The poster-tag tolerance only without a counter
"[ MacGyver.S01E03.DVDRip.XviD-MEDiEVAL ]-[04/38] - "sample.macgy.s01e03-med.avi""
shares one word with the sample's stem: 6225008 took the name for a tag and
split the episode 17 + 21. A counted subject names its upload.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Daan MeijerandClaude Opus 5.5 committed
1 parent
62250081ef
commit
0fafe71b21
3 files changed
+16
-10
No files matched your search
+9
-6
@@ -594,7 +594,7 @@ const TAG_WORDS: [&str; 21] = [
|
||||
];
|
||||
|
||||
/// isPosterTag: the text outside the quotes shares no word with the file's stem.
|
||||
pub fn is_poster_tag(remainder: &str, stem: &str) -> bool {
|
||||
pub fn is_poster_tag(remainder: &str, stem: &str, counted: bool) -> 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();
|
||||
@@ -621,8 +621,9 @@ pub fn is_poster_tag(remainder: &str, stem: &str) -> bool {
|
||||
shared += 1;
|
||||
}
|
||||
}
|
||||
// A tag of 4+ words may share one with the stem ("German Board - [Usenet4ever.info] und [SecretUsenet.com]").
|
||||
shared == 0 || (shared == 1 && significant >= 4)
|
||||
// Without a counter, a tag of 4+ words may share one with the stem ("German Board - [Usenet4ever.info] und [SecretUsenet.com]").
|
||||
// With one, that's a name ("[ MacGyver.S01E03.DVDRip.XviD-MEDiEVAL ]-[04/38] - "sample.macgy.s01e03-med.avi"").
|
||||
shared == 0 || (shared == 1 && significant >= 4 && !counted)
|
||||
}
|
||||
|
||||
/// releaseIdentity for a named (not weak) file: never null.
|
||||
@@ -697,7 +698,7 @@ pub fn release_identity(poster: &str, subject: &str, name: &str) -> Release {
|
||||
let hashed = file_total.is_none() && test(&HASHED, &stem);
|
||||
// A dated stem says which upload it is, short as it is ("09-30-2008_CTC.r00").
|
||||
let dated = date_led(stem.as_bytes());
|
||||
let tag = hashed || (!image && !numbered && !test(&AUDIO, name) && (describes(&stem) || dated) && !looks_random(&stem) && is_poster_tag(&remainder, &stem));
|
||||
let tag = hashed || (!image && !numbered && !test(&AUDIO, name) && (describes(&stem) || dated) && !looks_random(&stem) && is_poster_tag(&remainder, &stem, total.is_some()));
|
||||
let prefix = if image { image_prefix(&stem) } else { String::new() };
|
||||
// An image set's subject may be one word, the model's name: /^[A-Za-z]{3,}$/.
|
||||
let one_word = image && remainder.len() >= 3 && remainder.bytes().all(|c| c.is_ascii_alphabetic());
|
||||
@@ -1190,8 +1191,8 @@ mod tests {
|
||||
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"));
|
||||
assert!(is_poster_tag("PowBeertje80 Posted", "Bravo Hits Vol.14 2CD -1996-", false));
|
||||
assert!(!is_poster_tag("Audiofiles DIE ZEIT 35", "35_China_dl", false));
|
||||
}
|
||||
|
||||
#[test]
|
||||
@@ -1407,6 +1408,8 @@ mod image_set_tests {
|
||||
assert_ne!(gb("6c57319fbae505679737bd46a76193a1-u4e.part6.rar").key, gb("6fc0ff192a355703b3e81557893013b1-u4e.part6.rar").key);
|
||||
assert_ne!(gb("21.Jump.Street.2012.BDRip.AC3.German.XviD-POE.part01.rar").key, gb("Ice.Age.4.Voll.verschoben.2012.BDRip.AC3.German.XviD-POE.part01.rar").key);
|
||||
assert_eq!(gb("21.Jump.Street.2012.BDRip.AC3.German.XviD-POE.part01.rar").key, gb("21.Jump.Street.2012.BDRip.AC3.German.XviD-POE.par2").key);
|
||||
let mac = |i: &str, n: &str| release_identity("p", &format!(r#"[1598]-[FULL]-[#a.b.teevee@EFNet]-[ MacGyver.S01E03.DVDRip.XviD-MEDiEVAL ]-[{i}/38] - "{n}" yEnc"#), n);
|
||||
assert_eq!(mac("04", "sample.macgy.s01e03-med.avi").key, mac("05", "macgy.s01e03-med.r00").key);
|
||||
let pck = |i: &str, n: &str| release_identity("p", &format!(r#"(????) [{i}/2195] - "{n}" yEnc"#), n);
|
||||
assert_eq!(pck("0659", "DN.VPL PCK.1508").key, pck("0660", "DN.VPL PCK.1509").key);
|
||||
assert_eq!(pck("0659", "DN.VPL PCK.1508").name, "DN.VPL PCK.1508");
|
||||
|
||||
@@ -238,7 +238,7 @@ const TAG_WORDS = new Set(['www', 'http', 'https', 'com', 'net', 'org', 'info',
|
||||
* boilerplate doesn't count. A number of 5+ digits is a per-post id
|
||||
* ("FTD 769997", "(781442)"), so never a constant tag.
|
||||
*/
|
||||
export function isPosterTag(remainder: string, stem: string): boolean {
|
||||
export function isPosterTag(remainder: string, stem: string, counted = false): boolean {
|
||||
const words = stem.toLowerCase().replace(/[^a-z0-9]+/g, ' ').trim()
|
||||
const compact = words.replace(/ /g, '')
|
||||
const whole = new Set(words.split(' '))
|
||||
@@ -255,9 +255,10 @@ export function isPosterTag(remainder: string, stem: string): boolean {
|
||||
significant++
|
||||
if (word.length >= 4 ? compact.includes(word) : whole.has(word)) shared++
|
||||
}
|
||||
// A tag of 4+ words may share one with the stem ("German Board - [Usenet4ever.info] und [SecretUsenet.com] -
|
||||
// Without a counter, a tag of 4+ words may share one with the stem ("German Board - [Usenet4ever.info] und [SecretUsenet.com] -
|
||||
// "21.Jump.Street.2012.BDRip.AC3.German.XviD-POE.part01.rar"": the poster's uploads were releases of 1,535 files).
|
||||
return shared === 0 || (shared === 1 && significant >= 4)
|
||||
// With one, that's a name ("[ MacGyver.S01E03.DVDRip.XviD-MEDiEVAL ]-[04/38] - "sample.macgy.s01e03-med.avi"").
|
||||
return shared === 0 || (shared === 1 && significant >= 4 && !counted)
|
||||
}
|
||||
|
||||
/**
|
||||
@@ -344,7 +345,7 @@ export function releaseIdentity(input: ReleaseInput): ReleaseIdentity | null {
|
||||
// "055631a1_b6c820bc_a856086c_85112a5-u4e.part01.rar"": a day's uploads were one release; LOG.md 2026-10-12 (11)).
|
||||
const hashed = fileTotal === undefined && HASHED.test(stem)
|
||||
// A dated stem says which upload it is, short as it is ("09-30-2008_CTC.r00" under "#sterntuary - Coast To Coast Radio Show").
|
||||
const tag = hashed || (!image && !numbered && !AUDIO.test(input.name) && (describes(stem) || dated) && !looksRandom(stem) && isPosterTag(remainder, stem))
|
||||
const tag = hashed || (!image && !numbered && !AUDIO.test(input.name) && (describes(stem) || dated) && !looksRandom(stem) && isPosterTag(remainder, stem, fileTotal !== undefined))
|
||||
// An image set without a counter: the image's name up to its last number, without numbers
|
||||
// ("Kitchen Hottie 1 set233-76-lg" → "Kitchen Hottie set", "bm_jessica_blb_h087" → "bm jessica blb h").
|
||||
const prefix = image ? stem.replace(/\d+\D*$/, '').replace(/\d+/g, ' ').replace(/[\s._-]+/g, ' ').trim() : ''
|
||||
|
||||
@@ -333,6 +333,8 @@ test.group('release / par2 names', () => {
|
||||
assert.notEqual(gb('6c57319fbae505679737bd46a76193a1-u4e.part6.rar').key, gb('6fc0ff192a355703b3e81557893013b1-u4e.part6.rar').key)
|
||||
assert.notEqual(gb('21.Jump.Street.2012.BDRip.AC3.German.XviD-POE.part01.rar').key, gb('Ice.Age.4.Voll.verschoben.2012.BDRip.AC3.German.XviD-POE.part01.rar').key)
|
||||
assert.equal(gb('21.Jump.Street.2012.BDRip.AC3.German.XviD-POE.part01.rar').key, gb('21.Jump.Street.2012.BDRip.AC3.German.XviD-POE.par2').key)
|
||||
const mac = (i: string, n: string) => id(`[1598]-[FULL]-[#a.b.teevee@EFNet]-[ MacGyver.S01E03.DVDRip.XviD-MEDiEVAL ]-[${i}/38] - "${n}" yEnc`, n)
|
||||
assert.equal(mac('04', 'sample.macgy.s01e03-med.avi').key, mac('05', 'macgy.s01e03-med.r00').key)
|
||||
const pck = (i: string, n: string) => id(`(????) [${i}/2195] - "${n}" yEnc`, n)
|
||||
assert.equal(pck('0659', 'DN.VPL PCK.1508').key, pck('0660', 'DN.VPL PCK.1509').key)
|
||||
assert.equal(pck('0659', 'DN.VPL PCK.1508').name, 'DN.VPL PCK.1508')
|
||||
|
||||
Reference in new issue
Block a user