Empty Tiled Writer

[genematx]

This adds an EmptyTiledWriter callback that only creates a container for the Bluesky run, registers the Start and Stop documents, but does not write any data.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 69.94%. Comparing base (c17e852) to head (fcb2d57).

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #23      +/-   ##
==========================================
+ Coverage   68.85%   69.94%   +1.08%     
==========================================
  Files          14       14              
  Lines        1840     1853      +13     
==========================================
+ Hits         1267     1296      +29     
+ Misses        573      557      -16     

☔ 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.

[danielballan]

Makes sense to me. I'd be interested in @tacaswell's two cents on whether this is the best way to do this.

Another option: I thought of doing it on the RunEngine side, as in:

# Subscribe (standard) TiledWriter but only feed it 'start' and 'stop' documents.
# To TiledWriter (and thus, to Tiled) it will appear as if each run is empty.
RE.subscribe(tw, 'start')
RE.subscribe(tw, 'stop')

[genematx]

Ah yeah, that's even simpler! I forgot we can filter document types on subscription to RE. Might be tricky to remove unnecessary metadata from Start (if we want to be pedantic about not saving post-data-security data altogether in one place), but then we could also apply the access tags, which should work as usual.

[JunAishima]

The new class and tests look good. I would also welcome further input on whether this is the best way to proceed.

[danielballan]

Decision: Hold on this to see if RE.subscribe meets our requirements without introducing more code to maintain. TBD.