Skip to main content

Function: serializeHeapSnapshot()

serializeHeapSnapshot(snapshot, outputFile): void

Defined in: core/src/lib/HeapSerializer.ts:346

Write a parsed (and possibly modified) heap snapshot to a .heapsnapshot file that Chrome DevTools, memlab and any other V8 snapshot reader can open.

This is the inverse of parsing, and it makes "load a snapshot, change it, save it" a supported workflow — anonymizeHeapSnapshot is one such consumer.

Edits made through the snapshot's underlying data are picked up automatically: node names are read out of the string table on every access rather than cached, so a rewritten string table is already what the in-memory snapshot reports.

The file is streamed out through a single fixed buffer, one array element at a time. That is a hard requirement rather than an optimization: V8 caps a string at 512 MB (buffer.constants.MAX_STRING_LENGTH), so building the document — or even one of its larger arrays — as a string throws RangeError: Invalid string length on a large capture instead of merely running slowly.

Parameters​

snapshot​

IHeapSnapshot

the heap snapshot to write, as returned by getFullHeapFromFile or takeNodeMinimalHeap

outputFile​

string

absolute path of the file to create; an existing file at that path is overwritten

Returns​

void

this API returns void; the snapshot is written to outputFile

  • Examples:
import type {IHeapSnapshot} from '@memlab/core';
import {dumpNodeHeapSnapshot, serializeHeapSnapshot} from '@memlab/core';
import {getFullHeapFromFile} from '@memlab/heap-analysis';

(async function () {
  const file = dumpNodeHeapSnapshot();
  const heap: IHeapSnapshot = await getFullHeapFromFile(file);

  // ... modify the heap snapshot in memory ...

  serializeHeapSnapshot(heap, '/tmp/modified.heapsnapshot');
})();