Clean up of GTFS-RT adapter

[leonardehrenfried]

Summary

The GTFS-RT adapter has accumulated tech debt some of which this PR reduces. In particular it:

• moves the validation of the trip id and start/service date into a single place. If the TripUpdate is contructed it is validated and calls to tripId() and serviceDate() are safe and non-null.
• because trip updates are now validated, it's no longer necessary to pass along the id and service date with all method calls
• it flattens the deep nesting in some of the methods.
• timeZone and deduplicator, which never change, are passed to the contructor of TripTimesUpdater

Unit tests

One added.

Documentation

Javadoc.

Changelog

Skip.

Bumping the serialization version id

No.

[codecov]

Codecov Report

❌ Patch coverage is 88.35979% with 22 lines in your changes missing coverage. Please review.
✅ Project coverage is 72.16%. Comparing base (593abac) to head (5992956).
⚠️ Report is 23 commits behind head on dev-2.x.

Files with missing lines	Patch %	Lines
...dater/trip/gtfs/GtfsRealTimeTripUpdateAdapter.java	79.74%	11 Missing and 5 partials ⚠️
...ripplanner/updater/trip/gtfs/model/TripUpdate.java	91.30%	4 Missing ⚠️
...ripplanner/updater/trip/gtfs/TripTimesUpdater.java	90.00%	2 Missing ⚠️

Additional details and impacted files

@@              Coverage Diff              @@
##             dev-2.x    #7238      +/-   ##
=============================================
+ Coverage      72.13%   72.16%   +0.02%     
+ Complexity     21087    21085       -2     
=============================================
  Files           2297     2298       +1     
  Lines          85233    85184      -49     
  Branches        8489     8482       -7     
=============================================
- Hits           61484    61474      -10     
+ Misses         20770    20736      -34     
+ Partials        2979     2974       -5     

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

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[optionsome]

The changes are a bit hard to read, but I think I'll test locally a bit when this is otherwise ready.

[optionsome]

This logic could be "simplified" to be failuresByRelationship.merge(scheduleRelationship, 1, (a, b) -> a + b), but I don't know if it's more readable or not :D.

[optionsome]

Maybe failuresByRelationship.merge(scheduleRelationship, 1, Integer::sum) is even simpler.

[leonardehrenfried]

I was the one that originally added this logging but I actually don't think it's useful anymore. I can remove it altogether if you allow me.

[optionsome]

I would be fine with removing it. We still have the logging based on error types, right?

[leonardehrenfried]

And thanks for this tip - I didn't know about merge.

[leonardehrenfried]

Ok, will remove. Yes, we have logging based on the error type.

[leonardehrenfried]

I also moved the validation of the stop sequence to the validate() method of TripUpdate.