Initial snapshot before step 10 package split

2026-05-22 21:39:09 +02:00
commit e029e898e9
16 changed files with 3955 additions and 0 deletions
@@ -0,0 +1,22 @@
+{
+  "version": 1,
+  "domain": "query",
+  "description": "Page text / title -> canonical JAV ID. Consumed by the browser extension (content.js normalizeId). Difference from filename: looser context (sentences, mixed punctuation, site chrome). Includes forms (e.g. FC2PPV compact) that Python extract_id does NOT handle, by design — see fixtures/README.md.",
+  "case_schema": {
+    "name": "human label",
+    "input": "raw page text",
+    "expected": "canonical ID without part suffix (extension never emits #partN), or null when no ID found"
+  },
+  "cases": [
+    { "name": "title with site chrome",                "input": "SSIS-001 — JAV.tube",                                     "expected": "SSIS-001" },
+    { "name": "title with description",                "input": "Watch SSIS-001 1080p HD Online",                          "expected": "SSIS-001" },
+    { "name": "trailing letter variant",               "input": "IBW-902z Full Movie",                                     "expected": "IBW-902" },
+    { "name": "no hyphen in title",                    "input": "MVSD312 stream",                                          "expected": "MVSD-312" },
+    { "name": "FC2 PPV compact",                       "input": "FC2PPV-1841460 — preview",                                "expected": "FC2-PPV-1841460" },
+    { "name": "FC2 plain digits",                      "input": "FC2-1841460 thumbnail",                                   "expected": "FC2-PPV-1841460" },
+    { "name": "FC2-PPV explicit",                      "input": "FC2-PPV-1841460 Full",                                    "expected": "FC2-PPV-1841460" },
+    { "name": "leading zeros preserved",               "input": "ABF-042 — sample",                                        "expected": "ABF-042" },
+    { "name": "long numeric tail (7 digits)",          "input": "BLK-4748520 stream",                                      "expected": "BLK-4748520" },
+    { "name": "no ID present",                         "input": "JAV Database · home",                                     "expected": null }
+  ]
+}