New package: RMTool v0.1.0

[JuliaRegistrator]

• Registering package: RMTool
• Repository: https://github.com/rajraomichigan/RMTool.jl
• Created by: @rajraomichigan
• Version: v0.1.0
• Commit: f26b017bcc6122acb6497f56f054945cbe497359
• Reviewed by: @rajraomichigan
• Reference: rajraomichigan/RMTool.jl@f26b017#commitcomment-200077361
• Description: Symbolic random matrix eigenvalue distribution calculator

[github-actions]

Hello, I am an automated registration bot. I help manage the registration process by checking your registration against a set of AutoMerge guidelines. If all these guidelines are met, this pull request will be merged automatically, completing your registration. It is strongly recommended to follow the guidelines, since otherwise the pull request needs to be manually reviewed and merged by a human.

1. New package registration

Please make sure that you have read the package naming guidelines.

2. AutoMerge Guidelines are all met! ✅

Your new package registration met all of the guidelines for auto-merging and is scheduled to be merged when the mandatory waiting period (3 days) has elapsed.

3. To pause or stop registration

If you want to prevent this pull request from being auto-merged, simply leave a comment. If you want to post a comment without blocking auto-merging, you must include the text [noblock] in your comment.

Tip: You can edit blocking comments to add [noblock] in order to unblock auto-merging.

[goerz]

Thank you for submitting your package! The licensing information in the repository is contradictory: the LICENSE file is an MIT license (and that is what the registry will record), while the README and the front page of the documentation both state GPL-2.0-or-later, "as the original toolbox". Since you are the author of the original MATLAB toolbox, you are of course free to relicense the Julia port, but the repository should say the same thing in all three places. Could you settle that before this is merged?

The link from the README to the documentation is also broken: it points to https://rajraomichigan.github.io/RMTool/, but the deployed site is at https://rajraomichigan.github.io/RMTool.jl/. The installation instruction just above it has the same issue (add https://github.com/rajraomichigan/RMTool currently only works because GitHub redirects the old repository name).

One of the README examples is out of date, too: Lmz2MomS(b, 10) for b = wishartpol(1//2) is annotated as [1, 1, 3//2, 5//2, 37//8, ...], but it actually returns [1, 1, 3//2, 11//4, 45//8, ...] (which is what your own test suite asserts). Please make sure that the code and the output shown in the README actually match what the package does. You can add a test that dynamically extracts the examples from the README, writes them to a sandbox folder, and runs them.

Lastly, a recommendation rather than a registration concern: the __precompile__(false) in src/RMTool.jl means that neither RMTool nor any package depending on it can be precompiled, so every session pays the full load cost. The usual way around FLINT handles that cannot be serialized into a precompile cache is to construct the ring inside an __init__() function and assign it to a mutable global, rather than building it at load time.

[noblock]

[rajraomichigan]

Hi Michael, Thanks very much for the careful review — all four points were spot on. (I am new to package publishing and used Claude to help with the initial commit and also addressing your points -- so please forgive if I don't get something right away!) - *License:* settled as MIT. I'm the author of the original MATLAB toolbox and am happy to release the Julia port under MIT, which also matches what the registry recorded. The LICENSE file, the README, and the docs front page now all state MIT consistently. - *Broken links:* fixed. The README now points to https://rajraomichigan.github.io/RMTool.jl/ for the docs and https://github.com/rajraomichigan/RMTool.jl for installation, rather than relying on the old-name redirect. - *Stale example:* fixed. Lmz2MomS(wishartpol(1//2), …) is now annotated as [1, 1, 3//2, 11//4, 45//8, …], matching what the package returns and what the test suite asserts. I've also added a test that extracts the julia code blocks from the README, writes them to a sandbox, and runs them, so the documented output can't drift from the actual behavior again. - *__precompile__(false):* thanks for the pointer to the __init__() pattern — (Claude AI noted that) you're right that the FLINT handles are the reason. I'd like to land that as a follow-up patch (v0.1.1) so I can verify precompilation and load behavior properly against a live setup rather than rush it into this first release. It's a recommendation rather than a blocker, as you noted. The updated commit is on main. Happy to make any further changes. Thanks, Raj

…

On Fri, Sep 11, 2026 at 6:13 AM Michael Goerz ***@***.***> wrote: *goerz* left a comment (JuliaRegistries/General#167797) <#167797 (comment)> Thank you for submitting your package! The licensing information in the repository is contradictory: the LICENSE file is an MIT license (and that is what the registry will record), while the README and the front page of the documentation both state GPL-2.0-or-later, "as the original toolbox". Since you are the author of the original MATLAB toolbox, you are of course free to relicense the Julia port, but the repository should say the same thing in all three places. Could you settle that before this is merged? The link from the README to the documentation is also broken: it points to https://rajraomichigan.github.io/RMTool/, but the deployed site is at https://rajraomichigan.github.io/RMTool.jl/. The installation instruction just above it has the same issue (add https://github.com/rajraomichigan/RMTool currently only works because GitHub redirects the old repository name). One of the README examples is out of date, too: Lmz2MomS(b, 10) for b = wishartpol(1//2) is annotated as [1, 1, 3//2, 5//2, 37//8, ...], but it actually returns [1, 1, 3//2, 11//4, 45//8, ...] (which is what your own test suite asserts). Please make sure that the code *and* the output shown in the README actually match what the package does. You can add a test that dynamically extracts the examples from the README, writes them to a sandbox folder, and runs them. Lastly, a recommendation rather than a registration concern: the __precompile__(false) in src/RMTool.jl means that neither RMTool nor any package depending on it can be precompiled, so every session pays the full load cost. The usual way around FLINT handles that cannot be serialized into a precompile cache is to construct the ring inside an __init__() function and assign it to a mutable global, rather than building it at load time. — Reply to this email directly, view it on GitHub <#167797?email_source=notifications&email_token=AGGOI7U7QCG2X7ZGKIGW2UT5OPF3FA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTKNRTGI4TEMRRGE4KM4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLDGN5XXIZLSL5RWY2LDNM#issuecomment-5632922118>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AGGOI7RCN3YP5XFYWFK7WHT5OPF3FAVCNFSNUABFKJSXA33TNF2G64TZHMYTAMJYGA3TKMZRHNEXG43VMU5TKNBRG43TAMRTHA3KC5QC> . Triage notifications, keep track of coding agent tasks and review pull requests on the go with GitHub Mobile for iOS <https://github.com/notifications/mobile/ios/AGGOI7V7KIL5SZUW2V62UYL5OPF3FA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTKNRTGI4TEMRRGE4KM4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJKTGN5XXIZLSL5UW64Y> and Android <https://github.com/notifications/mobile/android/AGGOI7VH4VRYESZ2SKQTXUT5OPF3FA5CNFSNUABFM5UWIORPF5TWS5BNNB2WEL2JONZXKZKDN5WW2ZLOOQXTKNRTGI4TEMRRGE4KM4TFMFZW63VHNVSW45DJN5XKKZLWMVXHJLTGN5XXIZLSL5QW4ZDSN5UWI>. Download it today! You are receiving this because you were mentioned.Message ID: ***@***.***>

[goerz]

Make sure to retrigger the registration from the latest commit, keeping the version number that same, and add [noblock] to you own comment

[rajraomichigan]

Fixes pushed, license settled as MIT. [noblock]