I am using Bitbucket API to retrieve different information. However I am looking to do a request that retrieves the latest commit for a repository. I initially thought it would be done like this:
https://bitbucket.org/!api/2.0/repositories/xxxx/xxxx/commits?limit=1
This just showed all the commits as normal but I want to show the most recent one. From looking through the API documentation I can't find anything that shows about limiting the number of commits to show. So was wondering if anyone could point me in the right direction?
Okay this wasn't straight forward either and I spent a couple of hours looking for how to do this myself. It ended up being easy if not unfortunate. Simply put the available API will not return just a target commit (like the latest). You have do parse it yourself. The following api:
"https://api.bitbucket.org/2.0/repositories/<project>/<repo>/commits/<branch>?limit=1"
Will still return ALL the commits for that particular branch BUT in order. So you can simply just grab the first result on the first page that is returned and that's the most recent commit for that branch. Here is a basic python example:
import os
import requests
import json
headers = {"Content-Type": "application/json"}
USER = ""
PASS = ""
def get_bitbucket_credentials():
global USER, PASS
USER = "<user>"
PASS = "<pass>"
def get_commits(project, repo, branch):
return json.loads(call_url("https://api.bitbucket.org/2.0/repositories/%s/%s/commits/%s?limit=1" % (project, repo, branch)))
def get_modified_files(url):
data = json.loads(call_url(url))
file_paths = []
for value in data["values"]:
file_paths.append(value["new"]["path"])
return file_paths
def call_url(url):
global USER, PASS
response = requests.get(url, auth=(USER, PASS), headers=headers)
if response.status_code == requests.codes.ok:
return response.text
return ""
if __name__ == "__main__":
get_bitbucket_credentials()
data = get_commits("<project>","<repo>","<branch>")
for item in data["values"]:
print("Author Of Commit: "+item["author"]["raw"])
print("Commit Message: "+item["rendered"]["message"]["raw"])
print("List of Files Changed:")
print(get_modified_files(item["links"]["diff"]["href"].replace("/diff/","/diffstat/")))
break
You can run the above example:
python3 mysavedfile.py
and that will output like:
Author Of Commit: Persons Name <personemail#email.com>
Commit Message: my commit message
List of Files Changed:
['file1.yaml','file2.yaml']
There a simple way to do it, using Bitbucket pagination mechanism.
Like so:
https://bitbucket.org/!api/2.0/repositories/xxxx/xxxx/commits?pagelen=1
Normally you'll need to specify the page number, "page=1", but the default is 1 so...
Related
I am writing a REST API GET endpoint that needs to both return a response and store records to either GCP Cloud SQL (MySQL), but I want the return to not be dependent on completion of the writing of the records. Basically, my code will look like:
def predict():
req = request.json.get("instances")
resp = make_response(req)
write_to_bq(req)
write_to_bq(resp)
return resp
Is there any easy way to do this with Cloud SQL Client Library or something?
Turns our flask has a functionality that does what I require:
#app.route("predict", method=["GET"]):
def predict():
# do some stuff with the request.json object
return jsonify(response)
#app.after_request
def after_request_func(response):
# do anything you want that relies on context of predict()
#response.call_on_close
def persist():
# this will happen after response is sent,
# so even if this function fails, the predict()
# will still get it's response out
write_to_db()
return response
One important thing is that a method tagged with after_request must take an argument and return something of type flask.Response. Also I think if method has call_on_close tag, you cannot access from context of main method, so you need to define anything you want to use from the main method inside the after_request tagged method but outside (above) the call_on_close method.
I'm working in Flask and I want to allow users to enter information into a form without logging in but be required to login if they submit the form. After logging in, it should be as though a user just submitted the form (they shouldn't have to re-enter any information).
To store their information, I've used sessions like this. It works well:
if request.method == "POST":
if "arg1" not in session.keys() and "arg2" not in session.keys():
session["arg1"] = request.form.get('arg1')
session["arg2"] = request.form.get('arg2')
However, I'm having trouble with the login required part. I know I can use #login_required on the whole route but I just want #login_required to apply if the request is a post method. I've tried simply adding #login_required after checking if the method is a post request but it doesn't work.
My login route looks like this:
#app.route("/login", methods = ["POST", "GET"])
def login():
#log user in
return redirect(request.args.get("next") or url_for('index'))
It seems as though I need two things.
1: To apply #login_required solely to a post request.
2: To have request.args.get("next") call a post request, not get request
How could I go about doing these two things and achieve my goal?
Thank you!
Break out your routes. 1 for GET and 1 for POST
#app.route("/login", methods = ["GET"])
def get_login():
return stuff
#app.route("/login", methods = ["POST"])
#login_required
def post_login():
return stuff
There are a couple patterns that could be used here but this one is the most straight forward.
With reference to https://docs.pro.coinbase.com/#get-account-history
HTTP REQUEST
GET /accounts//holds
I am struggling to produce python code to get the account holds via API pagination request and I could not find any example of implementing it.
Could you please advise me on how to proceed with this one?
According to Coinbase Pro documentation, pagination works like so (example with the holds endpoint):
import requests
account_id = ...
url = f'https://api.pro.coinbase.com/accounts/{account_id}/holds'
response = requests.get(url)
assert response.ok
first_page = response.json() # here is the first page
cursor = response.headers['CB-AFTER']
response = requests.get(url, params={'after': cursor})
assert response.ok
second_page = response.json() # then the next one
cursor = response.headers['CB-AFTER']
# and so on, repeat until response.json() is an empty list
You should properly wrap this into helper functions or class, or, even better, use an existing library and save dramatic time.
This question already has an answer here:
Is there a way to update the headers in one feature file and use the Auth token from Karate.config.js?
(1 answer)
Closed 1 year ago.
I'm trying to setup a framework to run Graphql calls and create and E2E environment.
I've got the following setup so far but i can't seem to get the headers part of it working. i have managed to set the auth for each request and it all works but as it logs in for each request it doesn't really work as expected.
I want do the following steps:
run a login Test (different usernames valid/invalid)
run a logout test (Ensure token is removed)
Then login with correct user and extract the "set-cookie" header (to use globally for all future requests)
I was trying to use the following:
Karate-config.js
karate.callSingle('classpath:com/Auth/common-headers.feature', config);
headers.js
function fn() {
var headers = {}
headers["set-cookie"] = sessionAccessId
karate.log('Cookie Value: ', headers)
return headers
}
common-headers.feature
Feature: Login to Application and extract header
Background:
* url serverAuthenticateUri
* header Accept = 'application/json'
Scenario: 'Login to the system given credentials'
Given request { username: '#(username)', password: '#(password)'}
When method post
Then status 200
And match $.success == '#(result)'
And def myResult = response
* def sessionAccessId = responseHeaders['set-cookie'][0]
* configure headers = read('classpath:headers.js')
* print 'headers:', karate.prevRequest.headers
feature-file.feature
Feature: sample test script
Background:
* url serverBaseUri
* def caseResp = call read('classpath:com/E2E/POC/CommonFeatures/CreateCaseRequest.feature')
* def caseReqId = caseResp.response.data.createCaseAndRequest.siblings[0].id
* def caseId = caseResp.response.data.createCaseAndRequest.siblings[0].forensicCaseId
* def graphQlCallsPath = 'classpath:com/E2E/POC/GraphQl/intForensic/'
* def commmonFiles = 'classpath:E2E/CommonFiles/'
Scenario: TC1a - Request Server Details from Config DB (1st Run):
Should handle requesting Server Details Data from Config Database.
* def queryFile = graphQlCallsPath + '20-TC1a_req_req_valid_id.graphql'
* def responseFile = graphQlCallsPath + '20-TC1a_resp_req_valid_id.json'
Given def query = read(queryFile)
And replace query.reqId = caseReqId
And request { query: '#(query)' }
When method post
Then status 200
And json resp = read(responseFile)
And replace resp.reqId = caseReqId
And replace resp.caseID = caseId
And match resp == $
I can log in correctly and i get the set-cookie token but this isn't being passed on the feature-file.feature and i get an error saying "not logged in" in the response.
Any help appreciated! I might be looking at this totally wrong and i have tried to follow the shared-scope as much as i can but can't understand in.
Please make this change and hopefully that works !
headers["set-cookie"] = karate.get('sessionAccessId');
Why is explained here: (read the whole section carefully) https://github.com/intuit/karate#configure-headers
EDIT: one more suggestion:
var temp = karate.callSingle('classpath:com/Auth/common-headers.feature', config);
karate.configure('headers', { 'set-cookie': temp.sessionAccessId });
Some extra suggestions:
If you have just started with Karate - based on your question I would suggest you get one flow working in a single Scenario first without any use of call and with nothing whatsoever in karate-config.js. Hard-code everything and get it working first. Use the header keyword to set any headers you need. I also see you are trying to set a set-cookie header (which may work) but Karate has a special keyword for cookie.
And don't even think about callSingle() to start with :)
Once you get that first "hard-coded" flow working, then attempt to configure headers and then only finally try to do "framework" stuff. You seem to have jumped straight into super-complexity without getting the basics right.
Please read this other answer as well, because I suspect that you or someone in your team is attempting to introduce what I refer to as "too much re-use": https://stackoverflow.com/a/54126724/143475 - try not to do this.
Also note that your question is so complex that I have not been able to follow it, so please ask a simpler or more specifc question next time. If you still are stuck, kindly follow this process: https://github.com/intuit/karate/wiki/How-to-Submit-an-Issue
I'm searching for a way to let people open Trac ticket by email.
The only solution I've found so far is email2trac | https://oss.trac.surfsara.nl/email2trac/wiki The problem with this solution is that I don't want to install & setup a mailserver. I would like a less invasive solution.
I was thinking about a cron script that download messages from a POP3 account and open/update tickets by parsing the content.
Is this possible ?
I was thinking about a cron script that download messages from a POP3
account and open/update tickets by parsing the content. Is this
possible ?
I think it would be possible yes. Certainly once you had the data from a POP3 account, you could iterate over it and create/update tickets as appropriate with the Trac API.
For the data retrieval step, you could create a new plugin, with a Component which implements the IAdminCommandProvider interface. How you actually retrieve and parse the data is an implementation detail for you to decide, but you could probably use the email/poplib modules and follow some of the parsing structure from email2trac.
For some untested boilerplate to get you started...
from trac.admin import IAdminCommandProvider
from trac.core import Component, implements
from trac.ticket import Ticket
def EmailToTicket(Component):
implements(IAdminCommandProvider)
def get_admin_commands(self):
yield ('emailtoticket retrieve',
'Retrieve emails from a mail server.'
None, self._do_retrieve_email)
def _do_retrieve_email(self):
# TODO - log into the mail server, then parse data.
# It would be nice to have a tuple of dictionaries,
# with keys like id, summary, description etc
# iterate over the data and create/update tickets
for email in emails:
if 'id' in email: # assuming email is a dictionary
self._update_ticket(email)
else:
self._create_ticket(email)
def _update_ticket(self, data):
ticket = Ticket(self.env, data[id])
for field, value in data.iteritems():
ticket[field] = value
ticket.save_changes(author, comment, when)
def _create_ticket(self, data):
ticket = Ticket(self.env)
for field, value in data.iteritems():
ticket[field] = value
ticket.insert()
You could then have Cron tab execute this command via TracAdmin (the frequency is up to you - the below example runs every minute)
* * * * * trac-admin /path/to/projenv emailtoticket retrieve
The find out more about plugin development, you should read this Trac wiki page.