Add a real komoot GPX regression fixture (#38 partial, #6 fixture) #95

Merged

robert merged 1 commit from area/real-komoot-fixture into main 2026-09-04 16:59:41 +02:00

Conversation 0 Commits 1 Files changed 2 +11315

robert commented 2026-09-04 16:59:34 +02:00

Owner

Copy link

Robert rode and exported this GPX himself (Havelchaussee - Glienicker Brücke, Breitenbachplatz loop, ~119km) — the first real-world, not hand-built, fixture in this repo. Partial #38 progress, satisfies one of #6's own named fixture requirements (real komoot export >= 100km).

Real finding, not a bug: ascent computed on the RDP-simplified polyline (#24) is measurably lower than on the raw 2795 points — 515.6m raw vs 463.1m post-simplification, a ~10% reduction. RDP drops points by horizontal deviation only, so a dropped point's elevation reading (and any small GPS noise it captured) goes with it. Not fixed here — flagging as possibly worth a tracked issue if accurate elevation gain matters to a rider. Test tolerance reflects the real measured number, not padded to pass.

No turn-cue text in this file (11 named-landmark waypoints, not direction-plus-distance strings), so it exercises the importer and route metrics only, not GpxCueEnricher — real remaining #38/#6 scope.

Verified: ./gradlew :companion:core:test green with a real JDK 21.

Robert rode and exported this GPX himself (Havelchaussee - Glienicker Brücke, Breitenbachplatz loop, ~119km) — the first real-world, not hand-built, fixture in this repo. Partial #38 progress, satisfies one of #6's own named fixture requirements (real komoot export >= 100km). **Real finding, not a bug**: ascent computed on the RDP-simplified polyline (#24) is measurably lower than on the raw 2795 points — 515.6m raw vs 463.1m post-simplification, a ~10% reduction. RDP drops points by horizontal deviation only, so a dropped point's elevation reading (and any small GPS noise it captured) goes with it. Not fixed here — flagging as possibly worth a tracked issue if accurate elevation gain matters to a rider. Test tolerance reflects the real measured number, not padded to pass. No turn-cue text in this file (11 named-landmark waypoints, not direction-plus-distance strings), so it exercises the importer and route metrics only, not GpxCueEnricher — real remaining #38/#6 scope. Verified: `./gradlew :companion:core:test` green with a real JDK 21.

robert added 1 commit 2026-09-04 16:59:35 +02:00

Add a real ~119km komoot GPX as a regression fixture (#38 partial, #6 fixture 2) ... 4d14a9d551

Robert rode and exported this himself (Havelchaussee - Glienicker Brücke,
Breitenbachplatz loop) — the first real-world (not hand-built) fixture in
this repo. Partial progress on #38 ('real komoot GPX exports... a rural
stretch') and satisfies one of #6's own named fixture requirements ('one
komoot export >= 100 km' — this is ~119 km).

Real finding, not a bug: ascent computed on the RDP-simplified polyline
(#24) is measurably lower than ascent computed on the raw 2795 points —
515.6m raw vs 463.1m post-simplification on this exact file, a ~10%
reduction. RDP drops points by horizontal deviation only; a dropped
point's elevation reading goes with it, so small GPS elevation noise
between kept points is smoothed away. Not something this commit fixes —
worth a tracked issue if accurate elevation gain matters to a rider,
which it likely does. Test tolerance reflects the real number, not padded
to make it pass.

No turn-cue text in this file's waypoints (11 named landmarks, not
direction-plus-distance strings), so it says nothing about
GpxCueEnricher — that gap is real remaining #38/#6 scope.

Verified: ./gradlew :companion:core:test green, real JDK 21.

Claude-Session: https://claude.ai/code/session_01DAoXbRmJUf2uxNYBfdAXPt

robert merged commit 8a82128b2a into main 2026-09-04 16:59:41 +02:00

robert referenced this pull request from a commit 2026-09-04 16:59:42 +02:00

Merge pull request 'Add a real komoot GPX regression fixture (#38 partial, #6 fixture)' (#95) from area/real-komoot-fixture into main

robert referenced this pull request 2026-09-04 16:59:54 +02:00

Enrichment regression fixtures from real komoot GPX exports #38

robert referenced this pull request 2026-09-04 16:59:54 +02:00

Spike C: BRouter AIDL enrichment of a real komoot GPX #6

robert referenced this pull request 2026-09-04 20:30:12 +02:00

Audit issue #22 speed-pipeline test coverage, fill the real GPX-replay distance gap #103

Sign in to join this conversation.

Reviewers

No reviewers

Labels

Clear labels   

area:companion

Android companion app

  

area:docs

Documentation

  

area:shared

Message contract shared by both sides

  

area:tooling

Build tooling, SDK, dev scripts

  

area:watchapp

Pebble C watchapp

  

blocker

Blocks a large part of the project until resolved

  

kind:chore

Setup, scaffolding, maintenance

  

kind:feature

New functionality

  

kind:spike

Time-boxed investigation to de-risk a decision

  

kind:test

Tests and fixtures

No labels

area:companion

area:docs

area:shared

area:tooling

area:watchapp

blocker

kind:chore

kind:feature

kind:spike

kind:test

Milestone

Clear milestone

No items

No milestone

Projects

Clear projects

No items

No project

Assignees

Clear assignees

No assignees

1 participant

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference

robert/PedalPebble!95

No description provided.