Skip to content

Conversation

@atuleu
Copy link

@atuleu atuleu commented Mar 13, 2013

grack fails under non relative_url_root . I tried to fix by stripping relative root if its is set.

I tried to find spec / features for git smart http acces but did not found any.

refs #1228
refs #3153

@coveralls
Copy link

Coverage decreased (-22.95%) when pulling 8945c90aa4e38e87570d9dd58f7f02bd0c41ef9d on atuleu:issue-3153 into 366bc32 on gitlabhq:master.

View Details

@coveralls
Copy link

Coverage decreased (-0.12%) when pulling cb6febc46137db2faadda2c291b64f24f6a560ee on atuleu:issue-3153 into 2f6d289 on gitlabhq:master.

View Details

@coveralls
Copy link

Coverage decreased (-0.12%) when pulling e6159b8 on atuleu:issue-3153 into 2f6d289 on gitlabhq:master.

View Details

@atuleu
Copy link
Author

atuleu commented Mar 14, 2013

Finally itd does go green on travis.

All of my pushes were working on my gitlab-vagrant machine, I had to rebase several tuime against master to make it work .... Does anyone have any idea why ?

@jrmithdobbs
Copy link
Contributor

I'm not entirely sure this is the right way to fix this; however, without this http/https cloning is completely broken with a relative root set currently.

@jrmithdobbs
Copy link
Contributor

Related to fixing relative root support in general is also #3458 .

@atuleu
Copy link
Author

atuleu commented Mar 31, 2013

I agree that it may not be the better way to fix this, I just hacked it really quickly as I really needed http cloning for the intern gitlab of my lab.

@rbeuque74
Copy link

Works pretty well to fix http(s) cloning on 5.1-2 thanks

@richnou
Copy link

richnou commented May 27, 2013

Those two commits from @atuleu should be merged, they are still needed in gitlab 5.2
I guess this bothers a lot of users, is there any way to upvote pull requests?

Proposed fix #3458 from @jrmithdobbs did not work for me.

Cheers

@Razer6
Copy link
Member

Razer6 commented Jun 11, 2013

@randx can you take a look?

@dosire
Copy link
Member

dosire commented Jun 19, 2013

Also see http://feedback.gitlab.com/forums/176466-general/suggestions/4102121-relative-url-support that makes a request to properly support relative urls again.

@preparationh67
Copy link

I'd say please upvote the request in the feedback link to draw more attention to this as this feature is more useful to the "little guy"

@tknerr
Copy link

tknerr commented Jun 26, 2013

@randx will this be merged?

The HTTP Clone issue still persists with Gitlab 5.0 - 5.3 when Gitlab is run under a relative URL.

I'm currently on version 5.3 and this pull request resolves the issue => I can now happily clone and push via http with Gitlab running under the /gitlab path

Thanks @atuleu 👍

@atuleu
Copy link
Author

atuleu commented Jul 30, 2013

@dosire I have seen that today some changement in the policy of relative url has been made. Could you comment zyourself or ask another dev that knows better the matter the two following question I put first in this PR :

  • Is there any test already existing for http cloning ? It does not seems to me to have any kind of test (back in 4.2).
  • How could I set up a test environment with travis that test both for relative url and non relative url (basically running travis twice )

Thank you for your help

@dosire
Copy link
Member

dosire commented Jul 31, 2013

@atuleu Sorry, I don't know if there is a test for http cloning. Running the same travis test twice doesn't seem like a workable solution to me. Please discuss in http://feedback.gitlab.com/forums/176466-general/suggestions/4102121-relative-url-support

@atuleu
Copy link
Author

atuleu commented Jul 31, 2013

Hi,

I posted (almost) the same message on the issue and the feedback forum.

There is no test for that particular feature. However I just tried to run
rake gitlab:test with RAILS_RELATIVE_URL_ROOT sets, and test pass alreadz
from green to red.

I setted on atuleu/gitlabhq relative-url-root branch a travis environement
with the relative URL root, to CI this issues

bests

2013/7/31 Sytse Sijbrandij notifications@github.com

@atuleu https://github.com/atuleu Sorry, I don't know if there is a
test for http cloning. Running the same travis test twice doesn't seem like
a workable solution to me. Please discuss in
http://feedback.gitlab.com/forums/176466-general/suggestions/4102121-relative-url-support


Reply to this email directly or view it on GitHubhttps://github.com//pull/3204#issuecomment-21845177
.

@Razer6
Copy link
Member

Razer6 commented Sep 17, 2013

Fix included in #4670

@Razer6 Razer6 closed this Sep 17, 2013
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

9 participants