From 32e7a1a91365084a46532568743b05a3d1afe002 Mon Sep 17 00:00:00 2001
From: Sameh Khamis <sameh@halfpixel.ai>
Date: Tue, 10 Nov 2026 13:16:42 -0800
Subject: [PATCH] fix(shipwrecks): skip every wrecksite.eu row, even when the
 name comes after the 80-character cut or only in the link

The exclusion tested the provenance after it was cut to 80 characters, so a
long source list ending in wrecksite.eu kept the row, and a row whose only
trace was a wrecksite.eu page link was kept too. Both raw values are now
checked before anything is cut.
---
 src/layers/shipwrecks/records.js       | 12 ++++++++++--
 src/layers/shipwrecks/records.test.mjs | 12 ++++++++++++
 2 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/src/layers/shipwrecks/records.js b/src/layers/shipwrecks/records.js
index 7a7072205..c1d17e77b 100644
--- a/src/layers/shipwrecks/records.js
+++ b/src/layers/shipwrecks/records.js
@@ -108,9 +108,15 @@ function httpUrl(value) {
   }
 }
 
+/** Whether a provenance or link value points at wrecksite.eu, in any form. */
+function fromWrecksite(value) {
+  return typeof value === 'string' && /wrecksite/i.test(value);
+}
+
 /**
  * Normalize EMODnet's shipwreck GeoJSON. Records the portal took from
- * wrecksite.eu are skipped: that site's data is not openly licensed.
+ * wrecksite.eu, named in the provenance or linked as the record's page, are
+ * skipped: that site's data is not openly licensed.
  * @param {object} payload WFS GeoJSON FeatureCollection.
  * @returns {object[]} Shipwreck records.
  */
@@ -126,8 +132,10 @@ export function normalizeEmodnetFeatures(payload) {
     const lon = Number(coords[0]);
     const lat = Number(coords[1]);
     if (!validPoint(lat, lon)) continue;
+    // Checked on the raw values: the shown origin is cut to 80 characters,
+    // and a long provenance list can name wrecksite.eu after the cut.
+    if (fromWrecksite(p.source_inf) || fromWrecksite(p.website1)) continue;
     const origin = clean(p.source_inf, 80);
-    if (origin && /wrecksite/i.test(origin)) continue;
     const key = clean(String(feature.id ?? p.source_id ?? ''), 80);
     if (!key) continue;
     const id = `emodnet:${key}`;
diff --git a/src/layers/shipwrecks/records.test.mjs b/src/layers/shipwrecks/records.test.mjs
index a61154a5a..e65863331 100644
--- a/src/layers/shipwrecks/records.test.mjs
+++ b/src/layers/shipwrecks/records.test.mjs
@@ -47,6 +47,18 @@ test('EMODnet wrecks keep ship facts and drop text that names people', () => {
       feature('w.4', { name: 'Null island' }, [0, 0]),
     ],
   });
+  const ids = normalizeEmodnetFeatures({
+    features: [
+      feature('w.3', { name: 'Copied', source_inf: 'www.wrecksite.eu' }),
+      feature('w.5', {
+        source_inf: `${'SHOM; '.repeat(20)}wrecksite.eu`,
+      }),
+      feature('w.6', { website1: 'https://www.wrecksite.eu/wreck.aspx?1' }),
+      feature('w.7', { website1: 'wrecksite.eu/wreck/1' }),
+      feature('w.8', { name: 'Kept', source_inf: 'SHOM' }),
+    ],
+  }).map((record) => record.id);
+  assert.deepEqual(ids, ['emodnet:w.8'], 'wrecksite.eu rows are all skipped');
   assert.equal(wreck.id, 'emodnet:w.1');
   assert.equal(wreck.year, 1987);
   assert.equal(wreck.vesselType, 'Pusher or tug');