[BugFix] [API] Pull.API.Convert: Only try to get HeadBranch if HeadRepo exist

[6543]

close #10025

• remove one pr.Issue.LoadRepo (it is ececuted 2 times)
• prevent nil-pointer exeption on user.APIFormat()
• handle non existing HreadRepo

credits to @jolheiser @zeripath who found the witch part of the code i had to look at :)

[6543]

the diff looks weard on code bloc L101-129 but I only added a if statement before witch check if pr.HeadRepo != nil

[codecov-io]

Codecov Report

Merging #10029 into master will increase coverage by 0.07%.
The diff coverage is 60.86%.

@@            Coverage Diff             @@
##           master   #10029      +/-   ##
==========================================
+ Coverage   43.37%   43.45%   +0.07%     
==========================================
  Files         565      566       +1     
  Lines       78905    78987      +82     
==========================================
+ Hits        34226    34321      +95     
+ Misses      40469    40429      -40     
- Partials     4210     4237      +27

Impacted Files	Coverage Δ
modules/structs/repo.go	36.36% <ø> (ø)	⬆️
routers/repo/setting.go	15.01% <0%> (+0.02%)	⬆️
routers/api/v1/api.go	75.07% <100%> (+0.61%)	⬆️
services/repository/transfer.go	55.55% <28.57%> (-11.12%)	⬇️
routers/api/v1/repo/transfer.go	72.85% <72.85%> (ø)
services/pull/temp_repo.go	31.62% <0%> (-2.57%)	⬇️
models/unit.go	37.03% <0%> (-2.47%)	⬇️
routers/repo/view.go	38.59% <0%> (-0.88%)	⬇️
services/pull/patch.go	67.92% <0%> (ø)	⬆️
... and 6 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f2dd94f...10413ce. Read the comment docs.

[6543]

@lunny done

[lunny]

@6543 I think you could add a test append on function TestPullBranchDelete. I have tested a broken pull request there.

[6543]

@lunny test added
@lafriks changed

general Info: while I was creating the test i roun in a nil-pointer-exeption so i added a check on user.APIFormat() <- the test now works without this fix but it might prevent som 500 in the future

[lunny]

Only a nit #10029 (comment)

[6543]

by the way should we backport this?

[techknowlogick]

@6543 please backport :)