diff --git a/src/fastwq/libs/mdict/mdict_query.py b/src/fastwq/libs/mdict/mdict_query.py index 5db9933..afbb302 100644 --- a/src/fastwq/libs/mdict/mdict_query.py +++ b/src/fastwq/libs/mdict/mdict_query.py @@ -1,4 +1,8 @@ # -*- coding: utf-8 -*- +# version: python 3.5 + + +from readmdict import MDX, MDD from struct import pack, unpack from io import BytesIO import re @@ -6,8 +10,6 @@ import sys import os import sqlite3 import json -from aqt.utils import showInfo, showText, tooltip -from .readmdict import MDX, MDD # zlib compression is used for engine version >=2.0 import zlib @@ -26,11 +28,10 @@ version = '1.1' class IndexBuilder(object): - # todo: enable history - - def __init__(self, fname, encoding="", passcode=None, force_rebuild=False, - enable_history=False, sql_index=True, check=False): + #todo: enable history + def __init__(self, fname, encoding = "", passcode = None, force_rebuild = False, enable_history = False, sql_index = True, check = False): self._mdx_file = fname + self._mdd_file = "" self._encoding = '' self._stylesheet = {} self._title = '' @@ -38,62 +39,75 @@ class IndexBuilder(object): self._description = '' self._sql_index = sql_index self._check = check - self._force_rebuild = force_rebuild _filename, _file_extension = os.path.splitext(fname) - # assert(_file_extension == '.mdx') - # assert(os.path.isfile(fname)) + assert(_file_extension == '.mdx') + assert(os.path.isfile(fname)) self._mdx_db = _filename + ".mdx.db" - self._mdd_db = _filename + ".mdd.db" - self._mdd_file = _filename + ".mdd" - self.header_build_flag = False - - def get_header(self): - - def _(): - self.header_build_flag = True - mdx = MDX(self._mdx_file, only_header=True) - self._encoding = mdx.meta['encoding'] - self._stylesheet = json.loads(mdx.meta['stylesheet']) - self._title = mdx.meta['title'] - self._description = mdx.meta['description'] + # make index anyway + if force_rebuild: + self._make_mdx_index(self._mdx_db) + if os.path.isfile(_filename + '.mdd'): + self._mdd_file = _filename + ".mdd" + self._mdd_db = _filename + ".mdd.db" + self._make_mdd_index(self._mdd_db) if os.path.isfile(self._mdx_db): - # read from META table - try: - conn = sqlite3.connect(self._mdx_db) - #cursor = conn.execute("SELECT * FROM META") - cursor = conn.execute( - 'SELECT value FROM META WHERE key IN ("encoding","stylesheet","title","description","version")') - self._encoding, stylesheet,\ - self._title, self._description, self._version = ( - each[0] for each in cursor) - self._stylesheet = json.loads(stylesheet) + #read from META table + conn = sqlite3.connect(self._mdx_db) + #cursor = conn.execute("SELECT * FROM META") + cursor = conn.execute("SELECT * FROM META WHERE key = \"version\"") + #判断有无版本号 + for cc in cursor: + self._version = cc[1] + ################# if not version in fo ############# + if not self._version: + print("version info not found") conn.close() - if not self._version: - _() - except: - _() + self._make_mdx_index(self._mdx_db) + print("mdx.db rebuilt!") + if os.path.isfile(_filename + '.mdd'): + self._mdd_file = _filename + ".mdd" + self._mdd_db = _filename + ".mdd.db" + self._make_mdd_index(self._mdd_db) + print("mdd.db rebuilt!") + return None + cursor = conn.execute("SELECT * FROM META WHERE key = \"encoding\"") + for cc in cursor: + self._encoding = cc[1] + cursor = conn.execute("SELECT * FROM META WHERE key = \"stylesheet\"") + for cc in cursor: + self._stylesheet = json.loads(cc[1]) + + cursor = conn.execute("SELECT * FROM META WHERE key = \"title\"") + for cc in cursor: + self._title = cc[1] + + cursor = conn.execute("SELECT * FROM META WHERE key = \"description\"") + for cc in cursor: + self._description = cc[1] + + #for cc in cursor: + # if cc[0] == 'encoding': + # self._encoding = cc[1] + # continue + # if cc[0] == 'stylesheet': + # self._stylesheet = json.loads(cc[1]) + # continue + # if cc[0] == 'title': + # self._title = cc[1] + # continue + # if cc[0] == 'title': + # self._description = cc[1] else: - _() + self._make_mdx_index(self._mdx_db) - def rebuild(self): - self._make_mdx_index() - if os.path.isfile(self._mdd_file): - self._make_mdd_index() - - def check_build(self): - # check if the mdx.db and mdd.db file is available - if self.header_build_flag or not os.path.isfile(self._mdx_db): - self._make_mdx_index() - if os.path.isfile(self._mdd_file) and not os.path.isfile(self._mdd_db): - self._make_mdd_index() - self.header_build_flag = False - - @property - def meta(self): - return {'title': self._title, 'description': self._description, - 'encoding': self._encoding, 'version': self._version, - 'stylesheet': self._stylesheet} + if os.path.isfile(_filename + ".mdd"): + self._mdd_file = _filename + ".mdd" + self._mdd_db = _filename + ".mdd.db" + if not os.path.isfile(self._mdd_db): + self._make_mdd_index(self._mdd_db) + pass + def _replace_stylesheet(self, txt): # substitute stylesheet definition @@ -103,20 +117,19 @@ class IndexBuilder(object): for j, p in enumerate(txt_list[1:]): style = self._stylesheet[txt_tag[j][1:-1]] if p and p[-1] == '\n': - txt_styled = txt_styled + \ - style[0].encode('utf-8') + p.rstrip() + \ - style[1].encode('utf-8') + '\r\n' + txt_styled = txt_styled + style[0] + p.rstrip() + style[1] + '\r\n' else: - txt_styled = txt_styled + \ - style[0].encode('utf-8') + p + style[1].encode('utf-8') + txt_styled = txt_styled + style[0] + p + style[1] return txt_styled - def _make_mdx_index(self): - if os.path.exists(self._mdx_db): - os.remove(self._mdx_db) - mdx = MDX(self._mdx_file, only_header=False) - index_list = mdx.get_index(check_block=self._check) - conn = sqlite3.connect(self._mdx_db) + def _make_mdx_index(self, db_name): + if os.path.exists(db_name): + os.remove(db_name) + mdx = MDX(self._mdx_file) + self._mdx_db = db_name + returned_index = mdx.get_index(check_block = self._check) + index_list = returned_index['index_dict_list'] + conn = sqlite3.connect(db_name) c = conn.cursor() c.execute( ''' CREATE TABLE MDX_INDEX @@ -133,50 +146,65 @@ class IndexBuilder(object): tuple_list = [ (item['key_text'], - item['file_pos'], - item['compressed_size'], - item['decompressed_size'], - item['record_block_type'], - item['record_start'], - item['record_end'], - item['offset'] - ) + item['file_pos'], + item['compressed_size'], + item['decompressed_size'], + item['record_block_type'], + item['record_start'], + item['record_end'], + item['offset'] + ) for item in index_list - ] + ] c.executemany('INSERT INTO MDX_INDEX VALUES (?,?,?,?,?,?,?,?)', tuple_list) # build the metadata table + meta = returned_index['meta'] c.execute( '''CREATE TABLE META (key text, value text )''') + + #for k,v in meta: + # c.execute( + # 'INSERT INTO META VALUES (?,?)', + # (k, v) + # ) + c.executemany( - 'INSERT INTO META VALUES (?,?)', - [('encoding', self.meta['encoding']), - ('stylesheet', json.dumps(self.meta['stylesheet'])), - ('title', self.meta['title']), - ('description', self.meta['description']), + 'INSERT INTO META VALUES (?,?)', + [('encoding', meta['encoding']), + ('stylesheet', meta['stylesheet']), + ('title', meta['title']), + ('description', meta['description']), ('version', version) ] - ) - + ) + if self._sql_index: c.execute( ''' CREATE INDEX key_index ON MDX_INDEX (key_text) ''' - ) + ) conn.commit() conn.close() + #set class member + self._encoding = meta['encoding'] + self._stylesheet = json.loads(meta['stylesheet']) + self._title = meta['title'] + self._description = meta['description'] - def _make_mdd_index(self): - if os.path.exists(self._mdd_db): - os.remove(self._mdd_db) + + def _make_mdd_index(self, db_name): + if os.path.exists(db_name): + os.remove(db_name) mdd = MDD(self._mdd_file) - index_list = mdd.get_index(check_block=self._check) - conn = sqlite3.connect(self._mdd_db) + self._mdd_db = db_name + index_list = mdd.get_index(check_block = self._check) + conn = sqlite3.connect(db_name) c = conn.cursor() c.execute( ''' CREATE TABLE MDX_INDEX @@ -193,16 +221,16 @@ class IndexBuilder(object): tuple_list = [ (item['key_text'], - item['file_pos'], - item['compressed_size'], - item['decompressed_size'], - item['record_block_type'], - item['record_start'], - item['record_end'], - item['offset'] - ) + item['file_pos'], + item['compressed_size'], + item['decompressed_size'], + item['record_block_type'], + item['record_start'], + item['record_end'], + item['offset'] + ) for item in index_list - ] + ] c.executemany('INSERT INTO MDX_INDEX VALUES (?,?,?,?,?,?,?,?)', tuple_list) if self._sql_index: @@ -210,13 +238,12 @@ class IndexBuilder(object): ''' CREATE UNIQUE INDEX key_index ON MDX_INDEX (key_text) ''' - ) + ) conn.commit() conn.close() - @staticmethod - def get_data_by_index(fmdx, index): + def get_mdx_by_index(self, fmdx, index): fmdx.seek(index['file_pos']) record_block_compressed = fmdx.read(index['compressed_size']) record_block_type = record_block_compressed[:4] @@ -231,97 +258,111 @@ class IndexBuilder(object): print("LZO compression is not supported") # decompress header = b'\xf0' + pack('>I', index['decompressed_size']) - _record_block = lzo.decompress(record_block_compressed[ - 8:], initSize=decompressed_size, blockSize=1308672) - # zlib compression + _record_block = lzo.decompress(record_block_compressed[8:], initSize = decompressed_size, blockSize=1308672) + # zlib compression elif record_block_type == 2: # decompress _record_block = zlib.decompress(record_block_compressed[8:]) - data = _record_block[index['record_start'] - - index['offset']:index['record_end'] - index['offset']] - return data - - def get_mdx_by_index(self, fmdx, index): - data = self.get_data_by_index(fmdx, index) - record = data.decode(self._encoding, errors='ignore').strip( - u'\x00').encode('utf-8') + record = _record_block[index['record_start'] - index['offset']:index['record_end'] - index['offset']] + record = record = record.decode(self._encoding, errors='ignore').strip(u'\x00').encode('utf-8') if self._stylesheet: record = self._replace_stylesheet(record) record = record.decode('utf-8') return record def get_mdd_by_index(self, fmdx, index): - return self.get_data_by_index(fmdx, index) + fmdx.seek(index['file_pos']) + record_block_compressed = fmdx.read(index['compressed_size']) + record_block_type = record_block_compressed[:4] + record_block_type = index['record_block_type'] + decompressed_size = index['decompressed_size'] + #adler32 = unpack('>I', record_block_compressed[4:8])[0] + if record_block_type == 0: + _record_block = record_block_compressed[8:] + # lzo compression + elif record_block_type == 1: + if lzo is None: + print("LZO compression is not supported") + # decompress + header = b'\xf0' + pack('>I', index['decompressed_size']) + _record_block = lzo.decompress(record_block_compressed[8:], initSize = decompressed_size, blockSize=1308672) + # zlib compression + elif record_block_type == 2: + # decompress + _record_block = zlib.decompress(record_block_compressed[8:]) + data = _record_block[index['record_start'] - index['offset']:index['record_end'] - index['offset']] + return data - @staticmethod - def lookup_indexes(db, keyword, ignorecase=None): - indexes = [] - if ignorecase: - sql = u'SELECT * FROM MDX_INDEX WHERE lower(key_text) = lower("{}")'.format( - keyword) - else: - sql = u'SELECT * FROM MDX_INDEX WHERE key_text = "{}"'.format( - keyword) - with sqlite3.connect(db) as conn: - cursor = conn.execute(sql) - for result in cursor: - index = {} - index['file_pos'] = result[1] - index['compressed_size'] = result[2] - index['decompressed_size'] = result[3] - index['record_block_type'] = result[4] - index['record_start'] = result[5] - index['record_end'] = result[6] - index['offset'] = result[7] - indexes.append(index) - return indexes - - def mdx_lookup(self, keyword, ignorecase=None): + def mdx_lookup(self, keyword): + conn = sqlite3.connect(self._mdx_db) + cursor = conn.execute("SELECT * FROM MDX_INDEX WHERE key_text = " + "\"" + keyword + "\"") lookup_result_list = [] - indexes = self.lookup_indexes(self._mdx_db, keyword, ignorecase) - with open(self._mdx_file, 'rb') as mdx_file: - for index in indexes: - lookup_result_list.append( - self.get_mdx_by_index(mdx_file, index)) + mdx_file = open(self._mdx_file,'rb') + for result in cursor: + index = {} + index['file_pos'] = result[1] + index['compressed_size'] = result[2] + index['decompressed_size'] = result[3] + index['record_block_type'] = result[4] + index['record_start'] = result[5] + index['record_end'] = result[6] + index['offset'] = result[7] + lookup_result_list.append(self.get_mdx_by_index(mdx_file, index)) + conn.close() + mdx_file.close() + return lookup_result_list + + def mdd_lookup(self, keyword): + conn = sqlite3.connect(self._mdd_db) + cursor = conn.execute("SELECT * FROM MDX_INDEX WHERE key_text = " + "\"" + keyword + "\"") + lookup_result_list = [] + mdd_file = open(self._mdd_file,'rb') + for result in cursor: + index = {} + index['file_pos'] = result[1] + index['compressed_size'] = result[2] + index['decompressed_size'] = result[3] + index['record_block_type'] = result[4] + index['record_start'] = result[5] + index['record_end'] = result[6] + index['offset'] = result[7] + lookup_result_list.append(self.get_mdd_by_index(mdd_file, index)) + mdd_file.close() + conn.close() return lookup_result_list - def mdd_lookup(self, keyword, ignorecase=None): - lookup_result_list = [] - indexes = self.lookup_indexes(self._mdd_db, keyword, ignorecase) - with open(self._mdd_file, 'rb') as mdd_file: - for index in indexes: - lookup_result_list.append( - self.get_mdd_by_index(mdd_file, index)) - return lookup_result_list - - @staticmethod - def get_keys(db, query=''): - if not db: + def get_mdd_keys(self, query = ''): + if not self._mdd_db: return [] + conn = sqlite3.connect(self._mdd_db) if query: if '*' in query: - query = query.replace('*', '%') + query = query.replace('*','%') else: query = query + '%' - sql = 'SELECT key_text FROM MDX_INDEX WHERE key_text LIKE \"' + query + '\"' - else: - sql = 'SELECT key_text FROM MDX_INDEX' - with sqlite3.connect(db) as conn: - cursor = conn.execute(sql) + cursor = conn.execute('SELECT key_text FROM MDX_INDEX WHERE key_text LIKE \"' + query + '\"') keys = [item[0] for item in cursor] - return keys + else: + cursor = conn.execute('SELECT key_text FROM MDX_INDEX') + keys = [item[0] for item in cursor] + conn.close() + return keys - def get_mdd_keys(self, query=''): - try: - return self.get_keys(self._mdd_db, query) - except: - return [] + def get_mdx_keys(self, query = ''): + conn = sqlite3.connect(self._mdx_db) + if query: + if '*' in query: + query = query.replace('*','%') + else: + query = query + '%' + cursor = conn.execute('SELECT key_text FROM MDX_INDEX WHERE key_text LIKE \"' + query + '\"') + keys = [item[0] for item in cursor] + else: + cursor = conn.execute('SELECT key_text FROM MDX_INDEX') + keys = [item[0] for item in cursor] + conn.close() + return keys - def get_mdx_keys(self, query=''): - try: - return self.get_keys(self._mdx_db, query) - except: - return [] # mdx_builder = IndexBuilder("oald.mdx") diff --git a/src/fastwq/libs/mdict/readmdict.py b/src/fastwq/libs/mdict/readmdict.py index 2d79358..65b39b0 100644 --- a/src/fastwq/libs/mdict/readmdict.py +++ b/src/fastwq/libs/mdict/readmdict.py @@ -23,9 +23,8 @@ import re import sys import json -from .ripemd128 import ripemd128 -from .pureSalsa20 import Salsa20 -from aqt.utils import showInfo, showText, tooltip +from ripemd128 import ripemd128 +from pureSalsa20 import Salsa20 # zlib compression is used for engine version >=2.0 import zlib @@ -93,15 +92,12 @@ class MDict(object): Base class which reads in header and key block. It has no public methods and serves only as code sharing base class. """ - - def __init__(self, fname, encoding='', passcode=None, only_header=False): + def __init__(self, fname, encoding='', passcode=None): self._fname = fname self._encoding = encoding.upper() self._passcode = passcode self.header = self._read_header() - if only_header: - return try: self._key_list = self._read_keys() except: @@ -139,8 +135,7 @@ class MDict(object): assert(key_block_info_compressed[:4] == b'\x02\x00\x00\x00') # decrypt if needed if self._encrypt & 0x02: - key_block_info_compressed = _mdx_decrypt( - key_block_info_compressed) + key_block_info_compressed = _mdx_decrypt(key_block_info_compressed) # decompress key_block_info = zlib.decompress(key_block_info_compressed[8:]) # adler checksum @@ -164,12 +159,10 @@ class MDict(object): while i < len(key_block_info): # number of entries in current key block - num_entries += unpack(self._number_format, - key_block_info[i:i + self._number_width])[0] + num_entries += unpack(self._number_format, key_block_info[i:i + self._number_width])[0] i += self._number_width # text head size - text_head_size = unpack(byte_format, key_block_info[ - i:i + byte_width])[0] + text_head_size = unpack(byte_format, key_block_info[i:i + byte_width])[0] i += byte_width # text head if self._encoding != 'UTF-16': @@ -177,8 +170,7 @@ class MDict(object): else: i += (text_head_size + text_term) * 2 # text tail size - text_tail_size = unpack(byte_format, key_block_info[ - i:i + byte_width])[0] + text_tail_size = unpack(byte_format, key_block_info[i:i + byte_width])[0] i += byte_width # text tail if self._encoding != 'UTF-16': @@ -186,15 +178,12 @@ class MDict(object): else: i += (text_tail_size + text_term) * 2 # key block compressed size - key_block_compressed_size = unpack(self._number_format, key_block_info[ - i:i + self._number_width])[0] + key_block_compressed_size = unpack(self._number_format, key_block_info[i:i + self._number_width])[0] i += self._number_width # key block decompressed size - key_block_decompressed_size = unpack(self._number_format, key_block_info[ - i:i + self._number_width])[0] + key_block_decompressed_size = unpack(self._number_format, key_block_info[i:i + self._number_width])[0] i += self._number_width - key_block_info_list += [(key_block_compressed_size, - key_block_decompressed_size)] + key_block_info_list += [(key_block_compressed_size, key_block_decompressed_size)] assert(num_entries == self._num_entries) @@ -209,8 +198,7 @@ class MDict(object): # 4 bytes : compression type key_block_type = key_block_compressed[start:start + 4] # 4 bytes : adler checksum of decompressed key block - adler32 = unpack('>I', key_block_compressed[ - start + 4:start + 8])[0] + adler32 = unpack('>I', key_block_compressed[start + 4:start + 8])[0] if key_block_type == b'\x00\x00\x00\x00': key_block = key_block_compressed[start + 8:end] elif key_block_type == b'\x01\x00\x00\x00': @@ -219,12 +207,10 @@ class MDict(object): break # decompress key block header = b'\xf0' + pack('>I', decompressed_size) - key_block = lzo.decompress(key_block_compressed[ - start + 8:end], initSize=decompressed_size, blockSize=1308672) + key_block = lzo.decompress(key_block_compressed[start + 8:end], initSize = decompressed_size, blockSize=1308672) elif key_block_type == b'\x02\x00\x00\x00': # decompress key block - key_block = zlib.decompress( - key_block_compressed[start + 8:end]) + key_block = zlib.decompress(key_block_compressed[start + 8:end]) # extract one single key block into a key list key_list += self._split_key_block(key_block) # notice that adler32 returns signed value @@ -237,11 +223,9 @@ class MDict(object): key_list = [] key_start_index = 0 while key_start_index < len(key_block): - temp = key_block[ - key_start_index:key_start_index + self._number_width] + temp = key_block[key_start_index:key_start_index + self._number_width] # the corresponding record's offset in record block - key_id = unpack(self._number_format, key_block[ - key_start_index:key_start_index + self._number_width])[0] + key_id = unpack(self._number_format, key_block[key_start_index:key_start_index + self._number_width])[0] # key text ends with '\x00' if self._encoding == 'UTF-16': delimiter = b'\x00\x00' @@ -261,12 +245,6 @@ class MDict(object): key_list += [(key_id, key_text)] return key_list - @property - def meta(self): - return {'title': self._title, 'description': self._description, - 'encoding': self._encoding, 'version': self._version, - 'stylesheet': json.dumps(self._stylesheet)} - def _read_header(self): f = open(self._fname, 'rb') # number of bytes of header text @@ -349,8 +327,7 @@ class MDict(object): if self._encrypt & 1: if self._passcode is None: - raise RuntimeError( - 'user identification is needed to read encrypted file') + raise RuntimeError('user identification is needed to read encrypted file') regcode, userid = self._passcode if isinstance(userid, unicode): userid = userid.encode('utf8') @@ -388,8 +365,7 @@ class MDict(object): # read key block key_block_compressed = f.read(key_block_size) # extract key block - key_list = self._decode_key_block( - key_block_compressed, key_block_info_list) + key_list = self._decode_key_block(key_block_compressed, key_block_info_list) self._record_block_offset = f.tell() f.close() @@ -434,8 +410,7 @@ class MDict(object): # read key block key_block_compressed = f.read(key_block_size) # extract key block - key_list = self._decode_key_block( - key_block_compressed, key_block_info_list) + key_list = self._decode_key_block(key_block_compressed, key_block_info_list) self._record_block_offset = f.tell() f.close() @@ -453,7 +428,6 @@ class MDD(MDict): >>> for filename,content in mdd.items(): ... print filename, content[:10] """ - def __init__(self, fname, passcode=None): MDict.__init__(self, fname, encoding='UTF-16', passcode=passcode) @@ -500,8 +474,7 @@ class MDD(MDict): break # decompress header = b'\xf0' + pack('>I', decompressed_size) - record_block = lzo.decompress(record_block_compressed[ - start + 8:end], initSize=decompressed_size, blockSize=1308672) + record_block = lzo.decompress(record_block_compressed[start + 8:end], initSize = decompressed_size, blockSize=1308672) elif record_block_type == b'\x02\x00\x00\x00': # decompress record_block = zlib.decompress(record_block_compressed[8:]) @@ -530,16 +503,16 @@ class MDD(MDict): f.close() - # 获取 mdx 文件的索引列表，格式为 - # key_text(关键词，可以由后面的 keylist 得到) - # file_pos(record_block开始的位置) - # compressed_size(record_block压缩前的大小) - # decompressed_size(解压后的大小) - # record_block_type(record_block 的压缩类型) - # record_start (以下三个为从 record_block 中提取某一调记录需要的参数，可以直接保存） - # record_end - # offset - def get_index(self, check_block=True): + ### 获取 mdx 文件的索引列表，格式为 + ### key_text(关键词，可以由后面的 keylist 得到) + ### file_pos(record_block开始的位置) + ### compressed_size(record_block压缩前的大小) + ### decompressed_size(解压后的大小) + ### record_block_type(record_block 的压缩类型) + ### record_start (以下三个为从 record_block 中提取某一调记录需要的参数，可以直接保存） + ### record_end + ### offset + def get_index(self, check_block = True): f = open(self._fname, 'rb') index_dict_list = [] f.seek(self._record_block_offset) @@ -584,8 +557,7 @@ class MDD(MDict): # decompress header = b'\xf0' + pack('>I', decompressed_size) if check_block: - record_block = lzo.decompress(record_block_compressed[ - start + 8:end], initSize=decompressed_size, blockSize=1308672) + record_block = lzo.decompress(record_block_compressed[start + 8:end], initSize = decompressed_size, blockSize=1308672) elif record_block_type == b'\x02\x00\x00\x00': # decompress _type = 2 @@ -598,7 +570,7 @@ class MDD(MDict): assert(len(record_block) == decompressed_size) # split record block according to the offset info from key block while i < len(self._key_list): - # 用来保存索引信息的空字典 + ### 用来保存索引信息的空字典 index_dict = {} index_dict['file_pos'] = current_pos index_dict['compressed_size'] = compressed_size @@ -606,11 +578,10 @@ class MDD(MDict): index_dict['record_block_type'] = _type record_start, key_text = self._key_list[i] index_dict['record_start'] = record_start - index_dict['key_text'] = key_text.decode( - "utf-8", errors='ignore') + index_dict['key_text'] = key_text.decode("utf-8") index_dict['offset'] = offset # reach the end of current record block - if record_start - offset >= decompressed_size: + if record_start - offset >= decompressed_size: break # record end index if i < len(self._key_list) - 1: @@ -620,11 +591,10 @@ class MDD(MDict): index_dict['record_end'] = record_end i += 1 if check_block: - data = record_block[ - record_start - offset:record_end - offset] + data = record_block[record_start - offset:record_end - offset] index_dict_list.append(index_dict) - # yield key_text, data - offset += decompressed_size + #yield key_text, data + offset += decompressed_size size_counter += compressed_size assert(size_counter == record_block_size) f.close() @@ -640,9 +610,8 @@ class MDX(MDict): >>> for key,value in mdx.items(): ... print key, value[:10] """ - - def __init__(self, fname, encoding='', substyle=False, passcode=None, only_header=False): - MDict.__init__(self, fname, encoding, passcode, only_header) + def __init__(self, fname, encoding='', substyle=False, passcode=None): + MDict.__init__(self, fname, encoding, passcode) self._substyle = substyle def items(self): @@ -658,8 +627,7 @@ class MDX(MDict): for j, p in enumerate(txt_list[1:]): style = self._stylesheet[txt_tag[j][1:-1]] if p and p[-1] == '\n': - txt_styled = txt_styled + \ - style[0] + p.rstrip() + style[1] + '\r\n' + txt_styled = txt_styled + style[0] + p.rstrip() + style[1] + '\r\n' else: txt_styled = txt_styled + style[0] + p + style[1] return txt_styled @@ -688,20 +656,20 @@ class MDX(MDict): offset = 0 i = 0 size_counter = 0 - # 最后的索引表的格式为 - # key_text(关键词，可以由后面的 keylist 得到) - # file_pos(record_block开始的位置) - # compressed_size(record_block压缩前的大小) - # decompressed_size(解压后的大小) - # record_block_type(record_block 的压缩类型) - # record_start (以下三个为从 record_block 中提取某一调记录需要的参数，可以直接保存） - # record_end - # offset + ###最后的索引表的格式为 + ### key_text(关键词，可以由后面的 keylist 得到) + ### file_pos(record_block开始的位置) + ### compressed_size(record_block压缩前的大小) + ### decompressed_size(解压后的大小) + ### record_block_type(record_block 的压缩类型) + ### record_start (以下三个为从 record_block 中提取某一调记录需要的参数，可以直接保存） + ### record_end + ### offset for compressed_size, decompressed_size in record_block_info_list: record_block_compressed = f.read(compressed_size) - # 要得到 record_block_compressed 需要得到 compressed_size (这个可以直接记录） - # 另外还需要记录当前 f 对象的位置 - # 使用 f.tell() 命令/ 在建立索引是需要 f.seek() + ###### 要得到 record_block_compressed 需要得到 compressed_size (这个可以直接记录） + ###### 另外还需要记录当前 f 对象的位置 + ###### 使用 f.tell() 命令/ 在建立索引是需要 f.seek() # 4 bytes indicates block compression type record_block_type = record_block_compressed[:4] # 4 bytes adler checksum of uncompressed content @@ -716,16 +684,15 @@ class MDX(MDict): break # decompress header = b'\xf0' + pack('>I', decompressed_size) - record_block = lzo.decompress(record_block_compressed[ - 8:], initSize=decompressed_size, blockSize=1308672) + record_block = lzo.decompress(record_block_compressed[8:], initSize = decompressed_size, blockSize=1308672) # zlib compression elif record_block_type == b'\x02\x00\x00\x00': # decompress record_block = zlib.decompress(record_block_compressed[8:]) - # 这里比较重要的是先要得到 record_block, 而 record_block 是解压得到的，其中一共有三种解压方法 - # 需要的信息有 record_block_compressed, decompress_size, - # record_block_type - # 另外还需要校验信息 adler32 + ###### 这里比较重要的是先要得到 record_block, 而 record_block 是解压得到的，其中一共有三种解压方法 + ###### 需要的信息有 record_block_compressed, decompress_size, + ###### record_block_type + ###### 另外还需要校验信息 adler32 # notice that adler32 return signed value assert(adler32 == zlib.adler32(record_block) & 0xffffffff) @@ -742,15 +709,13 @@ class MDX(MDict): else: record_end = len(record_block) + offset i += 1 - # 需要得到 record_block , record_start, record_end, - # offset - record = record_block[ - record_start - offset:record_end - offset] + #############需要得到 record_block , record_start, record_end, + #############offset + record = record_block[record_start - offset:record_end - offset] # convert to utf-8 - record = record.decode(self._encoding, errors='ignore').strip( - u'\x00').encode('utf-8') + record = record.decode(self._encoding, errors='ignore').strip(u'\x00').encode('utf-8') # substitute styles - # 是否替换样式表 + #############是否替换样式表 if self._substyle and self._stylesheet: record = self._substitute_stylesheet(record) @@ -761,19 +726,19 @@ class MDX(MDict): f.close() - # 获取 mdx 文件的索引列表，格式为 - # key_text(关键词，可以由后面的 keylist 得到) - # file_pos(record_block开始的位置) - # compressed_size(record_block压缩前的大小) - # decompressed_size(解压后的大小) - # record_block_type(record_block 的压缩类型) - # record_start (以下三个为从 record_block 中提取某一调记录需要的参数，可以直接保存） - # record_end - # offset - # 所需 metadata - ### - def get_index(self, check_block=True): - # 索引列表 + ### 获取 mdx 文件的索引列表，格式为 + ### key_text(关键词，可以由后面的 keylist 得到) + ### file_pos(record_block开始的位置) + ### compressed_size(record_block压缩前的大小) + ### decompressed_size(解压后的大小) + ### record_block_type(record_block 的压缩类型) + ### record_start (以下三个为从 record_block 中提取某一调记录需要的参数，可以直接保存） + ### record_end + ### offset + ### 所需 metadata + ### + def get_index(self, check_block = True): + ### 索引列表 index_dict_list = [] f = open(self._fname, 'rb') f.seek(self._record_block_offset) @@ -798,21 +763,21 @@ class MDX(MDict): offset = 0 i = 0 size_counter = 0 - # 最后的索引表的格式为 - # key_text(关键词，可以由后面的 keylist 得到) - # file_pos(record_block开始的位置) - # compressed_size(record_block压缩前的大小) - # decompressed_size(解压后的大小) - # record_block_type(record_block 的压缩类型) - # record_start (以下三个为从 record_block 中提取某一调记录需要的参数，可以直接保存） - # record_end - # offset + ###最后的索引表的格式为 + ### key_text(关键词，可以由后面的 keylist 得到) + ### file_pos(record_block开始的位置) + ### compressed_size(record_block压缩前的大小) + ### decompressed_size(解压后的大小) + ### record_block_type(record_block 的压缩类型) + ### record_start (以下三个为从 record_block 中提取某一调记录需要的参数，可以直接保存） + ### record_end + ### offset for compressed_size, decompressed_size in record_block_info_list: current_pos = f.tell() record_block_compressed = f.read(compressed_size) - # 要得到 record_block_compressed 需要得到 compressed_size (这个可以直接记录） - # 另外还需要记录当前 f 对象的位置 - # 使用 f.tell() 命令/ 在建立索引是需要 f.seek() + ###### 要得到 record_block_compressed 需要得到 compressed_size (这个可以直接记录） + ###### 另外还需要记录当前 f 对象的位置 + ###### 使用 f.tell() 命令/ 在建立索引是需要 f.seek() # 4 bytes indicates block compression type record_block_type = record_block_compressed[:4] # 4 bytes adler checksum of uncompressed content @@ -830,25 +795,24 @@ class MDX(MDict): # decompress header = b'\xf0' + pack('>I', decompressed_size) if check_block: - record_block = lzo.decompress(record_block_compressed[ - 8:], initSize=decompressed_size, blockSize=1308672) + record_block = lzo.decompress(record_block_compressed[8:], initSize = decompressed_size, blockSize=1308672) # zlib compression elif record_block_type == b'\x02\x00\x00\x00': # decompress _type = 2 if check_block: record_block = zlib.decompress(record_block_compressed[8:]) - # 这里比较重要的是先要得到 record_block, 而 record_block 是解压得到的，其中一共有三种解压方法 - # 需要的信息有 record_block_compressed, decompress_size, - # record_block_type - # 另外还需要校验信息 adler32 + ###### 这里比较重要的是先要得到 record_block, 而 record_block 是解压得到的，其中一共有三种解压方法 + ###### 需要的信息有 record_block_compressed, decompress_size, + ###### record_block_type + ###### 另外还需要校验信息 adler32 # notice that adler32 return signed value if check_block: assert(adler32 == zlib.adler32(record_block) & 0xffffffff) assert(len(record_block) == decompressed_size) # split record block according to the offset info from key block while i < len(self._key_list): - # 用来保存索引信息的空字典 + ### 用来保存索引信息的空字典 index_dict = {} index_dict['file_pos'] = current_pos index_dict['compressed_size'] = compressed_size @@ -856,11 +820,10 @@ class MDX(MDict): index_dict['record_block_type'] = _type record_start, key_text = self._key_list[i] index_dict['record_start'] = record_start - index_dict['key_text'] = key_text.decode( - 'utf-8', errors='ignore') + index_dict['key_text'] = key_text.decode('utf-8') index_dict['offset'] = offset # reach the end of current record block - if record_start - offset >= decompressed_size: + if record_start - offset >= decompressed_size: break # record end index if i < len(self._key_list) - 1: @@ -869,27 +832,31 @@ class MDX(MDict): record_end = decompressed_size + offset index_dict['record_end'] = record_end i += 1 - # 需要得到 record_block , record_start, record_end, - # offset + #############需要得到 record_block , record_start, record_end, + #############offset if check_block: - record = record_block[ - record_start - offset:record_end - offset] + record = record_block[record_start - offset:record_end - offset] # convert to utf-8 - record = record.decode(self._encoding, errors='ignore').strip( - u'\x00').encode('utf-8') + record = record.decode(self._encoding, errors='ignore').strip(u'\x00').encode('utf-8') # substitute styles - # 是否替换样式表 + #############是否替换样式表 if self._substyle and self._stylesheet: record = self._substitute_stylesheet(record) index_dict_list.append(index_dict) - offset += decompressed_size + offset += decompressed_size size_counter += compressed_size - # todo: 注意！！！ - #assert(size_counter == record_block_size) + #todo: 注意！！！ + #assert(size_counter == record_block_size) f.close - return index_dict_list + #这里比 mdd 部分稍有不同，应该还需要传递编码以及样式表信息 + meta = {} + meta['encoding'] = self._encoding + meta['stylesheet'] = json.dumps(self._stylesheet) + meta['title'] = self._title + meta['description'] = self._description + return {"index_dict_list":index_dict_list, 'meta':meta} if __name__ == '__main__': import sys import os @@ -905,8 +872,7 @@ if __name__ == '__main__': try: regcode = codecs.decode(regcode, 'hex') except: - raise argparse.ArgumentTypeError( - "regcode must be a 32 bytes hexadecimal string") + raise argparse.ArgumentTypeError("regcode must be a 32 bytes hexadecimal string") return regcode, userid parser = argparse.ArgumentParser() @@ -987,8 +953,7 @@ if __name__ == '__main__': sf.close() # write out optional data files if mdd: - datafolder = os.path.join( - os.path.dirname(args.filename), args.datafolder) + datafolder = os.path.join(os.path.dirname(args.filename), args.datafolder) if not os.path.exists(datafolder): os.makedirs(datafolder) for key, value in mdd.items(): diff --git a/src/fastwq/query.py b/src/fastwq/query.py index d71203f..dd60e74 100644 --- a/src/fastwq/query.py +++ b/src/fastwq/query.py @@ -73,6 +73,7 @@ class QueryThread(QThread): super(QueryThread, self).__init__() self.index = 0 self.exit = False + self.finished = False self.manager = manager self.note_flush.connect(manager.handle_flush) @@ -96,6 +97,8 @@ class QueryThread(QThread): if self.manager: self.manager.queue.task_done() + + self.finished = True class QueryWorkerManager(object): @@ -122,11 +125,15 @@ class QueryWorkerManager(object): def start(self): self.total = self.queue.qsize() self.progress.start(self.total, min=0) - for x in range(0, min(config.thread_number, self.total)): - self.get_worker() - - for worker in self.workers: - worker.start() + if self.total > 1: + for x in range(0, min(config.thread_number, self.total)): + self.get_worker() + + for worker in self.workers: + worker.start() + else: + worker = self.get_worker() + worker.run() def update(self, note, results, success_num): self.mutex.lock() @@ -137,11 +144,15 @@ class QueryWorkerManager(object): val = update_note_fields(note, results) self.fields += val self.mutex.unlock() - return val > 0 + if self.total > 1: + return val > 0 + else: + self.handle_flush(note) + return False def join(self): for worker in self.workers: - while not worker.isFinished(): + while not worker.finished: if self.progress.abort(): worker.exit = True break @@ -323,22 +334,25 @@ def query_all_flds(note): dict_unique = each.get('dict_unique', '').strip() if dict_name and dict_name not in _sl('NOT_DICT_FIELD') and dict_field: s = services.get(dict_unique, None) - if s == None: - services[dict_unique] = service_pool.get(dict_unique)#service_manager.get_service(dict_unique) - tasks.append({'k': dict_unique, 'w': word, 'f': dict_field, 'i': i}) + if s is None: + s = service_pool.get(dict_unique) + if s.support: + services[dict_unique] = s + if s and s.support: + tasks.append({'k': dict_unique, 'w': word, 'f': dict_field, 'i': i}) success_num = 0 result = defaultdict(QueryResult) for task in tasks: - try: - service = services.get(task['k'], None) - qr = service.active(task['f'], task['w']) - if qr: - result.update({task['i']: qr}) - success_num += 1 - except: - showInfo(_("NO_QUERY_WORD")) - pass + #try: + service = services.get(task['k'], None) + qr = service.active(task['f'], task['w']) + if qr: + result.update({task['i']: qr}) + success_num += 1 + #except: + # showInfo(_("NO_QUERY_WORD")) + # pass for service in services.values(): service_pool.put(service) diff --git a/src/fastwq/service/LDOCE6.py b/src/fastwq/service/LDOCE6.py index aa39bd9..c6302bd 100644 --- a/src/fastwq/service/LDOCE6.py +++ b/src/fastwq/service/LDOCE6.py @@ -1,17 +1,22 @@ #-*- coding:utf-8 -*- import re +from .base import MdxService, export, register, with_styles, parseHtml -from aqt.utils import showInfo, showText -from .base import MdxService, export, register, with_styles +PATH = u'D:\\mdx_server\\mdx\\LDOCE6.mdx' -path = u'D:\\mdx_server\\mdx\\LDOCE6.mdx' +VOICE_PATTERN = r'' +MAPPINGS = [ + ['br', [re.compile(VOICE_PATTERN % r'r')]], + ['us', [re.compile(VOICE_PATTERN % r'b')]] +] +LANG_TO_REGEXPS = {lang: regexps for lang, regexps in MAPPINGS} @register(u'本地词典-LDOCE6') class Ldoce6(MdxService): def __init__(self): - super(Ldoce6, self).__init__(path) + super(Ldoce6, self).__init__(PATH) @property def unique(self): @@ -27,58 +32,74 @@ class Ldoce6(MdxService): m = re.search(r'(.*?)', html) if m: return m.groups()[0] + return '' - @export(u'Bre单词发音', 2) + def _fld_voice(self, html, voice): + """获取发音字段""" + from hashlib import sha1 + for regexp in LANG_TO_REGEXPS[voice]: + match = regexp.search(html) + if match: + val = '/' + match.group(1) + hex_digest = sha1( + val.encode('utf-8') if isinstance(val, unicode) + else val + ).hexdigest().lower() + + assert len(hex_digest) == 40, "unexpected output from hash library" + name = '.'.join([ + '-'.join([ + 'mdx', self.unique.lower(), hex_digest[:8], hex_digest[8:16], + hex_digest[16:24], hex_digest[24:32], hex_digest[32:], + ]), + 'mp3', + ]) + name = self.save_file(val, name) + if name: + return self.get_anki_label(name, 'audio') + return '' + + @export(u'英式发音', 2) def fld_voicebre(self): - html = self.get_html() - m = re.search(r'(.*?)', html) - if m: - return m.groups()[0] - return '' + return self._fld_voice(self.get_html(), 'br') - @export(u'Ame单词发音', 3) + @export(u'美式发音', 3) def fld_voiceame(self): - html = self.get_html() - m = re.search(r'(.*?)', html) - if m: - return m.groups()[0] - return '' + return self._fld_voice(self.get_html(), 'us') - @export(u'sentence', 4) + @export(u'例句', 4) def fld_sentence(self): - html = self.get_html() - m = re.search(r'(.*?)', html) + m = re.findall(r'\s*.*<\/span>', self.get_html()) if m: - return re.sub('', '', m.groups()[0]) - return '' - - @export(u'def', 5) - def fld_definate(self): - html = self.get_html() - m = re.search(r'(.*?)', html) - if m: - return m.groups()[0] - return '' - - @export(u'random_sentence', 6) - def fld_random_sentence(self): - html = self.get_html() - m = re.findall(r'(.*?)', html) - if m: - number = len(m) - index = random.randrange(0, number - 1, 1) - return re.sub('', '', m[index]) - return '' - - @export(u'all sentence', 7) - def fld_allsentence(self): - html = self.get_html() - m = re.findall( - r'(.+?.+?)', html) - if m: - items = 0 + soup = parseHtml(m[0]) + el_list = soup.findAll('span', {'class':'example'}) + if el_list: + maps = [u''.join(str(content).decode('utf-8') for content in element.contents) + for element in el_list] my_str = '' - for items in range(len(m)): - my_str = my_str + m[items] - return my_str + for i_str in maps: + i_str = re.sub(r']+?href=\"sound\:.*\.mp3\".*', '', i_str) + i_str = i_str.replace(' ', '') + my_str = my_str + '
  • ' + i_str + '
    • ' + return self._css(my_str) return '' + + @export(u'释义', 5) + def fld_definate(self): + m = m = re.findall(r'\s*.*<\/span>', self.get_html()) + if m: + soup = parseHtml(m[0]) + el_list = soup.findAll('span', {'class':'def'}) + if el_list: + maps = [u''.join(str(content).decode('utf-8') for content in element.contents) + for element in el_list] + my_str = '' + for i_str in maps: + my_str = my_str + '
  • ' + i_str + '
    • ' + return self._css(my_str) + return '' + + @with_styles(cssfile='_ldoce6.css') + def _css(self, val): + return val + \ No newline at end of file diff --git a/src/fastwq/service/base.py b/src/fastwq/service/base.py index 0715e71..589b515 100644 --- a/src/fastwq/service/base.py +++ b/src/fastwq/service/base.py @@ -1,8 +1,8 @@ #-*- coding:utf-8 -*- # -# Copyright © 2016–2017 Liang Feng +# Copyright © 2016–2017 ST.Huang # -# Support: Report an issue at https://github.com/finalion/WordQuery/issues +# Support: Report an issue at https://github.com/sth2018/FastWordQuery/issues # # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by @@ -29,17 +29,20 @@ import sqlite3 import urllib import urllib2 import zlib +import random from collections import defaultdict from functools import wraps import cookielib -from aqt import mw -from aqt.qt import QFileDialog -from aqt.utils import showInfo, showText from ..context import config -from ..lang import _ from ..libs import MdxBuilder, StardictBuilder from ..utils import MapDict, wrap_css +from ..libs.bs4 import BeautifulSoup + +try: + import threading as _threading +except ImportError: + import dummy_threading as _threading def register(label): @@ -120,17 +123,45 @@ def with_styles(**styles): return _deco return _with +# BS4资源锁，防止程序卡死 +BS_LOCKS = [_threading.Lock(), _threading.Lock()] + +def parseHtml(html): + ''' + 使用BS4解析html + ''' + lock = BS_LOCKS[random.randrange(0, len(BS_LOCKS) - 1, 1)] + lock.acquire() + soup = BeautifulSoup(html, 'html.parser') + lock.release() + return soup + class Service(object): '''service base class''' def __init__(self): + self.cache = defaultdict(defaultdict) self._exporters = self.get_exporters() self._fields, self._actions = zip(*self._exporters) \ if self._exporters else (None, None) # query interval: default 500ms self.query_interval = 0.5 + def cache_this(self, result): + self.cache[self.word].update(result) + return result + + def cached(self, key): + return (self.word in self.cache) and self.cache[self.word].has_key(key) + + def cache_result(self, key): + return self.cache[self.word].get(key, u'') + + @property + def support(self): + return True + @property def fields(self): return self._fields @@ -158,9 +189,9 @@ class Service(object): self.word = word # if the service instance is LocalService, # then have to build then index. - if isinstance(self, LocalService): - if isinstance(self, MdxService) or isinstance(self, StardictService): - self.builder.check_build() + #if isinstance(self, LocalService): + # if isinstance(self, MdxService) or isinstance(self, StardictService): + # self.builder.check_build() for each in self.exporters: if action_label == each[0]: @@ -180,21 +211,10 @@ class WebService(Service): def __init__(self): super(WebService, self).__init__() - self.cache = defaultdict(defaultdict) self._cookie = cookielib.CookieJar() self._opener = urllib2.build_opener( urllib2.HTTPCookieProcessor(self._cookie)) - self.query_interval = 1 - - def cache_this(self, result): - self.cache[self.word].update(result) - return result - - def cached(self, key): - return (self.word in self.cache) and self.cache[self.word].has_key(key) - - def cache_result(self, key): - return self.cache[self.word].get(key, u'') + self.query_interval = 1.0 @property def title(self): @@ -229,6 +249,9 @@ class WebService(Service): class LocalService(Service): + """ + 本地词典 + """ def __init__(self, dict_path): super(LocalService, self).__init__() @@ -236,6 +259,10 @@ class LocalService(Service): self.builder = None self.missed_css = set() + @property + def support(self): + return os.path.isfile(self.dict_path) + @property def unique(self): return self.dict_path @@ -248,28 +275,34 @@ class LocalService(Service): def _filename(self): return os.path.splitext(os.path.basename(self.dict_path))[0] +# mdx字典实例集 +mdx_builders = defaultdict(dict) class MdxService(LocalService): + """ + Mdx本地词典 + """ def __init__(self, dict_path): super(MdxService, self).__init__(dict_path) - self.media_cache = defaultdict(set) - self.cache = defaultdict(str) + self.html_cache = defaultdict(str) self.query_interval = 0.01 self.styles = [] - self.builder = MdxBuilder(dict_path) - self.builder.get_header() + if self.support: + if not mdx_builders.has_key(dict_path) or not mdx_builders[dict_path]: + mdx_builders[dict_path] = MdxBuilder(dict_path) + self.builder = mdx_builders[dict_path] - @staticmethod - def support(dict_path): - return os.path.isfile(dict_path) and dict_path.lower().endswith('.mdx') + @property + def support(self): + return os.path.isfile(self.dict_path) and self.dict_path.lower().endswith('.mdx') @property def title(self): if config.use_filename or not self.builder._title or self.builder._title.startswith('Title'): return self._filename else: - return self.builder.meta['title'] + return self.builder['_title'] @export(u"default", 0) def fld_whole(self): @@ -277,94 +310,46 @@ class MdxService(LocalService): js = re.findall(r'.*?', html, re.DOTALL) return QueryResult(result=html, js=u'\n'.join(js)) + def _get_definition_mdx(self): + """根据关键字得到MDX词典的解释""" + content = self.builder.mdx_lookup(self.word) + str_content = "" + if len(content) > 0: + for c in content: + str_content += c.replace("\r\n","").replace("entry:/","") + + return str_content + + def _get_definition_mdd(self, word): + """根据关键字得到MDX词典的媒体""" + word = word.replace('/', '\\') + content = self.builder.mdd_lookup(word) + if len(content) > 0: + return [content[0]] + else: + return [] + def get_html(self): - if not self.cache[self.word]: - html = '' - result = self.builder.mdx_lookup(self.word) # self.word: unicode - if result: - if result[0].upper().find(u"@@@LINK=") > -1: - # redirect to a new word behind the equal symol. - self.word = result[0][len(u"@@@LINK="):].strip() - return self.get_html() - else: - html = self.adapt_to_anki(result[0]) - self.cache[self.word] = html - return self.cache[self.word] + """取得self.word对应的html页面""" + if not self.html_cache[self.word]: + html = self._get_definition_mdx() + if html: + self.html_cache[self.word] = html + return self.html_cache[self.word] - def adapt_to_anki(self, html): - """ - 1. convert the media path to actual path in anki's collection media folder. - 2. remove the js codes (js inside will expires.) - """ - # convert media path, save media files - media_files_set = set() - mcss = re.findall(r'href="(\S+?\.css)"', html) - media_files_set.update(set(mcss)) - mjs = re.findall(r'src="([\w\./]\S+?\.js)"', html) - media_files_set.update(set(mjs)) - msrc = re.findall(r'', html) - media_files_set.update(set(msrc)) - msound = re.findall(r'href="sound:(.*?\.(?:mp3|wav))"', html) - if config.export_media: - media_files_set.update(set(msound)) - for each in media_files_set: - html = html.replace(each, u'_' + each.split('/')[-1]) - # find sounds - p = re.compile( - r']+?href=\"(sound:_.*?\.(?:mp3|wav))\"[^>]*?>(.*?)') - html = p.sub(u"[\\1]\\2", html) - self.save_media_files(media_files_set) - for cssfile in mcss: - cssfile = '_' + \ - os.path.basename(cssfile.replace('\\', os.path.sep)) - # if not exists the css file, the user can place the file to media - # folder first, and it will also execute the wrap process to generate - # the desired file. - if not os.path.exists(cssfile): - self.missed_css.add(cssfile[1:]) - new_css_file, wrap_class_name = wrap_css(cssfile) - html = html.replace(cssfile, new_css_file) - # add global div to the result html - html = u'
    {1}
    '.format( - wrap_class_name, html) - - return html - - def save_file(self, filepath_in_mdx, savepath=None): - basename = os.path.basename(filepath_in_mdx.replace('\\', os.path.sep)) - if savepath is None: - savepath = '_' + basename + def save_file(self, filepath_in_mdx, savepath): + """从mmd中取出filepath_in_mdx媒体文件并保存到savepath""" try: - bytes_list = self.builder.mdd_lookup(filepath_in_mdx) - if bytes_list and not os.path.exists(savepath): - with open(savepath, 'wb') as f: - f.write(bytes_list[0]) - return savepath + bytes_list = self._get_definition_mdd(filepath_in_mdx) + if bytes_list: + if not os.path.exists(savepath): + with open(savepath, 'wb') as f: + f.write(bytes_list[0]) + return savepath except sqlite3.OperationalError as e: - showInfo(str(e)) - - def save_media_files(self, data): - """ - get the necessary static files from local mdx dictionary - ** kwargs: data = list - """ - diff = data.difference(self.media_cache['files']) - self.media_cache['files'].update(diff) - lst, errors = list(), list() - wild = [ - '*' + os.path.basename(each.replace('\\', os.path.sep)) for each in diff] - try: - for each in wild: - keys = self.builder.get_mdd_keys(each) - if not keys: - errors.append(each) - lst.extend(keys) - for each in lst: - self.save_file(each) - except AttributeError: + #showInfo(str(e)) pass - - return errors + return '' class StardictService(LocalService): @@ -372,12 +357,13 @@ class StardictService(LocalService): def __init__(self, dict_path): super(StardictService, self).__init__(dict_path) self.query_interval = 0.05 - self.builder = StardictBuilder(self.dict_path, in_memory=False) - self.builder.get_header() + if self.support: + self.builder = StardictBuilder(self.dict_path, in_memory=False) + self.builder.get_header() - @staticmethod - def support(dict_path): - return os.path.isfile(dict_path) and dict_path.lower().endswith('.ifo') + @property + def support(self): + return os.path.isfile(self.dict_path) and self.dict_path.lower().endswith('.ifo') @property def title(self): @@ -388,7 +374,7 @@ class StardictService(LocalService): @export(u"default", 0) def fld_whole(self): - self.builder.check_build() + #self.builder.check_build() try: result = self.builder[self.word] result = result.strip().replace('\r\n', '
    ')\ diff --git a/src/fastwq/service/bing.py b/src/fastwq/service/bing.py index b3a8c86..66500d6 100644 --- a/src/fastwq/service/bing.py +++ b/src/fastwq/service/bing.py @@ -2,8 +2,7 @@ import re from aqt.utils import showInfo, showText -from BeautifulSoup import BeautifulSoup -from .base import WebService, export, register, with_styles +from .base import WebService, export, register, with_styles, parseHtml @register(u'Bing') @@ -15,55 +14,28 @@ class Bing(WebService): def _get_content(self): word = self.word.replace(' ', '_') data = self.get_response(u"http://cn.bing.com/dict/search?q={}&mkt=zh-cn".format(word)) - - soup = BeautifulSoup(data) - - def _get_element(soup, tag, id=None, class_=None, subtag=None): - # element = soup.find(tag, id=id, class_=class_) # bs4 - element = None - if id: - element = soup.find(tag, {"id": id}) - if class_: - element = soup.find(tag, {"class": class_}) - if subtag and element: - element = getattr(element, subtag, '') - return element - + soup = parseHtml(data) result = {} - element = _get_element(soup, 'div', class_='hd_prUS') + + element = soup.find('div', class_='hd_prUS') if element: result['phonitic_us'] = str(element).decode('utf-8') - element = _get_element(soup, 'div', class_='hd_pr') + + element = soup.find('div', class_='hd_pr') if element: result['phonitic_uk'] = str(element).decode('utf-8') - element = _get_element(soup, 'div', class_='hd_if') + + element = soup.find('div', class_='hd_if') if element: result['participle'] = str(element).decode('utf-8') - element = _get_element(soup, 'div', class_='qdef', subtag='ul') + + element = soup.find('div', class_='qdef') if element: - result['def'] = u''.join([str(content).decode('utf-8') - for content in element.contents]) - # for pair in pairs]) - # result = _get_from_element( - # result, 'advanced_ec', soup, 'div', id='authid') - # result = _get_from_element( - # result, 'ec', soup, 'div', id='crossid') - # result = _get_from_element( - # result, 'ee', soup, 'div', id='homoid') - # result = _get_from_element( - # result, 'web_definition', soup, 'div', id='webid') - # result = _get_from_element( - # result, 'collocation', soup, 'div', id='colid') - # result = _get_from_element( - # result, 'synonym', soup, 'div', id='synoid') - # result = _get_from_element( - # result, 'antonym', soup, 'div', id='antoid') - # result = _get_from_element( - # result, 'samples', soup, 'div', id='sentenceCon') + element = getattr(element, 'ul', '') + if element: + result['def'] = u''.join([str(content) for content in element.contents]) + return self.cache_this(result) - # except Exception as e: - # showInfo(str(e)) - # return {} def _get_field(self, key, default=u''): return self.cache_result(key) if self.cached(key) else self._get_content().get(key, default) @@ -90,35 +62,3 @@ class Bing(WebService): if val == None or val == '': return '' return self._css(val) - - # @export(u'权威英汉双解', 5) - # def fld_advanced_ec(self): - # return self._get_field('advanced_ec') - - # @export(u'英汉', 6) - # def fld_ec(self): - # return self._get_field('ec') - - # @export(u'英英', 7) - # def fld_ee(self): - # return self._get_field('ee') - - # @export(u'网络释义', 8) - # def fld_web_definition(self): - # return self._get_field('web_definition') - - # @export(u'搭配', 9) - # def fld_collocation(self): - # return self._get_field('collocation') - - # @export(u'同义词', 10) - # def fld_synonym(self): - # return self._get_field('synonym') - - # @export(u'反义词', 11) - # def fld_antonym(self): - # return self._get_field('antonym') - - # @export(u'例句', 12) - # def fld_samples(self): - # return self._get_field('samples') diff --git a/src/fastwq/service/longman.py b/src/fastwq/service/longman.py index 40dc25a..13800a1 100644 --- a/src/fastwq/service/longman.py +++ b/src/fastwq/service/longman.py @@ -8,9 +8,7 @@ Created: 12/20/2017 """ import os from warnings import filterwarnings - -import requests as rq -from bs4 import BeautifulSoup, Tag +from ..libs.bs4 import BeautifulSoup, Tag from .base import WebService, export, register, with_styles diff --git a/src/fastwq/service/manager.py b/src/fastwq/service/manager.py index e57e0e8..6546f1b 100644 --- a/src/fastwq/service/manager.py +++ b/src/fastwq/service/manager.py @@ -84,31 +84,31 @@ class ServiceManager(object): web_services, local_custom_services = set(), set() mypath = os.path.dirname(os.path.realpath(__file__)) files = [f for f in os.listdir(mypath) - if f not in ('__init__.py', 'base.py', 'manager.py', 'pool.py') and not f.endswith('.pyc')] + if f not in ('__init__.py', 'base.py', 'manager.py', 'pool.py') and not f.endswith('.pyc') and not os.path.isdir(mypath+os.sep+f)] base_class = (WebService, LocalService, MdxService, StardictService) for f in files: - try: - module = importlib.import_module( - '.%s' % os.path.splitext(f)[0], __package__) - for name, cls in inspect.getmembers(module, predicate=inspect.isclass): - if cls in base_class: - continue - #try: - #service = cls(*args) - service = service_wrap(cls, *args) - service.__unique__ = name - if issubclass(cls, WebService): - web_services.add(service) - # get the customized local services - if issubclass(cls, LocalService): - local_custom_services.add(service) - #except Exception: - # exclude the local service whose path has error. - # pass - except ImportError: - continue + #try: + module = importlib.import_module( + '.%s' % os.path.splitext(f)[0], __package__) + for name, cls in inspect.getmembers(module, predicate=inspect.isclass): + if cls in base_class: + continue + #try: + #service = cls(*args) + service = service_wrap(cls, *args) + service.__unique__ = name + if issubclass(cls, WebService): + web_services.add(service) + # get the customized local services + if issubclass(cls, LocalService): + local_custom_services.add(service) + #except Exception: + # exclude the local service whose path has error. + # pass + #except ImportError: + # continue return web_services, local_custom_services def _get_available_local_services(self): diff --git a/src/fastwq/service/oxford_learning.py b/src/fastwq/service/oxford_learning.py index b3ee7c8..3cc7995 100644 --- a/src/fastwq/service/oxford_learning.py +++ b/src/fastwq/service/oxford_learning.py @@ -1,47 +1,36 @@ # coding=utf-8 -from warnings import filterwarnings +#from warnings import filterwarnings +from ..libs.bs4 import Tag +from .base import WebService, export, register, with_styles, parseHtml -from bs4 import BeautifulSoup, Tag -from requests import Session - -from .base import WebService, export, register, with_styles - -filterwarnings('ignore') +#filterwarnings('ignore') import sys -reload(sys) -sys.setdefaultencoding('utf8') +#reload(sys) +#sys.setdefaultencoding('utf8') +BASE_URL = u'https://www.oxfordlearnersdictionaries.com/definition/english/{word}' @register(u'牛津学习词典') class OxfordLearning(WebService): - _base_url = 'https://www.oxfordlearnersdictionaries.com/definition/english/' - + def __init__(self): super(OxfordLearning, self).__init__() - self.s = Session() - self.s.headers = { - 'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 ' - '(KHTML, like Gecko) Chrome/31.0.1623.0 Safari/537.36' - } - self.s.get(self._base_url) - def query(self, word): """ - :param word: :rtype: WebWord """ - _qry_url = self._base_url + word + qry_url = BASE_URL.format(word=word) retried = 10 while retried: try: - rsp = self.s.get(_qry_url, ) - if rsp.status_code == 200: - return OxfordLearningDictWord(rsp.content.decode('utf-8')) + rsp = self.get_response(qry_url, timeout=15) + if rsp: + return OxfordLearningDictWord(rsp.decode('utf-8')) break except: retried -= 1 @@ -121,7 +110,7 @@ class OxfordLearningDictWord: if not markups: return self.markups = markups - self.bs = BeautifulSoup(self.markups, from_encoding="utf-8") + self.bs = parseHtml(self.markups) self._defs = [] self._defs_html = [] diff --git a/src/fastwq/service/static/_ldoce6.css b/src/fastwq/service/static/_ldoce6.css new file mode 100644 index 0000000..d855654 --- /dev/null +++ b/src/fastwq/service/static/_ldoce6.css @@ -0,0 +1,733 @@ +.entry { + line-height: 150%; + color: black; + display: block; +} +.abbr { + font-weight: bold; +} +.ac,.ac { + padding-left: 2px; + padding-right: 2px; + border-radius: 2px 2px 2px 2px; + border-style: solid; + border-width: 1px; + font-variant: small-caps; + font-size: 80%; + font-weight: bold; + color: blue; +} +.amequiv { + font-weight: bold; +} +.brequiv { + font-weight: bold; +} +.collo { + font-weight: bold; +} +.colloexa { + display: block; +} +.colloinexa { + font-style: italic; + font-weight: bold; +} +.comp { + font-weight: bold; +} +.deriv { + font-weight: bold; + color: blue; +} +.errorbox { + display: block; +} +.etymsense { + display: block; +} +.etymrefhwd { + font-style: italic; +} +.etymrefhom { + font-size: 80%; + vertical-align: super; + font-style: normal; +} +.etymorigin { + font-style: italic; +} +.etymtran { + font-weight: bold; +} +.etymbox { + margin-top: 1em; + display: block; +} +.example { + font-style: italic; + display: block; + color: blue; +} +.freq,.freq { + padding-left: 2px; + padding-right: 2px; + border-radius: 2px 2px 2px 2px; + border-style: solid; + border-width: 1px; + font-variant: small-caps; + font-size: 80%; + font-weight: bold; + color: red; +} +.level { + color: red; + font-size: 160%; +} +.fullform { + font-weight: bold; +} +.geo,span.geo { + font-weight: normal; + font-style: italic; + color: purple; +} +.gloss,.collgloss { + font-weight: normal; + font-style: normal; + color: black; +} +.gram { + color: green; +} +.hintbold { + font-weight: bold; +} +.hintitalic { + font-style: italic; +} +hinttitle { + font-weight: bold; +} +.homnum { + vertical-align: super; + font-size: 8pt; + color: blue; + font-weight: bold; +} +.hwd { + display: none; +} +.hyphenation { + font-weight: bold; + font-size: 120%; + color: blue; +} +.frequent { + color: red; +} +.lexunit { + font-weight: bold; +} +.lexvar { + font-weight: bold; +} +.linkword { + font-style: italic; +} +note { + display: block; +} +object { + font-weight: normal; +} +.opp { + font-weight: bold; +} +.orthvar { + font-weight: bold; +} +.pastpart { + font-weight: bold; +} +pastpartx { + font-weight: bold; +} +.pasttense { + font-weight: bold; +} +pasttensex { + font-weight: bold; +} +.phrvbentry { + display: block; + margin-top: 10px; + margin-left: 10px; +} +.phrvbhwd { + font-weight: bold; + color: blue; +} +.pluralform { + font-weight: bold; +} +.pos { + font-style: italic; + color: green; + font-weight: normal; +} +.prespart { + font-weight: bold; +} +prespartx { + font-weight: bold; +} +.propform { + font-weight: bold; + display: block; +} +.propformprep { + font-weight: bold; + display: block; +} +.ptandpp { + font-weight: bold; +} +ptandppx { + font-weight: bold; +} +.refhomnum { + vertical-align: super; + font-size: 60%; +} +.refhwd,.refsensenum,refsense { + font-style: normal; + font-variant: small-caps; + text-transform: lowercase; +} +.refsensenum { + font-size: 80%; +} +crossrefto .reflex { + display: none; +} +.reflex { + font-weight: bold; +} +.registerlab { + font-style: italic; + color: purple; +} +.relatedwd { + font-weight: bold; +} +.runon { + display: block; + margin-top: 8px; +} +.sense { + display: block; + margin-bottom: 14px; + margin-top: 6px; +} +.signpost { + color: white; + background-color: #35A3FF; + margin-left: .5em; + font-weight: bold; + font-variant: small-caps; + text-transform: uppercase; + font-size: 80%; + padding: 1px 2px 1px 1px; +} +.spokensect { + border-top: solid; + border-bottom: solid; + border-width: 2px; + display: block; + margin-bottom: 1ex; + clear: both; + margin-top: 1ex; + border-color: #00A2E8; + padding: 3px; +} +.spokensecthead { + display: block; + color: #00A2E8; + font-weight: bold; +} +.pronstrong { + font-style: italic; +} +.subsense { + display: block; +} +.superl { + font-weight: bold; +} +.syn { + font-weight: bold; +} +.t3perssing { + font-weight: bold; +} +t3perssingx { + font-weight: bold; +} +unclassified { + font-weight: bold; +} +span.neutral { + font-style: normal; + font-weight: normal; + font-variant: normal; + color: black; + text-decoration: none; +} +.cross { + color: red; + font-weight: bold; +} +span.italic { + color: black; + font-style: italic; + font-weight: normal; +} +.badexa { + text-decoration: line-through; + font-style: italic; +} +.hint .expl { + display: inline; +} +span.infllab { + font-style: italic; + font-weight: normal; +} +span.warning { + font-style: normal; + font-weight: bold; + color: red; +} +span.sensenum { + font-style: normal; + font-weight: bold; + margin-right: 3px; + color: blue; +} +span.synopp { + padding-left: 3px; + padding-right: 3px; + border-radius: 2px 2px 2px 2px; + border-style: solid; + border-width: 1px; + font-variant: small-caps; + font-size: 80%; + font-weight: bold; + color: blue; +} +.subheading,.secheading { + display: block; + font-weight: bold; + font-weight: bold; + color: white; + background-color: #8187BF; + margin-left: -3px; + margin-right: -3px; + font-variant: small-caps; + padding-left: 3px; +} +.collocate,.exponent { + display: block; + padding-bottom: 5px; + margin-top: 5px; +} +.collocate.inline { + display: inline; +} +.expl { + display: block; + padding: 0 3px; +} +.colloc { + font-weight: bold; +} +.exp { + font-weight: bold; +} +.expr { + font-weight: bold; +} +.colloc.key { + color: blue; +} +span.keycollo { + font-weight: bold; + color: blue; +} +.thespropform { + font-weight: bold; +} +collexa { + font-style: italic; +} +collexa .colloinexa { + font-weight: normal; +} +thesexa { + font-style: italic; +} +learneritem { + display: block; +} +.goodcollo { + font-weight: bold; +} +.badcollo { + text-decoration: line-through; +} +.defbold { + font-weight: bold; +} +topic { + font-variant: small-caps; + color: blue; +} +.thesref.newline { + display: block; +} +.heading.newline { + display: block; +} +.thesref span.thesaurus { + color: blue; + font-variant: small-caps; +} +.thesref .refhwd,.thesref .refhomnum { + color: blue; + font-weight: bold; +} +i { + font-style: italic; +} +.imgholder { + cursor: pointer; + /*float: right;*/ + display: block; + margin-bottom: 1ex; + padding: 2px; + clear: both} +.imgholder img { + border: 1px solid #DDD; +} +.buttons { + display: block; +} +.popup-button { + background-color: #f0f2fc; + border-radius: 2px; + border: 1px solid #7e92c7; + color: #7e92c7; + text-transform: uppercase; + font-size: 66%; + padding: 2px 3px; + text-decoration: none; +} +.popup-button-hover { + background-color: #dfe3f8; + cursor: pointer; +} +.popverbs { + display: block; + color: white; + font-weight: bold; + background-color: #ec008d; + padding-left: 3px; + margin-bottom: 5px; +} +.verbtable .lemma { + color: blue; + font-size: 120%; + font-weight: bold; +} +.verbtable table { + border-collapse: separate; + border-spacing: 1px; + margin-top: 10px; +} +.verbtable td { + padding: 0 5px 0 2px; + border-style: solid; + border-width: 1px; + border-color: #D2D2D2; +} +.header { + font-weight: bold; + font-variant: small-caps; +} +.verbtable td.col1 { + font-weight: bold; +} +.verbtable td.col2 { + font-style: italic; +} +.verbtable .geo { + font-style: italic; + color: #000; + font-size: normal; + font-weight: normal; +} +.verbtable .aux { + font-weight: bold; +} +.verbtable .verb_form { + color: blue; + font-weight: bold; +} +.collocations .last { + margin-bottom: 20px; +} +.collocations .colloc { + display: inline-block; + font-weight: bold; + margin-left: -2px; +} +.collobox,.thesbox,.usagebox,.grambox,spoken,.f2nbox { + border-radius: 9px 9px 9px 9px; + border-style: solid; + border-width: 2px; + display: block; + margin-bottom: 1ex; + clear: both; + margin-top: 1ex; +} +.f2nbox { + border-color: #00A2E8; +} +.f2nbox .heading { + color: #00A2E8; +} +.heading { + color: white; + font-weight: bold; + line-height: 100%; + padding: 3px; +} +.section { + display: block; + padding: 0 3px; +} +.last { + border-bottom-left-radius: 9px; + border-bottom-right-radius: 9px; +} +.thesbox { + background-color: #652D91; + border-color: #652D91; +} +.thesbox .section { + background-color: #E8E2F0; +} +.collobox { + background-color: #1D3E99; + border-color: #1D3E99; +} +.collobox .section { + background-color: #E2F4FD; +} +.usagebox { + background-color: #00A2E8; + border-color: #00A2E8; +} +.usagebox .expl { + background-color: #E2F4FD; +} +.grambox { + background-color: #00ADEE; + border-color: #00ADEE; +} +.grambox .expl,.grambox .compareword { + background-color: #E2F4FD; + display: block; + padding-left: 3px; +} +.compareword { + padding-top: 5px; + padding-bottom: 5px; +} +.gramrefcont { + display: block; + text-transform: lowercase; + font-variant: small-caps; + padding-left: 3px; + padding-bottom: 1px; +} +.thesaurus .sense { + padding-left: 3px; + padding-right: 3px; + margin-top: 0; +} +.thesaurus .section,.collocations .section { + margin-bottom: 10px; +} +.thesaurus .secheading,.thesbox .secheading { + background-color: #A186BD; +} +.add_exa { + font-style: italic; + display: block; +} +.nodeword { + color: blue; + font-weight: bold; +} +.phrase { + display: block; +} +.phrasetext { + font-weight: bold; +} +.expandable { + cursor: pointer; +} +.entry div.content { + display: none; + margin-bottom: 10px; +} +.group,.w { + display: block; +} +.group .pos { + font-weight: bold; + display: block; +} +.item { + display: block; +} +.w { + font-weight: bold; +} +.popheader { + display: block; + color: white; + font-weight: bold; + padding-left: 3px; +} +.popheader.popexa { + background-color: red; +} +.popheader.popphrase { + background-color: purple; +} +.popheader.popcollo { + background-color: #1D3E99; +} +.popheader.popthes { + background-color: #652D91; +} +.popheader.popetym { + background-color: green; +} +.popheader.popwf { + background-color: #FFD448; +} +.popheader.pope_menu { + background-color: #57C0E0; +} +.ws-head { + font-weight: bold; + color: blue; + font-size: larger; +} +.ws-head.ref { + display: block; + margin-top: 10px; +} +.wswd { + font-weight: bold; + display: block; +} +.cyan { + color: #00A2E8; + font-size: larger; +} +.menuitem { + display: block; +} +ul.exas li { + font-style: italic; + color: blue; + margin-bottom: 3px; +} +.grammar { + display: block; +} +.str { + font-size: large; + font-weight: bold; +} +.etymology { + margin-top: 10px; +} +.group { + margin-top: 10px; +} +.menuitem .signpost { + background-color: #FFF; + color: black; + margin-left: 0; + font-size: 90%; +} +.phrvbs { + display: block; + margin-top: 10px; + margin-left: 5px; +} +.phrvbs .heading { + display: block; + color: blue; + font-weight: bold; +} +.phrv { + font-weight: bold; +} +.secheading,.subheading { + font-variant: normal; + text-transform: uppercase; +} +.secheading.no_convert { + text-transform: none; +} +.grambox .heading.newline { + background-color: #6CD0F6; +} +img[src*="img/spkr_"] { + margin-bottom: -5px; +} +.chwd { + margin-top: 10px; + margin-left: 8px; + margin-right: 8px; +} +.chwd a { + display: inline-block; + background-color: green; + color: white; + padding-top: 2px; + padding-bottom: 3px; + padding-left: 5px; + padding-right: 5px; + margin-right: 3px; + margin-bottom: 3px; + text-decoration: none; + font-size: 80%; + border-radius: 2px; +} +.chwd .hw { + color: white; +} +.entry, .entry.lozenge, .verbtable { + margin-top: 10px; + margin-left: 10px; + margin-right: 10px; +} diff --git a/src/fastwq/ui.py b/src/fastwq/ui.py index 09b2700..64a0277 100644 --- a/src/fastwq/ui.py +++ b/src/fastwq/ui.py @@ -345,24 +345,29 @@ class OptionsDialog(QDialog): for cls in service_manager.local_services: # combo_data.insert("data", each.label) service = service_pool.get(cls.__unique__) - dict_combo.addItem( - service.title, userData=service.unique) - service_pool.put(service) + if service and service.support: + dict_combo.addItem( + service.title, userData=service.unique) + service_pool.put(service) dict_combo.insertSeparator(dict_combo.count()) for cls in service_manager.web_services: service = service_pool.get(cls.__unique__) - dict_combo.addItem( - service.title, userData=service.unique) - service_pool.put(service) + if service and service.support: + dict_combo.addItem( + service.title, userData=service.unique) + service_pool.put(service) def set_dict_combo_index(): - dict_combo.setCurrentIndex(-1) + #dict_combo.setCurrentIndex(-1) for i in range(dict_combo.count()): if current_text in _sl('NOT_DICT_FIELD'): dict_combo.setCurrentIndex(0) + return if dict_combo.itemText(i) == current_text: dict_combo.setCurrentIndex(i) + return + dict_combo.setCurrentIndex(0) set_dict_combo_index() @@ -376,15 +381,19 @@ class OptionsDialog(QDialog): field_combo.setFocus(Qt.MouseFocusReason) # MouseFocusReason else: field_text = field_combo.currentText() - service_unique = dict_combo_itemdata - current_service = service_pool.get(service_unique) + unique = dict_combo_itemdata + service = service_pool.get(unique) + text = '' # problem - if current_service and current_service.fields: - for each in current_service.fields: + if service and service.support and service.fields: + for each in service.fields: field_combo.addItem(each) if each == field_text: - field_combo.setEditText(field_text) - service_pool.put(current_service) + text = each + + field_combo.setEditText(text) + field_combo.setEnabled(text != '') + service_pool.put(service) def radio_btn_checked(self): rbs = self.findChildren(QRadioButton)