| Summary: | [ews] use -o while unzipping layout-test-results.zip | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Aakash Jain <aakash_jain> | ||||
| Component: | Tools / Tests | Assignee: | Aakash Jain <aakash_jain> | ||||
| Status: | RESOLVED FIXED | ||||||
| Severity: | Normal | CC: | jbedard, lingcherd_ho, lingho, ryanhaddad, webkit-bug-importer | ||||
| Priority: | P2 | Keywords: | InRadar | ||||
| Version: | Other | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Bug Depends on: | |||||||
| Bug Blocks: | 175056 | ||||||
| Attachments: |
|
||||||
|
Description
Aakash Jain
2020-10-19 11:26:03 PDT
Created attachment 411767 [details]
Patch
Comment on attachment 411767 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=411767&action=review > Tools/BuildSlaveSupport/ews-build/steps.py:2740 > unzip says that -o overwrites files without prompting....how did we get away with this in the first place? In case of EWS, the file shouldn't exist previously. The uploaded file name is unique for each build. This is to eliminate the unecessary four lstat and stat calls everytime a file is getting extracted. But I haven't verified if -o indeed does that. Aakash have you? Committed r268689: <https://trac.webkit.org/changeset/268689> All reviewed patches have been landed. Closing bug and clearing flags on attachment 411767 [details]. Looks like adding -o won't help. Unzip is still calling lstat and stat (twice each) before creating a file. So don't expect any performance improvement. Discussed with Ling offline. As per him: I guess there is no harm leaving -o there. Might be the right thing to do because if for whatever reason the files were there before, we should overwrite. |