[Bug]: trace stop_chunk. the network in chunk includes the network data from the previous chunk

[bingosam]

Version

1.43.0

Steps to reproduce

with sync_playwright() as playwright:
    browser = playwright.chromium.launch(channel="chrome")
    context = browser.new_context()
    context.tracing.start(name='trace')

    context.tracing.start_chunk(name='trace', title='case1')
    page = context.new_page()
    page.goto('url of case 1')
    context.tracing.stop_chunk(path='case1.zip')

    context.tracing.start_chunk(name='trace', title='case2')
    page = context.new_page()
    page.goto('url of case 2')
    context.tracing.stop_chunk(path='case2.zip')

Expected behavior

The network data in case2.zip should not include the network data from case1.zip

Actual behavior

The network data in case2.zip include the network data from case1.zip.

Additional context

No response

Environment

- Operating System: [Windows 11]
- Browser: [Chrome]
- Python Version: [3.9.16]

[mxschmitt]

Was able to reproduce it with

import { chromium } from '@playwright/test';

(async () => {
  const browser = await chromium.launch();
  const context = await browser.newContext();
  await context.tracing.start({ name: 'trace', snapshots: true });

  await context.tracing.startChunk();
  let page = await context.newPage();
  await page.goto('https://www.microsoft.com/pt-pt/');
  await context.tracing.stopChunk({ path: 'case1.zip' });

  await context.tracing.startChunk();
  page = await context.newPage();
  await page.goto('https://example.com');
  await context.tracing.stopChunk({ path: 'case2.zip' });

  await browser.close();
})();

Looks like this is working as intended:

playwright/packages/playwright-core/src/server/trace/recorder/tracing.ts

Line 287 in 3b7c4fa

// Network file survives across chunks, make a snapshot before returning the resulting entries.

Moving upstream.