Identity: other quoted text, text separators, ids in image names, more

- Quoted text other than the file name stays in the remainder: ">>> CREATIVE
  COMMONS NZB <<< "0190 alarm 5.0 " - File 1 of 9: "…rar"" made 12,605
  files of all uploads one release.
- namesRelease accepts words joined by , & : ; \ / + ("Posting & deleting,
  read 0 file "E:\VladModels\y038_Irina\185" …"): such image sets fell to
  the stem rule ("P8250001.JPG"), one release over all folders.
- imageSet: a 5+ digit number left in the set is a per-image id
  ("sss-70556-3474182-23-10825418-pc", "Alena p 26663_…"): no set. Copy
  markers may repeat ("diana-050-052[1][1]").
- unquotedName: an image after numbered text without a separator is the
  last word ("<abm> … AB: part2 0573-1002 043.jpg").

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

edge/src/indexer.rs

@@ -296,6 +296,12 @@ pub fn name_beside_quotes(quoted: &str, before_counter: &str) -> Option<String>
 }
 
 /// unquotedName (lib/headers/subject.ts): the file name of a subject without a quoted one, from the text before its counter.
+/// /\.(jpe?g|png|gif|bmp|webp|tiff?)$/i
+fn is_image_name(name: &str) -> bool {
+    let lower = name.to_ascii_lowercase();
+    [".jpg", ".jpeg", ".png", ".gif", ".bmp", ".webp", ".tif", ".tiff"].iter().any(|e| lower.ends_with(e))
+}
+
 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") {
@@ -340,7 +346,17 @@ pub fn unquoted_name(before_counter: &str) -> Option<String> {
         let dash = first_track(text).or_else(|| text.rfind(" - ")).map_or(0, |i| i + 3);
         // Or after an arrow ("… -> Exclusive-tgirls_000000-lg.jpg").
         let arrow = text.rfind("-> ").map_or(0, |i| i + 3);
-        let name = js::trim(&text[bracket.max(dash).max(arrow)..]);
+        let cut = bracket.max(dash).max(arrow);
+        let name = js::trim(&text[cut..]);
+        // An image after numbered text without a separator: its last word ("<abm> … AB: part2 0573-1002 043.jpg").
+        // Not an index alone ("Teen Tease Self Shooter sad1 (10).jpg").
+        if cut == 0 && is_image_name(name) && name.contains(' ') {
+            let space = name.rfind(' ').unwrap();
+            let last = &name[space + 1..];
+            if name[..space].trim_end_matches(js::is_space).ends_with(|c: char| c.is_ascii_digit()) && !last.starts_with('(') {
+                return Some(last.to_string());
+            }
+        }
         return (!name.is_empty() && has_extension(name)).then(|| name.to_string());
     }
     text.split(' ').find(|w| has_extension(w)).map(str::to_string)
@@ -913,6 +929,7 @@ mod unquoted_tests {
             ("pal!    [Mivina high(1).wmv\" yEnc] ", Some("Mivina high(1).wmv")),
             ("[03/21 Geroellheimer - S03E16.part01.rar\" yEnc ", Some("Geroellheimer - S03E16.part01.rar")),
             ("<Group> FP -> {EXL01~P~004} -> Alexia (3) -> VF -> [0001 of 0076] -> Exclusive-tgirls_000000-lg.jpg ", Some("Exclusive-tgirls_000000-lg.jpg")),
+            ("<abm> <Duke> here's a little thanksgiving present: AB: part2 0573-1002 043.jpg ", Some("043.jpg")),
             ("[7w] - 9BbEV2MrTi59GBTEFlVrssBrXNwXf2SK9cQEeDjg yEnc ", None),
             ("Re: Will somebody answer the question? ", None),
         ];

edge/src/names.rs

@@ -465,13 +465,15 @@ pub fn image_set(stem: &str) -> String {
             break;
         }
     }
-    match s.iter().rposition(u8::is_ascii_digit) {
-        Some(last) => {
-            let start = s[..=last].iter().rposition(|c| !c.is_ascii_digit()).map_or(0, |i| i + 1);
-            String::from_utf8_lossy(&s[..start]).into_owned()
-        }
-        None => String::new(),
+    let set = match s.iter().rposition(u8::is_ascii_digit) {
+        Some(last) => &s[..s[..=last].iter().rposition(|c| !c.is_ascii_digit()).map_or(0, |i| i + 1)],
+        None => return String::new(),
+    };
+    // A number of 5+ digits left in it is an id, per image ("sss-70556-3474182-23-10825418-pc"): no set.
+    if set.split(|c| !c.is_ascii_digit()).any(|run| trim_zeros(run).len() >= 5) {
+        return String::new();
     }
+    String::from_utf8_lossy(set).into_owned()
 }
 
 /// Number(digits) equality for digit strings: without leading zeros (all zeros: "0").
@@ -481,7 +483,7 @@ fn trim_zeros(d: &[u8]) -> &[u8] {
 }
 
 static TIMESTAMP: LazyLock<Regex> = LazyLock::new(|| re(r"[-_. ]?(?:19|20)[0-9]{2}[-._ ](?:0?[1-9]|1[0-2])[-._ ][0-9]{1,2}(?:[-._ ][0-9]{1,2})*$"));
-static COPY_MARK: LazyLock<Regex> = LazyLock::new(|| re(r"SP*[\[(][0-9]+[\])]$"));
+static COPY_MARK: LazyLock<Regex> = LazyLock::new(|| re(r"(?:SP*[\[(][0-9]+[\])])+$"));
 static RESOLUTION: LazyLock<Regex> = LazyLock::new(|| re(r"(?i)[_-][0-9]{3,4}(?:x[0-9]{3,4}|px)$"));
 static IMAGEVENUE: LazyLock<Regex> = LazyLock::new(|| re(r"(?i)_123_[0-9]+lo$"));
 static ZERO_SUFFIX: LazyLock<Regex> = LazyLock::new(|| re(r"[_-]0+$"));
@@ -637,8 +639,10 @@ pub fn release_identity(poster: &str, subject: &str, name: &str) -> Release {
     let stem = file_stem(name);
     // A name that isn't quoted (unquoted_name) is taken out as the quotes would have been.
     // The subject here has its spaces collapsed; the name may not.
-    let named = if subject.contains(&format!("\"{name}\"")) {
-        subject.to_string()
+    // Other quoted text stays: it names the release (">>> CREATIVE COMMONS NZB <<< "0190 alarm 5.0 " - File 1 of 9: …").
+    let quoted = format!("\"{name}\"");
+    let named = if subject.contains(&quoted) {
+        subject.replacen(&quoted, " ", 1).replace('"', " ")
     } else {
         let collapsed = js::collapse_spaces(name);
         let collapsed = js::trim(&collapsed);
@@ -916,7 +920,7 @@ pub fn upload_tool(message_id: &str) -> &'static str {
 
 /// [A-Za-z]{2,}[._ -]+[A-Za-z0-9]{2,} anywhere.
 fn name_like(b: &[u8]) -> bool {
-    let sep = |c: u8| matches!(c, b'.' | b'_' | b' ' | b'-');
+    let sep = |c: u8| matches!(c, b'.' | b'_' | b' ' | b'-' | b',' | b'&' | b':' | b';' | b'\\' | b'/' | b'+');
     (2..b.len()).any(|k| {
         if !(sep(b[k]) && b[k - 1].is_ascii_alphabetic() && b[k - 2].is_ascii_alphabetic()) {
             return false;
@@ -1221,6 +1225,9 @@ mod image_set_tests {
             ("092_081_2000x1500", "092_"),
             ("KP0027_197_00", "KP"),
             ("01079-106-2048px", "01079-"),
+            ("diana-050-052[1][1]", "diana-050-"),
+            ("sss-70556-3474182-23-10825418-pc", ""),
+            ("Alena p 26663_alena-p-2-8323_123_39lo", ""),
             ("she has sensual legs 172 _ 172", "she has sensual legs "),
             ("jamie.lee.curtis_the.tailor.of.panama_17_123_244lo", "jamie.lee.curtis_the.tailor.of.panama_"),
             ("JimAndJoey Kindly Husband-to-Be", ""),
@@ -1238,6 +1245,15 @@ mod image_set_tests {
         assert_eq!(mint("JimAndJoey Kindly Husband-to-Be.JPG").key, mint("Oops OnlyOneMore.JPG").key);
     }
 
+    #[test]
+    fn other_quoted_text_and_text_separators_name_the_release() {
+        let cc = |name: &str, n: &str| release_identity("p", &format!(r#">>> CREATIVE COMMONS NZB <<< "{name} " - File 1 of 9: "creative_commons_nzb_{n}.rar" yEnc"#), &format!("creative_commons_nzb_{n}.rar"));
+        assert_ne!(cc("0190 alarm 5.0", "0190_alarm_5.0_").key, cc("portable irfanview 4.22", "portable_irfanview_4.22").key);
+        let vlad = |dir: &str, total: u32, n: &str| release_identity("p", &format!(r#"Posting & deleting, read 0 file "E:\VladModels\y038_Irina\{dir}" Day 5/6 [01/{total}] - {n} yEnc"#), n);
+        assert_eq!(vlad("185", 77, "P8250001.JPG").source, "subject");
+        assert_ne!(vlad("185", 77, "P8250001.JPG").key, vlad("234", 77, "P5300085.JPG").key);
+    }
+
     #[test]
     fn images_without_words_by_their_set() {
         let id = |n: &str, c: &str| release_identity("p", &format!(r#"##[ "{n}" ({c})"#), n);