Outside of the hostnames previously fixed, I talked with @williammartin about what to do with the remaining references and the concern about the mixed format.
One suggestion for the error / test use case was to remove the extraneous text all together.
There is still a bit of work to get the gh pr comment tests in order, however this goes a way towards fixing the issue along with acceptance tests.
Also, it turns out some of the issue acceptance tests were really running `pr` tests.
The previous commit failed to update the 2nd set of tests to rely upon GH_REPO instead of --repo flag.
These have been tested against `trunk` and fail because not being in a git directory when they should pass.
This commit builds on top of work previously done within acceptance tests around gh secret and remote disambiguation.
The choice of expanding this existing test rather than creating a new test was in order to keep context together within a single acceptance test rather than splitting it over multiple.
`gh repo rename` can only rename a repo, but not change its owner
(i.e. transfer it).
As flagged in #10034, at the moment, the CLI behaves like it can do
this, and produces weird results.
If I ask to rename my repo to `polyseam/frappe-containers`:
```bash
gh repo rename polyseam/frappe-containers
```
..the preview suggests that it will do the right thing:
```
? Rename polyseam/containers to polyseam/frappe-containers? Yes
```
...but the repo gets renamed wrongly:
```
✓ Renamed repository polyseam/polyseam-frappe-containers
```
This adds an upfront validation that looks for a slash and tells
you that the command can't transfer a repo.
Fixes#10034.
These are acceptance tests based on @williammartin work within various issues associated with #9208.
These need to be enhanced to account for `gh pr status`, however that should be quicker given these pass based on the updated branch and upstream fix for cross-repo same org changes.
Based on insights gained from reviewing conventions in #9815 with @jtmcg, I'm renaming this testscript to keep consistent with `gpg-key`, `label`, `ssh-key`, etc.
As this testscript creates a Bash-based script extension, the testscript should be skipped if it isn't on the path and executable.
Ideally, we would refactor this test to isolate that portion of the tests OR switch to a Go-based extension that can be compiled and run everywhere.
Similar to `gpg-key`, `label`, `ssh-key`, this coarse grained testscript should be named after the commandset given it isn't a collection of targeted scenarios.
