Leetcode 题目数据分析

data

import requests
import os
import json
import re
import time
import html
import html2text

root_url = 'https://leetcode.com'
contest_info_api = 'https://leetcode.com/contest/api/info/weekly-contest-'
question_urls = 'https://leetcode.com/graphql'

save_json = True

leet_dict = {}

raw_data = 'raw_data'


def get_problems(res_json):
    ret_dict = {}
    probs = res_json['stat_status_pairs']
    for prob in probs:
        # 跳过付费
        if prob['paid_only']:
            continue
        front_id = prob['stat']['frontend_question_id']
        # 访问problem需要的是titleSlug
        ret_dict[front_id] = {}
        ret_dict[front_id]['titleSlug'] = prob['stat']['question__title_slug']
    return ret_dict


def get_problems_raw():
    all_questions_url = 'https://leetcode.com/api/problems/all'
    req = requests.get(url=all_questions_url)
    htmlstr = req.text
    problems_json = json.loads(htmlstr)
    if save_json:
        with open('problems.json', 'w') as f:
            json.dump(problems_json, f, indent=4)
    return problems_json, get_problems(problems_json)


def get_one_problem_detail(res_json):
    prob_dict = {}
    question = res_json['data']['question']
    prob_dict['questionFrontendId'] = question['questionFrontendId']
    prob_dict['title'] = question['title']
    prob_dict['titleSlug'] = question['titleSlug']
    prob_dict['difficulty'] = question['difficulty']

    prob_dict['likes'] = question['likes']
    prob_dict['dislikes'] = question['dislikes']

    # prob_dict['similarQuestions'] = question['similarQuestions']
    prob_dict['similarQuestions'] = []
    similar_json = json.loads(question['similarQuestions'])
    for similar in similar_json:
        prob_dict['similarQuestions'] .append(similar['titleSlug'])

    # prob_dict['topicTags'] = question['topicTags']
    prob_dict['topicTags'] = []

    for tag in question['topicTags']:
        prob_dict['topicTags'].append(tag['name'])

    # prob_dict['stats'] = question['stats']
    stats = json.loads(question['stats'])
    prob_dict['totalAcceptedRaw'] = stats['totalAcceptedRaw']
    prob_dict['totalSubmissionRaw'] = stats['totalSubmissionRaw']
    prob_dict['acceptRatio'] = prob_dict['totalAcceptedRaw'] / \
        prob_dict['totalSubmissionRaw']

    prob_dict['hints'] = question['hints']

    content = question['content']
    content = content.replace('\t', '').replace('\n\n', '\n')
    content = html2text.html2text(content)
    content = content.replace('**Input:**', 'Input:').replace('**Output:**',
                                                            'Output:').replace('**Explanation:**', 'Explanation:')
    prob_dict['content'] = content

    with open('markdown/{}.md'.format(prob_dict['questionFrontendId']), 'w') as f:
        f.write(content)

    return prob_dict


def get_one_problem_raw(question_url):
    if question_url[-1] == '/':
        question_url = question_url[:-1]
    question_str = question_url.split('/')[-1]
    req_json = {"operationName": "questionData", "variables": {"titleSlug": "check-if-the-sentence-is-pangram"},
                "query": "query questionData($titleSlug: String!) {\n  question(titleSlug: $titleSlug) {\n    questionId\n    questionFrontendId\n    boundTopicId\n    title\n    titleSlug\n    content\n    translatedTitle\n    translatedContent\n    isPaidOnly\n    difficulty\n    likes\n    dislikes\n    isLiked\n    similarQuestions\n    exampleTestcases\n    contributors {\n      username\n      profileUrl\n      avatarUrl\n      __typename\n    }\n    topicTags {\n      name\n      slug\n      translatedName\n      __typename\n    }\n    companyTagStats\n    codeSnippets {\n      lang\n      langSlug\n      code\n      __typename\n    }\n    stats\n    hints\n    solution {\n      id\n      canSeeDetail\n      paidOnly\n      hasVideoSolution\n      paidOnlyVideo\n      __typename\n    }\n    status\n    sampleTestCase\n    metaData\n    judgerAvailable\n    judgeType\n    mysqlSchemas\n    enableRunCode\n    enableTestMode\n    enableDebugger\n    envInfo\n    libraryUrl\n    adminUrl\n    __typename\n  }\n}\n"}

    req_json['variables']['titleSlug'] = question_str
    headers = {'Content-Type': 'application/json;charset=UTF-8'}
    res = requests.request("post", question_urls,
                           json=req_json, headers=headers, timeout=60)

    res_json = json.loads(res.text)

    if save_json:
        front_id = res_json['data']['question']['questionFrontendId']
        filename = 'raw_data/{}.json'.format(front_id)
        with open(filename, 'w') as f:
            json.dump(res_json, f, indent=4)

    return res_json


_, prob_dict = get_problems_raw()

output_dict = {}
for front_id, problem_detail in prob_dict.items():
    if os.path.exists('raw_data/{}.json'.format(front_id)):
        with open('raw_data/{}.json'.format(front_id)) as f:
            raw = json.load(f)
        # continue
    else:
        titleSlug = problem_detail['titleSlug']
        question_url = 'https://leetcode.com/problems/' + titleSlug
        raw = get_one_problem_raw(question_url)
    data = get_one_problem_detail(raw)
    output_dict[front_id] = data

with open('output.json', 'w')as f:
    json.dump(output_dict, f, indent=4)

process

import json

with open('output.json') as f:
    problems = json.load(f)
    datas = []
    for id,problem in problems.items():
        if problem['dislikes'] ==0:
            problem['dislikes'] = 1
        # likeRatio = problem['likes']/ (problem['likes'] + problem['dislikes'])
        likeRatio = problem['likes']/ ( problem['dislikes'])
        data = {}
        data['id'] = id
        data['titleSlug']= problem['titleSlug']
        data['likeRatio'] = likeRatio
        data['acceptRatio'] = problem['acceptRatio']
        data['difficulty'] = problem['difficulty']
        data['title'] = problem['title']

        datas.append(data)

    datas = sorted(datas,reverse= True,key = lambda data:data['acceptRatio'])

    steps = [10,20,30,40,50,60,70,80,90,100]

    for step in steps:
        count = 0
        for data in datas:
            count = count +1
            if  count  >= len(datas) * step / 100:
                # print( "{}%".format(step), data['acceptRatio'])
                break

    newdatas = []
    for data in datas:
        if data['acceptRatio'] > 0.5824890307140008 or data['acceptRatio'] < 0.4303226863502119:
            continue
        if data['likeRatio'] < 2:
            continue
        newdatas.append(data)
    newdatas = sorted(newdatas,reverse=True,key= lambda data:data['likeRatio'])
    # print(newdatas)
    for n in newdatas:
        mstr = '[{} {}. {}](https://leetcode.com/problems/{})\n'.format(n['id'],n['difficulty'],n['title'],n['titleSlug'])
        print(mstr)

正确率分数段

10% 0.7095958690429233
20% 0.6340542320781645
30% 0.5824890307140008
40% 0.5411619367900896
50% 0.5014095348054072
60% 0.46498277841561425
70% 0.4303226863502119
80% 0.38708126302292034
90% 0.3427132331515631
100% 0.1391385259566549

990 Medium. Satisfiability of Equality Equations

1579 Hard. Remove Max Number of Edges to Keep Graph Fully Traversable

1235 Hard. Maximum Profit in Job Scheduling

1368 Hard. Minimum Cost to Make at Least One Valid Path in a Grid

72 Hard. Edit Distance

958 Medium. Check Completeness of a Binary Tree

1713 Hard. Minimum Operations to Make a Subsequence

1388 Hard. Pizza With 3n Slices

1546 Medium. Maximum Number of Non-Overlapping Subarrays With Sum Equals Target

42 Hard. Trapping Rain Water

1671 Hard. Minimum Number of Removals to Make Mountain Array

1643 Hard. Kth Smallest Instructions