# Help regenerating lost summary\_requests data

**URL:** <https://discuss.httparchive.org/t/help-regenerating-lost-summary-requests-data/1616>\
**Category:** Analysis\
**Created:** [March 28, 2019, 4:58pm UTC](https://discuss.httparchive.org/t/help-regenerating-lost-summary-requests-data/1616 "2019-03-28T16:58:39Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![rviscomi](https://yyz1.discourse-cdn.com/flex035/user_avatar/discuss.httparchive.org/rviscomi/32/835_2.png) [@rviscomi](https://discuss.httparchive.org/u/rviscomi)\
**Post date:** [March 28, 2019, 4:58pm UTC](https://discuss.httparchive.org/t/help-regenerating-lost-summary-requests-data/1616/1 "2019-03-28T16:58:39Z")

</div>

We had a [data loss issue](https://github.com/HTTPArchive/httparchive.org/issues/135) with the March desktop requests. The good news is that the HAR-based JSON data is all there, but our processed summary data didn’t make it. So I have a challenge for anyone looking to sharpen their SQL skills.

As part of our goal to deprecate the legacy PHP/mysql system, we need to come up with a way to summarize the HAR-based JSON data into a format identical (or similar) to the mysql-based CSV data, eg extracting high-level request and response headers, HTTP status, etc for the summary\_requests tables. Most if not all of the fields in the summary CSV-based tables should be included in the HAR data.

So: given [this sample data](https://bigquery.cloud.google.com/table/httparchive:scratchspace.sample_requests_2019_03_01_desktop?pli=1&tab=preview) could you produce results that fit [the schema of the summary\_requests tables](https://bigquery.cloud.google.com/table/httparchive:summary_requests.2019_03_01_desktop?pli=1&tab=schema)?

For more info see [this issue](https://github.com/HTTPArchive/httparchive.org/issues/23).