VobSub

VobSub is a DVD subtitle format using .idx (index) and .sub (bitmap data).

Timing

• Index file timestamps in HH:MM:SS:ms

Images

• Default parsing decodes bitmap data into event.image (format: 'indexed').
• Use decode: 'rle' to keep raw RLE data and skip bitmap decoding.
• Use decode: 'none' to skip sub packet parsing and keep only timing (no image data).

Parsing

import { parseVobSub } from 'subforge/vobsub'
import { unwrap } from 'subforge/core'

const idx = await fetch('/subs.idx').then(r => r.text())
const sub = new Uint8Array(await fetch('/subs.sub').then(r => r.arrayBuffer()))
const doc = unwrap(parseVobSub(idx, sub))

Fast metadata parse (skip bitmap decode):

import { parseVobSub } from 'subforge/vobsub'
import { unwrap } from 'subforge/core'

const idx = await fetch('/subs.idx').then(r => r.text())
const sub = new Uint8Array(await fetch('/subs.sub').then(r => r.arrayBuffer()))
const doc = unwrap(parseVobSub(idx, sub, { decode: 'rle' }))

Timing-only parse (skip sub packets):

import { parseVobSub } from 'subforge/vobsub'
import { unwrap } from 'subforge/core'

const idx = await fetch('/subs.idx').then(r => r.text())
const sub = new Uint8Array(await fetch('/subs.sub').then(r => r.arrayBuffer()))
const doc = unwrap(parseVobSub(idx, sub, { decode: 'none' }))

If you already parsed the .idx, you can reuse it:

import { parseIdx, parseVobSub } from 'subforge/vobsub'
import { unwrap } from 'subforge/core'

const index = parseIdx(idx)
const doc = unwrap(parseVobSub(index, sub))

Serialization

import { toVobSub } from 'subforge/vobsub'

const { idx, sub } = toVobSub(doc)

Notes

• Full parsing requires both .idx and .sub.
• For large timelines, use decode: 'none' to keep parsing under the 20ms/100k target and defer image work.
• VobSub stores image data on event.image and metadata on event.vobsub for performance; segments can be empty.