[Baspa]

What

Every forecast(), actualForecastForStation(), stations(), nearestStation(), sunrise() and sunset() call used to download and decode the whole JSON feed again (~35 KB).

• The decoded feed is reused by the same instance for 60 seconds. That's long enough to collapse the calls in one request/command, and short enough that a long-running process (queue worker, daemon) still sees fresh data; Buienradar updates the measurements every 10 minutes.
• refresh() forces a new download ($buienradar->refresh()->forecast()).
• Optional second constructor argument cacheSeconds (default 60, 0 turns caching off). Non-breaking: existing new Buienradar() / new Buienradar($client) calls behave the same apart from the reuse.
• A failed download is not cached, so the next call retries.
• rainForecast() is not cached (different feed, per coordinate).
• README: "Caching" section.
• Review follow-ups: expiry uses the monotonic hrtime() clock (a wall-clock jump can't stretch or cut the window), a negative cacheSeconds throws InvalidArgumentException, and the README states that failed downloads and rainForecast() aren't cached and that refresh() affects the next download (call forecast() again to reload get()/report()).

Usage

$buienradar = new Buienradar(cacheSeconds: 300);

$buienradar->forecast()->report();     // downloads the feed
$buienradar->forecast()->shortTerm();  // reuses it
$buienradar->refresh()->forecast();    // downloads again

Tests

53 tests / 102 assertions pass, PHPStan level 8 clean. New tests/FeedCacheTest.php: several different calls on one instance make one request and return the fixture data; the cache expires after cacheSeconds; refresh() makes a second; cacheSeconds: 0 disables reuse; a failed download is retried on the next call; a negative cache time is rejected. Live check: forecast(), stations(), sunrise() and a station lookup on one instance take 0.9 s in total (one download).

Closes #42

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 87.72%. Comparing base (79bf4ca) to head (2c67645).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff              @@
##               main      #53      +/-   ##
============================================
+ Coverage     84.74%   87.72%   +2.98%     
- Complexity       50       68      +18     
============================================
  Files             4        4              
  Lines           177      220      +43     
============================================
+ Hits            150      193      +43     
  Misses           27       27              

☔ View full report in Codecov by Harness.
📢 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.