KM_LC.py

#!/opt/bin/python3

# pip install markdownify

import os
import requests
import markdownify

question_id = os.environ.get('KMVAR_LeetCodeID', 'add-two-numbers')

session = ''

cookies={
    "LEETCODE_SESSION": session
}

def get_csrf_cookie() -> str:
    response = requests.get(
        "https://leetcode.com/",
        cookies=cookies,
    )

    return response.cookies["csrftoken"]

HEADERS = {
    'x-csrftoken': get_csrf_cookie(),
}

def find_until(lines, prefix) -> int:
  for i, line in enumerate(lines):
    if line.startswith(prefix):
      return i

  return len(lines)


def format_lines(lines):
  result = []
  while len(lines) != 0:
    line = lines[0]

    if line.strip() == "":
      lines = lines[1:]
      continue

    if line.startswith('<p><strong>') and line.endswith('</strong></p>'):
      result.append('- ' + markdownify.markdownify(line).strip())

      inner_line_count = find_until(lines[1:], '<p><strong>')

      inner_result = format_lines(lines[1:1+inner_line_count])
      for r in inner_result:
        result.append('\t' + r)

      lines = lines[1+inner_line_count:]
      continue

    if line.startswith("<pre>"):
      #result.append('- ```')
      inner_line_count = find_until(lines[1:], '</pre>')

      for line in lines[1:1+inner_line_count]:
        result.append('- '+markdownify.markdownify(line))

      #result.append('```')

      lines = lines[2+inner_line_count:]
      continue

    if line.startswith("<ul>"):
      inner_line_count = find_until(lines[1:], '</ul>')

      for line in lines[1:1+inner_line_count]:
        result.append(markdownify.markdownify(line).strip())

      lines = lines[2+inner_line_count:]
      continue

    if line.startswith("<ol>"):
      inner_line_count = find_until(lines[1:], '</ol>')

      for line in lines[1:1+inner_line_count]:
        result.append(markdownify.markdownify(line).strip())

      lines = lines[2+inner_line_count:]
      continue

    if line.startswith('<p>') and line.endswith('</p>'):
      result.append('- ' + markdownify.markdownify(line).strip())
      lines = lines[1:]
      continue

    if line.startswith('<img ') and line.endswith(' />'):
      result.append('- ' + markdownify.markdownify(line).strip())
      lines = lines[1:]
      continue


    result.append(lines[0])
    lines = lines[1:]

  return result

def html_to_markdown(html):
  return '\n'.join(format_lines(html.replace('\xa0', ' ').strip().split('\n')))


def get_leetcode_question_detail(title_slug: str):
    query = '''query questionData($titleSlug: String!) {
      question(titleSlug: $titleSlug) {
        questionId
        questionFrontendId
        categoryTitle
        boundTopicId
        title
        titleSlug
        content
        translatedTitle
        translatedContent
        isPaidOnly
        difficulty
        similarQuestions
        langToValidPlayground
        companyTagStats
        codeSnippets {
          lang
          langSlug
          code
          __typename
        }
        topicTags {
          name
          slug
          translatedName
          __typename
        }
        stats
        hints
        status
        sampleTestCase
        metaData
        judgerAvailable
        judgeType
        mysqlSchemas
        enableRunCode
        isDailyQuestion
        dailyRecordStatus
        editorType
        ugcQuestionId
        style
        exampleTestcases
        __typename
      }
    }
    '''

    request_body = {
        'operationName': "questionData",
        'query': query,
        'variables': {
            'titleSlug': title_slug,
        }
    }


    response = requests.post('https://leetcode-cn.com/graphql/', headers=HEADERS, cookies=cookies, json=request_body)

    detail = response.json()['data']['question']

    # detail['topicTagsLogseq'] = ''.join([' #[['+x['name']+']]' for x in detail['topicTags']])
    # detail['contentMarkdown'] = markdownify.markdownify(detail['content'])
    # detail['translatedContentMarkdown'] = markdownify.markdownify(detail['translatedContent'])
    # detail['translatedContentMarkdownLogseq'] = '\n'.join(['\t'+ x for x in html_to_markdown(detail['translatedContent']).split('\n')])


    return detail


def get_lintcode_question_detail(title_slug: str):
    response = requests.get('https://www.lintcode.com/problem/{}/'.format(title_slug), allow_redirects=False)
    if response.status_code != 301:
      raise RuntimeError('invalid response: status_code = {}'.format(response.status_code))

    location = response.headers['location']

    if location == 'https://www.lintcode.com/problem/':
        return None

    return {
        'questionId': int(location.split('/')[-2]),
    }


def get_question_detail(title_slug: str):
    leetcode_detail = get_leetcode_question_detail(title_slug)
    lintcode_detail = get_lintcode_question_detail(title_slug)

    detail = {
        'titleSlug': title_slug,

        'leetcode': leetcode_detail,
        'lintcode': lintcode_detail,
    }

    detail['title'] = leetcode_detail['title']
    detail['chinese-title'] = leetcode_detail['translatedTitle']
    detail['difficulty'] = leetcode_detail['difficulty']

    detail['content'] = '\n'.join(['\t'+ x for x in html_to_markdown(leetcode_detail['translatedContent']).split('\n')])

    tags = {x['translatedName'] for x in leetcode_detail['topicTags']}
    detail['tags'] = ' '.join(['#[[' + x + ']]' for x in tags])


    detail['leetcodeQuestionId'] = leetcode_detail['questionFrontendId']
    detail['leetcodeLine'] = '#Leetcode #Leetcode-{leetcodeQuestionId} [LC国际](https://leetcode.com/problems/{titleSlug}/) [LC中国](https://leetcode.cn/problems/{titleSlug}/)'.format(**detail)
    detail['ojLine'] = '- ' + detail['leetcodeLine']

    if lintcode_detail is not None:
        detail['lintcodeQuestionId'] = lintcode_detail['questionId']
        detail['lintcodeLine'] = '#lintcode #Lintcode-{lintcodeQuestionId} [LintCode](https://www.lintcode.com/problem/{lintcodeQuestionId}/)'.format(**detail)
        detail['ojLine'] += '\n- ' + detail['lintcodeLine']
        

    return detail


detail = get_question_detail(question_id)

template = '''
- alias:: {title}
- #Algorithm #{difficulty} {tags}
{ojLine}
- # 题目
{content}

'''


print(template.format(**detail).strip())