[CI:DOCS] go-md2man: use vendored-in version, not system

[edsantiago]

go-md2man is fragile, especially around tables (#18678, #19278).
Podman man pages are finely tuned to look OK using v2.02, which
is what we vendor in test/tools, so we should really use it
instead of whatever is installed on the system.

This fixes 'make docs' on RHEL8, broken as of #19278.

Signed-off-by: Ed Santiago [email protected]

Fixes broken tables on some RHEL8 man pages

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: edsantiago

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [edsantiago]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[edsantiago]

Well, I guess now we know why we require system go-md2man. Giving up for today.

[Luap99]

I am pretty sure you have to remove the single quotes as they are part of the variable

[edsantiago]

It does not seem to be the quotes.

OSX Cross:

go build -o build/go-md2man ./vendor/github.com/cpuguy83/go-md2man/v2
go: creating work dir: stat podman_tmp_SzEc: no such file or directory

and Windows Cross:

/usr/bin/bash: line 6: ./test/tools/build/go-md2man: cannot execute binary file: Exec format error

[Luap99]

The windows one sets GOOS=windows GOARCH=amd64 so it cannot execute the windows binary on linux.

Note sure on OSX, that error makes no sense to me.

[edsantiago]

The podman_tmp_XXXX one on OSX looks suspiciously like something related to the TMPDIR situation I try to address in #19281: I think make podman-release is being called before make docs, and the bogus TMPDIR is causing harm.

[rhatdan]

LGTM

[lsm5]

since we're building from vendored sources, using vendored go-md2man is cool. Packit tasks are happy except the expected rawhide and el9 failures, which are unrelated issues.

[lsm5]

/lgtm
/hold