def get_kiso(filter_regex):
wiki_text = get_text()
pattern = r'{{基礎情報(.+)\n}}\n'
compiled_string = re.compile(pattern, flags=(re.MULTILINE | re.DOTALL))
finded_result = compiled_string.findall(wiki_text)
if filter_regex != None:
replaced_str = re.sub(filter_regex, "", finded_result[0])
else:
replaced_str = finded_result[0]
finded_list = replaced_str.split('\n|')
result_dict = {}
for s in finded_list:
each_dict = s.split(" = ")
if len(each_dict) > 1:
result_dict[each_dict[0]] = each_dict[1]
else:
result_dict["self"] = each_dict[0]
return result_dict
import re
from knock20 import get_text
# 正規表現
# MULTILINE: 複数行マッチング
# DOTALL: .を改行以外のあらゆる文字と解釈する
# 1. \|(.+?) |***から始まる
pattern = re.compile(r'^\|(.+?)\s=\s(.+?)(?=\n(\||\}))',
re.MULTILINE | re.DOTALL)
basic_info = {}
s = get_text()
for match in pattern.finditer(s):
basic_info[match.group(1)] = match.group(2)
for (key, value) in basic_info.items():
print("{}: {}".format(key, value))
import re
from knock20 import get_text
# 正規表現
# 1. 行頭が"[[Category:"で始まる
# 2. (.+?)は任意の文字列を表す
# 3. (\|.+)?で"|***"となる部分を0以上繰り返す
# 4. 最後に]]で閉じられる
pattern = re.compile(r'^\[\[Category:(.+?)(\|.+)?\]\]$')
for s in get_text().split("\n"):
# 各行で該当箇所を探す
text = pattern.search(s)
# 該当箇所が見つかった場合
if text is not None:
# 1つ目(上の説明で言う2に該当する部分)を抜き出す
print(text.group(1))
import re
from knock20 import get_text
wiki_text = get_text()
level_dict = {}
for i in range(2,6):
regex_str = '\n={%d}([^=]+)={%d}\n' % (i,i)
level_list = re.findall(regex_str, wiki_text)
level_dict[i] = level_list
print(level_dict)