Olivier Rojon writes:
> I have set the INFOPATH so that it points -- among others -- to the
> path of my current profile which is the canonical
> `.config/guix/current/share/info`.
What happens when "${GUIX_PROFILE}/share/info" is added to the INFOPATH?
Where ${GUIX_PROFILE} (in my case) is "${H
Olivier Rojon writes:
> I am beginning to suspect this is an emacs-specific issue.
Perhaps. FWIW, I am using Guix on a foreign distribution (openSUSE
Tumbleweed) and am able to view the info page for recutils from within
Emacs (29.3).
In my case the info page is installed under "${GUIX_PROFILE
Olivier Rojon writes:
> Do you mind sharing the values of both your `INFOPATH`
"${HOME}/.guix-profile/share/info:${HOME}/.config/guix/current/share/info:"
> and your `Info[-]?(additional|default)?-directory-list`?
Info-additional-directory-list is nil. Info-default-directory-list is
nil. Inf
Olivier Rojon writes:
> now the problem is solved.
Closing the issue.
--
Suhail
Liliana Marie Prikler writes:
> this is a somewhat experimental patch, which reduces Vcomp_native_version_dir
> to simply Vcomp_abi_hash. Note, that this is not enough to address the
> issues currently noticed with Emacs native compilation, as Vcomp_abi_hash
> is unstable between 29.3 and 29.4.
Liliana Marie Prikler writes:
> +;; Note: When using (replacement …), ensure that comp-native-version-dir
> +;; stays the same across grafts.
Perhaps an acknowledgment that we don't yet know how to accomplish that
would be helpful? Unless I have misunderstood the current state and am
mi
Liliana Marie Prikler writes:
> It could possibly allow us some flexibility in future changes, but at
> the very least it doesn't work right now. Hence why it's RFC while
> the other patch isn't.
Thank you for clarifying. It is clear that different versions of Emacs
store the natively compiled
Liliana Marie Prikler writes:
> with the grafting of Emacs 29.3 to 29.4, we see that Emacs itself is
> still correctly loaded, but Emacs libraries (e.g. dash) aren't.
>
> (comp-el-to-eln-filename (expand-file-name "…/dash.el"))
> => $HOME/.config/emacs/eln-cache/29.4-46e5bcbe/dash-2.19.1/dash.eln
Liliana Marie Prikler writes:
>> IIUC, the issue is that replacement packages are grafted post-build.
>> This means that when emacs-dash is built, its AOT native-compilation
>> happens with Emacs 29.3. However, at run-time Emacs 29.4 gets
>> grafted in.
> Nitpick: Emacs 29.4 gets grafted in at p
Liliana Marie Prikler writes:
> Am Sonntag, dem 14.07.2024 um 12:27 -0400 schrieb Suhail Singh:
>> However, perhaps there is a way to ensure that the proposed
>> replacement doesn't have a different ABI_VERSION. Could this be
>> caught by a test or "sanity che
Suhail Singh writes:
> Liliana Marie Prikler writes:
>
>> Am Sonntag, dem 14.07.2024 um 12:27 -0400 schrieb Suhail Singh:
>>> However, perhaps there is a way to ensure that the proposed
>>> replacement doesn't have a different ABI_VERSION. Could this be
>
Liliana Marie Prikler writes:
> + (test-equal "native-comp-dir"
> +(emacs-native-comp-dir
> + #$(file-append old-emacs "/bin/emacs"))
> +(emacs-native-comp-dir
> + #$(file-append new-emacs "/bin/emacs")))
I like that there is a test that f
Liliana Marie Prikler writes:
> +;; Run `make check-system TESTS=emacs-native-comp' to ensure that grafts
> +;; can meaningfully be applied.
Either this isn't the correct invocation or I am doing something wrong.
Running that command in guix checkout seems to run many tests (possibly
all
Liliana Marie Prikler writes:
> Selected 1 system tests...
I get this line as well early on. However, after that lots more things
happen.
Could you please confirm the guix checkout commit you're applying your
patch on?
--
Suhail
Liliana Marie Prikler writes:
> Commit 9724e61cda80e4c59a2eb419a453887ecc551b9a
Okay, I can confirm that it's working when applied on that commit.
I am not sure whether the issue I was observing on
24163eea584663568b68e19f364256fc7396b61f were due to incorrect
configuration on my part or someth
Suhail Singh writes:
> I am not sure whether the issue I was observing on
> 24163eea584663568b68e19f364256fc7396b61f
Ah. I was simply unlucky. The failing tests were related to the build
of guile 3.0.10. This commit has since been reverted in master.
Apologies for the noise.
--
Suhail
Saku Laesvuori via Bug reports for GNU Guix writes:
> Anyway, I am not opposed to this change. The only effects for my use
> cases are positive (nicer UI with the --allow flag). I just want to
> point out that I don't think this makes any attacks significantly
> harder.
FWIW, this summarizes my
* gnu/packages/emacs-xyz.scm (emacs-debbugs): Update to 0.42.
Change-Id: Id386890143aa6c2efe6e61c928fb6f9dba6a6e98
---
gnu/packages/emacs-xyz.scm | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/gnu/packages/emacs-xyz.scm b/gnu/packages/emacs-xyz.scm
index 59c804066a..411ae
Nicolas Graves writes:
> Building a package yes, but you can have external commands in a
> manifest.scm or guix.scm.
>
> ...
>
> What I was saying is that we could restrain recording `guix shell --allow`
> only if the manifest builds properly containerized and without network
> access (outside pa
Simon Tournier writes:
> However, I am proposing: Aside the profile ’manifest’ file (lossy
> process), we could store all the manifests provided by the command
> line. Something as:
>
> /gnu/store/-profile/manifest
> /gnu/store/
Nicolas Graves writes:
> My last message to Saku basically agreed to this ;)
Yes, my bad for only noticing that message after having sent mine.
Whoops.
> I'm actually willing to improve that patch series if you have better
> ideas/implementations, I was just building on what I know
> (direnv/.d
Christopher Howard writes:
> Recipe:
> 1. M-x guix
> 2. c (for "guix shell commands)
In case relevant, on a foreign distribution with on Emacs 29.4 and
emacs-guix installed via MELPA (using the same commit emacs-guix uses
for upstream) I am unable to reproduce the issue.
However, the below allo
Cayetano Santos via Bug reports for GNU Guix writes:
> No problem here. So the issue is related with the guix shell container,
> in my opinion (unless there is something else in my default profile).
>
> Could someone confirm ?
M-x guix-packages-by-name works fine in a guix shell container. I.e.
Suhail Singh writes:
> #+begin_src example
> $> gx_shell bash-minimal coreutils-minimal emacs emacs-guix guix guile
> [env]$> emacs
>
>
>
> $> type gx_shell
> gx_shell is a function
> gx_shell ()
> {
> if ob_shell_sessio
Christopher Howard writes:
> So, my experiment was to see if I could run the scheme command «(guix-command
> "pull")» from the console instead, just using guix repl. However, since guix
> repl will not load any emacs-guix scheme code, I had to add
> "/home/christopher/Repos/emacs-guix/scheme" to
Christopher Howard writes:
> I discovered that if I go to Gnome console, set the GUILE_LOAD_PATH
> appropriately for the emacs-guix scheme code, run «,re (emacs-guix),»
> and then run (guix-command "pull") that the pull completes without any
> issues.
Out of curiosity, does the below raise an as
Christopher Howard writes:
> Hi, this is a little confusing since my example involved working outside
> of Emacs, running Scheme from the guix repl.
I am confused (as well). I thought you were running the command via the
"*Guix REPL*" buffer created by emacs-guix. Perhaps not? I'll share
belo
Simon Tournier writes:
>> For --save-provenance, is your proposal that in addition to saving the
>> additional manifests, the commandline that was used would also be saved?
>
> We could save the command line too using another file. :-)
That would be helpful :)
Given an image generated with --sa
Closing this, since an equivalent patch was committed by Maxim in
2e8a8b3ddb4ca31b918d63ff2957c6da1424a2a9 .
--
Suhail
* gnu/packages/patches/racket-launcher-config-dir.patch: New file. With this
patch, packages installed at runtime via raco have configuration directory
correctly set.
* gnu/local.mk (dist_patch_DATA): Register it.
* gnu/packages/racket.scm (%racket-origin): Use it.
Change-Id: Ibba2d73b72a66e58a3c
Liliana Marie Prikler writes:
> * `guix shell emacs emacs-transient --with-input=emacs-minimal=emacs` will
> load the native-compiled emacs
Hm there may be a bug somewhere (hopefully not in my testing).
I created a profile using:
#+begin_src sh
./pre-inst-env guix package
--profile=${HOME
Hello Guix,
Currently, binaries installed via "raco" (included in Guix's "racket"
package) don't include the -G / --config configuration option specifying
the "Main configuration directory". This is problematic.
Specifically, when installing "rash" via raco using the below:
#+begin_src sh
rac
V2 looks good; works as expected. Tested with emacs-org, emacs-magit,
emacs-transient.
--
Suhail
Ludovic Courtès writes:
> Long story short: it would be great if invalid keyboard layouts were
> caught when the system is instantiated. It could be ‘ckbcomp’ errorring
> out, or we could have additional code that browses xkeyboard-config’s
> ‘base.xml’ file.
Ludo, this issue currently has the
45mg <45mg.wri...@gmail.com> writes:
> It looks like this isn't the only REPL command that doesn't work in
> `guix repl`. I haven't done anything close to a full investigation (I am
> new to using Guile!), but many other commands simply don't work.
Thank you for your investigation. I am cc-ing g
Suhail Singh writes:
>> It looks like this isn't the only REPL command that doesn't work in
>> `guix repl`. I haven't done anything close to a full investigation (I am
>> new to using Guile!), but many other commands simply don't work.
>
> Thank y
* gnu/packages/patches/racket-launcher-config-dir.patch: New file. With this
patch, packages installed at runtime via raco have configuration directory
correctly set.
* gnu/local.mk (dist_patch_DATA): Register it.
* gnu/packages/racket.scm (%racket-origin): Use it.
Change-Id: Ibba2d73b72a66e58a3c
37 matches
Mail list logo
