Compare a Lossless HAR Archive with a Cleaned Copy
Admin ·
Use compression to preserve a complete capture for later analysis. Use a separately labeled cleaned copy when the recipient only needs part of the HTTP record.
Compare the two exports from one practice capture
Choose Load practice HAR. The summary contains two entries. The POST request to /api/draft has the largest saved response body at 2,060 bytes.
Leave all removal options clear and prepare GZIP. The report says Original HAR bytes kept. Next select Remove response bodies and Remove cookies and choose HAR JSON. Save the smaller copy and check that POST /api/draft still has time 20.
| Check | Lossless GZIP | Cleaned HAR copy |
|---|---|---|
| Entries | 2 | 2 |
| POST response content text | Present | Removed |
| POST request cookie array | Contains practice cookie | Empty |
| POST time | 20 ms | 20 ms |
Open the HAR File CompressorHistorical HTTP Archive format draftRFC 1952 GZIP file format
Reopen the HAR from the selected container
Decompress .har.gz with a GZIP reader or open the single .har member in the ZIP. Compare its byte count with HAR inside on the result card.
Read log.entries in a JSON viewer or import the inner HAR in browser developer tools. Review the request URL, response status and timing fields you need for your comparison.
RFC 1952 GZIP file formatChrome DevTools network HAR reference
Inspect fields that remain before sharing
Request and response cookie arrays are separate from recorded HTTP headers. URLs, query strings and vendor fields can also hold sensitive text. Numeric values in fields you keep retain their original digits when the cleaned JSON is compacted. Review the actual exported file for the intended recipient.
Leave Reset timing values clear for performance analysis. If selected, the tool sets request timing numbers to zero while keeping capture timestamps.
Historical HTTP Archive format draftChrome DevTools network HAR reference
Common questions
Why can GZIP save space without removing a HAR body?
GZIP replaces repeats with a compressed representation. A compatible decoder recovers the complete input bytes.
Why can a compressed file be larger than its input?
Very small or already compact files can gain more from archive headers than they save in compressed data. The result card shows the measured difference.