Enforce flex rule: at least two stop times per trip

[leonardehrenfried]

Summary

Both the GTFS and NeTEx specs require flex trips, just like regular ones, to have at least 2 stop times/passing times.

Even though our flex routing engine doesn't produce a result when you have a trip with a single stop time, we nevertheless import the data anyway and then do nothing with it. It may end up in API response which can be quite confusing.

For this reason, this PR rejects flex trips with fewer than 2 stop times and adds an issue to the store.

Issue

No.

Unit tests

Added.

[codecov]

Codecov Report

❌ Patch coverage is 80.00000% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 72.08%. Comparing base (c5fd182) to head (4fc2ef3).
⚠️ Report is 200 commits behind head on dev-2.x.

Files with missing lines	Patch %	Lines
.../org/opentripplanner/ext/flex/FlexTripsMapper.java	80.00%	0 Missing and 1 partial ⚠️
.../main/java/org/opentripplanner/model/StopTime.java	0.00%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@              Coverage Diff              @@
##             dev-2.x    #6828      +/-   ##
=============================================
+ Coverage      71.99%   72.08%   +0.08%     
- Complexity     19437    19521      +84     
=============================================
  Files           2100     2105       +5     
  Lines          78774    78972     +198     
  Branches        7963     8009      +46     
=============================================
+ Hits           56717    56927     +210     
+ Misses         19249    19221      -28     
- Partials        2808     2824      +16     

☔ 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]

What is the need for this check? Can these stoptimes be any stop times or are they filtered somewhere to be just "flex" stop times? Should this check be moved to be the first if so you don't have to account for this edge case in the other checks?

[leonardehrenfried]

These are all stop times, not just the flex ones therefore you need something like that.

If I took your comment literally, then I can move the longish if statement to the top and that would look like that: leonardehrenfried@ddf302a

However, I don't find this any better.

Thinking a bit bigger, we could have a first step where I extract only the flex stop times and in a second one we sort it into the two types of flex trips and add the issues.

What do you think of that?

[vpaturet]

Remark: Some flex trips consist in a single area where passengers can be picked up and then dropped off. In practice these trips are modeled with 2 flexible stops that refer to the same area. In this case, the invariant "at least 2 stops" still holds true.