Rebuild: a poster's constant tag no longer makes one release of all their uploads

Release 145637 held four albums: the poster puts '[KiLL.dA.FuCkIn.
EaStErRaBbIt]----' before every upload, and the text outside the quotes
(when it names a release) was the key with the poster. The same made
one release of a poster's albums under '(PowBeertje80 Posted)' (a post
id, with the same [i/M]), of masterposts' films, and of one volume of
every set in a DVD9 post.

Now, when the file's stem describes something itself (12+ letters and
digits, two words of 3+ letters) and the text outside the quotes shares
no word with it (4+ letters as part of the stem, 3-letter words and
numbers whole; boilerplate like www/net/posted not counted; no 5+ digit
per-post id), that text is taken for the poster's tag and the stem
says which release the file is. Albums of track files ('FTA … Ketchup'),
'DIE_ZEIT_35' with '35_China_dl', 'FTD769997' posts, scene names with
abbreviated file names ('grp-sm') and generic sidecars ('file_id.diz')
keep their grouping.

releases:regroup re-keys a sample of the catalog with the current rule:
on homelab's test data 145637 → 4 albums, 111566 → 17 albums of 27
files, the rest of the subject releases that change mostly lose an ad
image ('Flyer-download-2-Day.gif'); stem releases unchanged. Ported to
edge-bench-rs: check.ts still finds all captures identical (ath 2013:
16 → 76 releases on both sides).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

tools/edge-bench-rs/src/names.rs

@@ -348,15 +348,76 @@ fn hash(parts: &[&str]) -> String {
     sha1_hex(&parts.join("\n"))
 }
 
+/// A name that says what it is: 12+ letters and digits, two words of 3+ letters (`describes`).
+fn describes(stem: &str) -> bool {
+    let b = stem.as_bytes();
+    if b.iter().filter(|c| c.is_ascii_alphanumeric()).count() < 12 {
+        return false;
+    }
+    // [A-Za-z]{3,}[^A-Za-z0-9]+[A-Za-z]{3,}
+    let mut i = 0;
+    while i < b.len() {
+        let letters = b[i..].iter().take_while(|c| c.is_ascii_alphabetic()).count();
+        if letters == 0 {
+            i += 1;
+            continue;
+        }
+        let end = i + letters;
+        if letters >= 3 {
+            let gap = b[end..].iter().take_while(|c| !c.is_ascii_alphanumeric()).count();
+            if gap > 0 && b[end + gap..].iter().take_while(|c| c.is_ascii_alphabetic()).count() >= 3 {
+                return true;
+            }
+        }
+        i = end;
+    }
+    false
+}
+
+const TAG_WORDS: [&str; 21] = [
+    "www", "http", "https", "com", "net", "org", "info", "the", "and", "presents", "posted", "post", "powered", "yenc", "part", "rar", "par2", "nzb", "voor", "for", "from",
+];
+
+/// isPosterTag: the text outside the quotes shares no word with the file's stem.
+pub fn is_poster_tag(remainder: &str, stem: &str) -> bool {
+    let lower = stem.to_lowercase();
+    let words: Vec<&str> = lower.split(|c: char| !(c.is_ascii_lowercase() || c.is_ascii_digit())).filter(|w| !w.is_empty()).collect();
+    let compact: String = words.concat();
+    let whole: std::collections::HashSet<&str> = words.iter().copied().collect();
+    // A per-post id, also inside a word ("FTD772544").
+    let b = remainder.as_bytes();
+    if b.windows(5).any(|w| w.iter().all(u8::is_ascii_digit)) {
+        return false;
+    }
+    let rem = remainder.to_lowercase();
+    for word in rem.split(|c: char| !(c.is_ascii_lowercase() || c.is_ascii_digit())) {
+        if !word.is_empty() && word.bytes().all(|c| c.is_ascii_digit()) {
+            if word.len() >= 2 && whole.contains(word) {
+                return false;
+            }
+            continue;
+        }
+        if word.len() < 3 || TAG_WORDS.contains(&word) {
+            continue;
+        }
+        if if word.len() >= 4 { compact.contains(word) } else { whole.contains(word) } {
+            return false;
+        }
+    }
+    true
+}
+
 /// releaseIdentity for a named (not weak) file: never null.
 pub fn release_identity(poster: &str, subject: &str, name: &str) -> Release {
     let file_total = file_counter_total(subject);
     let stem = file_stem(name);
     let remainder = subject_remainder(subject);
-    if names_release(&remainder) {
+    // The poster's constant tag outside the quotes: the file's stem says which upload it is.
+    let tag = describes(&stem) && !looks_random(&stem) && is_poster_tag(&remainder, &stem);
+    if !tag && names_release(&remainder) {
         return Release { key: hash(&["subject", poster, &remainder]), name: remainder, source: "subject", file_total };
     }
-    if let Some(total) = file_total.filter(|&n| n != 0) {
+    if let Some(total) = file_total.filter(|&n| n != 0 && !tag) {
         if is_post_token(&remainder) {
             return Release { key: hash(&["post", poster, &remainder, &total.to_string()]), name: remainder, source: "post", file_total };
         }
@@ -722,6 +783,24 @@ mod tests {
         }
     }
 
+    #[test]
+    fn poster_tags_as_in_the_typescript() {
+        let kill = "www.ggb.6x.to <darkness@earth.de>";
+        let a = release_identity(kill, r#"[KiLL.dA.FuCkIn.EaStErRaBbIt]---- (03/10) "Cephalgy.-.Weisses.Fleisch.2003[[g.6x.to]].r00" yEnc"#, "Cephalgy.-.Weisses.Fleisch.2003[[g.6x.to]].r00");
+        let nfo = release_identity(kill, r#"[KiLL.dA.FuCkIn.EaStErRaBbIt]---- (01/10) "Cephalgy.-.Weisses.Fleisch.2003[[g.6x.to]].nfo" yEnc"#, "Cephalgy.-.Weisses.Fleisch.2003[[g.6x.to]].nfo");
+        let b = release_identity(kill, r#"[KiLL.dA.FuCkIn.EaStErRaBbIt]---- (03/10) "Shnarph.-.Lautlos.2008[[g.6x.to]].r00" yEnc"#, "Shnarph.-.Lautlos.2008[[g.6x.to]].r00");
+        assert_eq!(a.key, nfo.key);
+        assert_ne!(a.key, b.key);
+        assert_eq!(a.source, "stem");
+        let ftd1 = release_identity("p", r#"FTD769997 [01/23] - "Prison.Break.S04E02.720p.part01.rar" yEnc"#, "Prison.Break.S04E02.720p.part01.rar");
+        let ftd2 = release_identity("p", r#"FTD769997 [23/23] - "Prison.Break.S04E02.720p.part.par2" yEnc"#, "Prison.Break.S04E02.720p.part.par2");
+        assert_eq!(ftd1.key, ftd2.key);
+        let scene = release_identity("p", r#"Some.Movie.2008.DVDRip.XviD-GRP [01/40] - "grp-sm.part01.rar" yEnc"#, "grp-sm.part01.rar");
+        assert_eq!(scene.name, "Some Movie 2008 DVDRip XviD GRP");
+        assert!(is_poster_tag("PowBeertje80 Posted", "Bravo Hits Vol.14 2CD -1996-"));
+        assert!(!is_poster_tag("Audiofiles DIE ZEIT 35", "35_China_dl"));
+    }
+
     #[test]
     fn random_names() {
         assert!(looks_random("prvjr7hapd0i1ny14pz4"));