perf: use sha1 instead of sha256 for hashing#13421
Conversation
|
I'm not very attached or convinced by this PR, but I don't also see how it can necessarily be harmful? I trust your judgement on the merge-ability of this. |
a496155 to
9007c31
Compare
|
From jshttp/etag#17 (comment), is sha1 gonna be denied soon? |
|
Hmm, that's fair, though I'm not aware of the approximate time frame before FIPS indeed excludes sha1. Of course, for our purposes, we aren't interested in the security properties of our hashing algorithm - though FIPS environments don't know that. Perhaps the alternate move is to expose this as a configuration option? Allow choosing between md5 and sha256? |
|
I'm fine with landing this. If it breaks FIPS, then we'll change in the future. Up until we can actually run tests in FIPS systems, regressions are bound to sneak in. But in this case it should be supported just fine (in som potential future it'll be unsupported, but no reason to deal with that now) |
|
CI is failing, tho 🙂 |
3be5c47 to
6f68fb1
Compare
This should be more performant while still being FIPS compliant (see jestjs#12722). sha1 isn't as secure as sha256, but since the usage context is just "has this file changed? 🤔", this should be an acceptable degredation. pu Signed-off-by: Mitchell Hentges <mhentges@spotify.com>
6f68fb1 to
c4d07a9
Compare
Signed-off-by: Mitchell Hentges <mhentges@spotify.com>
|
The only failure here seems to be the same ( |
- Here is the [PR](jestjs/jest#13421) - Update snapshots regarding the [snapshotFormat change](https://jestjs.io/blog/2022/08/25/jest-29)
* chore(Jest): upgrade jest form v28 to v29 for faster executions - Here is the [PR](jestjs/jest#13421) - Update snapshots regarding the [snapshotFormat change](https://jestjs.io/blog/2022/08/25/jest-29) * Update snapshots
|
This pull request has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs. |
This should be more performant while still being FIPS compliant (see #12722).
sha1 isn't as secure as sha256, but since the usage context is just "has this file changed? 🤔", this should be an acceptable degredation.