Created
September 12, 2015 13:10
-
-
Save Konctantin/094bddb98569caa2328a to your computer and use it in GitHub Desktop.
wowhead parser
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import sys | |
from urllib.request import urlopen | |
from time import sleep | |
def clenuap(content): | |
return content.strip() \ | |
.replace("<br />", "\\n") \ | |
.replace("<", "<") \ | |
.replace(">", ">") \ | |
.replace(" ", " ") \ | |
.replace(""", "\"") \ | |
.replace("&", "&") \ | |
.replace("'", "\\'") \ | |
.replace("\"", "\\\"") \ | |
.replace("<name>", "$N") \ | |
.replace("<имя>", "$N") \ | |
.replace("<класс>", "$C") \ | |
.replace("<class>", "$C") \ | |
.replace("<race>", "$R") \ | |
.replace("<раса>", "$R") \ | |
.replace("|3-6(Воин)", "|3-6($C)") \ | |
.replace("|3-6(Паладин)", "|3-6($C)") \ | |
.replace("|3-6(Охотник)", "|3-6($C)") \ | |
.replace("|3-6(Разбойник)", "|3-6($C)") \ | |
.replace("|3-6(Жрец)", "|3-6($C)") \ | |
.replace("|3-6(Рыцарь смерти)","|3-6($C)") \ | |
.replace("|3-6(Маг)", "|3-6($C)") \ | |
.replace("|3-6(Чернокнижник)", "|3-6($C)") \ | |
.replace("|3-6(Монах)", "|3-6($C)") \ | |
.replace("|3-6(Друид)", "|3-6($C)") \ | |
.replace("|3-6(Человек)", "|3-6($R)") \ | |
.replace("|3-6(Гном)", "|3-6($R)") \ | |
.replace("|3-6(Дреней)", "|3-6($R)") \ | |
.replace("|3-6(Дворф)", "|3-6($R)") \ | |
.replace("|3-6(Орк)", "|3-6($R)") \ | |
.replace("|3-6(Таурен)", "|3-6($R)") \ | |
.replace("|3-6(Троль)", "|3-6($R)") \ | |
.replace("|3-6(Нежить)", "|3-6($R)") \ | |
.replace("|3-6(Эльф крови)", "|3-6($R)") \ | |
.replace("|3-6(Ночной эльф)", "|3-6($R)") \ | |
with open("output.sql", "w") as sql_file: | |
def parse(entry, table_name, content, start_pattern, end_pattern): | |
spos = content.find(start_pattern) | |
endpos = content.find(end_pattern, spos) | |
if (spos > -1 and endpos > -1): | |
start = spos + len(start_pattern) | |
text = content[start:endpos] | |
sql = "INSERT INTO `%s` (entry, text) VALUES (%d, '%s');\n" % (table_name, entry, clenuap(text)) | |
sql_file.writelines(sql) | |
sql_file.flush() | |
print(sql) | |
else: | |
print("No matches found for quest: %d -> '%s'" % (entry, start_pattern)) | |
pass | |
sql_file.writelines( | |
"DROP TABLE IF EXISTS `RequestItemsText`;\n"+\ | |
"CREATE TABLE `RequestItemsText` (\n"+\ | |
" `entry` MEDIUMINT(8) UNSIGNED NOT NULL,\n"+\ | |
" `text` TEXT DEFAULT NULL\n"+\ | |
") ENGINE=MyISAM DEFAULT CHARSET=utf8 ROW_FORMAT=FIXED;\n\n"+\ | |
"DROP TABLE IF EXISTS `OfferRewardText`;\n"+\ | |
"CREATE TABLE `OfferRewardText` (\n"+\ | |
" `entry` MEDIUMINT(8) UNSIGNED NOT NULL,\n"+\ | |
" `text` TEXT DEFAULT NULL\n"+\ | |
") ENGINE=MyISAM DEFAULT CHARSET=utf8 ROW_FORMAT=FIXED;\n\n") | |
try: | |
entryList = open("list2.txt", "r").readlines() | |
print("Use defined list!"); | |
except: | |
entryList = range(1, 60000) | |
print("Use range!") | |
for entry in entryList: | |
entry = int(entry) | |
url = "http://ru.wowhead.com/quest=%d" % entry | |
try: | |
content = urlopen(url)\ | |
.read()\ | |
.decode("utf-8") | |
parse(entry, "RequestItemsText", content, "progress\" style=\"display: none\">", "</div>") | |
parse(entry, "OfferRewardText", content, "completion\" style=\"display: none\">", "</div>") | |
except Exception as ex: | |
print("Err: %d (%s)" % (entry, ex)) | |
sleep(1) | |
print("Done!") |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment