JPLSpec is even more down#3547
Conversation
|
This needs a changelog. And maybe separating out into two PRs, one for the backup url (I expect that to get merged quickly); and the other for the new kwarg (if it needs more work)? |
|
The backup doesn't do anything without the new kwarg, so these should go together. I will fix the failures tonight - this was an emergency patch with my student, since she needs access now. |
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #3547 +/- ##
==========================================
- Coverage 73.22% 73.14% -0.09%
==========================================
Files 226 226
Lines 21000 21060 +60
==========================================
+ Hits 15378 15405 +27
- Misses 5622 5655 +33 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
bsipocz
left a comment
bsipocz
left a comment
There was a problem hiding this comment.
jplspec seems to be back right now, do you still want to go ahead with the wayback machine addition?
| by the query service. This workaround is needed while JPLSpec's query | ||
| tool is broken. | ||
|
|
||
| use_getmolecule is an option to force the query to use get_molecule. |
There was a problem hiding this comment.
outside of scope here, but these methods should have proper docstrings
|
are you sure it's back? If you run a query, you probably get |
All tests pass locally on |
|
Huh, you're right. I'll dig a little more, maybe I can default the fallback to 'off'. |
…w, may never come back. wayback machine URL added to config
|
@keflavich - can you look at the test failures? |
|
Yep, I'll dig in - I clearly (re?)introduced a bug. |
|
I just got an e-mail from the JPL folks saying they had fixed the server! I'll try to verify that later. |
- Fix E131 hanging-indent in jplspec/__init__.py - Drop unused EmptyResponseError import from test_jplspec_remote - Add CHANGES.rst entry for astropy#3547 Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
bsipocz
left a comment
bsipocz
left a comment
There was a problem hiding this comment.
overall looks good. There is one remote test failure though with the documentation, if you could also fix that up before merging that would be awesome.
there's no duplicated logic
bsipocz
left a comment
bsipocz
left a comment
There was a problem hiding this comment.
OK, let's merge this. I have one minor comment and pushed a doctest fix, but it should all be good to go now.
2 participants
This is a harder workaround for jplspec being down: it switches to using the wayback machine instead of jpl's server, which has gone totally unresponsive.
This raises bigger questions about data preservation/curation. Can't answer those here, but here's a 🩹 .