sub-merge.ts

#!/usr/bin/env -S bun run

import fs from 'node:fs';

const indexRegexp = new RegExp(/^\d+$/);
const timeRegexp =  new RegExp(/^(\d\d:\d\d:\d\d,\d\d\d) --> (\d\d:\d\d:\d\d,\d\d\d)$/);

const file_a = Bun.argv[2];
const file_b = Bun.argv[3];

interface SubEntry {
	begin: string;
	end: string;
	text: string;
}

function parseSub (file: string): SubEntry[] {
	const out: SubEntry[] = [];
	let begin: string = '';
	let end: string = '';
	let text: string[] = [];
	const input = fs.readFileSync(file).toString().replace(/\r\n/g, '\n').split('\n');
	for (let i = 0; i < input.length; i++) {
		if (indexRegexp.test(input[i]) && timeRegexp.test(input[i+1])) {
			const item = {
				begin,
				end,
				text: text.join('\n'),
			};
			if (text.length > 0) {
				out.push(item);
			}
			i += 1;
			let matches = input[i].match(timeRegexp);
			if (matches == null) {
				throw "Woops, that's not supposed to happen";
			} else {
				begin = matches[1];
				end = matches[2];
				text = [];
			}
		} else {
			text.push(input[i]);
		}
	}
	if (text.length > 0) {
		out.push({
			begin,
			end,
			text: text.join('\n'),
		});
	}
	return out;
}

function LT (a: string, b: string): boolean {
	a = a.replace(/\D/g, '');
	b = b.replace(/\D/g, '');
	return (+a < +b);
}

function LTE (a: string, b: string): boolean {
	a = a.replace(/\D/g, '');
	b = b.replace(/\D/g, '');
	return (+a <= +b);
}

function duration (begin: string, end: string): number {
	// in ms
	begin = begin.replace(/\D/g, '');
	end = end.replace(/\D/g, '');
	return (+end - +begin);
}

const print = (() => {
	let outIndex: number = 1;
	const minTime = 250;
	return function (e: SubEntry): void {
		if (duration(e.begin, e.end) < minTime) {
			// too short
			return;
		}
		console.log(outIndex++);
		console.log(`${e.begin} --> ${e.end}`);
		console.log(e.text.replace(/\n\n+/g, "\n\n").trim() + "\n");
	}
})();

function merge(a: SubEntry[], b: SubEntry[]): void {
	let ai = 0, bi = 0;
	let time: string = "0";
	const amax = a.length - 1;
	const bmax = b.length - 1;
	while (1) {
		if (ai > amax && bi > bmax) {
			// done
			break;
		}
		if (ai > amax && bi <= bmax) {
			print (b[bi]);
			bi++;
			continue;
		}
		if (bi > bmax && ai <= amax) {
			print (a[ai]);
			ai++;
			continue;
		}

		if (LT(a[ai].begin, b[bi].begin)) {
			// a begins first
			if (LTE(a[ai].end, time)) {
				// already ended
				ai++;
				continue;
			}
			if (LT(a[ai].begin, time)) {
				// truncate begin time
				a[ai].begin = time;
			}

			if (LT(a[ai].end, b[bi].begin)) {
				// a ends before b begins
				print (a[ai]);
				ai++;
				continue;
			} else {
				// b begins before a ends, truncate a and keep it
				print ({
					...a[ai],
					end: b[bi].begin,
				});
				print (b[bi]);
				time = b[bi].end;
				bi++;
				continue;
			}
		} else {
			// b begins first, so takes priority
			print (b[bi]);
			time = b[bi].end;
			bi++;
			continue;
		}
	}
}

const subs_a = parseSub(file_a);
const subs_b = parseSub(file_b);

merge(subs_a, subs_b);