from . import tool import datetime import html import re def table_parser(data, cel_data, start_data, num = 0): table_class = 'class="' all_table = 'style="' cel_style = 'style="' row_style = 'style="' row = '' cel = '' table_width = re.search("<table ?width=((?:(?!>).)*)>", data) if table_width: all_table += 'width: ' + table_width.groups()[0] + 'px;' table_height = re.search("<table ?height=((?:(?!>).)*)>", data) if table_height: all_table += 'height: ' + table_height.groups()[0] + 'px;' table_align = re.search("<table ?align=((?:(?!>).)*)>", data) if table_align: if table_align.groups()[0] == 'right': all_table += 'float: right;' elif table_align.groups()[0] == 'center': all_table += 'margin: auto;' table_text_align = re.search("<table ?textalign=((?:(?!>).)*)>", data) if table_text_align: num = 1 if table_text_align.groups()[0] == 'right': all_table += 'text-align: right;' elif table_text_align.groups()[0] == 'center': all_table += 'text-align: center;' row_t_a = re.search("<row ?textalign=((?:(?!>).)*)>", data) if row_t_a: if row_t_a.groups()[0] == 'right': row_style += 'text-align: right;' elif row_t_a.groups()[0] == 'center': row_style += 'text-align: center;' else: row_style += 'text-align: left;' table_cel = re.search("<-((?:(?!>).)*)>", data) if table_cel: cel = 'colspan="' + table_cel.groups()[0] + '"' else: cel = 'colspan="' + str(round(len(start_data) / 2)) + '"' table_row = re.search("<\|((?:(?!>).)*)>", data) if table_row: row = 'rowspan="' + table_row.groups()[0] + '"' row_bgcolor = re.search("<rowbgcolor=(#(?:[0-9a-f-A-F]{3}){1,2}|\w+)>", data) if row_bgcolor: row_style += 'background: ' + row_bgcolor.groups()[0] + ';' table_border = re.search("<table ?bordercolor=(#(?:[0-9a-f-A-F]{3}){1,2}|\w+)>", data) if table_border: all_table += 'border: ' + table_border.groups()[0] + ' 2px solid;' table_bgcolor = re.search("<table ?bgcolor=(#(?:[0-9a-f-A-F]{3}){1,2}|\w+)>", data) if table_bgcolor: all_table += 'background: ' + table_bgcolor.groups()[0] + ';' bgcolor = re.search("<(?:bgcolor=)?(#(?:[0-9a-f-A-F]{3}){1,2}|\w+)>", data) if bgcolor: cel_style += 'background: ' + bgcolor.groups()[0] + ';' cel_width = re.search("<width=((?:(?!>).)*)>", data) if cel_width: cel_style += 'width: ' + cel_width.groups()[0] + 'px;' cel_height = re.search("<height=((?:(?!>).)*)>", data) if cel_height: cel_style += 'height: ' + cel_height.groups()[0] + 'px;' text_right = re.search("<\)>", data) text_center = re.search("<:>", data) text_left = re.search("<\(>", data) if text_right: cel_style += 'text-align: right;' elif text_center: cel_style += 'text-align: center;' elif text_left: cel_style += 'text-align: left;' elif num == 0: if re.search('^ (.*) $', cel_data): cel_style += 'text-align: center;' elif re.search('^ (.*)$', cel_data): cel_style += 'text-align: right;' elif re.search('^(.*) $', cel_data): cel_style += 'text-align: left;' text_class = re.search("<table ?class=((?:(?!>).)+)>", data) if text_class: table_class += text_class.groups()[0] all_table += '"' cel_style += '"' row_style += '"' table_class += '"' return [all_table, row_style, cel_style, row, cel, table_class, num] def start(conn, data, title): # DB 지정 curs = conn.cursor() # 초기 설정 data = '\n' + data + '\n' # 포함 문법 처리 while 1: include = re.search('\[include\(((?:(?!\)\]).)+)\)\]', data) if include: include = include.groups()[0] include_data = re.search('^((?:(?!,).)+)', include) if include_data: include_data = include_data.groups()[0] else: include_data = 'Test' include_link = include include = re.sub('^((?:(?!,).)+)', '', include) curs.execute("select data from data where title = ?", [include_data]) include_data = curs.fetchall() if include_data: include_parser = include_data[0][0] while 1: include_plus = re.search(', ?((?:(?!=).)+)=((?:(?!,).)+)', include) if include_plus: include_plus = include_plus.groups() include_parser = re.sub('@' + include_plus[0] + '@', include_plus[1], include_parser) include = re.sub(', ?((?:(?!=).)+)=((?:(?!,).)+)', '', include, 1) else: break include_parser = re.sub('\[\[분류:(((?!\]\]|#include).)+)\]\]', '', include_parser) data = re.sub('\[include\(((?:(?!\)\]).)+)\)\]', '\n' + include_parser + '\n', data, 1) else: include_link = re.sub('(, ?(?:(?:(?!,).)+))+$', '', include_link) data = re.sub('\[include\(((?:(?!\)\]).)+)\)\]', '[[' + include_link + ']]', data, 1) else: break # 개행 정리 data = re.sub('\r\n', '\n', data) # 표 정리 data = re.sub('\n( +)\|\|', '\n||', data) data = re.sub('\|\|( +)\n', '||\n', data) # XSS 이스케이프 data = html.escape(data) # 추가 데이터 지정 plus_data = '' # 수식 처리 first = 0 while 1: math = re.search('<math>((?:(?!<\/math>).)+)<\/math>', data) if math: if first == 0: plus_data += '' math = math.groups()[0] first += 1 data = re.sub('<math>((?:(?!<\/math>).)+)<\/math>', '', data, 1) plus_data += '' else: break # 추가 이스케이프 data = data.replace('\\', '\') # 텍스트 꾸미기 문법 data = re.sub(''''(?P((?!''').)+)'''', '\g', data) data = re.sub('''(?P((?!'').)+)''', '\g', data) data = re.sub('~~(?P(?:(?!~~).)+)~~', '\g', data) data = re.sub('--(?P(?:(?!~~).)+)--', '\g', data) data = re.sub('__(?P(?:(?!__).)+)__', '\g', data) data = re.sub('^^(?P(?:(?!^^).)+)^^', '\g', data) data = re.sub(',,(?P(?:(?!,,).)+),,', '\g', data) # 넘겨주기 변환 data = re.sub('\n#(?:redirect|넘겨주기) (?P(?:(?!\n).)+)\n', '', data) # [목차(없음)] 처리 if not re.search('\[목차\(없음\)\]\n', data): if not re.search('\[목차\]', data): data = re.sub('\n(?P={1,6}) ?(?P(?:(?!=).)+) ?={1,6}\n', '\n[목차]\n\g \g \g\n', data, 1) else: data = re.sub('\[목차\(없음\)\]\n', '', data) # 문단 문법 toc_full = 0 toc_top_stack = 6 toc_stack = [0, 0, 0, 0, 0, 0] edit_number = 0 toc_data = '
목차\n\n' while 1: toc = re.search('\n(={1,6}) ?((?:(?!=).)+) ?={1,6}\n', data) if toc: toc = toc.groups() toc_number = len(toc[0]) edit_number += 1 # 더 크면 그 전 스택은 초기화 if toc_full > toc_number: for i in range(toc_number, 6): toc_stack[i] = 0 if toc_top_stack > toc_number: toc_top_stack = toc_number toc_full = toc_number toc_stack[toc_number - 1] += 1 toc_number = str(toc_number) all_stack = '' # 스택 합치기 for i in range(0, 6): all_stack += str(toc_stack[i]) + '.' all_stack = re.sub('0.', '', all_stack) data = re.sub('\n(={1,6}) ?((?:(?!=).)+) ?={1,6}\n', '\n' + all_stack + ' ' + toc[1] + ' (편집)\n', data, 1) toc_data += '' + all_stack + ' ' + toc[1] + '\n' else: break toc_data += '
' data = re.sub('\[목차\]', toc_data, data) while 1: hr = re.search('\n-{4,9}\n', data) if hr: data = re.sub('\n-{4,9}\n', '\n
\n', data, 1) else: break data += '\n' # 중괄호 문법 처리 middle_stack = 0 middle_list = [] middle_number = 0 while 1: middle_data = re.search('(?:{{{((?:(?! |{{{).)*)|(}}}))', data) if middle_data: middle_data = middle_data.groups() if not middle_data[1]: if middle_stack > 0: middle_stack += 1 data = re.sub('(?:{{{((?:(?! |{{{).)*)|(}}}))', '{{{' + middle_data[0], data, 1) else: check = 0 middle_search = re.search('^(#(?:[0-9a-f-A-F]{3}){1,2})', middle_data[0]) if middle_search and check == 0: check = 1 middle_search = middle_search.groups()[0] middle_list += ['span'] data = re.sub('(?:{{{((?:(?! |{{{).)*)|(}}}))', '', data, 1) middle_search = re.search('^(?:#(\w+))', middle_data[0]) if middle_search and check == 0: check = 1 middle_search = middle_search.groups()[0] middle_list += ['span'] data = re.sub('(?:{{{((?:(?! |{{{).)*)|(}}}))', '', data, 1) middle_search = re.search('^(?:@((?:[0-9a-f-A-F]{3}){1,2}))', middle_data[0]) if middle_search and check == 0: check = 1 middle_search = middle_search.groups()[0] middle_list += ['span'] data = re.sub('(?:{{{((?:(?! |{{{).)*)|(}}}))', '', data, 1) middle_search = re.search('^(?:@(\w+))', middle_data[0]) if middle_search and check == 0: check = 1 middle_search = middle_search.groups()[0] middle_list += ['span'] data = re.sub('(?:{{{((?:(?! |{{{).)*)|(}}}))', '', data, 1) middle_search = re.search('^(\+|-)([1-5])', middle_data[0]) if middle_search and check == 0: check = 1 middle_search = middle_search.groups() if middle_search[0] == '+': font_size = str(int(middle_search[1]) * 20 + 100) else: font_size = str(100 - int(middle_search[1]) * 10) middle_list += ['span'] data = re.sub('(?:{{{((?:(?! |{{{).)*)|(}}}))', '', data, 1) if not check == 1: middle_list += ['code'] middle_stack += 1 data = re.sub('(?:{{{((?:(?! |{{{).)*)|(}}}))', '' + middle_data[0], data, 1) middle_number += 1 else: if middle_stack > 0: middle_stack -= 1 if middle_stack > 0: data = re.sub('(?:{{{((?:(?! |{{{).)*)|(}}}))', '}}}', data, 1) else: if middle_number > 0: middle_number -= 1 data = re.sub('(?:{{{((?:(?! |{{{).)*)|(}}}))', '', data, 1) del(middle_list[middle_number]) else: break # 일부 매크로 처리 data = tool.savemark(data) data = re.sub("\[br\]", '\n', data) data = re.sub("\[anchor\((?P(?:(?!\)\]).)+)\)\]", '', data) data = re.sub("\[nicovideo\((?P(?:(?!,|\)\]).)+)(?:(?:(?!\)\]).)*)\)\]", "[[http://embed.nicovideo.jp/watch/\g|\g]]", data) data = re.sub('\[ruby\((?P(?:(?!,).)+)\, ?(?P(?:(?!\)\]).)+)\)\]', '\g(\g)', data) # 원래 코드 재탕 now_time = tool.get_time() data = re.sub('\[date\]', now_time, data) time_data = re.search('^([0-9]{4}-[0-9]{2}-[0-9]{2})', now_time) time = time_data.groups() age_data = re.findall('\[age\(([0-9]{4}-[0-9]{2}-[0-9]{2})\)\]', data) for age in age_data: old = datetime.datetime.strptime(time[0], '%Y-%m-%d') will = datetime.datetime.strptime(age, '%Y-%m-%d') e_data = old - will data = re.sub('\[age\(([0-9]{4})-([0-9]{2})-([0-9]{2})\)\]', str(int(int(e_data.days) / 365)), data, 1) dday_data = re.findall('\[dday\(([0-9]{4}-[0-9]{2}-[0-9]{2})\)\]', data) for dday in dday_data: old = datetime.datetime.strptime(time[0], '%Y-%m-%d') will = datetime.datetime.strptime(dday, '%Y-%m-%d') e_data = old - will if re.search('^-', str(e_data.days)): e_day = str(e_data.days) else: e_day = '+' + str(e_data.days) data = re.sub('\[dday\(([0-9]{4}-[0-9]{2}-[0-9]{2})\)\]', e_day, data, 1) # 유튜브, 카카오 티비 처리 while 1: video = re.search('\[(youtube|kakaotv)\(((?:(?!\)\]).)+)\)\]', data) if video: video = video.groups() width = re.search(', ?width=((?:(?!,).)+)', video[1]) if width: video_width = width.groups()[0] else: video_width = '560' height = re.search(', ?height=((?:(?!,).)+)', video[1]) if height: video_height = height.groups()[0] else: video_height = '315' code = re.search('^(((?!,).)+)', video[1]) if code: video_code = code.groups()[0] else: if video[0] == 'youtube': video_code = 'BQ5PcIUcdUE' else: video_code = '66861302' if video[0] == 'youtube': video_code = re.sub('^https:\/\/www\.youtube\.com\/watch\?v=', '', video_code) video_code = re.sub('^https:\/\/youtu\.be\/', '', video_code) video_src = 'https://www.youtube.com/embed/' + video_code else: video_code = re.sub('^https:\/\/tv\.kakao\.com\/channel\/9262\/cliplink\/', '', video_code) video_code = re.sub('^http:\/\/tv\.kakao\.com\/v\/', '', video_code) video_src = 'https://tv.kakao.com/embed/player/cliplink/' + video_code +'?service=kakao_tv' data = re.sub('\[(youtube|kakaotv)\(((?:(?!\)\]).)+)\)\]', '', data, 1) else: break # 인용문 구현 while 1: block = re.search('(\n(?:> ?(?:(?:(?!\n).)+)?\n)+)', data) if block: block = block.groups()[0] block = re.sub('^\n> ?', '', block) block = re.sub('\n> ?', '\n', block) block = re.sub('\n$', '', block) data = re.sub('(\n(?:> ?(?:(?:(?!\n).)+)?\n)+)', '\n
' + block + '
\n', data, 1) else: break # 리스트 구현 while 1: li = re.search('(\n(?:(?: *)\* ?(?:(?:(?!\n).)+)\n)+)', data) if li: li = li.groups()[0] while 1: sub_li = re.search('\n(?:( *)\* ?((?:(?!\n).)+))', li) if sub_li: sub_li = sub_li.groups() # 앞의 공백 만큼 margin 먹임 if len(sub_li[0]) == 0: margin = 20 else: margin = len(sub_li[0]) * 20 li = re.sub('\n(?:( *)\* ?((?:(?!\n).)+))', '
  • ' + sub_li[1] + '
    • ', li, 1) else: break data = re.sub('(\n(?:(?: *)\* ?(?:(?:(?!\n).)+)\n)+)', '\n\n
      ' + li + '
    \n', data, 1) else: break # 들여쓰기 구현 while 1: indent = re.search('\n( +)', data) if indent: indent = len(indent.groups()[0]) # 앞에 공백 만큼 margin 먹임 margin = '' * indent data = re.sub('\n( +)', '\n' + margin, data, 1) else: break # 표 처리 while 1: table = re.search('\n((?:(?:(?:(?:\|\|)+(?:(?:(?!\|\|).(?:\n)*)+))+)\|\|(?:\n)?)+)', data) if table: table = table.groups()[0] # return [all_table, row_style, cel_style, row, cel, table_class, num] while 1: all_table = re.search('^((?:\|\|)+)((?:<(?:(?:(?!>).)+)>)*)((?:(?!\|\||<\/td>).)+)', table) if all_table: all_table = all_table.groups() return_table = table_parser(all_table[1], all_table[2], all_table[0]) number = return_table[6] table = re.sub('^((?:\|\|)+)((?:<(?:(?:(?!>).)+)>)*)', '\n
    \n', table, 1) else: break table = re.sub('\|\|\n$', '
    ', table) while 1: row_table = re.search('\|\|\n((?:\|\|)+)((?:<(?:(?:(?!>).)+)>)*)((?:(?!\|\||<\/td>).)+)', table) if row_table: row_table = row_table.groups() return_table = table_parser(row_table[1], row_table[2], row_table[0], number) table = re.sub('\|\|\n((?:\|\|)+)((?:<(?:(?:(?!>).)+)>)*)', '', table, 1) else: break while 1: cel_table = re.search('((?:\|\|)+)((?:<(?:(?:(?!>).)+)>)*)((?:(?:(?!\|\||<\/td>).)|\n)+)', table) if cel_table: cel_table = cel_table.groups() return_table = table_parser(cel_table[1], re.sub('\n', ' ', cel_table[2]), cel_table[0], number) table = re.sub('((?:\|\|)+)((?:<(?:(?:(?!>).)+)>)*)', '', table, 1) else: break data = re.sub('\n((?:(?:(?:(?:\|\|)+(?:(?:(?!\|\|).(?:\n)*)+))+)\|\|(?:\n)?)+)', table, data, 1) else: break # 링크 관련 문법 구현 category = '\n
    분류: ' while 1: link = re.search('\[\[((?:(?!\[\[|\]\]).)+)\]\]', data) if link: link = link.groups()[0] link_split = re.search('((?:(?!\|).)+)(?:\|((?:(?!\|).)+))', link) if link_split: link_split = link_split.groups() main_link = link_split[0] see_link = link_split[1] else: main_link = link see_link = link if re.search('^(파일|외부):', main_link): width = re.search('width=((?:(?!,).)+)', see_link) if width: file_width = width.groups()[0] else: file_width = 'auto' height = re.search('height=((?:(?!,).)+)', see_link) if height: file_height = height.groups()[0] else: file_height = 'auto' align = re.search('align=((?:(?!,).)+)', see_link) if align: file_align = align.groups()[0] if file_align == 'center': file_align = 'display: block; text-align: center;' else: file_align = 'float: ' + file_align + ';' else: file_align = '' if re.search('^외부:', main_link): file_src = re.sub('^외부:', '', main_link) file_alt = main_link else: file_data = re.search('^파일:((?:(?!\.).)+)\.(.+)$', main_link) if file_data: file_data = file_data.groups() file_name = file_data[0] file_end = file_data[1] else: file_name = 'TEST' file_end = 'jpg' file_src = '/image/' + tool.sha224(file_name) + '.' + file_end file_alt = '파일:' + file_name + '.' + file_end data = re.sub('\[\[((?:(?!\[\[|\]\]).)+)\]\]', '' + file_alt + '', data, 1) elif re.search('^분류:', main_link): see_link = re.sub('#include', '', see_link) main_link = re.sub('#include', '', main_link) if re.search('#blur', main_link): see_link = '스포일러' link_id = 'id="inside"' main_link = re.sub('#blur', '', main_link) else: link_id = '' category += '' + re.sub('^분류:', '', see_link) + ' / ' data = re.sub('\[\[((?:(?!\[\[|\]\]).)+)\]\]', '', data, 1) elif re.search('^wiki:', main_link): data = re.sub('\[\[((?:(?!\[\[|\]\]).)+)\]\]', '' + see_link + '', data, 1) elif re.search('^http(s)?:\/\/', main_link): data = re.sub('\[\[((?:(?!\[\[|\]\]).)+)\]\]', '' + see_link + '', data, 1) else: if re.search('^:', main_link): main_link = re.sub('^:', '', main_link) curs.execute("select title from data where title = ?", [main_link]) if not curs.fetchall(): link_id = 'id="not_thing"' else: link_id = '' data = re.sub('\[\[((?:(?!\[\[|\]\]).)+)\]\]', '' + see_link + '', data, 1) else: break # 각주 처리 footnote_number = 0 footnote_all = '\n
      ' footnote_dict = {} while 1: footnote = re.search('(?:\[\*((?:(?! |\]).)*)(?: ((?:(?!\]).)+))?\]|(\[각주\]))', data) if footnote: footnote_data = footnote.groups() if footnote_data[2]: footnote_all += '
    ' data = re.sub('(?:\[\*((?:(?! ).)*) ((?:(?!\]).)+)\]|(\[각주\]))', footnote_all, data, 1) footnote_all = '\n
      ' else: footnote = footnote_data[1] footnote_name = footnote_data[0] if footnote_name and not footnote: data = re.sub('(?:\[\*((?:(?! |\]).)*)(?: ((?:(?!\]).)+))?\]|(\[각주\]))', '(' + footnote_name + ')', data, 1) else: footnote_number += 1 if not footnote_name: footnote_name = str(footnote_number) else: footnote_dict.update({ footnote_name : str(footnote_number) }) footnote_all += '
    • (' + footnote_name + ') ' + footnote + '
      • ' data = re.sub('(?:\[\*((?:(?! |\]).)*)(?: ((?:(?!\]).)+))?\]|(\[각주\]))', '(' + footnote_name + ')', data, 1) else: break footnote_all += '
    ' if footnote_all == '\n
      ': footnote_all = '' data = re.sub('\n$', footnote_all, data, 1) # 분류 마지막 처리 category += '
      ' category = re.sub(' / <\/div>$', '', category) if category == '\n
      분류:
      ': category = '' data += category # 마지막 처리 data = re.sub('(?P<\/h[0-9]>)(\n)+', '\g', data) data = re.sub('<\/ul>\n\n', '\n', data) data = re.sub('^(\n)+', '', data) data = re.sub('(\n)+
        ', '
          ', data) data = re.sub('(?P).)*)>)\n', '\g', data) data = re.sub('(\n)?
          (\n)?', '
          ', data) data = re.sub('\n', '
          ', data) if data == '': data = '문서가 없습니다.' return [data, plus_data]