silo: add patch for constant namescheme issue

[markcmiller86]

We discovered Silo 4.12.0 has an issue with handling of delimiter characters in constant nameschemes.

This PR adds a patch for that.

[markcmiller86]

@cyrush or @JustinPrivitera when you have a chance, can you pls take a look at this.

[alecbcs]

Quick suggestion to reference the upstream merge commit instead of copying the patch into the spack-packages repository. Otherwise looks good to me. Thanks @markcmiller86!

[alecbcs]

Looks good to me. Thanks @markcmiller86!

[markcmiller86]

@alecbcs thanks for quick response and merge.

[haampie]

@alecbcs / @markcmiller86 did you try this? The patch does not apply.

[markcmiller86]

Well...now that I think about it ..I may have tried only the file based one before switching to GitHub ref. What are you seeing? Can you post details.

[cyrush]

Wasn’t that pr into the rc - not develop/main. Could that be the issue?

…

-Cyrus

________________________________ From: Mark C. Miller ***@***.***> Sent: Monday, February 9, 2026 10:02:32 AM To: spack/spack-packages ***@***.***> Cc: Harrison, Cyrus D. ***@***.***>; Mention ***@***.***> Subject: Re: [spack/spack-packages] silo: add patch for constant namescheme issue (PR #3252) [https://avatars.githubusercontent.com/u/5720676?s=20&v=4]markcmiller86 left a comment (spack/spack-packages#3252)<https://urldefense.us/v3/__https://github.com/spack/spack-packages/pull/3252*issuecomment-3873122912__;Iw!!G2kpM7uM-TzIFchu!zjfNko9_SwelIW5NeBEES2UukWNwhixbepg1Cqm9ji1Kgl9EWJpnZ5krBpg7GmGAFcMqIuZ0BONn1GtmtbhuXybB$> Well...now that I think about it ..I may have tried only the file based one before switching to GitHub ref. What are you seeing? Can you post details. — Reply to this email directly, view it on GitHub<https://urldefense.us/v3/__https://github.com/spack/spack-packages/pull/3252*issuecomment-3873122912__;Iw!!G2kpM7uM-TzIFchu!zjfNko9_SwelIW5NeBEES2UukWNwhixbepg1Cqm9ji1Kgl9EWJpnZ5krBpg7GmGAFcMqIuZ0BONn1GtmtbhuXybB$>, or unsubscribe<https://urldefense.us/v3/__https://github.com/notifications/unsubscribe-auth/AAJDUHW6GDDWTHKQVGHW2AD4LDDTRAVCNFSM6AAAAACT7ON32WVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTQNZTGEZDEOJRGI__;!!G2kpM7uM-TzIFchu!zjfNko9_SwelIW5NeBEES2UukWNwhixbepg1Cqm9ji1Kgl9EWJpnZ5krBpg7GmGAFcMqIuZ0BONn1GtmtQ_6fDgK$>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[alecbcs]

That sounds likely that'd be the issue. Apologies for not noticing that before I suggested the merge commit.

@haampie stepped in and fixed the issue (saved the day?) by using a commit from within the PR. Not usually my preference, but it does work here fairly well.

[markcmiller86]

That sounds likely that'd be the issue. Apologies for not noticing that before I suggested the merge commit.

@haampie stepped in and fixed the issue (saved the day?) by using a commit from within the PR. Not usually my preference, but it does work here fairly well.

Ah, ok, now I understand. That PR had work related to fixing Silo testing on macOS as well and only one specific commit from it was the patch to the source code itself to fix the constant namescheme bug. Apologies for not paying attention to that and just assuming it was the same commit hash for that one change in the PR and not the whole PR. The correct commit hash for that one change is 43a52d788a3c15bee3b9391906e8ed276c5a456c.

Thanks @haampie so much for unraveling this to the correct commit hash.

[markcmiller86]

The correct commit hash for that one change is 43a52d788a3c15bee3b9391906e8ed276c5a456c.

which I confirm is now on develop.