-
-
Notifications
You must be signed in to change notification settings - Fork 13.6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
sasquatch: 4.4 -> 4.5.1-1 #219169
base: master
Are you sure you want to change the base?
sasquatch: 4.4 -> 4.5.1-1 #219169
Conversation
Ah, figured keeping the existing test doesn't make sense as it will just use squashfs-tools anyway. |
Any news, any blockage ? Or any review from @Pamplemousse if he is still the right contact for this PR ? |
Some questions got answerred: I shouldn't depend on squashFsTools at all, as it got updated to 4.6.1, and it is unlikely I want to rebase sasquatch the near future. |
As |
@AkechiShiro Sorry I missed your message earlier. I have stopped using this package in years... I am not well placed to test. |
Asked for their removal in NixOS#219169 (comment)
Done |
This PR has been stuck for the last 6 months waiting for a review from someone who explicitly requested to be removed from the maintainers list. Who should review this then ? How can we make it move forward ? Asking cause it's the last item missing in out meta-issue to get unblob in nixpkgs #217836 |
This PR needs a rebase, first of all on master @vlaci could you please rebase? |
Asked for their removal in NixOS#219169 (comment)
Asked for their removal in NixOS#219169 (comment)
Happy new year everyone :) |
Thanks @vlaci, there seems to be a typo in the meta attribute of the package, please fix it, so that the ofborg evaluation test passes, see more details : https://gist.github.com/GrahamcOfBorg/6d11645476eaaef9b7fc87c573d4a188, I cannot help further as I'm inexperienced in reviews @qkaiser, maybe mentioning the pull request as ready for review here, should bring this PR to more experienced reviewers : https://discourse.nixos.org/t/prs-ready-for-review and hopefully help. |
Asked for their removal in NixOS#219169 (comment)
@AkechiShiro thanks for the heads up, I sent a reply to the Discourse thread asking for a review on this PR |
This pull request has been mentioned on NixOS Discourse. There might be relevant details there: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I ran nixpkgs-review and sucessfully used both x86_64 and aarch64 to extract an example squashfs I found somewhere on my disk.
For me the diffoscope tests are broken on master on x86_64 even without this patch.
Result of nixpkgs-review pr 219169
run on x86_64-linux 1
3 packages failed to build:
- diffoscope
- diffoscope.dist
- diffoscope.man
10 packages built:
- binwalk (python311Packages.binwalk)
- binwalk.dist (python311Packages.binwalk.dist)
- python310Packages.binwalk
- python310Packages.binwalk-full
- python310Packages.binwalk-full.dist
- python310Packages.binwalk.dist
- python311Packages.binwalk-full
- python311Packages.binwalk-full.dist
- sasquatch
- sasquatch-v4be
Result of nixpkgs-review pr 219169
run on aarch64-linux 1
13 packages built:
- binwalk (python311Packages.binwalk)
- binwalk.dist (python311Packages.binwalk.dist)
- diffoscope
- diffoscope.dist
- diffoscope.man
- python310Packages.binwalk
- python310Packages.binwalk-full
- python310Packages.binwalk-full.dist
- python310Packages.binwalk.dist
- python311Packages.binwalk-full
- python311Packages.binwalk-full.dist
- sasquatch
- sasquatch-v4be
What does the bigendian variant do? Do we need both? |
pname = "sasquatch"; | ||
version = "4.4"; | ||
drv = stdenv.mkDerivation rec { | ||
pname = "sasquatch${lib.optionalString bigEndian "-v4be"}"; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This repeats the version number (v4), that's already appended to pname, so the full name would be sasquatch-v4be-4.5.1-4
. Perhaps it should just be sasquatch-be
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The v4be
indicates the squashfs filesystem version while the 4.5.1-4
indicates the squashfs-tools version sasquatch is based on. These versions are unrelated.
sasquatch is built on top of squashfs-tools to support non standard squashfs filesystems. By the standard, squashfs version 4.0 filesystems are fixed little-endian. There are no big-endian 4.0 filesystems. However, embedded vendors love to mess with standards. So instead of adapting squashfs-tools code, we build a version with More information about the big-endian variant can be found at onekey-sec/sasquatch@0c65f00 Here's a small preview of what vendors do with squashfs. Cells with a star (*) are standard.
|
What's the status of this? Can it be merged? |
Hello @smancill I think there was a review that approved the package, I guess running I don't know however if a rebase is needed to make sure that the current package still build with the latest upstream in Nixpkgs. |
@vlaci there might be a dead link as a source for the current version : I cannot find the repo on GitHub, maybe there is new unblob that is being worked on that doesn't depend on I only found this repo : https://github.com/onekey-sec/sasquatch EDIT :
|
It must have been a github api error, as the source is there... |
4832e2d
to
f8b0cee
Compare
Asked for their removal in NixOS#219169 (comment)
IDK if I should move these from |
I believe there is a typo in the package @vlaci error: evaluation aborted with the following error message: 'Failed to evaluate sasquatch-4.5.1-4: «unknown-meta»: has an invalid meta attrset:
- key 'meta.mainprogram' is unrecognized; expected one of:
['available', 'badPlatforms', 'branch', 'broken', 'changelog', 'description', 'downloadPage', 'executables', 'homepage', 'hydraPlatforms', 'insecure', 'isBuildPythonPackage', 'isFcitxEngine', 'isGutenprint', 'isHydraChannel', 'isIbusEngine', 'knownVulnerabilities', 'license', 'longDescription', 'mainProgram', 'maintainers', 'maxSilent', 'name', 'outputsToInstall', 'pkgConfigModules', 'platforms', 'position', 'priority', 'schedulingPriority', 'sourceProvenance', 'tag', 'tests', 'timeout', 'unfree', 'unsupported', 'version']
'
|
f1438af
to
b5f6276
Compare
Asked for their removal in NixOS#219169 (comment)
Thanks, this error did not emerge on my Nix version. |
Tested a build ( Output of
I ran the build manually. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Just waiting for sasquatch-v4be
to be moved from top-level/all-packages.nix
into by-name/sa/sasquatch-v4be
, then I will approve this PR on my side, I can't really test however on nix darwin
b5f6276
to
1a2f7f1
Compare
NOTE: as of now, it requires overriding nixpkgs to NixOS/nixpkgs#219169
I've created a draft PR for unblob pointing to this branch, so its tests will run for darwin as well: https://github.com/onekey-sec/unblob/pull/946/checks |
Changed using ONEKEY fork of sasquatch which is up-to-date with squashfs-tools 4.5.1 and the Darwin compatibility patch used to be shipped in Nixpkgs applies cleanly as well. The updated version also provides a big-endian compatible build as well to support even more insane squashfs dialects.
Asked for their removal in NixOS#219169 (comment)
1a2f7f1
to
465008e
Compare
I don't think the
|
This pull request has been mentioned on NixOS Discourse. There might be relevant details there: https://discourse.nixos.org/t/prs-ready-for-review/3032/4576 |
Required for #217836
Description of changes
Changed using ONEKEY fork of sasquatch which is up-to-date with squashfs-tools 4.5.1 and the Darwin compatibility patch shipped in Nixpkgs applies cleanly as well.
I have chosen to override the
squashfsTools
derivation, as this fork is entirely compatible with that version. I am open to critique on this. I wasn't sure copying from squashfstools or lifting ageneric.nix
base derivation out of it is the better solution. Existing NixOS test is also nice to have as a validation, although it may be overkill here.Things done
sandbox = true
set innix.conf
? (See Nix manual)nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD"
. Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/
)