diff mbox series

bug#60266: [PATCH] gnu: Add form.

Message ID 07209d425a70f9e21169478617feba541de5ca13.camel@posteo.net
State New
Headers show
Series bug#60266: [PATCH] gnu: Add form. | expand

Commit Message

Eric Bavier March 22, 2023, 4:54 a.m. UTC 
Antero,

Thank you for your effort in this review.

On Tue, 2023-03-21 at 18:53 +0000, Antero Mejr wrote:
> > > > Guix usually assumes `SSE2` capabilities for x86_64 targets, iirc. 
> > It does, but not for i386 to i686, which also often see sse2 flags in 
> > build systems.
> 
> I did some benchmarks on a few workloads with millions of terms. The
> "tunable" optimizations didn't consistently improve the
> performance. Sometimes it would be a couple percent faster, sometimes
> slower. So I didn't mark the package as tunable.

I was watching my system load while `parform` was running its tests, and things did not
seem compute bound, so the package being not tunable seems fine to me.

> > > I used the other homepage so people can skip the weird "license
> > > agreement" page at https://www.nikhef.nl/~form/.
> > Calling the GPL a license agreement is weird, and it doesn't help that 
> > https://www.nikhef.nl/~form/maindir/ also refers to it.  Perhaps we can
> > raise an issue about that upstream?
> 
> I would prefer not to bother upstream further regarding the website,
> since I already had to ask them to change an invalid license statement
> in a file header (which they did).

Version 1 of the GPL used the term "license agreement" when describing itself, so the
phrasing is not *super* weird, just maybe a bit outdated.  There are also plenty of free
software projects, e.g. nodejs, that present users with an "accept this license agreement"
prompt in their M$ installers.

I've pushed these patches in 

  15ee08ebf23ea6bc5be9fa157889419aa079076a and
  31bed61b8278d3736cc29055d60db068a1160071

with a small adjustment to appease `guix lint`:


Thanks again,
`~Eric
diff mbox series

Patch

--- a/gnu/packages/algebra.scm
+++ b/gnu/packages/algebra.scm
@@ -1873,6 +1873,9 @@  (define-public form
 answers are returned in a textual mathematical representation.  The size of
 the considered expressions in FORM is only limited by the available disk space
 and not by the available RAM.")
+      ;; XXX: Ignore this CVE to work around a name clash with the unrelated
+      ;; "neos/forms" package.
+      (properties '((lint-hidden-cve . ("CVE-2021-32697"))))
       ;; x86_64 only due to test failures on other platforms.
       ;; Developers say other platforms are not "tier 1" supported:
       ;; https://github.com/vermaseren/form/issues/426