, 5) ], list(reader.get_all(('z', 'c')))) self.assert_equal([], list(reader.get_all(('l', 'm')))) self.assert_equal(2, reader.count(('a', 'b'))) self.assert_equal(1, reader.count(('z', 'c'))) self.assert_equal(0, reader.count(('l', 'm')))
def test_write_then_read(self): writer = IndexWriter('a') writer.init() address = writer.add('full', 'time', 3) writer.index([('a', 'b')], address) address = writer.add(2, 7, 5) writer.index([('a', 'b'), ('z', 'c')], address) writer.terminate() reader = IndexFile('a') self.assert_equal(['time', 7], list(reader.get(('a', 'b'), 1))) self.assert_equal([7], list(reader.get(('z', 'c'), 1))) self.assert_equal([], list(reader.get(('l', 'm'), 1))) self.assert_equal([('full', 'time', 3), (2, 7, 5)], list(reader.get_all(('a', 'b')))) self.assert_equal([(2, 7, 5)], list(reader.get_all(('z', 'c')))) self.assert_equal([], list(reader.get_all(('l', 'm')))) self.assert_equal(2, reader.count(('a', 'b'))) self.assert_equal(1, reader.count(('z', 'c'))) self.assert_equal(0, reader.count(('l', 'm')))