1c7 / srt-parser-2 Goto Github PK

While working on a bug, I had some suggestions I'd like to implement in this lib.

• Replace mocha / chai with vitest. Vitest is a modern test runner powered by Vite. It runs TS easily.
• Replace all var statements with let/const
• Add code documentation (JSDocs) (to see comments if you hover the function)

Are these changes of any interest? If so, I'd be happy to help and open PRs from time to time implementing them.

I use this awesome lib in a personal project. I have only used intalled in the project, today I was trying to use the CLI version and noticed it was not working.

Command I ran:

npx srt-parser-2 -i input.srt -o output.json --minify

Output:

(node:9952) Warning: To load an ES module, set "type": "module" in the package.json or use the .mjs extension.
(Use `node --trace-warnings ...` to show where the warning was created)
C:\Users\ppedr\AppData\Local\npm-cache\_npx\b731975f552886f1\node_modules\srt-parser-2\bin\index.js:4
import { readFile, writeFile } from "fs/promises";
^^^^^^

SyntaxError: Cannot use import statement outside a module
    at internalCompileFunction (node:internal/vm:73:18)
    at wrapSafe (node:internal/modules/cjs/loader:1176:20)
    at Module._compile (node:internal/modules/cjs/loader:1218:27)
    at Module._extensions..js (node:internal/modules/cjs/loader:1308:10)
    at Module.load (node:internal/modules/cjs/loader:1117:32)
    at Module._load (node:internal/modules/cjs/loader:958:12)
    at Function.executeUserEntryPoint [as runMain] (node:internal/modules/run_main:81:12)
    at node:internal/main/run_main_module:23:47

Machine information:
OS: Windows 10
Node: v18.16.0

I'm trying to fix this error and will open a PR if a manage to do it.

I have an SRT file I wanted to quickly parse and output as a JSON.
I was hoping for something like this without creating a new project

npm install -g srt-parser-2
srt-parser-2 -i input.srt -o output.json

It's possible to do this with a bin entry in package.json pointing to a node script.
https://docs.npmjs.com/cli/v7/configuring-npm/package-json#bin

Is this a planned feature or would you be interested in adding it to the package? I can try make a PR for it if needed

Please could you update srt-parser-2 to make it compatible with ES6-pure implementations which would use
import srt from 'srt-parser-2' to import the parser.

can use this:

const srtTimestampToSeconds = (srtTimestamp) => {
  const [rest, millisecondsString] = srtTimestamp.split(",");
  const milliseconds = parseInt(millisecondsString);
  const [hours, minutes, seconds] = rest.split(":").map(x => parseInt(x));
  return milliseconds * 0.001 + seconds + 60 * minutes + 3600 * hours;
};

Thanks for the nice library. It'd be great if you could include type definitions.

As a work around, I'm creating a file in my project called @types/srt-parser-2.d.ts with the following code in it.

export interface Line {
  id: string;
  startTime: string;
  endTime: string;
  text: string;
}
export default class Parser {
  seperator: string;
  correctFormat(time: string): string;
  private fixed_str_digit;
  private tryComma;
  private tryDot;
  fromSrt(data: string): {
      id: string;
      startTime: string;
      endTime: string;
      text: string;
  }[];
  toSrt(data: Array<Line>): string;
}

I can't seem to get a result for chinese words

Hi,

I would love to use your library but I noticed some odd javascript rounding behaviour.

In this function

const timestampToSeconds = (srtTimestamp: string) => {
  const [rest, millisecondsString] = srtTimestamp.split(",");
  const milliseconds = parseInt(millisecondsString);
  const [hours, minutes, seconds] = rest.split(":").map((x) => parseInt(x));
  return milliseconds * 0.001 + seconds + 60 * minutes + 3600 * hours;
};

When using the timestamp:

timestampToSeconds('00:01:20,460');

returns: 80.46000000000001

Actual calc is 460 * 0.001 + 20 + 60, which returns 80.46000000000001 in JS

I guess this is a JS perk. But can be overcome by doing some extra rouding in this function.
e.g. Math.round(80.46000000000001 * 1000) / 1000

Another example is this timestamp: 00:01:32,020

Issue #13 asked for implementing parsing timestamps in seconds, but there is an issue with that: the documentation shows an example of that, but do not explain which of startTime or startSeconds are used when converting back to a string.

One simple improvement is fixing the documentation by making this behavior clear.

Other complex improvement is also allowing to choose between one and other (maybe something like parser.toSrt(srt_array, {useSeconds: true})) and/or exposing a secondsToTimestamp method.