ECC-2220 Fix setting string keys in BUFR messages

[oiffrig]

Description

Setting a single-valued string key in a BUFR message results in the value being truncated, e.g.:

from eccodes import BUFRMessage
bufr = BUFRMessage("BUFR4")
bufr["unexpandedDescriptors"] = [1025]
bufr["stormIdentifier"] = "70A"
assert bufr["stormIdentifier"] == "70A", "Fail before pack()"
bufr.pack()
print("Storm identifier:", bufr["stormIdentifier"])
assert bufr["stormIdentifier"] == "70A", "Fail after pack()"

Fails on the second assert. This fixes the issue to my understanding, but please check it is appropriate as I am not totally familiar with the high-level BUFR interface.

Contributor Declaration

By opening this pull request, I affirm the following:

• All authors agree to the Contributor License Agreement.
• The code follows the project's coding standards.
• I have performed self-review and added comments where needed.
• I have added or updated tests to verify that my changes are effective and functional.
• I have run all existing tests and confirmed they pass.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 0% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 30.01%. Comparing base (46a0885) to head (8e0f2fe).
⚠️ Report is 2 commits behind head on develop.

Files with missing lines	Patch %	Lines
eccodes/highlevel/_bufr/coder.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff            @@
##           develop     #154   +/-   ##
========================================
  Coverage    30.01%   30.01%           
========================================
  Files           34       34           
  Lines         6180     6180           
  Branches       743      743           
========================================
  Hits          1855     1855           
  Misses        4296     4296           
  Partials        29       29           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[tomas-kral]

Hi @oiffrig, you've just uncovered a bug in my botched attempt at a workaround for an old bug in the eccodes library that had already been fixed quite some time ago (see ECC-1623). My workaround is botched in a sense that it only works correctly for compressed messages (compressedData=1); whereas, as you found out, it doesn't do the right thing when the message is uncompressed (as is the case in your example). Unfortunately, with your change in place, this will now only work correctly for uncompressed messages, and will fail with the compressed ones. :)

In any case, given that the root cause had been resolved since eccodes 2.31.0, which is quite old version by now, I think it's safe to drop the problematic workaround entirely. The generic code that follows after my workaround should now work seamlessly for both compressed and uncompressed messages.

If you are OK with that, I suggest we close this PR, and I will submit a new one with the problematic code removed, and maybe will add some tests as well so that we can catch any issues with this in the future.

How does it sound?

[oiffrig]

Sounds good to me. Thanks for looking into it!