fix(i18n): refuse a translation another entry already holds

What is wrong with the answers the collapse check reports is not their
shape. A label is caught by its length, but `sq` shares a translation with
four sources at 0.84 of their length, `ga` at 0.73 and `he` at 0.62: full
renderings, spent twice. No length rule separates those from a language
that is simply terse, because the fault is that another entry already says
it, which a predicate reading one entry cannot see.

The writer now carries the translations a catalog already spends on a
source long enough to deserve its own, and turns down a repeat. It lives
where the whole catalog is in hand; `apply` alone sees one chunk. Both
arguments default to None, so a caller without the catalog behaves as
before.

The floor is now inclusive, matching the check it mirrors. `az` collapsed
on sources of exactly forty characters, which the check counts and the
first version of this guard skipped.

`criteria` is threaded through so the run names what it refused. A
duplicate is turned down here rather than in the client, and the summary
reported the discard with no reason for it.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

tests/unit/python/utils/i18n/test_translate.py

@@ -20,7 +20,14 @@ from utils.i18n.catalog import (
 )
 from utils.i18n.libretranslate import LibreTranslate
 from utils.i18n.placeholders import Rejected
-from utils.i18n.translate import URL_SUPPRESSION, apply, damaged, pending, retry
+from utils.i18n.translate import (
+    URL_SUPPRESSION,
+    apply,
+    damaged,
+    pending,
+    retry,
+    taken_translations,
+)
 
 TOKEN = re.compile(r'<x id="\d+"></x>')
 
@@ -318,5 +325,50 @@ class TestDamagedMarkup(unittest.TestCase):
         self.assertEqual(damaged(catalog), [])
 
 
+class TestDuplicateTranslation(unittest.TestCase):
+    LONG_A = "Central identity provider issuing tokens for every service"
+    LONG_B = "Issue and project tracking platform for software teams today"
+    SHORT_A = "Save the file"
+    SHORT_B = "Open the file"
+
+    def _messages(self, *sources: str) -> list:
+        catalog = Catalog(locale="de")
+        for source in sources:
+            catalog.add(source, "")
+        return [message for message in catalog if message.id]
+
+    def test_a_second_long_source_may_not_reuse_a_translation(self) -> None:
+        messages = self._messages(self.LONG_A, self.LONG_B)
+
+        discarded = apply(
+            messages,
+            ["Beschreibung eines Dienstes", "Beschreibung eines Dienstes"],
+            set(),
+        )
+
+        self.assertEqual(discarded, 1)
+        self.assertEqual(str(messages[1].string), "")
+
+    def test_short_sources_may_share_one(self) -> None:
+        messages = self._messages(self.SHORT_A, self.SHORT_B)
+
+        discarded = apply(messages, ["Speichern", "Speichern"], set())
+
+        self.assertEqual(discarded, 0)
+        self.assertEqual(str(messages[1].string), "Speichern")
+
+    def test_without_the_set_the_check_is_skipped(self) -> None:
+        messages = self._messages(self.LONG_A, self.LONG_B)
+
+        self.assertEqual(apply(messages, ["Gleich", "Gleich"], None), 0)
+
+    def test_what_the_catalog_already_spends_is_seeded(self) -> None:
+        catalog = Catalog(locale="de")
+        catalog.add(self.LONG_A, "Schon vergeben")
+        catalog.add(self.SHORT_A, "Kurz")
+
+        self.assertEqual(taken_translations(catalog), {"Schon vergeben"})
+
+
 if __name__ == "__main__":
     unittest.main()