Omaha crate improvements by james-parky · Pull Request #88 · flatcar/ue-rs

james-parky · 2025-09-12T20:20:26Z

General Improvements to the `omaha` Crate

The following is a slightly ammended collection of the commit bodies from the constituent commits of this PR. They serve not only as a description of all changes, but also (if read in written order) may serve to explain the justification behind the changes.

Hashing Algorithms

Rename HashAlgo trait to Hasher since that is what structs implementing it do.
Rework of core Hash struct and Hasher trait. Rather than provide Hasher (prev. HashAlgo) impl structs to users that can then use them in conjunction with the sha1/sha2 etc. hasher that they are meant to represent, allow the structs implementing Hasher (Sha1 and Sha2) to perform hashing logic themselves. Therefore add three new trait methods: new(), update(), and finalize(). Other methods can be added later if required, but for now this is sufficient.
Remove Hasher::from_boxed(). There is no reason for it to exist if you encapsulate the hashing logic (update and finalize) into the trait itself.
Add associated trait method Hasher::try_from_hex_string() for use in the pre-existing sha256_hex mod (for use in hard_xml reading). Note I rename this module to sha256_from_str since that it literally all it is for, and the word "hex" isn't used consistently across the module.
Add new sha1_from_str module that mimics the sha256 one. Move this and the other one to hash_types.rs to be defined together with the types they use.
Rather than having a Hash struct that has some methods, just define two type aliases Sha1Digest and Sha256Digest when the output of these hashing algorithms is needed (where you could otherwise use H::Output where H is a struct implementing Hasher.
Declare that Hasher::Output must implement PartialEq so that we can compare hash outputs (and therefor compare the above mentioned ShaXXXDigest types).
Simplify download.rs:hash_on_disk() with the fact that the trait implementations encapsulate their own hashing logic.
Temporarily switch to printing all instances of a Hasher::Output to a Debug print.
Introduce associated const FINGERPRINT_SIZE to Hasher trait and comment about that there is a better way to potentially represent the output type of these structs in the future if a particular RFC stabilises.
Doc strings for contained structs, traits and functions.
Add various unit tests.

Requests and Responses

Remove #[rustfmt::skip]s in response.rs. rustfmt almost always produces better formatting than doing it manually.
Introduce preliminary custom Error and Result types -- planning on removing reliance on anyhow crate just for contexts. We can create custom error strings using a crate-specific Error enum. This also removes the need for the ct-codecs dependency.
Remove now unused dependencies anyhow and digest.
Remove Uuid type and provide unit tests for xml read/write. This was documenting as existing to provide serialisation of the underlying inner value of type uuid::Uuid with surrounding braces. Whenever uuids are used throughout the codebase I image they are used just as uuid::Uuids, therefore we don't need a type just for (de)seriailsation purposes -- just provide a with = attribute for XmlRead and XmlWrite impls.
Also provide here preliminary unit tests for reading/writing some omaha structs using the base uuid::Uuid type to prove this is sufficient.
Introduce util function for creating XmlError values.
Reduce code by making better use of derive(XmlRead). There were some structs that had custom implementation of XmlRead when it could have been derived, so derive instead. Since we cannot derive or manually implement XmlRead for url::Url, we must create a newtype Urls(Vec<Url>) and implement XmlRead for that (based off the original code).
Write unit tests for XmlRead-ing new Urls struct.
Write some doc strings and remove now redundant comment.
Write unit tests for XmlRead-ing Package struct with a generic helper function. This also requires going through and deriving PartialEq on the response field structs. This is done behind a cfg_attr(test, ) though since there is currently no business-logic requirement for it to be derived fully.
Re-order struct defs in a top-down way to make generated docs read better when they are written.

Other

Remove type.rs and therefore the FileSize struct. It doesn't appear to have a need to exist; using a FileSize in places that could just be a usize doesn't provide any extra type-safety, and doesn't really provide any extra semantic meaning to a dev. Wherever it was used it was just being converted back and forth from a usize anyway. Furthermore, there is no special (de)serialisation logic that is required for it -- it is also just a usize in the XML.

Notes

Further additions/changes e.g. docstrings and more structs in the request.rs and response.rs files will require conversation with someone more familiar with ue, but happy to do the work.

dongsupark

First of all, thank you for huge effort you put into this PR.

I like the approach in general. I am still in middle of reading, but I wrote several comments below. Will read the rest tomorrow.

The thing is, if a PR has over 10 commits that have own different purposes, it becomes hard for reviewers to look into each commit, and follow what you want to do.

In general, I would recommend creating multiple PRs that have own purposes. For example one PR for trivial changes(renaming variables/functions, reordering, cleaning up), another PR for dealing with improving xml parts, and another PR for rewriting hasher structure like the beginning of this PR. Then you could get the easiest part merged immediately, and after that rebase harder PRs on top of that, and go through review cycles etc. That would be really helpful for reviewers.

omaha/src/hash_types.rs

james-parky · 2025-09-16T17:50:06Z

The thing is, if a PR has over 10 commits that have own different purposes, it becomes hard for reviewers to look into each commit, and follow what you want to do.

Yes fair enough. I know that this PR seems almost random and I suppose it kind of is. I was just changing things as I thought of them, without small initial goals for refactoring -- so things ended up order of a more logical order. Will keep to smaller PRs in the future though.

Since that is what structs implementing it do. This will make more sense with further commits. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

- Rather than provide Hashers (prev. HashAlgo) impl structs to users that can then use them in conjunction with the sha1/sha2 etc. hasher that they are meant to represent, allow the structs implementing Hasher (Sha1 and Sha2) to perform hashing logic themselves. Therefore add three new trait methods: new(), update(), and finalize(). - Remove Hasher::from_boxed(). There is no reason for it to exist if you encapsulate the hashing logic (update and finalize) into the trait itself. - Add associated trait method Hasher::try_from_hex_string() for use in the pre-existing sha256_hex mod (for use in hard_xml Reading). Note I rename this module to sha256_from_str since that it literally all it is for, and the word "hex" isn't used consistently across the module. - Add new sha1_from_str module that mimics the sha256 one. - Rather than having a Hash struct that has some methods, just define two type aliases Sha1Digest and Sha256Digest when the output of these hashing algorithms is needed (where you could otherwise use H::Output where H is a struct implementing Hasher. - Declare that Hasher::Output must implement PartialEq so that we can compare hash outputs (and therefor compare the above mentioned ShaXXXDigest types). - Simplify download.rs:hash_on_disk() with the fact that the trait implementations encapsulate their own hashing logic. - Temporarily switch to printing all instances of a Hasher::Output to a Debug print. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

And comment about that there is a better way to potentially do this in the future if a particular RFC stabilises. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

rustfmt almost always produces better formatting than doing it manually. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

Planning on removing reliance on anyhow crate just for contexts. We can create custom error strings using a crate-specific Error enum. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

anyhow and digest Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

This was documenting as existing to provide serialisation of the underlying inner value of type uuid::Uuid with surrounding braces. Whenever uuids are used throughout the codebase I image they are used just as uuid::Uuids, therefore we don't need a type just for (de)seriailsation purposes -- just provide a `with = ` attribute for XmlRead and XmlWrite. Also provide here preliminary unit tests for reading/writing some omaha structs using the base uuid::Uuid type to prove this is sufficient. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

To be logically close to the types they are using. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

There were some structs that had custom implementation of XmlRead when it could have been derived, so derive instead. Since we cannot derive or manually implement XmlRead for url::Url, we must create a newtype Urls(Vec<Url>) and implement XmlRead for that (using the original code). Note that I am fairly sure that this doesn't change behaviour but I am yet to write unit tests. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

This also requires going through and deriving PartialEq on the response field structs. This is done behind a cfg_attr(test, ) though since there is currently no business-logic requirement for it to be derived fully. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

In a top-down way to make generated docs read better when they are written. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

It doesn't appear to have a need to exist; using a FileSize in places that could just be a usize doesn't provide any extra type-safety, and doesn't really provide any extra semantic meaning to a dev. Wherever it was used it was just being converted back and forth from a usize anyway. Furthermore, there is no special (de)serialisation logic that is required for it -- it is also just a usize in the XML. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

PR flatcar#87 created the download crate which caused merge conflicts with flatcar#88.

PR flatcar#87 created the download crate which caused merge conflicts with flatcar#88. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

james-parky · 2025-09-18T20:19:45Z

Just rebased this onto new main branch after #87 was merged.

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

dongsupark

Our CI fails at clippy warnings by default. Can you please address them?

error: using `clone` on type `Option<[u8; 32]>` which implements the `Copy` trait
   --> src/download/package.rs:104:13
    |
104 |             self.hash_sha256.clone(),
    |             ^^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `self.hash_sha256`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
    = note: `-D clippy::clone-on-copy` implied by `-D warnings`
    = help: to override `-D warnings` add `#[allow(clippy::clone_on_copy)]`
error: using `clone` on type `Option<[u8; 20]>` which implements the `Copy` trait
   --> src/download/package.rs:105:13
    |
105 |             self.hash_sha1.clone(),
    |             ^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `self.hash_sha1`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `[u8; 32]` which implements the `Copy` trait
   --> src/download/package.rs:122:70
    |
122 |         debug!("    sha256 match?      {}", self.hash_sha256 == Some(calculated_sha256.clone()));
    |                                                                      ^^^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `calculated_sha256`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `[u8; 20]` which implements the `Copy` trait
   --> src/download/package.rs:125:66
    |
125 |         debug!("    sha1 match?      {}", self.hash_sha1 == Some(calculated_sha1.clone()));
    |                                                                  ^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `calculated_sha1`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `[u8; 32]` which implements the `Copy` trait
   --> src/download/package.rs:127:67
    |
127 | ... self.hash_sha256 != Some(calculated_sha256.clone()) || self.hash_sha1.is_some() && self.hash_sha1 != Some(calculated_sha1.clone()) {
    |                              ^^^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `calculated_sha256`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `[u8; 20]` which implements the `Copy` trait
   --> src/download/package.rs:127:148
    |
127 | ...) && self.hash_sha1 != Some(calculated_sha1.clone()) {
    |                                ^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `calculated_sha1`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `[u8; 32]` which implements the `Copy` trait
   --> src/download/package.rs:155:34
    |
155 |         let hdhashvec: Vec<u8> = hdhash.clone().into();
    |                                  ^^^^^^^^^^^^^^ help: try removing the `clone` call: `hdhash`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `[u8; 32]` which implements the `Copy` trait
   --> src/download/mod.rs:117:65
    |
117 |     debug!("    sha256 match?      {}", expected_sha256 == Some(calculated_sha256.clone()));
    |                                                                 ^^^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `calculated_sha256`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `[u8; 20]` which implements the `Copy` trait
   --> src/download/mod.rs:120:61
    |
120 |     debug!("    sha1 match?      {}", expected_sha1 == Some(calculated_sha1.clone()));
    |                                                             ^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `calculated_sha1`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `[u8; 32]` which implements the `Copy` trait
   --> src/download/mod.rs:122:61
    |
122 |     if expected_sha256.is_some() && expected_sha256 != Some(calculated_sha256.clone()) {
    |                                                             ^^^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `calculated_sha256`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `[u8; 20]` which implements the `Copy` trait
   --> src/download/mod.rs:125:57
    |
125 |     if expected_sha1.is_some() && expected_sha1 != Some(calculated_sha1.clone()) {
    |                                                         ^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `calculated_sha1`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `Option<[u8; 32]>` which implements the `Copy` trait
   --> src/download/mod.rs:142:60
    |
142 |         || do_download_and_hash(client, url.clone(), path, expected_sha256.clone(), expected_sha1.clone()),
    |                                                            ^^^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `expected_sha256`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy
error: using `clone` on type `Option<[u8; 20]>` which implements the `Copy` trait
   --> src/download/mod.rs:142:85
    |
142 |         || do_download_and_hash(client, url.clone(), path, expected_sha256.clone(), expected_sha1.clone()),
    |                                                                                     ^^^^^^^^^^^^^^^^^^^^^ help: try removing the `clone` call: `expected_sha1`
    |
    = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#clone_on_copy

Pulls in flatcar/ue-rs#72, flatcar/ue-rs#74, flatcar/ue-rs#84, flatcar/ue-rs#85, flatcar/ue-rs#87, flatcar/ue-rs#88. Signed-off-by: Dongsu Park <dongsu@dpark.io>

Pulls in flatcar/ue-rs#72, flatcar/ue-rs#84, flatcar/ue-rs#85, flatcar/ue-rs#87, flatcar/ue-rs#88. Signed-off-by: Dongsu Park <dongsu@dpark.io>

Pulls in flatcar/ue-rs#72, flatcar/ue-rs#84, flatcar/ue-rs#85, flatcar/ue-rs#87, flatcar/ue-rs#88, flatcar/ue-rs#90. Signed-off-by: Dongsu Park <dongsu@dpark.io>

dongsupark

Looks good. Thanks.
CI of flatcar/scripts#3310 passed, which means no regression with this PR.
Once you address clippy issues, I will merge it.

james-parky · 2025-09-29T17:17:15Z

Once you address clippy issues, I will merge it.

Yes I will do that later today, thanks

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

james-parky · 2025-09-29T20:53:40Z

@dongsupark fixed clippy lint warnings

Pulls in flatcar/ue-rs#72, flatcar/ue-rs#84, flatcar/ue-rs#85, flatcar/ue-rs#87, flatcar/ue-rs#88, flatcar/ue-rs#90. Signed-off-by: Dongsu Park <dongsu@dpark.io>

PR #87 created the download crate which caused merge conflicts with #88. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

james-parky requested a review from a team as a code owner September 12, 2025 20:20

james-parky force-pushed the dev-omaha-improvements branch from 6e9da5d to fd49c6b Compare September 12, 2025 20:24

james-parky mentioned this pull request Sep 12, 2025

Split download verify parts of download_sysext into separate module #87

Merged

james-parky changed the title ~~WIP: Omaha crate improvements~~ Omaha crate improvements Sep 12, 2025

dongsupark reviewed Sep 16, 2025

View reviewed changes

omaha/src/hash_types.rs Outdated Show resolved Hide resolved

omaha/src/hash_types.rs Outdated Show resolved Hide resolved

james-parky added 24 commits September 18, 2025 21:03

omaha: rename HashAlgo trait to Hasher

a4d366e

Since that is what structs implementing it do. This will make more sense with further commits. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: introduce associated const to Hasher trait

8a6e4a7

And comment about that there is a better way to potentially do this in the future if a particular RFC stabilises. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: genericise and extract Hasher::try_from_hex_string() into own fn

7fb7b53

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: doc strings for contained structs, traits and functions

bda229a

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: various unit tests

2f15dcf

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: response: remove #[rustfmt::skip]s

90a441a

rustfmt almost always produces better formatting than doing it manually. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: preliminary custom Error and Result types

8a3d031

Planning on removing reliance on anyhow crate just for contexts. We can create custom error strings using a crate-specific Error enum. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: remove now unused dependencies

53bd9e8

anyhow and digest Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: move shax_from_str modules to hash_types.rs

216768b

To be logically close to the types they are using. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: response: introduce util function for creating XmlError values

05f6a50

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: response: unit tests for XmlRead-ing new Urls struct

90b3b39

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: response: refine Urls XmlRead implementation

27e6497

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: response: doc strings

f6a6b73

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: remove now redundant comment

8092ca1

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: response: unit tests for XmlRead-ing Package struct

6f45d74

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: response further improve unit tests making use of .. operator

8294d71

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: response: re-order struct defs

9c397b1

In a top-down way to make generated docs read better when they are written. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: response: deduplicate values used in unit tests

3659818

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: remove ct-codecs dep; no not needed with custom Error enum

3bee9f3

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

omaha: request: reorder struct defs

78b710f

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

james-parky added a commit to james-parky/ue-rs that referenced this pull request Sep 18, 2025

download: fix new download crate after rebasing this branch onto main

7d2e60e

PR flatcar#87 created the download crate which caused merge conflicts with flatcar#88.

james-parky force-pushed the dev-omaha-improvements branch from fd49c6b to 7d2e60e Compare September 18, 2025 20:18

download: fix new download crate after rebasing this branch onto main

2f69107

PR flatcar#87 created the download crate which caused merge conflicts with flatcar#88. Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

james-parky force-pushed the dev-omaha-improvements branch from 7d2e60e to 2f69107 Compare September 18, 2025 20:19

omaha: simplify try_from_hex_string() func with for loop and comments

753c210

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

james-parky force-pushed the dev-omaha-improvements branch from b748c4d to 753c210 Compare September 19, 2025 16:37

james-parky mentioned this pull request Sep 19, 2025

[RFE] Audit anyhow Usage #89

Closed

dongsupark reviewed Sep 25, 2025

View reviewed changes

github-actions bot mentioned this pull request Sep 26, 2025

Monthly contributions report 2025-08-22 - 2025-09-21 flatcar/Flatcar#1899

Closed

dongsupark added a commit to flatcar/scripts that referenced this pull request Sep 26, 2025

overlay ue-rs: update to 2025-09-26

a10111f

Pulls in flatcar/ue-rs#72, flatcar/ue-rs#84, flatcar/ue-rs#85, flatcar/ue-rs#87, flatcar/ue-rs#88. Signed-off-by: Dongsu Park <dongsu@dpark.io>

dongsupark mentioned this pull request Sep 26, 2025

overlay ue-rs: update to 2025-09-26 flatcar/scripts#3310

Merged

1 task

dongsupark approved these changes Sep 29, 2025

View reviewed changes

clippy: fix clippy lint warning for cloning structs that impl Copy

29a23ad

Signed-off-by: james-parky <72101291+james-parky@users.noreply.github.com>

james-parky requested a review from dongsupark September 29, 2025 20:53

dongsupark approved these changes Sep 30, 2025

View reviewed changes

dongsupark merged commit 5d4db08 into flatcar:main Sep 30, 2025
1 check passed

dongsupark mentioned this pull request Oct 14, 2025

Processing passed XML for payload download is broken #92

Closed

github-actions bot mentioned this pull request Oct 22, 2025

Monthly contributions report 2025-09-22 - 2025-10-21 flatcar/Flatcar#1927

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Omaha crate improvements#88

Omaha crate improvements#88
dongsupark merged 31 commits intoflatcar:mainfrom
james-parky:dev-omaha-improvements

james-parky commented Sep 12, 2025 •

edited

Loading

Uh oh!

dongsupark left a comment

Uh oh!

Uh oh!

Uh oh!

james-parky commented Sep 16, 2025

Uh oh!

james-parky commented Sep 18, 2025

Uh oh!

dongsupark left a comment

Uh oh!

dongsupark left a comment

Uh oh!

james-parky commented Sep 29, 2025

Uh oh!

james-parky commented Sep 29, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

james-parky commented Sep 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

General Improvements to the omaha Crate

Hashing Algorithms

Requests and Responses

Other

Notes

Uh oh!

dongsupark left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

james-parky commented Sep 16, 2025

Uh oh!

james-parky commented Sep 18, 2025

Uh oh!

dongsupark left a comment

Choose a reason for hiding this comment

Uh oh!

dongsupark left a comment

Choose a reason for hiding this comment

Uh oh!

james-parky commented Sep 29, 2025

Uh oh!

james-parky commented Sep 29, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

james-parky commented Sep 12, 2025 •

edited

Loading

General Improvements to the `omaha` Crate