Python 从GitHub API提取的提交与我从git.Repo为同一项目获得的提交不同
我试图做的是:我想提取在拉请求中修改过的类的名称。为此,我采取以下措施: 从GitHub API:Python 从GitHub API提取的提交与我从git.Repo为同一项目获得的提交不同,python,git,commit,pull-request,Python,Git,Commit,Pull Request,我试图做的是:我想提取在拉请求中修改过的类的名称。为此,我采取以下措施: 从GitHub API: projects = [ {'owner':'google', 'repo':'gson', 'pull_requests': []}] def get(url): global nb PARAMS = { 'client_id': '----my_client_id---', 'client_secret': '---my_client_secret---'
projects = [ {'owner':'google', 'repo':'gson', 'pull_requests': []}]
def get(url):
global nb
PARAMS = {
'client_id': '----my_client_id---',
'client_secret': '---my_client_secret---',
'per_page': 100,
'state': 'all' #open, closed, all
}
result = requests.get(url = url, params=PARAMS)
nb+=1
if(not (result.status_code in [200, 304])):
raise Exception('request error', url, result, result.headers)
data = result.json()
while 'next' in result.links.keys():
result = requests.get(url = result.links['next']['url'],
params=PARAMS)
data.extend(result.json())
nb+=1
return data
def get_pull_requests(repo):
url = 'https://api.github.com/repos/{}/pulls'.format(repo)
result = get(url)
return result
def get_commits(url):
result = get(url)
return result
for i,project in enumerate(projects):
project['pull_requests'] =
get_pull_requests('{}/{}'.format(project['owner'],project['repo']))
for p in project['pull_requests']:
p['commits'] = get_commits(p['commits_url'])
print('{}/{}'.format(project['owner'],project['repo']), ':',
len(project['pull_requests']))
1) 我提取一个项目的所有请求
2) 我提取每个拉请求的所有提交
3) 对于每个请求,我只保留第一次提交和最后一次提交
因为此时,我不知道如何提取这两个提交请求之间的修改类列表,所以我使用'git'包,如下所示:
我在D:\\projects\\gson
import git
repo = git.Repo("D:\\projects\\gson")
commits_list = list(repo.iter_commits())
temp = []
for x in commits_list[0].diff(commits_list[-1]):
if (x.a_path == x.b_path):
if x.a_path.endswith('.java'):
temp.append(x.a_path)
else:
if x.b_path.endswith('.java'):
temp.append(x.b_path)
以下是我如何从GitHub API提取提交:
projects = [ {'owner':'google', 'repo':'gson', 'pull_requests': []}]
def get(url):
global nb
PARAMS = {
'client_id': '----my_client_id---',
'client_secret': '---my_client_secret---',
'per_page': 100,
'state': 'all' #open, closed, all
}
result = requests.get(url = url, params=PARAMS)
nb+=1
if(not (result.status_code in [200, 304])):
raise Exception('request error', url, result, result.headers)
data = result.json()
while 'next' in result.links.keys():
result = requests.get(url = result.links['next']['url'],
params=PARAMS)
data.extend(result.json())
nb+=1
return data
def get_pull_requests(repo):
url = 'https://api.github.com/repos/{}/pulls'.format(repo)
result = get(url)
return result
def get_commits(url):
result = get(url)
return result
for i,project in enumerate(projects):
project['pull_requests'] =
get_pull_requests('{}/{}'.format(project['owner'],project['repo']))
for p in project['pull_requests']:
p['commits'] = get_commits(p['commits_url'])
print('{}/{}'.format(project['owner'],project['repo']), ':',
len(project['pull_requests']))
这两种代码都有效。问题是,我从GitHub API获得了287次提交,但对于同一个项目,从git.Repo获得的提交只有86次。当我尝试匹配这些提交时,少于40个提交匹配
问题:
1) 为什么我在同一个项目中得到不同的承诺
2) 哪一个是正确的,我应该使用
3) 有没有一种方法可以让我知道使用Git.Repo提交的请求是什么
4) 有没有办法在GithubAPI中的两次提交之间提取修改过的类
5) 有没有人知道更好的方法来提取每个拉请求修改过的类
我知道这是一篇很长的帖子,但我在这里尽量说得具体一些。如果您能回答这些问题,我们将不胜感激