implemented a lot of features

tests/unit/subject.spec.ts

@@ -0,0 +1,108 @@
+import { test } from '@japa/runner'
+import { fileKey, parseSubject } from '#lib/subject/index'
+import { parseXrefGroups } from '#lib/nntp/xref'
+import { inferTotalParts } from '#lib/yenc/index'
+
+test.group('subject / parseSubject', () => {
+  test('parses file counter, filename and segment counter', ({ assert }) => {
+    assert.deepEqual(parseSubject('. - [09/51] - "obrMkrDeLmbhaouwYO34.part08.rar" yEnc (56/274)'), {
+      name: 'obrMkrDeLmbhaouwYO34.part08.rar',
+      segment: 56,
+      totalSegments: 274,
+      fileIndex: 9,
+      fileTotal: 51,
+      base: '. - [09/51] - "obrMkrDeLmbhaouwYO34.part08.rar" yEnc',
+    })
+  })
+
+  test('handles a file size after the segment counter', ({ assert }) => {
+    const subject = '[8/9] - "x.tar.zst.vol063+064.par2" yEnc (37/72) 51315352'
+    const parsed = parseSubject(subject)!
+    assert.equal(parsed.segment, 37)
+    assert.equal(parsed.totalSegments, 72)
+    assert.equal(parsed.base, '[8/9] - "x.tar.zst.vol063+064.par2" yEnc 51315352')
+  })
+
+  test('ignores counters inside the filename', ({ assert }) => {
+    const parsed = parseSubject('"Show (1/2) Special.mkv" yEnc (3/10)')!
+    assert.equal(parsed.name, 'Show (1/2) Special.mkv')
+    assert.equal(parsed.segment, 3)
+    assert.equal(parsed.totalSegments, 10)
+  })
+
+  test('the base is equal for all segments of one file', ({ assert }) => {
+    const a = parseSubject('[1/9] - "a.tar.zst" yEnc (253/1000) 716554110')!
+    const b = parseSubject('[1/9] - "a.tar.zst" yEnc (254/1000) 716554110')!
+    assert.equal(a.base, b.base)
+  })
+
+  test('returns null for obfuscated or invalid subjects', ({ assert }) => {
+    assert.isNull(parseSubject('GZFDJc0hB3Y0FwwAznkk'))
+    assert.isNull(parseSubject('"name.rar" yEnc'))
+    assert.isNull(parseSubject('"name.rar" yEnc (0/5)'))
+    assert.isNull(parseSubject('"name.rar" yEnc (6/5)'))
+    assert.isNull(parseSubject('"" yEnc (1/5)'))
+  })
+})
+
+test.group('subject / fileKey', () => {
+  test('differs per poster and per segment count, is stable otherwise', ({ assert }) => {
+    const key = fileKey('poster@a', '"x.rar" yEnc', 10)
+    assert.lengthOf(key, 40)
+    assert.equal(key, fileKey('poster@a', '"x.rar" yEnc', 10))
+    assert.notEqual(key, fileKey('poster@b', '"x.rar" yEnc', 10))
+    assert.notEqual(key, fileKey('poster@a', '"x.rar" yEnc', 11))
+  })
+})
+
+test.group('nntp / parseXrefGroups', () => {
+  test('extracts crossposted groups', ({ assert }) => {
+    assert.deepEqual(
+      parseXrefGroups('news.eweka.nl alt.binaries.nl:35801424104 alt.binaries.test:27275430913'),
+      ['alt.binaries.nl', 'alt.binaries.test']
+    )
+    assert.deepEqual(parseXrefGroups(undefined), [])
+  })
+})
+
+test.group('yenc / inferTotalParts', () => {
+  const size = 104857600
+  const partSize = 384000
+
+  test('uses total= when present', ({ assert }) => {
+    assert.equal(inferTotalParts({ begin: { part: 2, total: 7, line: 128, size, name: 'a' } }), 7)
+  })
+
+  test('derives the total from the part size for middle and last parts', ({ assert }) => {
+    const expected = Math.ceil(size / partSize) // 274, as in the subject "(56/274)"
+    assert.equal(expected, 274)
+    assert.equal(
+      inferTotalParts({
+        begin: { part: 56, line: 128, size, name: 'a' },
+        part: { begin: 21120001, end: 21504000 },
+      }),
+      expected
+    )
+    assert.equal(
+      inferTotalParts({
+        begin: { part: 274, line: 128, size, name: 'a' },
+        part: { begin: 273 * partSize + 1, end: size },
+      }),
+      274
+    )
+  })
+
+  test('single-part posts have one part', ({ assert }) => {
+    assert.equal(inferTotalParts({ begin: { line: 128, size: 10, name: 'a' } }), 1)
+  })
+})
+
+test.group('indexer / skipBodyMessageIds', () => {
+  test('matches the multi-group uploader, not other posts', async ({ assert }) => {
+    const { default: indexerConfig } = await import('#config/indexer')
+    const skip = (id: string) => indexerConfig.skipBodyMessageIds.some((re) => re.test(id))
+    assert.isTrue(skip('<b800526$4db823e$6544066@d810c77fc0.d15fa>'))
+    assert.isFalse(skip('<Part56of274.4B8CB0C9360F4548AB800FEDCB39CE07@1790532933.local>'))
+    assert.isFalse(skip('<a6a525b170f44d68a756f1ba6fbd4c60@ngPost>'))
+  })
+})