Subjects without a quoted file name: the name from the rest of the subject

Most "obfuscated" subjects of older and smaller groups name their file
without quotes ("DVDFab 8.2.0.5.rar (006/332)", fta's "q3B9…part023.rar
(104/137)", "New_Super_Mario_Bros._[Keygen].exe (1/2)") or with one
quote ("[keira_s04_0037_l.jpg\" yEnc] (2/2)"): 100% of ath 2013's and
fta's, 94% of models' (LOG.md 2026-10-11 (5)). With a segment counter and
no quoted name, unquotedName takes the name before a lone quote (after a
"[", else after the last space), else the text when it ends in a file
name (after "] " or " - "), else its first word that is one; a file
counter stuck in a bracket is dropped. The release identity takes such a
name out of the subject as it would quotes (else every image of a set
was a release: 9,130 instead of 267 on the models fixture).

Both implementations; a fourth fixture (fta, 25k lines) and tests; check
identical on all four. Truly random subjects (ngPost hex = message-id,
random tokens) stay obfuscated.

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

edge/src/indexer.rs

@@ -89,8 +89,8 @@ struct Counter {
     total: i64,
 }
 
-/// The last "(k/N)" after `after`, scanning back from the end (findCounter).
-fn find_counter(subject: &[u8], after: usize) -> Option<Counter> {
+/// The last "(k/N)" after `after` (anywhere when None: JS -1), scanning back from the end (findCounter).
+fn find_counter(subject: &[u8], after: Option<usize>) -> Option<Counter> {
     let mut open = subject.len();
     loop {
         // lastIndexOf('(', open - 1): at most open - 1.
@@ -98,7 +98,7 @@ fn find_counter(subject: &[u8], after: usize) -> Option<Counter> {
             return None;
         }
         open = subject[..open].iter().rposition(|&c| c == b'(')?;
-        if open <= after {
+        if after.is_some_and(|a| open <= a) {
             return None;
         }
         let mut i = open + 1;
@@ -188,10 +188,8 @@ impl Scanner {
             return None;
         }
         let sb = subject.as_bytes();
-        let name_end = subject.rfind('"')?;
-        if name_end == 0 {
-            return None;
-        }
+        // After the last quote; without one, anywhere (unquoted_name finds the file name).
+        let name_end = subject.rfind('"');
         let counter = find_counter(sb, name_end)?;
         self.key.clear();
         self.key.push_str(from);
@@ -223,9 +221,16 @@ impl Scanner {
     }
 }
 
-fn describe(subject: &str, name_end: usize, counter: &Counter, from: &str, xref: &str, group: &str) -> Option<FileMeta> {
-    let name_start = subject[..name_end].rfind('"')?;
-    let name = &subject[name_start + 1..name_end];
+fn describe(subject: &str, name_end: Option<usize>, counter: &Counter, from: &str, xref: &str, group: &str) -> Option<FileMeta> {
+    let quoted = match name_end {
+        Some(end) if end > 0 => subject[..end].rfind('"').map(|start| subject[start + 1..end].to_string()),
+        _ => None,
+    };
+    let name = match quoted {
+        Some(name) => name,
+        None => unquoted_name(&subject[..counter.index])?,
+    };
+    let name = name.as_str();
     if name.is_empty() || counter.total < 1 {
         return None;
     }
@@ -244,6 +249,64 @@ fn describe(subject: &str, name_end: usize, counter: &Counter, from: &str, xref:
     })
 }
 
+/// EXTENSION: /[^.\s]\.[A-Za-z0-9]{2,4}$/
+fn has_extension(s: &str) -> bool {
+    let Some(dot) = s.rfind('.') else { return false };
+    let ext = &s.as_bytes()[dot + 1..];
+    if !(2..=4).contains(&ext.len()) || !ext.iter().all(|c| c.is_ascii_alphanumeric()) {
+        return false;
+    }
+    s[..dot].chars().last().is_some_and(|c| c != '.' && !js::is_space(c))
+}
+
+/// unquotedName (lib/headers/subject.ts): the file name of a subject without a quoted one, from the text before its counter.
+pub fn unquoted_name(before_counter: &str) -> Option<String> {
+    let mut text = before_counter.trim_end_matches(js::is_space);
+    if text.len() >= 4 && text.is_char_boundary(text.len() - 4) && text[text.len() - 4..].eq_ignore_ascii_case("yenc") {
+        text = text[..text.len() - 4].trim_end_matches(js::is_space);
+    }
+    if let Some(quote) = text.find('"') {
+        if text[quote + 1..].contains('"') {
+            return None;
+        }
+        let before = &text[..quote];
+        if let Some(bracket) = before.rfind('[') {
+            let inside = js::trim(&before[bracket + 1..]);
+            // replace(/^\d+\/\d+ +/, '')
+            let b = inside.as_bytes();
+            let mut i = 0;
+            while i < b.len() && b[i].is_ascii_digit() {
+                i += 1;
+            }
+            let mut stripped = inside;
+            if i > 0 && b.get(i) == Some(&b'/') {
+                let mut j = i + 1;
+                while j < b.len() && b[j].is_ascii_digit() {
+                    j += 1;
+                }
+                if j > i + 1 && b.get(j) == Some(&b' ') {
+                    while j < b.len() && b[j] == b' ' {
+                        j += 1;
+                    }
+                    stripped = &inside[j..];
+                }
+            }
+            if has_extension(stripped) {
+                return Some(stripped.to_string());
+            }
+        }
+        let word = &before[before.rfind(' ').map_or(0, |i| i + 1)..];
+        return has_extension(word).then(|| word.to_string());
+    }
+    if has_extension(text) {
+        let bracket = text.rfind("] ").map_or(0, |i| i + 2);
+        let dash = text.rfind(" - ").map_or(0, |i| i + 3);
+        let name = js::trim(&text[bracket.max(dash)..]);
+        return (!name.is_empty() && has_extension(name)).then(|| name.to_string());
+    }
+    text.split(' ').find(|w| has_extension(w)).map(str::to_string)
+}
+
 /// One line of an XOVER reply: number, subject, from, date, message-id, bytes, xref (scanOverviewText).
 fn scan_overview_text(text: &str, mut line: impl FnMut(Option<i64>, &str, &str, &str, &str, &str, &str)) {
     let b = text.as_bytes();
@@ -794,4 +857,27 @@ mod tests {
         assert_eq!(first[0].iter().map(|s| s.message_id.as_str()).collect::<Vec<_>>(), vec!["a1", "a2", "a3"]);
         assert!(agg.open.is_empty());
     }
+
+#[cfg(test)]
+mod unquoted_tests {
+    use super::unquoted_name;
+
+    #[test]
+    fn file_names_of_subjects_without_a_quoted_one() {
+        let cases = [
+            ("DVDFab 8.2.0.5.rar ", Some("DVDFab 8.2.0.5.rar")),
+            ("q3B9OrpKiCxUnzgtUGnp.part023.rar ", Some("q3B9OrpKiCxUnzgtUGnp.part023.rar")),
+            ("New_Super_Mario_Bros._[Keygen].exe ", Some("New_Super_Mario_Bros._[Keygen].exe")),
+            ("Emily_18_-_Bed.part01.rar Emily 18 bed video FULL NUDE ", Some("Emily_18_-_Bed.part01.rar")),
+            ("[33/117] 674a37548b09.dat ", Some("674a37548b09.dat")),
+            ("pal!    [Mivina high(1).wmv\" yEnc] ", Some("Mivina high(1).wmv")),
+            ("[03/21 Geroellheimer - S03E16.part01.rar\" yEnc ", Some("Geroellheimer - S03E16.part01.rar")),
+            ("[7w] - 9BbEV2MrTi59GBTEFlVrssBrXNwXf2SK9cQEeDjg yEnc ", None),
+            ("Re: Will somebody answer the question? ", None),
+        ];
+        for (text, want) in cases {
+            assert_eq!(unquoted_name(text).as_deref(), want, "{text}");
+        }
+    }
+}
 }

edge/src/names.rs

@@ -514,7 +514,8 @@ pub fn release_identity(poster: &str, subject: &str, name: &str) -> Release {
     let post_total = file_counter_total(subject);
     let file_total = counter_total(subject).and_then(|t| t.parse::<i64>().ok());
     let stem = file_stem(name);
-    let remainder = subject_remainder(subject);
+    // A name that isn't quoted (unquoted_name) is taken out as the quotes would have been.
+    let remainder = if subject.contains(&format!("\"{name}\"")) { subject_remainder(subject) } else { subject_remainder(&subject.replacen(name, " ", 1).replace('"', " ")) };
     // A document or comic is a release of its own.
     if test(&DOCUMENT, name) {
         let shown = replace(&DOCUMENT, &stem, "", false);