Closed (fixed)
Project:
Drupal core
Version:
8.0.x-dev
Component:
openid.module
Priority:
Normal
Category:
Task
Assigned:
Unassigned
Reporter:
Created:
13 Dec 2012 at 20:42 UTC
Updated:
29 Jul 2014 at 21:39 UTC
Jump to comment: Most recent file
Comments
Comment #1
berdirThis one is a bit more interesting.
Was able to remove the @todo about #1096890: drupal_http_request should return error if reaches max allowed redirects, Guzzle throws an exception here. Speaking of exceptions, not sure what to do with them. openid.module currently doesn't do any kind of logging on errors.
Comment #3
berdir#1: openid-guzzle-1866124-1.patch queued for re-testing.
Comment #5
berdir#1: openid-guzzle-1866124-1.patch queued for re-testing.
Comment #6
yesct commented#1: openid-guzzle-1866124-1.patch queued for re-testing.
Comment #7
sutharsan commented$response->getLocation() returns the requested location. Only $response->getPreviousResponse()->getLocation() returns the redirection location.
Without any knowledge of the process, we should ignore the notification and open a follow-up issue to evaluate the exception handling.
See above.
Comment #8
berdirRe-roll for the improved exceptions and using Drupal::httpClient()
Comment #9
yesct commented#576248: [policy] Coding Standards Update for PHP5
And, #1862538-10: Convert drupal_http_request usage in update.fetch.inc to Guzzle
Comment #10
ParisLiakos commentedbesides the two following minors seems good
we are only catching RequestExceptions so this use is not needed
missed this catch
Comment #11
berdirThanks, fixed that.
Comment #12
ParisLiakos commentedif testbot disagrees it ll let us know:)
thanks!
Comment #14
berdir#11: openid-guzzle-1866124-11.patch queued for re-testing.
Comment #15
berdirThat was a weird already-installed random fail. Let's try that again. Back to RTBC.
Comment #16
webchickThis is perhaps the most confusing set of comment/test pairs I've ever seen, since it *looks* like the comments are +1 what the code is doing. However, this is a pre-existing issue, so I guess they must be zero-based?
However, overall, this is eliminating tons of detailed-ly commented code in favour of pretty standard stuff. Looks good here, thanks!
Committed and pushed to 8.x.