Re: git-fetch without a hash
Once we update the ‘guix’ package, the daemon will no longer accept ‘url-fetch’ downloads with hash = #f. Ok. Is there something like (git-checkout) we can use that will work instead of an (origin) here for url downloads?
Re: git-fetch without a hash
However, there's no real reason that git-fetch *needs* to be fixed-output in terms of having a hash pre-defined, at least for local development and other purposes. So is there a way around this? • write (package (source (git-checkout …)) …) This works well. Now I'm curious how to know what can go in the (source) field? Obviously not just (origin)... If having a way around it is not desirable should url-fetch consider this an error as well? I’m not sure; do you have an example where it’s not behaving as expected? Yes. When using (sha256 #f) url-fetch still has network access and works to download things, which is inconsistent vs other fetchers. signature.asc Description: PGP signature
git-fetch without a hash
Hello Guix! It seem that url-fetch will work without a hash (that is, with (sha256 #f)) but git-fetch will not. As near as I can tell this is because git-fetch uses a fixed derivation build going via nix/build.cc stuff which contains this line: if (i.second.hash == "") fixedOutput = false; And this results in /etc/resolv.conf not getting written and DNS resolution failing. Now, I *think* by my reading that it is intended for *only* fixed-output derivations to have network access, because otherwise the operation would be impure? And then url-fetch is just not actually ending up in a build container and so it works even when it isn't fixed-output? However, there's no real reason that git-fetch *needs* to be fixed-output in terms of having a hash pre-defined, at least for local development and other purposes. So is there a way around this? If having a way around it is not desirable should url-fetch consider this an error as well? Finally, *if* git-fetch should not allow this, the current error message is beyond confusing (DNS resolution fails during git fetch and it tries to fall back to SWH). So should git-fetch check for hash of #f and raise at that point with a better error message in that case? signature.asc Description: PGP signature
Getting GEM_PATH right for ruby scripts
Hello everyone! I've finally been diving deep into the Guix and doing a lot of packaging. One thing that has now come to my attention is the way ruby-build-system set GEM_PATH for wrapped scripts. It takes the whole GEM_PATH that is present at build time. This means that all those packages end up keeping store references to build-only (indeed, unit-test-only!) dependencies such as rake and minitest. Of course the problem is that the package specification does not have any way to know which inputs are going to be needed past build time. In the ruby packages, convention is to use propogated-inputs for this, but even so the GEM_PATH we need would be for all the implicitly propogated inputs. I guess there should be some way to get this, since it's what GEM_PATH will be in a profile that the package is installed to? I'm sure topics like this have been beaten to death, but I'd like to understand how things are and if they can be improved as much as will fit in my head. Thanks, signature.asc Description: PGP signature
Re: Test failures on new build
Could you attach tests/guix-git-authenticate.log?
Likewise.
Likewise.
Likewise.
Attached.
+ '[' -d /home/singpolyma/src/guix/.git ']'
+ guile -c '(use-modules (git))
(member "refs/heads/keyring" (branch-list (repository-open ".")))'
+ intro_commit=9edb3f66fd807b096b48283debdcddccfea34bad
+ intro_signer='BBB0 2DDF 2CEA F6A8 0D1D E643 A2A0 6DF2 A33A 54FA'
+ cache_key=test-4756
+ guix git authenticate 9edb3f66fd807b096b48283debdcddccfea34bad 'BBB0 2DDF
2CEA F6A8 0D1D E643 A2A0 6DF2 A33A 54FA' --cache-key=test-4756 --stats
--end=9549f0283a78fe36f2d4ff2a04ef8ad6b0c02604
guix git: error: Git error: cannot locate remote-tracking branch 'keyring'
FAIL tests/guix-git-authenticate.sh (exit status: 1)
substitute: guix substitute: warning: ACL for archive imports seems to be
uninitialized, substitutes may be unavailable
substitute: guix substitute: warning: authentication and authorization of
substitutes disabled!
waiting for locks or build slots...
Backtrace:
8 (primitive-load-path "tests/derivations.scm")
In ice-9/eval.scm:
626:19 7 (_ #)
293:34 6 (_ #(#(#) #))
In ice-9/boot-9.scm:
1736:10 5 (with-exception-handler _ _ #:unwind? _ # _)
In guix/store.scm:
636:37 4 (thunk)
2066:24 3 (run-with-store # ?)
In guix/tests.scm:
122:18 2 (_ _)
In guix/ui.scm:
566:4 1 (_ system-error "copy-file" _ _ _)
In ice-9/boot-9.scm:
1669:16 0 (raise-exception _ #:continuable? _)
ice-9/boot-9.scm:1669:16: In procedure raise-exception:
In procedure copy-file: Permission denied:
"/home/singpolyma/src/guix/gnu/packages/bootstrap/i686-linux/bash"
test-name: alist-cons-before
location: /home/singpolyma/src/guix/tests/build-utils.scm:37
source:
+ (test-equal
+ "alist-cons-before"
+ '((a . 1) (x . 42) (b . 2) (c . 3))
+ (alist-cons-before
+ 'b
+ 'x
+ 42
+ '((a . 1) (b . 2) (c . 3
expected-value: ((a . 1) (x . 42) (b . 2) (c . 3))
actual-value: ((a . 1) (x . 42) (b . 2) (c . 3))
result: PASS
test-name: alist-cons-before, reference not found
location: /home/singpolyma/src/guix/tests/build-utils.scm:41
source:
+ (test-equal
+ "alist-cons-before, reference not found"
+ '((a . 1) (b . 2) (c . 3) (x . 42))
+ (alist-cons-before
+ 'z
+ 'x
+ 42
+ '((a . 1) (b . 2) (c . 3
expected-value: ((a . 1) (b . 2) (c . 3) (x . 42))
actual-value: ((a . 1) (b . 2) (c . 3) (x . 42))
result: PASS
test-name: alist-cons-after
location: /home/singpolyma/src/guix/tests/build-utils.scm:45
source:
+ (test-equal
+ "alist-cons-after"
+ '((a . 1) (b . 2) (x . 42) (c . 3))
+ (alist-cons-after
+ 'b
+ 'x
+ 42
+ '((a . 1) (b . 2) (c . 3
expected-value: ((a . 1) (b . 2) (x . 42) (c . 3))
actual-value: ((a . 1) (b . 2) (x . 42) (c . 3))
result: PASS
test-name: alist-cons-after, reference not found
location: /home/singpolyma/src/guix/tests/build-utils.scm:49
source:
+ (test-equal
+ "alist-cons-after, reference not found"
+ '((a . 1) (b . 2) (c . 3) (x . 42))
+ (alist-cons-after
+ 'z
+ 'x
+ 42
+ '((a . 1) (b . 2) (c . 3
expected-value: ((a . 1) (b . 2) (c . 3) (x . 42))
actual-value: ((a . 1) (b . 2) (c . 3) (x . 42))
result: PASS
test-name: alist-replace
location: /home/singpolyma/src/guix/tests/build-utils.scm:53
source:
+ (test-equal
+ "alist-replace"
+ '((a . 1) (b . 77) (c . 3))
+ (alist-replace 'b 77 '((a . 1) (b . 2) (c . 3
expected-value: ((a . 1) (b . 77) (c . 3))
actual-value: ((a . 1) (b . 77) (c . 3))
result: PASS
test-name: alist-replace, key not found
location: /home/singpolyma/src/guix/tests/build-utils.scm:57
source:
+ (test-assert
+ "alist-replace, key not found"
+ (not (false-if-exception
+ (alist-replace 'z 77 '((a . 1) (b . 2) (c . 3))
actual-value: #t
result: PASS
test-name: fold-port-matches
location: /home/singpolyma/src/guix/tests/build-utils.scm:61
source:
+ (test-equal
+ "fold-port-matches"
+ (make-list 3 "Guix")
+ (call-with-input-string
+ "Guix is cool, Guix rocks, and it uses Guile, Guix!"
+ (lambda (port)
+ (fold-port-matches cons '() "Guix" port
expected-value: ("Guix" "Guix" "Guix")
actual-value: ("Guix" "Guix" "Guix")
result: PASS
test-name: fold-port-matches, trickier
location: /home/singpolyma/src/guix/tests/build-utils.scm:67
source:
+ (test-equal
+ "fold-port-matches, trickier"
+ (reverse '("Guix" "guix" "Guix" "guiX" "Guix"))
+ (call-with-input-string
+ "Guix, guix, GuiGuixguiX, Guix"
+ (lambda (port)
+ (fold-port-matches
+ cons
+ '()
+ (list (char-set #\G #\g)
+ (char-set #\u)
+ (char-set #\i)
+ (char-set #\x #\X))
+ port
expected-value: ("Guix" "guiX" "Guix" "guix" "Guix")
actual-value: ("Guix" "guiX" "Guix" "guix" "Guix")
result: PASS
test-name: fold-port-matches, with unmatched chars
location: /home/singpolyma/src/guix/tests/build-utils.scm:78
source:
+ (test-equal
+ "fold-port-matches, with unmatched chars"
+ '("Guix"
+ #\,
Test failures on new build
Trying to get an environment for hacking on Guix going on my new computer. Installed guix on my Debian stable using guix-install.sh then ran guix pull. Did a fresh git clone on guix then guix environment guix --pure ./bootstrap ./configure --localstatedir=/var make -j24 make -j24 check mostly works, but I get these failures: FAIL: tests/guix-git-authenticate.sh PASS: tests/workers.scm make[4]: *** [Makefile:5605: tests/derivations.log] Error 1 make[4]: *** Waiting for unfinished jobs FAIL: tests/build-utils.scm make[4]: *** [Makefile:5605: tests/grafts.log] Error 1 any thoughts? signature.asc Description: PGP signature
Re: Racket packages / build system
Thanks for the notes. I've skimmed through them and they seem sensible. I'll look at how other build systems are written as a first step, then get my hands wet. Do I open an issue for this I've recently become interested in this as well. I pinged cwebber asking for the best place to start and got shown this thread. Have you made any progress in thinking about how to go about this? Did an issue get created (I don't see one)? Any place I can jump in to help out? Just don't want to start mucking about and duplicating effort if it's not needed :) signature.asc Description: PGP signature
