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>

lib/edge/scan.ts

@@ -1,4 +1,4 @@
-import { fileKey } from '../headers/index.js'
+import { fileKey, unquotedName } from '../headers/index.js'
 import { releaseIdentity, type ReleaseIdentity } from '../release/index.js'
 
 /** A file's details, shared by all of its segments (one object per file, not per article). */
@@ -112,8 +112,8 @@ export class SegmentScanner {
   /** A segment from an article's fields, or null when its subject is obfuscated. */
   segment(articleNumber: number, subject: string, from: string, date: string, messageId: string, bytes: string, xref: string, group: string): EdgeSegment | null {
     if (!subject || !messageId) return null
+    // After the last quote; without one, anywhere (unquotedName finds the file name).
     const nameEnd = subject.lastIndexOf('"')
-    if (nameEnd <= 0) return null
     const counter = findCounter(subject, nameEnd)
     if (!counter) return null
     const key = from + '\u0001' + subject.slice(0, counter.index) + '\u0000' + subject.slice(counter.index + counter.length)
@@ -137,10 +137,10 @@ export class SegmentScanner {
 
   /** A file's details from its first article seen (as parseSubject and segmentFromHeader). */
   private describe(subject: string, nameEnd: number, counter: Counter, from: string, xref: string, group: string): FileMeta | null {
-    const nameStart = subject.lastIndexOf('"', nameEnd - 1)
-    if (nameStart === -1) return null
-    const name = own(subject.slice(nameStart + 1, nameEnd))
-    if (!name || counter.total < 1) return null
+    const nameStart = nameEnd > 0 ? subject.lastIndexOf('"', nameEnd - 1) : -1
+    const found = nameStart !== -1 ? subject.slice(nameStart + 1, nameEnd) : unquotedName(subject.slice(0, counter.index))
+    if (!found || counter.total < 1) return null
+    const name = own(found)
     const base = own((subject.slice(0, counter.index) + subject.slice(counter.index + counter.length)).replace(/\s+/g, ' ').trim())
     const poster = own(from)
     return {

lib/headers/index.ts

@@ -1,3 +1,3 @@
-export { parseSubject, type ParsedSubject } from './subject.js'
+export { parseSubject, unquotedName, type ParsedSubject } from './subject.js'
 export { fileKey, weakFileKey, counterFileKey, looksLikeFilename } from './file_key.js'
 export { segmentFromHeader, type IngestSegment, type OverviewHeader, type IdentityCache } from './segment.js'

lib/headers/subject.ts

@@ -16,6 +16,46 @@ export interface ParsedSubject {
 }
 
 const SEGMENT_COUNTER = /\((\d+)\/(\d+)\)/g
+/** A file name's extension: a dot and 2 to 4 letters or digits at the end. */
+const EXTENSION = /[^.\s]\.[A-Za-z0-9]{2,4}$/
+
+/**
+ * The file name of a subject without a quoted one, from the text before its
+ * segment counter (LOG.md 2026-10-11 (5): most "obfuscated" subjects of older
+ * groups): with one quote only, the name before it (after a "[", else after
+ * the last space: `[keira_s04_0037_l.jpg" yEnc] (2/2)`); without quotes, the
+ * text when it ends in a file name (after a "] " or " - ": `DVDFab
+ * 8.2.0.5.rar (006/332)`, `[33/117] 674a37548b09.dat (1/1)`), else its
+ * first word that is one (`Emily_18_-_Bed.part01.rar Emily 18 bed video
+ * (01/24)`). Null when no file name is found: the subject stays obfuscated.
+ * Words are split on spaces only (both implementations alike).
+ */
+export function unquotedName(beforeCounter: string): string | null {
+  let text = beforeCounter.trimEnd()
+  if (/yenc$/i.test(text)) text = text.slice(0, -4).trimEnd()
+  const quote = text.indexOf('"')
+  if (quote !== -1) {
+    if (text.indexOf('"', quote + 1) !== -1) return null
+    const before = text.slice(0, quote)
+    const bracket = before.lastIndexOf('[')
+    if (bracket !== -1) {
+      // A file counter that lost its closing bracket ("[03/21 Name.part01.rar") isn't part of the name.
+      const inBracket = before.slice(bracket + 1).trim().replace(/^\d+\/\d+ +/, '')
+      if (EXTENSION.test(inBracket)) return inBracket
+    }
+    const word = before.slice(before.lastIndexOf(' ') + 1)
+    return EXTENSION.test(word) ? word : null
+  }
+  if (EXTENSION.test(text)) {
+    const bracket = text.lastIndexOf('] ')
+    const dash = text.lastIndexOf(' - ')
+    const cut = Math.max(bracket === -1 ? 0 : bracket + 2, dash === -1 ? 0 : dash + 3)
+    const name = text.slice(cut).trim()
+    return name && EXTENSION.test(name) ? name : null
+  }
+  for (const word of text.split(' ')) if (EXTENSION.test(word)) return word
+  return null
+}
 const FILE_COUNTER = /\[(\d+)\/(\d+)\]/
 
 /**
@@ -28,16 +68,15 @@ const FILE_COUNTER = /\[(\d+)\/(\d+)\]/
 export function parseSubject(subject: string): ParsedSubject | null {
   const nameEnd = subject.lastIndexOf('"')
   const nameStart = nameEnd > 0 ? subject.lastIndexOf('"', nameEnd - 1) : -1
-  if (nameStart === -1) return null
-  const name = subject.slice(nameStart + 1, nameEnd)
-  if (!name) return null
 
-  // The segment counter is the last "(k/N)" after the filename; a file size may follow it.
+  // The segment counter is the last "(k/N)" after the filename (anywhere without a quote); a file size may follow it.
   let counter: RegExpExecArray | undefined
   for (const match of subject.matchAll(SEGMENT_COUNTER)) {
     if (match.index > nameEnd) counter = match
   }
   if (!counter) return null
+  const name = nameStart !== -1 ? subject.slice(nameStart + 1, nameEnd) : unquotedName(subject.slice(0, counter.index))
+  if (!name) return null
 
   const segment = Number.parseInt(counter[1], 10)
   const totalSegments = Number.parseInt(counter[2], 10)
@@ -47,7 +86,7 @@ export function parseSubject(subject: string): ParsedSubject | null {
     .replace(/\s+/g, ' ')
     .trim()
 
-  const fileCounter = FILE_COUNTER.exec(subject.slice(0, nameStart))
+  const fileCounter = FILE_COUNTER.exec(subject.slice(0, nameStart !== -1 ? nameStart : counter.index))
   return {
     name,
     segment,

lib/release/identity.ts

@@ -197,8 +197,10 @@ export function releaseIdentity(input: ReleaseInput): ReleaseIdentity | null {
     return { key: hash('weak', stem), name: stem, source: 'weak-stem' }
   }
 
-  // Text outside the quotes that names the release, e.g. "Some.Release [01/45] - "x.rar"".
-  const remainder = subjectRemainder(input.subject)
+  // Text outside the quotes that names the release, e.g. "Some.Release [01/45] - "x.rar"". A name that isn't
+  // quoted (unquotedName: "DVDFab 8.2.0.5.rar (006/332)", "[keira_s04_0037_l.jpg" yEnc] (2/2)") is taken out
+  // as the quotes would have been: otherwise every file of a set would name a release of its own.
+  const remainder = subjectRemainder(input.subject.includes(`"${input.name}"`) ? input.subject : input.subject.replace(input.name, ' ').replace(/"/g, ' '))
   // A document or comic is a release of its own (collections post hundreds under one subject).
   if (DOCUMENT.test(input.name)) return { key: hash('stem', input.poster, stem), name: stem.replace(DOCUMENT, '') || stem, fileTotal, source: 'stem' }
   const image = IMAGE.test(input.name)