Fixup integration tests - #10
Merged
Merged
Conversation
subset_gtfs.py slices a full GTFS feed down to one route with referential closure (routes → trips → stop_times → stops → shapes → calendar). find_detour_candidates.py ranks AVL traces by how detour- like they look by measuring each vehicle's maximum deviation from the route shape. Together they support the capture → promote pipeline documented in integration-tests.md.
WMATA's June 2025 Better Bus Network redesign renamed every route to a <zone-letter><number> scheme, so the 2016-era S2/3T/5A fixtures could not simply be re-captured. Replaced with fresh traces from equivalent current routes: - S2 (16th St trunk) → D40 "7 St-Georgia Av", vehicle 5506 - 3T (VA crosstown) → A40 "Columbia Pike-National Landing", vehicle 3151 - plus D20 "H Street", vehicle 7198, kept as a spare for future tests 5A fixtures are left in place since their two tests still pass. RecoverFromDetourTest is un-ignored and passing against A40_3151 (vehicle exits layover, final schedule adherence -7.5 min within the 10-min threshold). PredictionAccuracyIntegrationTest has its path constants updated but remains @ignored pending regeneration of pred/D40_5506.csv as its regression baseline. integration-tests.md documents the end-to-end refresh workflow for future captures. Fixtures were produced via tools/wmata_capture plus the new subset_gtfs.py and find_detour_candidates.py helpers. Refs #7 (detour test, now re-enabled) Refs #8 (prediction accuracy, baseline pending)
Tried to regenerate pred/D40_5506.csv by dumping the current predictor's output against the refreshed fixtures, intending to un-ignore testPredictions. Discovered that PlaybackModule.runTrace is non-deterministic across JVMs: replaying the same AVL trace produces prediction counts that vary by >20% (observed 2155, 2696, 2118 across three runs) and ArrivalDeparture counts that vary by >2x. A frozen CSV baseline therefore cannot serve as a regression signal — the test's newTotalError/oldTotalError comparison fails on run-to-run variance, not on predictor regression. Update the @ignore message to name the real blocker and extend integration-tests.md with two suggested paths for re-enabling (deterministic replay, or tolerance-based assertions). RecoverFromDetourTest remains live and passing; no regression. Refs #8
CLAUDE.md: point at integration-tests.md for fixture-refresh work and record that PlaybackModule.runTrace is non-deterministic — needed so future test design doesn't assume frozen CSV baselines will work. tools/wmata_capture/README.md: the promotion recipe still pointed at 2016 S2 paths and didn't know about the GTFS-subsetting step. Rewrite that section to use subset_gtfs.py and find_detour_candidates.py, drop the now-obsolete block_id caveat (100% coverage post-BBN redesign), note that the pred baseline path is currently blocked on non-determinism, and add the route-ID rename caveat so future captures know the 2016 short names are gone.
|
Caution Review failedPull request was closed or merged during review 📝 WalkthroughWalkthroughRefreshes integration-test fixtures (remove 3T/S2, add A40/D20/D40), adds GTFS helper CLIs for subsetting and detour candidate ranking, updates integration tests to use refreshed fixtures (un-ignore RecoverFromDetourTest, keep PredictionAccuracyIntegrationTest ignored with documented non-determinism), and documents an end-to-end fixture-refresh workflow and caveats. Changes
Sequence Diagram(s)sequenceDiagram
participant Dev as Developer
participant Capture as wmata_capture tools
participant Subset as subset_gtfs.py
participant Finder as find_detour_candidates.py
participant Repo as Repo test resources
participant Tests as Integration tests
Dev->>Capture: run capture.py to pull GTFS/AVL
Capture->>Repo: write raw captures
Dev->>Subset: run subset_gtfs.py (--route)
Subset->>Repo: write per-route GTFS subset into test resources
Dev->>Finder: run find_detour_candidates.py (--avl-dir)
Finder->>Dev: return ranked detour candidate traces
Dev->>Repo: promote selected AVL CSVs into fixtures
Dev->>Tests: run integration tests against refreshed fixtures
Tests->>Repo: read GTFS/AVL fixtures
Tests-->>Dev: pass/fail (note: PredictionAccuracyIntegrationTest ignored due to non-determinism)
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~22 minutes Possibly related issues
Possibly related PRs
🚥 Pre-merge checks | ✅ 3 | ❌ 2❌ Failed checks (1 warning, 1 inconclusive)
✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 7
🧹 Nitpick comments (2)
transitclockIntegration/src/test/java/org/transitclock/integration_tests/RecoverFromDetourTest.java (1)
27-29: Assert vehicle presence before dereferencing it.If playback/fixture drift causes lookup to miss, this currently fails as a generic NPE instead of a clear test failure reason.
Suggested fix
import static org.junit.Assert.assertFalse; +import static org.junit.Assert.assertNotNull; import static org.junit.Assert.assertTrue; @@ PlaybackModule.runTrace(GTFS, AVL); IpcVehicleComplete v = VehicleDataCache.getInstance().getVehicle(VEHICLE); + assertNotNull("Expected vehicle to be present after playback", v); assertFalse(v.isLayover()); int adh = Math.abs(v.getRealTimeSchedAdh().getTemporalDifference());🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@transitclockIntegration/src/test/java/org/transitclock/integration_tests/RecoverFromDetourTest.java` around lines 27 - 29, The test dereferences the result of VehicleDataCache.getInstance().getVehicle(VEHICLE) (stored in IpcVehicleComplete v) without checking for null, causing a NPE on v.isLayover() when the fixture misses; add an explicit assertion that v is not null (e.g., assertNotNull with a clear message referencing VEHICLE) immediately after retrieving v, then proceed with assertFalse(v.isLayover()) and the adh calculation using v.getRealTimeSchedAdh().getTemporalDifference().transitclockIntegration/src/test/resources/gtfs/D40/trips.txt (1)
1-1610: Add automated GTFS fixture integrity checks for this large trips dataset.Given the size of Line 1–Line 1610, please add a lightweight validation step to catch orphaned
service_id,route_id, and optionalshape_idreferences before tests run.#!/bin/bash set -euo pipefail python - <<'PY' import csv from pathlib import Path root = Path("transitclockIntegration/src/test/resources/gtfs") def read_csv(path): with path.open(newline='', encoding='utf-8') as f: return list(csv.DictReader(f)) def validate_dataset(ds): p = root / ds trips_p = p / "trips.txt" cal_p = p / "calendar.txt" routes_p = p / "routes.txt" shapes_p = p / "shapes.txt" trips = read_csv(trips_p) calendar = read_csv(cal_p) routes = read_csv(routes_p) cal_ids = {r["service_id"] for r in calendar} route_ids = {r["route_id"] for r in routes} shape_ids = set() if shapes_p.exists(): shape_ids = {r["shape_id"] for r in read_csv(shapes_p)} missing_service = sorted({t["service_id"] for t in trips if t["service_id"] not in cal_ids}) missing_route = sorted({t["route_id"] for t in trips if t["route_id"] not in route_ids}) missing_shape = sorted({t["shape_id"] for t in trips if shape_ids and t["shape_id"] not in shape_ids}) print(f"[{ds}] trips={len(trips)}") print(f"[{ds}] missing service_ids: {missing_service[:10]}{' ...' if len(missing_service)>10 else ''}") print(f"[{ds}] missing route_ids: {missing_route[:10]}{' ...' if len(missing_route)>10 else ''}") if shapes_p.exists(): print(f"[{ds}] missing shape_ids: {missing_shape[:10]}{' ...' if len(missing_shape)>10 else ''}") # non-zero exit if core referential integrity fails if missing_service or missing_route or (shapes_p.exists() and missing_shape): raise SystemExit(1) for ds in ("D40",): validate_dataset(ds) PYAlso run the integration suite with the intended profile to validate fixture behavior end-to-end:
mvn install -P include-integration-testsBased on learnings: Run integration tests with
mvn install -P include-integration-tests; they live in thetransitclockIntegrationmodule and are excluded by default.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@transitclockIntegration/src/test/resources/gtfs/D40/trips.txt` around lines 1 - 1610, Add a lightweight GTFS referential-integrity check that runs before integration tests: implement the provided validate_dataset flow (use functions read_csv and validate_dataset, root pointing at transitclockIntegration/src/test/resources/gtfs) to load trips.txt, calendar.txt, routes.txt and optional shapes.txt and fail fast if any trip references a missing service_id, route_id, or shape_id; wire this script into the test setup (pre-test step for the transitclockIntegration suite so it runs before mvn install -P include-integration-tests) and ensure the script exits non-zero when missing_service, missing_route, or missing_shape are non-empty.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@docs/integration-tests.md`:
- Around line 78-87: The document contains conflicting guidance about the status
of PredictionAccuracyIntegrationTest (and similarly referenced at the block
around lines 226-231): one section lists it as `@Ignore` due to nondeterministic
baseline, while another instructs removing `@Ignore`; reconcile by updating the
runbook so a single canonical status and action are stated for
PredictionAccuracyIntegrationTest (and RecoverFromDetourTest if applicable):
explicitly mark it as “blocked by nondeterminism” or “ready to un-ignore”,
remove the contradictory instruction, and add a short rationale and next step
(e.g., reproduce/fix nondeterminism or a link to issue `#8`) so readers have
unambiguous state/action guidance referencing the test name
PredictionAccuracyIntegrationTest and the related issue `#8`.
- Around line 254-256: The fenced code block containing the CSV header line
"id,affectedByWaitStop,avlTime,configRev,creationTime,gtfsStopSeq,isArrival,predictionTime,routeId,schedBasedPred,stopId,tripId,vehicleId"
is missing a language identifier and triggers markdownlint MD040; update the
opening fence (the triple backticks immediately before that header) to include a
language such as csv or text (e.g., change ``` to ```csv) so the block is
properly identified.
- Line 137: The runbook contains a machine-specific absolute path ("cd
/Users/aaron/repos/onebusaway/transitime/tools/wmata_capture") which is not
portable; replace this hardcoded path with a relative repo path or a
parameterized/env-var placeholder (e.g., use "./tools/wmata_capture" or
"${REPO_ROOT}/tools/wmata_capture") and update the surrounding instructions to
document how to set REPO_ROOT or run from the repo root so contributors and CI
can run the command without editing the file.
- Around line 45-47: The docs currently show the wrong integration-test
invocation ("mvn -P include-integration-tests test"); update the text to use the
project's documented command by replacing that invocation with "mvn install -P
include-integration-tests" and add a brief note that these tests reside in the
transitclockIntegration module and are excluded by default so readers know why
the profile is required.
In `@tools/wmata_capture/find_detour_candidates.py`:
- Around line 129-140: Filter the accumulated results to actual detour
candidates before sorting/printing by keeping only entries where
longest_bracketed_off_run > 0 and ends_on_route is True (this is the candidate
definition used by score_vehicle results); do this right after the loop that
builds results (the loop iterating over avl_files that calls score_vehicle) and
before the results.sort and the print loop, then sort/print only the filtered
list (use args.top as before).
- Line 96: The median calculation for "median_dist_m" in
find_detour_candidates.py currently uses sorted(distances)[len(distances)//2],
which biases toward the upper middle for even-length lists; update the
assignment to compute a true median by either calling
statistics.median(distances) or by sorting distances and, if len(distances) is
even, averaging the two middle values (sorted_dist[n//2 - 1] and
sorted_dist[n//2]), otherwise taking the middle element—ensure you reference the
same "distances" sequence and replace the existing "median_dist_m" expression
accordingly.
In `@tools/wmata_capture/subset_gtfs.py`:
- Around line 73-81: After computing the GTFS subsets (kept_trips,
kept_trip_ids, kept_service_ids, kept_shape_ids and kept_stop_times), add a
fast-fail check that if kept_trips is empty or kept_stop_times is empty you
abort with a clear error (raise an exception or call sys.exit(1)) including the
route_id and which collection is empty; update the code around the kept_trips /
kept_stop_times computation (the variables kept_trips, kept_trip_ids,
kept_stop_times) to perform this validation before proceeding to write out files
so no empty slice is written.
---
Nitpick comments:
In
`@transitclockIntegration/src/test/java/org/transitclock/integration_tests/RecoverFromDetourTest.java`:
- Around line 27-29: The test dereferences the result of
VehicleDataCache.getInstance().getVehicle(VEHICLE) (stored in IpcVehicleComplete
v) without checking for null, causing a NPE on v.isLayover() when the fixture
misses; add an explicit assertion that v is not null (e.g., assertNotNull with a
clear message referencing VEHICLE) immediately after retrieving v, then proceed
with assertFalse(v.isLayover()) and the adh calculation using
v.getRealTimeSchedAdh().getTemporalDifference().
In `@transitclockIntegration/src/test/resources/gtfs/D40/trips.txt`:
- Around line 1-1610: Add a lightweight GTFS referential-integrity check that
runs before integration tests: implement the provided validate_dataset flow (use
functions read_csv and validate_dataset, root pointing at
transitclockIntegration/src/test/resources/gtfs) to load trips.txt,
calendar.txt, routes.txt and optional shapes.txt and fail fast if any trip
references a missing service_id, route_id, or shape_id; wire this script into
the test setup (pre-test step for the transitclockIntegration suite so it runs
before mvn install -P include-integration-tests) and ensure the script exits
non-zero when missing_service, missing_route, or missing_shape are non-empty.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: f0b0aa65-cceb-4aac-94e9-3a59d709002f
⛔ Files ignored due to path filters (6)
transitclockIntegration/src/test/resources/avl/3T_3757.csvis excluded by!**/*.csvtransitclockIntegration/src/test/resources/avl/A40_3151.csvis excluded by!**/*.csvtransitclockIntegration/src/test/resources/avl/D20_7198.csvis excluded by!**/*.csvtransitclockIntegration/src/test/resources/avl/D40_5506.csvis excluded by!**/*.csvtransitclockIntegration/src/test/resources/avl/S2_2113.csvis excluded by!**/*.csvtransitclockIntegration/src/test/resources/pred/S2_2113.csvis excluded by!**/*.csv
📒 Files selected for processing (52)
CLAUDE.mdREADME.mddocs/integration-tests.mdtools/wmata_capture/README.mdtools/wmata_capture/find_detour_candidates.pytools/wmata_capture/subset_gtfs.pytransitclockIntegration/src/test/java/org/transitclock/integration_tests/RecoverFromDetourTest.javatransitclockIntegration/src/test/java/org/transitclock/integration_tests/prediction/PredictionAccuracyIntegrationTest.javatransitclockIntegration/src/test/resources/gtfs/3T/agency.txttransitclockIntegration/src/test/resources/gtfs/3T/calendar.txttransitclockIntegration/src/test/resources/gtfs/3T/calendar_dates.txttransitclockIntegration/src/test/resources/gtfs/3T/routes.txttransitclockIntegration/src/test/resources/gtfs/3T/shapes.txttransitclockIntegration/src/test/resources/gtfs/3T/stop_times.txttransitclockIntegration/src/test/resources/gtfs/3T/stops.txttransitclockIntegration/src/test/resources/gtfs/3T/trips.txttransitclockIntegration/src/test/resources/gtfs/A40/agency.txttransitclockIntegration/src/test/resources/gtfs/A40/calendar.txttransitclockIntegration/src/test/resources/gtfs/A40/calendar_dates.txttransitclockIntegration/src/test/resources/gtfs/A40/feed_info.txttransitclockIntegration/src/test/resources/gtfs/A40/routes.txttransitclockIntegration/src/test/resources/gtfs/A40/shapes.txttransitclockIntegration/src/test/resources/gtfs/A40/stop_times.txttransitclockIntegration/src/test/resources/gtfs/A40/stops.txttransitclockIntegration/src/test/resources/gtfs/A40/trips.txttransitclockIntegration/src/test/resources/gtfs/D20/agency.txttransitclockIntegration/src/test/resources/gtfs/D20/calendar.txttransitclockIntegration/src/test/resources/gtfs/D20/calendar_dates.txttransitclockIntegration/src/test/resources/gtfs/D20/feed_info.txttransitclockIntegration/src/test/resources/gtfs/D20/routes.txttransitclockIntegration/src/test/resources/gtfs/D20/shapes.txttransitclockIntegration/src/test/resources/gtfs/D20/stop_times.txttransitclockIntegration/src/test/resources/gtfs/D20/stops.txttransitclockIntegration/src/test/resources/gtfs/D20/trips.txttransitclockIntegration/src/test/resources/gtfs/D40/agency.txttransitclockIntegration/src/test/resources/gtfs/D40/calendar.txttransitclockIntegration/src/test/resources/gtfs/D40/calendar_dates.txttransitclockIntegration/src/test/resources/gtfs/D40/feed_info.txttransitclockIntegration/src/test/resources/gtfs/D40/routes.txttransitclockIntegration/src/test/resources/gtfs/D40/shapes.txttransitclockIntegration/src/test/resources/gtfs/D40/stop_times.txttransitclockIntegration/src/test/resources/gtfs/D40/stops.txttransitclockIntegration/src/test/resources/gtfs/D40/trips.txttransitclockIntegration/src/test/resources/gtfs/S2/agency.txttransitclockIntegration/src/test/resources/gtfs/S2/calendar.txttransitclockIntegration/src/test/resources/gtfs/S2/calendar_dates.txttransitclockIntegration/src/test/resources/gtfs/S2/routes.txttransitclockIntegration/src/test/resources/gtfs/S2/shapes.txttransitclockIntegration/src/test/resources/gtfs/S2/stop_times.txttransitclockIntegration/src/test/resources/gtfs/S2/stops.txttransitclockIntegration/src/test/resources/gtfs/S2/t.txttransitclockIntegration/src/test/resources/gtfs/S2/trips.txt
💤 Files with no reviewable changes (11)
- transitclockIntegration/src/test/resources/gtfs/S2/t.txt
- transitclockIntegration/src/test/resources/gtfs/S2/agency.txt
- transitclockIntegration/src/test/resources/gtfs/3T/agency.txt
- transitclockIntegration/src/test/resources/gtfs/S2/routes.txt
- transitclockIntegration/src/test/resources/gtfs/3T/calendar.txt
- transitclockIntegration/src/test/resources/gtfs/S2/calendar.txt
- transitclockIntegration/src/test/resources/gtfs/3T/stops.txt
- transitclockIntegration/src/test/resources/gtfs/3T/routes.txt
- transitclockIntegration/src/test/resources/gtfs/S2/stops.txt
- transitclockIntegration/src/test/resources/gtfs/3T/trips.txt
- transitclockIntegration/src/test/resources/gtfs/S2/trips.txt
Comment on lines
+45
to
+47
| - **Step 5 validate — DONE (for non-ignored tests).** `mvn -P | ||
| include-integration-tests test` on the module passes: 3 tests green | ||
| (detour + two 5A), 1 skipped (testPredictions). Matches the pre- |
There was a problem hiding this comment.
Use the documented integration-test invocation.
Line 45-47 should use the project’s integration-test command to avoid confusion in future reruns.
Based on learnings: Run integration tests with mvn install -P include-integration-tests; they live in the transitclockIntegration module and are excluded by default.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@docs/integration-tests.md` around lines 45 - 47, The docs currently show the
wrong integration-test invocation ("mvn -P include-integration-tests test");
update the text to use the project's documented command by replacing that
invocation with "mvn install -P include-integration-tests" and add a brief note
that these tests reside in the transitclockIntegration module and are excluded
by default so readers know why the profile is required.
| "vehicle": avl_path.stem, | ||
| "rows": len(rows), | ||
| "max_dist_m": max(distances), | ||
| "median_dist_m": sorted(distances)[len(distances) // 2], |
There was a problem hiding this comment.
median_dist_m is not a true median for even-length traces.
Selecting sorted(distances)[len//2] biases to the upper middle value when row count is even.
Suggested fix
import argparse
import csv
import math
+import statistics
import sys
from pathlib import Path
@@
- "median_dist_m": sorted(distances)[len(distances) // 2],
+ "median_dist_m": statistics.median(distances),🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@tools/wmata_capture/find_detour_candidates.py` at line 96, The median
calculation for "median_dist_m" in find_detour_candidates.py currently uses
sorted(distances)[len(distances)//2], which biases toward the upper middle for
even-length lists; update the assignment to compute a true median by either
calling statistics.median(distances) or by sorting distances and, if
len(distances) is even, averaging the two middle values (sorted_dist[n//2 - 1]
and sorted_dist[n//2]), otherwise taking the middle element—ensure you reference
the same "distances" sequence and replace the existing "median_dist_m"
expression accordingly.
Comment on lines
+73
to
+81
| trips_fields, trips_rows = read_csv(input_dir / "trips.txt") | ||
| kept_trips = [t for t in trips_rows if t["route_id"] == route_id] | ||
| kept_trip_ids = {t["trip_id"] for t in kept_trips} | ||
| kept_service_ids = {t["service_id"] for t in kept_trips} | ||
| kept_shape_ids = {t["shape_id"] for t in kept_trips if t.get("shape_id")} | ||
|
|
||
| st_fields, st_rows = read_csv(input_dir / "stop_times.txt") | ||
| kept_stop_times = [r for r in st_rows if r["trip_id"] in kept_trip_ids] | ||
| kept_stop_ids = {r["stop_id"] for r in kept_stop_times} |
There was a problem hiding this comment.
Fail fast when subsetting yields no usable trips/stop_times.
Right now an empty route slice still writes files, which can hide bad input/capture windows until later tests fail.
Suggested safeguard
trips_fields, trips_rows = read_csv(input_dir / "trips.txt")
kept_trips = [t for t in trips_rows if t["route_id"] == route_id]
+ if not kept_trips:
+ raise SystemExit(
+ f"No trips found for route_short_name={route_short_name!r} (route_id={route_id!r})"
+ )
kept_trip_ids = {t["trip_id"] for t in kept_trips}
kept_service_ids = {t["service_id"] for t in kept_trips}
kept_shape_ids = {t["shape_id"] for t in kept_trips if t.get("shape_id")}
st_fields, st_rows = read_csv(input_dir / "stop_times.txt")
kept_stop_times = [r for r in st_rows if r["trip_id"] in kept_trip_ids]
+ if not kept_stop_times:
+ raise SystemExit(
+ f"No stop_times found for route_short_name={route_short_name!r} (route_id={route_id!r})"
+ )🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@tools/wmata_capture/subset_gtfs.py` around lines 73 - 81, After computing the
GTFS subsets (kept_trips, kept_trip_ids, kept_service_ids, kept_shape_ids and
kept_stop_times), add a fast-fail check that if kept_trips is empty or
kept_stop_times is empty you abort with a clear error (raise an exception or
call sys.exit(1)) including the route_id and which collection is empty; update
the code around the kept_trips / kept_stop_times computation (the variables
kept_trips, kept_trip_ids, kept_stop_times) to perform this validation before
proceeding to write out files so no empty slice is written.
- docs/integration-tests.md: update "Current state" table and fixture layout to reflect the completed refresh (A40/D40/D20, D40_5506 avl, RecoverFromDetourTest passing). Annotate Step 4 edits with what's already done, and gate the pred-baseline regeneration section behind the non-determinism blocker so readers don't un-ignore the test prematurely. Drop the hardcoded /Users/ path in Step 1. - find_detour_candidates.py: filter results to actual detour candidates (bracketed off-route run that returns on-route) before sorting and printing, so the top-N list can't surface non-candidates.
0x3mr
pushed a commit
to 0x3mr/trip-updates
that referenced
this pull request
Apr 30, 2026
jQuery UI was vendored in two locations (~250KB of JS+CSS+images) but
only two of its widgets were actually used:
- .datepicker() in two report-param JSPs — replaced with air-datepicker v2
(already vendored at javascript/air-datepicker/, never wired up before).
Same jQuery-plugin call shape ($("#x").datepicker({...})). Cross-field
min/maxDate constraint is now wired via onSelect callbacks.
- .tooltip() — global init in transitime.js plus per-page wirings on
Select2 containers in 8 form-param JSPs. Dropped entirely; native
browser tooltips on the underlying <input>/<select> remain. The two
"tooltip-disable" workarounds in maps/map.jsp and synoptic/index.jsp
existed solely to suppress side effects of the global init and
disappear with it.
Also:
- $('#mapTitle').hide('fade', 1000) — a jQuery UI effect — replaced
with core jQuery $.fadeOut(1000).
- transitime.js now has a cache-buster query string, mirroring the
existing pattern for general.css.
- <br>-based HTML in date/time tooltip title attrs converted to
&OneBusAway#10; line-break entities so native browser tooltips render
multi-line content correctly.
Deleted:
- transitclockWebapp/src/main/webapp/jquery-ui/ (full vendored library
with 12 files including 10 PNG images, full CSS+JS, and a demo).
- transitclockWebapp/src/main/webapp/javascript/jquery-ui.{js,min.js}
(duplicate vendored copy in the javascript/ dir).
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary by CodeRabbit
Documentation
New Features
Tests