An example on using the Github GraphQL API with Python 3

graphql_example.py

# An example to get the remaining rate limit using the Github GraphQL API.

import requests

headers = {"Authorization": "Bearer YOUR API KEY"}


def run_query(query): # A simple function to use requests.post to make the API call. Note the json= section.
    request = requests.post('https://api.github.com/graphql', json={'query': query}, headers=headers)
    if request.status_code == 200:
        return request.json()
    else:
        raise Exception("Query failed to run by returning code of {}. {}".format(request.status_code, query))

        
# The GraphQL query (with a few aditional bits included) itself defined as a multi-line string.       
query = """
{
  viewer {
    login
  }
  rateLimit {
    limit
    cost
    remaining
    resetAt
  }
}
"""

result = run_query(query) # Execute the query
remaining_rate_limit = result["data"]["rateLimit"]["remaining"] # Drill down the dictionary
print("Remaining rate limit - {}".format(remaining_rate_limit))

Very helpful, thank you!

If you're using a Personal Access Token (PAT), you need to put the word "token" or "bearer" before the actual token:
headers = {"Authorization": "token XYZ123TOKENTOKENTOKEN"}

Very helpful demonstration 👏

Great demo! I only had to tweak the url which, for GitHub Enterprise, seems to have a slightly different scheme: e.g. https://github.mycompany.com/api/graphql

Very helpful! Thank you!

Thanks! :D

If you get a 401 code, do what @YakDriver says

Thanks!

Thanks!

we can also do a get request right with python request's module
to consume the api and use it to display the data right something like this.

r = requests.get()

i thought there are no status codes returned from GraphQL server other than 200. You may need to inspect the result to contain error field.

How to pass variables?

How to pass variables?

Same question here. Do you know how to pass variables?

To pass variables---here queryString and maxItems---you can use format, e.g.

query = """
{{
   search(query: "{queryString}", type: REPOSITORY, first: {maxItems}) {{
     repositoryCount
     edges {{
       node {{
         ... on Repository {{
           name
           url
           stargazers {{ totalCount }}
         }}
       }}
    }}
  }}
}}
"""
variables = {
   'queryString' : 'is:public stars:>9999',
   'maxItems' : '5'
}

result = run_query(query.format(**variables))

To pass variables---here queryString and maxItems---you can use format, e.g.

query = """
{{
   search(query: "{queryString}", type: REPOSITORY, first: {maxItems}) {{
     repositoryCount
     edges {{
       node {{
         ... on Repository {{
           name
           url
           stargazers {{ totalCount }}
         }}
       }}
    }}
  }}
}}
"""
variables = {
   'queryString' : 'is:public stars:>9999',
   'maxItems' : '5'
}

result = run_query(query.format(**variables))

This unfortunately gives me a "KeyError: '\n trip(\n from'" when parsing this data: https://hastebin.com/buxivileha.makefile

Any idea why?

Because, you don't need to format the string .. you can pass both query and variables as separate keys in the json and the graphql endpoint can figure out what goes were

For anyone trying to pass variables for these queries, I suggest looking at string.Template in python:

from string Import Template
queryTemplate = Template(
"""{
  viewer {
    repositories(first: $num) {
      pageInfo {
        hasNextPage
        endCursor
      }
      edges {
        node {
          name
        }
      }
    }
  }
}"""
)

query = queryTemplate.substitute(num=n)

@Zylvian @deepaksisodiya

To pass in variables to the query you need to:

• add variables parameter in your run_query function.
• add variables key to the json you that is going to be appended to the graphql endpoint
• create the variables dictionary with the parameters you want to pass

my python is a bit rusty but this works

import requests

class graphQL:

	headers = {"Authorization": "token <token_here>"}

	"""docstring for graphQL"""
	def __init__(self):
		super(graphQL, self).__init__()

	def run_query(self, query, variables): # A simple function to use requests.post to make the API call. Note the json= section.
		request = requests.post('https://api.github.com/graphql', json={'query': query, 'variables': variables}, headers=self.headers)
		if request.status_code == 200:
			return request.json()
		else:
			raise Exception("Query failed to run by returning code of {}. {}".format(request.status_code, query))
	


	def getClosedIssuesActors(self):
		listOfNames = []
		query = '''
			query($owner: String!, $name: String!) { 
				repository(owner: $owner, name: $name){
				  issues(states: CLOSED, first:10){
					edges{
					  node{
						... on Issue{
						  timeline(last: 100){
							edges{
							  node{
								__typename
								... on ClosedEvent{
								  actor{
									login
								  }
								}
							  }
							}
						  }
						}
					  }
					}
				  }
				}
			}'''

		variables = {
  			"owner": "tatmush",
  			"name": "Saturday-THORN-Dev-Rank"
		}	

		result = self.run_query(query, variables) #execute query
		print(result)

don't forget to declare the variables in the query itself.

Thankkk youu!!!!!!!!!!!!!! You saved me:)

Here is an example using generators to overcome GitHub's a limit of 100 items/request.

from requests import exceptions, request


class GitHubQuery:
    BASE_URL = "https://api.github.com/graphql"
    ORGANIZATION = "org_name"

    def __init__(
        self,
        github_token=None,
        query=None,
        query_params=None,
        additional_headers=None
    ):
        self.github_token = github_token
        self.query = query
        self.query_params = query_params
        self.additional_headers = additional_headers or dict()

    @property
    def headers(self):
        default_headers = dict(
            Authorization=f"token {self.github_token}",
        )
        return {
            **default_headers,
            **self.additional_headers
        }

    def generator(self):
        while(True):
            try:
                yield request(
                    'post',
                    GitHubQuery.BASE_URL,
                    headers=self.headers,
                    json=dict(query=self.query.format_map(self.query_params))
                ).json()
            except exceptions.HTTPError as http_err:
                raise http_err
            except Exception as err:
                raise err

    def iterator(self):
        pass

import GitHubQuery


class VulnurabilityReport(GitHubQuery):
    VULNERABILITY_QUERY = """
        {{
            organization(login: "org_name") {{
                repositories(first: 100, after: {after}) {{
                    pageInfo{{
                        hasNextPage
                        endCursor
                    }}
                    totalCount
                    nodes {{
                        ... on Repository {{
                        id
                        name
                        vulnerabilityAlerts(first: 2) {{
                            totalCount
                            nodes {{
                                id
                            }}
                        }}
                        }}
                    }}
                }}
            }}
        }}
        """
    QUERY_PARAMS = dict(after='')

    ADDITIONAL_HEADERS = dict(
        Accept="application/vnd.github.vixen-preview+json",
    )

    def __init__(self, github_token):
        super().__init__(
            github_token=github_token,
            query=VulnurabilityReport.VULNERABILITY_QUERY,
            query_params=VulnurabilityReport.QUERY_PARAMS,
            additional_headers=VulnurabilityReport.ADDITIONAL_HEADERS
        )

    def iterator(self):
        generator = self.generator()
        hasNextPage = True
        repos_vulnerabilities = []
        while(hasNextPage):
            response = next(generator)
            endCursor = response["data"]["organization"]["repositories"]["pageInfo"]["endCursor"]
            self.query_params = dict(after=endCursor)
            repos_vulnerabilities.extend(response["data"]["organization"]["repositories"]["nodes"])
            hasNextPage = response["data"]["organization"]["repositories"]["pageInfo"]["hasNextPage"]
        return (repos_vulnerabilities)

I just noticed @tatmush use of variables. I used python's format_map, but variables keep the query string simpler.

Hi, same in Python2 (here to check which users as enabled Single Sign-On on an Organization)
Credits for @tatmush and @BarakBD-Globality

class GitHubGraphQLQuery(object):

    BASE_URL = "https://api.github.com/graphql"

    def __init__(self, token, query, variables=None, additional_headers=None):
        self._token = token
        self._query = query
        self._variables = variables or dict()
        self._additional_headers = additional_headers or dict()

    @property
    def headers(self):
        default_headers = dict(Authorization="token {}".format(self._token))
        return dict(default_headers.items() + self._additional_headers.items())

    def generator(self):
        while True:
            try:
                yield requests.request(
                    "post",
                    GitHubGraphQLQuery.BASE_URL,
                    headers=self.headers,
                    json={"query": self._query, "variables": self._variables},
                ).json()
            except requests.exceptions.HTTPError as http_err:
                raise http_err
            except Exception as err:
                raise err

    def iterator(self):
        pass

class GithubSAMLIdentityProvider(GitHubGraphQLQuery):

    QUERY = """
        query($org: String!, $after: String) {
            organization(login: $org) {
                samlIdentityProvider {
                    ssoUrl,
                    externalIdentities(first: 100, after: $after) {
                        pageInfo {
                            hasNextPage
                            endCursor
                        }
                        edges {
                            node {
                                guid,
                                samlIdentity {
                                    nameId
                                }
                                user {
                                    login
                                }
                            }
                        }
                    }
                }
            }
        }
    """
    ADDITIONAL_HEADERS = dict(Accept="application/vnd.github.vixen-preview+json")

    def __init__(self, organization_name, token):
        super(GithubSAMLIdentityProvider, self).__init__(
            token=token,
            query=GithubSAMLIdentityProvider.QUERY,
            variables=dict(org=organization_name, after=None),
            additional_headers=GithubSAMLIdentityProvider.ADDITIONAL_HEADERS,
        )
        self._identities = list()

    def iterator(self):
        generator = self.generator()
        hasNextPage = True
        saml_identities = list()
        while hasNextPage:
            response = next(generator)
            endCursor = response["data"]["organization"]["samlIdentityProvider"]["externalIdentities"]["pageInfo"]["endCursor"]
            self._variables["after"] = endCursor
            saml_identities.extend(
                response["data"]["organization"]["samlIdentityProvider"]["externalIdentities"]["edges"]
            )
            hasNextPage = response["data"]["organization"]["samlIdentityProvider"]["externalIdentities"]["pageInfo"]["hasNextPage"]
        return saml_identities

Does anyone if it's possible to define variables inside fragments (see example below)?

        fragment repoQuery on Repository($states: [String!]) {
            pullRequests(first: 100, states: $states) {
                totalCount
                    nodes{
                        title

                }
            }
        }

Saved me hours of frustration.... Thank you!

Thanks, exactly what I needed! 👍

👍

Thank you!

Awesome, thanks!

If you're using a Personal Access Token (PAT), you need to put the word "token" or "bearer" before the actual token:
headers = {"Authorization": "token XYZ123TOKENTOKENTOKEN"}

Thanks!

Hey, thanks for this post. Can we enable auto completion or graphiql like features while making these kinds of requests from notebook?

Mutations are the same.

May I mention that the part where it says json={'query': query} does not vary regardless of whether it is a query or mutation. So, yes, it does end up with {"query": { "query"... or {"query": {"mutation":...

👍 amazing thank you...just what I was looking for.

This is just what I needed - thanks!

so cool, exactly what I need, thank you!

Easy to read, flexible and still works in 2021, thanks for this!!!!