mirror of
https://github.com/mauriceboe/TREK.git
synced 2026-10-19 13:36:48 +00:00
Maurice
4ceade9d8b
Cloud models got the first 4,000 characters, so totals and lodging sections further down never reached them, without a word. They now get 24,000 with the start and the end kept and a warning when the middle is left out; a server whose context is too small is asked once more with 8,000. A total the model repeated on every flight of one booking stays on the first flight only, when the document prints it once as a total, and the prompt says so too.
46 lines
2.0 KiB
TypeScript
46 lines
2.0 KiB
TypeScript
import {
|
|
CLOUD_TEXT_WINDOW,
|
|
fitTextWindow,
|
|
LOCAL_FLIGHT_TEXT_WINDOW,
|
|
LOCAL_TEXT_WINDOW,
|
|
} from '../../../../src/nest/llm-parse/text-window';
|
|
|
|
import { describe, it, expect } from 'vitest';
|
|
|
|
describe('fitTextWindow', () => {
|
|
it('leaves a text that fits as it is', () => {
|
|
expect(fitTextWindow('short', 10)).toEqual({ text: 'short', omitted: 0 });
|
|
expect(fitTextWindow('x'.repeat(10), 10)).toEqual({ text: 'x'.repeat(10), omitted: 0 });
|
|
});
|
|
|
|
it('keeps three quarters from the start and a quarter from the end around a marker', () => {
|
|
const text = 'a'.repeat(5000) + 'b'.repeat(5000);
|
|
const { text: out, omitted } = fitTextWindow(text, 4000);
|
|
expect(omitted).toBe(6000);
|
|
expect(out).toBe(`${'a'.repeat(3000)}\n[... 6000 characters omitted ...]\n${'b'.repeat(1000)}`);
|
|
});
|
|
|
|
it('cuts on a nearby line break, so a line at either cut stays whole', () => {
|
|
const lines = Array.from({ length: 400 }, (_, i) => `Line ${String(i).padStart(3, '0')}: Total paid USD 200.00`);
|
|
const text = lines.join('\n');
|
|
const { text: out, omitted } = fitTextWindow(text, 4000);
|
|
const [head, tail] = out.split(/\n\[\.\.\. \d+ characters omitted \.\.\.\]\n/);
|
|
for (const part of [head, tail]) for (const line of part.split('\n')) expect(lines).toContain(line);
|
|
expect(head.length + tail.length).toBeLessThanOrEqual(4000);
|
|
expect(head.length + tail.length + omitted).toBe(text.length);
|
|
expect(tail.endsWith(lines[lines.length - 1])).toBe(true);
|
|
});
|
|
|
|
it('cuts inside a line when no line break is close', () => {
|
|
const text = `${'a'.repeat(2000)}\n${'b'.repeat(8000)}`;
|
|
const { text: out } = fitTextWindow(text, 4000);
|
|
expect(out.startsWith(`${'a'.repeat(2000)}\n${'b'.repeat(999)}\n[...`)).toBe(true);
|
|
});
|
|
|
|
it('gives a cloud model a far wider window than a local one, and keeps the local ones', () => {
|
|
expect(CLOUD_TEXT_WINDOW).toBe(24_000);
|
|
expect(LOCAL_FLIGHT_TEXT_WINDOW).toBe(16_000);
|
|
expect(LOCAL_TEXT_WINDOW).toBe(6_000);
|
|
});
|
|
});
|