Repository navigation
TAPHandle is CDS/VOFrance - #118
Conversation
JeremyMcCormick
left a comment
JeremyMcCormick
left a comment
There was a problem hiding this comment.
Suggested a few minor changes.
|
@msdemlei I have invited you to the Web Editors team so that you can make PRs on branches going forward. This is preferred because our CI which deploys the test site doesn't work on forks due to authentication issues. The fork is fine for this PR since it is such a minor change, but please use branches on the main repo going forward to suggest updates. |
|
TAPHandle has been developed by the SSC XMM-Newton team in Strasbourg. |
Is there an appropriate acronym that can be used for this? The other organizations in this column are quite short as they are all acronyms. (This page really needs a glossary for these, but that's another issue.) |
|
ObAS |
e4d0c9f to
cf1e139
Compare
|
This CI failed because the SFTP configuration is incorrect for the preview of a branch. It has been correctly configured for the @molinaro-m , @JeremyMcCormick , @msdemlei Is it correct if I copy-paste the configuration (username, password) from deploy-main.yaml into deploy-version.yaml? Of course, I won't change the target directory. |
Its author (Laurent Michel) suggested "Strasbourg Observatory - SSC XMM-Newton". However, a shorter version was preferred: "ObAS".
Also write the complete affiliation. It is an important product from SSC XMM-Newton and is made by the Observatory of Strasbourg (not CDS)....so, both deserve to be equally highlighted.
…path, and renaming port and key ones
|
Hi @gmantele, @msdemlei, @JeremyMcCormick , adding @szautkin. I think the issue is this branch being on a fork. If I remember correctly, the workflows were set up to work from branches on the ivoa/ivoa-web repo. Not sure if this is actually the issue, nor how to solve it, besides moving the branch into this repo. |
|
One could try to move the branch in the |
|
On Wed, Sep 02, 2026 at 01:34:26AM -0700, Grégory Mantelet wrote:
gmantele left a comment (ivoa/ivoa-web#118)
One could try to move the branch in the `ivoa` organization and let's see.
@msdemlei , since it is your PullRequest, do you want to?
Hm. I'm not really sure what "upload the branch version" is supposed
to do, but frankly I think we shouldn't do uploads from within a PR's
CI. To my instinct, there's nothing wrong with building the page and
perhaps run a few tests on the result (perhaps "new broken links",
although I'd really limit that to internal links), but that's it.
Was the idea to have a preview rendering somewhere?
Anyway, I've burned far too much thoughts and time trying to fiddle
with tricking github's permission system into moving a bug recently
that I'm reluctant to dive into this again.
I think it's pretty clear what the change does. Why don't we just
back out d523b9f and merge?
|
|
OK, I take the risk on me, since it's definitely minimal, a change on 1 line, not affecting other stuff. Anyway, yes, the idea is to have a staging preview before putting things in production, that's why we are locked in the PRs from branches on the primary origin repo. |
|
merged |
|
On Wed, Sep 02, 2026 at 03:02:32AM -0700, Marco Molinaro wrote:
molinaro-m left a comment (ivoa/ivoa-web#118)
Anyway, yes, the idea is to have a staging preview before putting
things in production, that's why we are locked in the PRs from
branches on the primary origin repo.
Ok, if that's the way we want it in the repo, let's document it
somewhere (in the README, I suppose, which then should also say where
the staging server is; while doing that, updating the current banner
("re-write the current") could be updated, too).
For the record, I'd still accept (and encourage) PRs from people with
no write access; if we *really* want to have CI staging for these as
well, I think we should find out a workflow where someone with write
access (but I personally would not sweat that).
|
The idea of this CI is exactly that: to build the website and to upload it somewhere so that we can preview the result before merging and publishing it for everyone. But yes, I totally agree that this PR was a lot of interactions just to change a simple label and URL.
I agree with @molinaro-m , we need to test the staging setup in order to avoid a so long process the next time.
That is actually written in the CONTRIBUTING.md file. Or would you mean to write something else? |
|
On Wed, Sep 02, 2026 at 04:35:06AM -0700, Grégory Mantelet wrote:
That is actually written in the [CONTRIBUTING.md](https://github.com/ivoa/ivoa-web/blob/main/CONTRIBUTING.md) file. Or would you mean to write something else?
<blush> ok, I should have read this before (or even noticed its
existence).
On its contents: I'm not sure whether we should have this level of
detail on basic git operations; it might be preferable to just have
"general git operations is explained at <x>, and here are the special
rules for this repo".
In PR#160, I'm trying to stress that non-fork contributions are
supposed to be the default. I'm also updating the README, and I'm
cleaning away the legacy HTML. I'd back that removal out if people
agree they should stay around, but they've already confused me twice
("is that the build?").
|
No description provided.