bug#55344: ftw thinks readable directories are unreadable if "other" don't have read permission

ftw incorrectly thinks that if other users don't have the read permission then the owner can't read it either. This also affects ntfw. Guile version: 3.0.8 Machine type: x86_64-pc-linux-gnu guile@3.0.8 from Guix To reproduce: Create 2 directories, one with rwxrwxrwx permissions and another

bug#54915: [PATCH] Guile rejects empty vendor in GNU triplets, as used by NetBSD

NetBSD uses an empty vendor string in a GNU target triplet, like `x86_64--netbsd' or `armv7--netbsdelf-eabihf'; it's been like this since the mid-'90s. (Before that, NetBSD omitted the vendor altogether, so if x86_64 existed then it might have been `x86_64-netbsd'.) This runs afoul of a rule in

bug#54911: Missing modules argument for coverage-data->lcov

Hello, I noticed the documentation specify that the `coverage-data->lcov` function should be able to take modules to limit the scope of the data that's coverted to lcov. From the docs: -- Scheme Procedure: coverage-data->lcov data port #:key modules Traverse code coverage information

bug#31878: Module autoloading is not thread safe

Hi Ludo and Mark, > However peg.test fails and it may be related The patch fails when a module has one of its submodules in its uses list, and the program starts loading the module before it starts loading the submodule. In the test case, (ice-9 peg) uses (ice-9 peg codegen) and the test starts

bug#54622: [PATCH] Find unidata_to_charset.awk in $(srcdir)

* libguile/Makefile.am (srfi-14.i.c): Prepend $(srcdir). --- libguile/Makefile.am | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libguile/Makefile.am b/libguile/Makefile.am index bb1ba3613..e3e5f8ed8 100644 --- a/libguile/Makefile.am +++ b/libguile/Makefile.am @@ -463,7

bug#50068: [PATCH] In curried definitions, move docstrings to outermost lambda

Hi, Please consider the following patch, fixing bug #50068. Best regards, Jean From 79552d2974e9cbcfcf01960aab68cb6824c88972 Mon Sep 17 00:00:00 2001 From: Jean Abou Samra Date: Tue, 29 Mar 2022 00:14:45 +0200 Subject: [PATCH] In curried definitions, move docstrings to outermost lambda

bug#54478: unwanted source references

I think the error is in use of "src" arg to make-seq in the following loop, in ice-9/psyntax.scm, starting at line 1633: (let lp ((var-ids var-ids) (vars vars) (vals vals) (tail (expand-tail-expr))) (cond ((null?

bug#54566: Can't build with `--disable-regex`

Hello there, While trying to build a minimal version of Guile I ran into some errors which seem to stem from the `--disable-regex` flag. I was building from the guile-3.0.8.tar.gz tarball on an updated amd64 Arch GNU/Linux system. After extracting I ran the following commands: ``` ./configure

bug#46258: bug#54485: [PATCH] gnu: Add guile-with-openat.

Ludovic Courtès schreef op do 24-03-2022 om 12:37 [+0100]: > Instead, I would prefer to actually review and apply the patches > upstream.  I’ve been meaning to do that, really, but it requires a chunk > of time and focus, which keeps leading me to postpone. OK. I can relate, w.r.t. the IPFS

bug#54538: (ice-9 suspendable-ports) doesn’t implement the 'escape conversion strategy

‘peek-char’ in (ice-9 suspendable-ports) checks for the 'substitute port conversion strategy, but it doesn’t check for 'escape: (if (eq? (port-conversion-strategy port) 'substitute) (values #\xFFFD buf (+ cur len)) (decoding-error

bug#43521: ports.test "non-revealed port is closed" breaks other tests

Looks like it has been resolved now, see the commit by Rob Browing: Do we close the issue now, or only at the next release? Thanks, Maxime. signature.asc Description: This is a digitally signed

bug#54478: unwanted source references showing up w/ -O0 (with patch)

In guile-3.0.8, when I compile w/ optimiation-level 0, I am seeing references to the beginning let-form. The following illustrates the issue. Consider foo.scm: (define *a* (make-parameter 0)) (define (foo . args) (let* ((b (+ (*a*) 1)) (c (+ (*a*) b 2))) (simple-format #t

bug#54444: wrong .pc file for `configure --disable-shared`

[guile 2.2.7] If I configure guile with ``` configure --disable-shared ``` the created `guile-2.2.pc` is incorrect: Instead of ``` Libs: -L${libdir} -lguile-2.2 -lgc Libs.private: -lgmp -lltdl -L/usr/lib64/../lib64 -lffi \ -lunistring -lcrypt -lm ``` it should have ```

bug#54276: Bug#990250: guile-2.2 FTBFS on musl: dh_missing complains about charset.alias

Hi Rob, On Sun, Mar 06, 2022 at 12:04:20PM -0600, Rob Browning wrote: > Hmm, it looks like this may have changed in more recent 3.0 releases > (i.e. need_charset_alias is no longer mentioned anywhere in the tree). > I wonder if that means we need a different patch, or perhaps the problem > has

bug#54276: Bug#990250: guile-2.2 FTBFS on musl: dh_missing complains about charset.alias

Helmut Grohne writes: > --- a/lib/Makefile.am > +++ b/lib/Makefile.am > @@ -1043,7 +1043,7 @@ install-exec-localcharset: all-local > case '$(host_os)' in \ > darwin[56]*) \ > need_charset_alias=true ;; \ > - darwin* | cygwin* | mingw* | pw32* | cegcc*) \ >

bug#54276: Bug#990250: guile-2.2 FTBFS on musl: dh_missing complains about charset.alias

[If possible, please preserve at least the -forwarded address in any replies.] It looks like a file might be missing from installs when building for linux-musl: Helmut Grohne writes: > Source: guile-2.2 > Version: 2.2.7+1-6 > > guile-2.2 fails to build from source on musl-linux-any, because

bug#54272: Avoid producing ? in locales with too few characters.

This seems difficult to test. Any ideas? From 8f9a8e56cc3e8496fdfed3f889cff9fca19b3663 Mon Sep 17 00:00:00 2001 From: Maxime Devos Date: Sun, 6 Mar 2022 12:51:33 + Subject: [PATCH 2/2] Deprecate non-functional bind-textdomain-codeset. TODO: this only deprecated it in the documentation, it

bug#54211: Simplified version of scheme

Dear Sir/Madam, I would like to use the Guile framework for a company's CSR assessment. Would it be possible to get a link for a simplified version of the scheme, please? Your help is much appreciated. Kind regards, D. Rimal

bug#54171: [PATCH] web: default to INADDR_ANY instead of INADDR_LOOPBACK

Bon dia! Aleix Conchillo Flaqué skribis: > Using INADDR_ANY instead of INADDR_LOOPBACK makes it convenient when > starting the web server inside containers without the need to having to > specify INADDR_ANY all the time. This is the default in most libraries > and languages. > > This doesn't

bug#54198: Guile 3.0.8 cross-compiled to i586-pc-gnu crashes

Hi, Ludovic Courtès skribis: > #11 0x010b68dc in range_error (bad_val=bad_val@entry=0x6, min=0x2, > max=max@entry=0xfffe) at numbers.c:6611 > #12 0x010b6b8f in inum_in_range (max=-1, min=0, x=0x6) at numbers.c:6630 > #13 scm_to_uint32 (arg=0x6) at numbers.c:6787 Turns out we had

bug#54198: Guile 3.0.8 cross-compiled to i586-pc-gnu crashes

Hi, Guix’s Guile 3.0.8 package cross-compiled to i586-pc-gnu (GNU/Hurd) crashes when starting (3.0.7 was fine): --8<---cut here---start->8--- ludo@childhurd ~$ /gnu/store/19dfmmsy41614n9nxkbpn73sr8xhccqy-guile-3.0.7/bin/guile --version guile (GNU Guile) 3.0.7

bug#54172: [PATCH] vectors: add (vector-last) support

* libguile/vectors.c: add (vector-last) support. * libguile/vectors.h: define scm_vector_last and scm_c_vector_last. * doc/ref/api-data.texi (Vector Accessors): add documentation for (vector-last). --- doc/ref/api-data.texi | 10 ++ libguile/vectors.c| 30

bug#54171: [PATCH] web: default to INADDR_ANY instead of INADDR_LOOPBACK

Using INADDR_ANY instead of INADDR_LOOPBACK makes it convenient when starting the web server inside containers without the need to having to specify INADDR_ANY all the time. This is the default in most libraries and languages. This doesn't break backwards compatibility since INADDR_LOOPBACK is

bug#54163: guile 3.0.8 'make check' fails

> FAIL: test-out-of-memory It looks like this test has not been very functional for a while, rlb has marked it as an expected failure for Debian. Link to his patch:

bug#54163: guile 3.0.8 'make check' fails

I just got the 3.0.8 tarball and did a build on Fedora 35 x64. I used no parameters to 'configure', except for '--prefix'. Then I did 'make' and 'make check'. One test failed. The output ends with: wrote

bug#54147: [PATCH] doc: fix web (run-server) examples

* doc/ref/web.texi (Web Server): need quasiquote to in order to evaluate AF_INET6. --- doc/ref/web.texi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/doc/ref/web.texi b/doc/ref/web.texi index 93cd0214f..49a09d0ca 100644 --- a/doc/ref/web.texi +++ b/doc/ref/web.texi @@

bug#54143: [PATCH] Add string-split-substring to ice-9 string-fun

Hello, I've attached a patch adding a new function, (string-split-substring str substr), that splits a string into a list of multiple strings delimited by a substring. The behavior matches (string-split str char/pred) exactly. ~ Vijay >From 3c0940082ff49695ac9c2147c900b959be5f8e70 Mon Sep 17

bug#54142: [PATCH] Add srfi 214 - flexvectors

Hello, I've attached a patch that adds support for dynamic vectors, aka flexvectors. Tests and documentation included. ~ Vijay >From 42206dec4d5e9ae51665c6e98ef07715b89b12fe Mon Sep 17 00:00:00 2001 From: Vijay Marupudi Date: Tue, 18 Jan 2022 20:52:08 -0500 Subject: [PATCH] Added srfi-214:

bug#54141: [PATCH] Allow utf[8/16/32]->string functions to take start and ends bounds

Hello, I have attached a patch that extend the bytevector->string functions to take an start and end range. The second patch changes the r7rs compatibility layer to use this new functionality instead of the making a new bytevector as an intermediate step. ~ Vijay >From

bug#52230: 'guild compile' and C(++) extensions (in the context of LilyPond)

Le 19/02/2022 à 22:25, Olivier Dion a écrit : On Sat, 19 Feb 2022, Jean Abou Samra wrote: I had similar problem with Jami. I added C++ primitives to Guile, but these were not load using the foreign function interface. Note, I'm using Guile 3.0.8, but I think the same could be done for

bug#52230: 'guild compile' and C(++) extensions (in the context of LilyPond)

On Sat, 19 Feb 2022, Jean Abou Samra wrote: I had similar problem with Jami. I added C++ primitives to Guile, but these were not load using the foreign function interface. Note, I'm using Guile 3.0.8, but I think the same could be done for Guile 2.0. Basically what I do is to add a `compile`

bug#52230: 'guild compile' and C(++) extensions (in the context of LilyPond)

Hi, (Cross-posted to guile-user, guile-devel and the debbugs issue, I'm unsure where this should go.) In LilyPond, we have now made a development release with binaries using Guile 2.2. However, a major problem is that we don't ship Guile bytecode yet. Notably, one problem to get the bytecode in

bug#53928: [PATCH] Allow null bytes in UNIX sockets.

The current socket address constructors all assume, that there are no null bytes in the socket path. This assumption does not hold in Linux, which uses an initial null byte to demarcate abstract sockets and ignores all further null bytes [1]. [1]

bug#53898: The module name for using weak vectors is undocumented.

Hi, Weak vectors are documented in ‘6.17.3.2 Weak vectors’. However, they are not available from the (guile) module and that section does not document which module needs to be imported to use the procedures from that section. I only found the required module (ice-9 weak-vector) by doing $ git

bug#52835: [PATCH v3] Fix child spawning closing standard fds prematurely

* libguile/posix.c (renumber_file_descriptor): Refactor it as dup_handle_error. (dup_handle_error, dup2_handle_error): New functions that wrap around dup and dup2 by retrying on EINTR or EBUSY, as well as erroring out on other errors. (start_child): Close standard file descriptors only after all

bug#53201: string->uri-reference rejects domain names with final ‘.’

New patch. Now with 3 test cases! -Dale From f4eece6395e75197030bff42a583e847e5a34e15 Mon Sep 17 00:00:00 2001 From: "Dale P. Smith" Date: Thu, 27 Jan 2022 19:20:57 -0500 Subject: [PATCH] Allow trailing "." in urls bug #53201 --- module/web/uri.scm| 17 ++---

bug#53201: string->uri-reference rejects domain names with final ‘.’

Probably not the best fix. Seems to work. Includes a few tests. -Dale diff --git a/module/web/uri.scm b/module/web/uri.scm index 8e0b9bee7..d6758fcc6 100644 --- a/module/web/uri.scm +++ b/module/web/uri.scm @@ -212,7 +212,9 @@ for ‘build-uri’ except there is no scheme." (and (regexp-exec

bug#53408: guile site broken link

Hi ! The link for Greg Badros' Scheme Lecture Notes on the page https://www.gnu.org/software/guile/learn/ is broken. Seems that there's just a / missing : https://courses.cs.washington.edu/courses/cse341/99su/lectures/scheme/ works. Cheers,

bug#52559: guix pull fails with `Unknown command: nix fish: nix show-derivation "~/.fr-sqOEpp/....-module-import-compiled.drv"

reopen 52559 reassign 52559 guile retitle 52559 Guile doesn't like carriage returns after \ thanks (I thought I renamed it already?) Grigory Shepelev schreef op ma 17-01-2022 om 22:37 [+0300]: > Same problem is: > Computing Guix derivation for 'x86_64-linux'... |ice- > 9/read.scm:126:4: In

bug#53244: Guile 3.0.7.14-118ee re-export bug

There's a difference in exports depending on if you use use-modules vs resolve-module. If a module re-exports a function and the module is imported using use-modules the re-export seems to work as one would expect as the documentation says, the cdr of the pair is the member of the public

bug#53201: string->uri-reference rejects domain names with final ‘.’

Guilers, What the subject says :-) Omitting the final dot is optional (and common), not mandatory. scheme@(guile-user)> (string->uri-reference "http://x.org;) $1 = #< … host: "x.org" …> scheme@(guile-user)> (string->uri-reference "http://x.org.;) $2 = #f ; wrong! This actually breaks

bug#53139: sleep return value error, with pselect6 system call on Linux

The sleep procedure is documented as returning 0 when it's not interrupted by a signal. This behaviour seems to have changed in guile as distributed by GNU Guix (see the time-machine commands below). Previously, guile used the select system call which seemed to change the timeout argument to the

bug#53125: Missing abstract unix socket support

Currently, It's not valid to create abstract unix socket in Guile via `connect` procedure, because it rejects string with leading NUL. Abstract unix socket is widely used in D-Bus. Missing support for it hurts the ability of Guile in D-Bus programming. I see a patch at

bug#53023: A global variable gets an erroneous value

This bug is present only in hppa and m68k architectures, see https://buildd.debian.org/status/fetch.php?pkg=theme-d=hppa=4.1.0-2=1640945222=0 and https://buildd.debian.org/status/fetch.php?pkg=theme-d=m68k=4.1.0-2=1640945363=0 The backtrace in those logs contains the following lines: In

bug#52835: [PATCH 0/2] Fix spawning a child not setting standard fds properly

Hello Timothy, Timothy Sample writes: > One thing that ‘renumber_file_descriptor’ does that we seem to be losing > here is error checking. To my eye, the old code will try and warn the > user if they run out of file descriptors, but the new code will not. I may have been too hasty on that

bug#52835: [PATCH 0/2] Fix spawning a child not setting standard fds properly

Hey Josselin, Thanks for finding this bug! I have one concern about your patches: Josselin Poiret writes: > The second patch removes renumber_file_descriptor, as it is no longer > used. One thing that ‘renumber_file_descriptor’ does that we seem to be losing here is error checking. To my

bug#52835: [PATCH v2 2/2] Remove unused renumber_file_descriptor

* libguile/posix.c (renumber_file_descriptor): Remove it. --- Sorry for the noise, but I just saw that this patch omitted a closing */, here is a fixed version. libguile/posix.c | 30 -- 1 file changed, 30 deletions(-) diff --git a/libguile/posix.c b/libguile/posix.c

bug#52835: [PATCH 2/2] Remove unused renumber_file_descriptor

* libguile/posix.c (renumber_file_descriptor): Remove it. --- libguile/posix.c | 32 +--- 1 file changed, 1 insertion(+), 31 deletions(-) diff --git a/libguile/posix.c b/libguile/posix.c index 148ebeb3d..2624d07c2 100644 --- a/libguile/posix.c +++ b/libguile/posix.c

bug#52835: [PATCH 1/2] Fix child spawning closing standard fds prematurely

* libguile/posix.c (start_child): Close standard file descriptors only after all of them have been dup2'd. --- libguile/posix.c | 40 +++- 1 file changed, 19 insertions(+), 21 deletions(-) diff --git a/libguile/posix.c b/libguile/posix.c index

bug#52835: [PATCH 0/2] Fix spawning a child not setting standard fds properly

Hello, While working on the Guix installer (that needs to use system* quite a lot), I've noticed that output/error redirection doesn't behave as intended when combined with system*. Here's a test you can try at home: --8<---cut here---start->8---

bug#52646: GC thread freeze

Hey! > This is happening in this order, right? Right, plus I don't see the following warning: --8<---cut here---start->8--- warning: call to primitive-fork while multiple threads are running; further behavior unspecified. See "Processes" in the

bug#52646: GC thread freeze

Hello! Mathieu Othacehe skribis: > I experiment a strange behaviour with this Guile 3.0.7 process: > https://git.savannah.gnu.org/cgit/guix/guix-cuirass.git/tree/src/cuirass/scripts/remote-worker.scm. > > The process is forking N processes that in turn start 4 threads. This is happening in

bug#52646: GC thread freeze

Hello, I experiment a strange behaviour with this Guile 3.0.7 process: https://git.savannah.gnu.org/cgit/guix/guix-cuirass.git/tree/src/cuirass/scripts/remote-worker.scm. The process is forking N processes that in turn start 4 threads. On aarch64 machines specifically, some of those threads

bug#52482: [PATCH] doc/ref/texinfo.texi: corrected typo generare->generate

Signed-off-by: Blake Shaw --- doc/ref/texinfo.texi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/ref/texinfo.texi b/doc/ref/texinfo.texi index 64a866b23..59df7f8e7 100644 --- a/doc/ref/texinfo.texi +++ b/doc/ref/texinfo.texi @@ -472,7 +472,7 @@ Serialize the stexi

bug#52436: [patch] autoconf: Check for gperf if running from git

Hi, the attached patch adds a check for gperf. From 478eb18a38eb8d15cc96fde148842cf0329bfba5 Mon Sep 17 00:00:00 2001 From: Arne Babenhauserheide Date: Mon, 11 Oct 2021 07:33:23 +0200 Subject: [PATCH] autoconf: Check for gperf if running from git this avoids breaking at runtime. ---

bug#52434: [patch] [doc] add http-request example to web examples

Hi, the attached patch adds a client example that access the hello-world server. From 7d41deec7a550a186fb5e653a7007ae38fd62dbd Mon Sep 17 00:00:00 2001 From: Arne Babenhauserheide Date: Sat, 11 Dec 2021 15:43:21 +0100 Subject: [PATCH] doc: Provide full server<->client example *

bug#52218: Unable to define more than one R7RS library per file

It is the same bug. The r7rs define-library expands to a r6rs library, which expands to a guile module form that lacks support for multiple libraries. The solution would probably include a non-trivial (but probably not very big) change to guile's module system. First you would have to save and

bug#52239: (No Subject)

fixed in https://debbugs.gnu.org/cgi/bugreport.cgi?bug=40252

bug#52239: (No Subject)

I do not reproduce with guile (GNU Guile) 3.0.7.70-6af33

bug#52239: R7RS define-library does not support cond-expand

workaround: #;> find live -type f -exec sh -c "echo \";;; cat {}\"; cat {}" \; ;;; cat live/hello/body.scm (define (hello name) (display "Hello schemer ") (display name) (display "!") (newline)) ;;; cat live/hello.scm (define-library (live hello) (import (scheme base) (scheme

bug#52218: Unable to define more than one R7RS library per file

On 2021-12-01 11:00 +0100, Linus Björnstam wrote: > This is in line with the limitations of guile's R6RS implementation. I > think this is a long standing, low priority bug. It has to do with the > expansion of library (or define-library in this case) not properly > resetting the current module. >

bug#52230: Variable not visible to compiler causes compilation failure when importing module

In experiments with using guild for byte compilation in LilyPond, I am hitting what seems like a bug. When a variable is not defined declaratively in a way that makes the compiler aware of it, the module it is defined in can compile, but modules using it cannot. Here is an example. fail.scm

bug#52218: Unable to define more than one R7RS library per file

Hi Sam! I am not a maintainer , nor a very proficient guile hacker, but I do have some input. This is in line with the limitations of guile's R6RS implementation. I think this is a long standing, low priority bug. It has to do with the expansion of library (or define-library in this case) not

bug#52218: Unable to define more than one R7RS library per file

Guile is unable to handle more than one define-library per file. For example, this is "mylibs.sld", which contains two define-library: (define-library (example mylib) (import (scheme base)) (export f) (begin (define (f x) (+ x 1 (define-library (example main) (import

bug#52112: Compilation of macro returning record type fails

Thanks for your reply. Le 26/11/2021 à 19:54, Taylan Kammer a écrit : I don't think that's supposed to work. I find it surprising that record-type objects are self-evaluating, and in any case they can't be turned into bytecode. Your macro could return code that evaluates to a record

bug#52112: Compilation of macro returning record type fails

On 25.11.2021 18:45, Jean Abou Samra wrote: > Record types are self-evaluating: > > scheme@(guile-user)> (eval (make-record-type 'my-record '()) (current-module)) > $1 = # > > Yet, when a macro returns a record type, Guile fails > to byte-compile the code, even though the execution > (via the

bug#52112: Compilation of macro returning record type fails

Record types are self-evaluating: scheme@(guile-user)> (eval (make-record-type 'my-record '()) (current-module)) $1 = # Yet, when a macro returns a record type, Guile fails to byte-compile the code, even though the execution (via the evaluator) works fine. Here is a simple example:

bug#39947: [3.0.0] Compiler mishandles literal bignums passed to

Related: https://debbugs.gnu.org/50609

bug#50609: number overflow

I've applied the patch in c6b1171c6b5632ac04120f482af786444e17d3fe. Thanks for the report!

bug#48150: breaking gmp

Fixed in bf9d30f3c3fd03cb37d604f1927e9ce5c699338b. Thanks!

bug#50609: number overflow

Not quite comfortable poking around in the compiler, but find a patch attached. I also found this related bug: https://debbugs.gnu.org/cgi/bugreport.cgi?bug=32644. 0001-Limit-the-range-of-ash-round-ash-count-argument-to-I.patch Description: Binary data

bug#45595: recvfrom! optional start and end parameter invalid

Hi, Your patch didn't allow for start == end, which is valid as far as I can tell. With that amended, applied in 1a8294f495cb202f8fcd0f260627c58e7a4c4d10. Thanks!

bug#50127: [PATCH] doc: Fix typos.

Patch applied in c5f443de79d1cb16ddf34873adb185599836a91b. Thanks!

bug#43987: strerror return value

Patch applied in ab9c0c9ca787da60ac9b956fdbcf4c1fa68c3300. Thanks!

bug#48816: guile: Add ARC support

Applied in 492bd638835c4c00623239190032770583e09714. Thanks!

bug#39601: srfi library naming in r7rs

Well, as someone who has written a lot of macros https://srfi.schemers.org/srfi-206/ is a fantastic utility SRFI. It allows several libraries to define the same aux syntax (say like srfi-26's <>) without having collisions. That is not strictly what you were looking for if I understand the

bug#39601: srfi library naming in r7rs

Applied in a960d7869bc82bb56d5446ece01b8efff9b15fef. Thank you!

bug#39601: srfi library naming in r7rs

On 01.11.2021 18:52, lloda wrote: > > Hi Taylan, > > Your patch leaks a bunch of identifiers, could you fix that? > > thanks > > Daniel > Apparently anything defined in boot-9 is implicitly made public in the (guile) module, I wasn't aware of that. Is there a work-around that allows

bug#39601: srfi library naming in r7rs

Hi Taylan, Your patch leaks a bunch of identifiers, could you fix that? thanks Daniel

bug#51445: (no subject)

This is more of a feature request. Apologies if this is an annoying request. So I made a pretty noobie-like mistake a few days ago using (guix records). When one tries to create a , he invariably creates an infinite number of records. The guile compiler eventually runs out of memory and stops

bug#51264: Calling ‘texi-fragment->stexi’ in parallel leads to crashes

Ludovic Courtès skribis: > I just stumbled upon this bug (here I use Guix to feed Texinfo strings > but I suppose we could reduce the test case to be Guix-less.) Here’s a standalone reproducer: (use-modules (texinfo) (sxml simple) (ice-9 threads)) (define sequential?

bug#51276: Problems with format and scaling floats

Nice catch :) On +2021-10-18 17:22:49 -0400, Timothy Sample wrote: > Hi Guilers, > > It turns out there’s a little blunder in ‘format’ (from ‘ice-9’). Look > at what happens when using the SCALE argument to format a fixed-point > float (this is Guile from the Git repo at the time of writing): >

bug#51276: Problems with format and scaling floats

Thank you very much, patch applied! Now that you've dived into (ice-9 format), what about this one: (format #t "~,,2f~%" 0.01) => 1.0 (format #t "~,,3f~%" 0.01) ; !! => 010.0 (format #t "~,,3f~%" 0.001) => 1.0 (format #t "~,,4f~%" 0.001) ; !! => 0010.0 (format #t "~,,4f~%" 0.0001) => 1.0

bug#51276: Problems with format and scaling floats

Hi Guilers, It turns out there’s a little blunder in ‘format’ (from ‘ice-9’). Look at what happens when using the SCALE argument to format a fixed-point float (this is Guile from the Git repo at the time of writing): GNU Guile 3.0.7.6-22120 Copyright (C) 1995-2021 Free Software

bug#51264: Calling ‘texi-fragment->stexi’ in parallel leads to crashes

Hello! I just stumbled upon this bug (here I use Guix to feed Texinfo strings but I suppose we could reduce the test case to be Guix-less.) Test case: (use-modules (guix) (gnu) (texinfo) (ice-9 threads)) (define sequential? (getenv "SEQUENTIAL")) (define (for-each/maybe-parallel

bug#51133: [PATCH 1/1] Tolerate http response line without reason phrase

jaku...@riseup.net writes: > You're right but you didn't address my second point. > The fact that with this patch > > (call-with-input-string "HTTP/1.1 \n" > (lambda (port) (read-response-line port))) I see, my bad, thanks! Please find a newly attached patch. I added a test for such a case,

bug#51133: [PATCH 1/1] Tolerate http response line without reason phrase

You're right but you didn't address my second point. The fact that with this patch (call-with-input-string "HTTP/1.1 \n" (lambda (port) (read-response-line port))) passes the check for 'bad-response error inside read-response-line. It throws 'bad-header-component from non-negative-integer

bug#51133: [PATCH 1/1] Tolerate http response line without reason phrase

On Tue, Oct 12, 2021 at 10:26:22AM +0200, Alexey Abramov via Bug reports for GUILE, GNU's Ubiquitous Extension Language wrote: > Hi, > > I agree that it is not a complient http response, According to The Book [1] ;-) there should be at least one space (SP) (as far as I understand this is really

bug#51133: [PATCH 1/1] Tolerate http response line without reason phrase

Hi, I agree that it is not a complient http response, but unfortunately I came across with some http services (redfish, cimc from Cisco ), where they don't put a reason phrase. As you can see the difference is that response line doesn't have a space after the response code, that is why it raise

bug#51133: [PATCH 1/1] Tolerate http response line without reason phrase

I don't think the reason phrase is optional, even though it can be just a whitespace. Even if I'm not mistaken, however, I don't see the reason for Guile not to be able to parse the status line without a space at the end. Aside from that consider a string "HTTP/1.1 ", which should be a bad

bug#51133: [PATCH 1/1] Tolerate http response line without reason phrase

* module/web/http.scm (read-response-line): Use the end of the string, in case a line doesn't have char-set:whitespace at the end. * test-suite/tests/web-http.test ("read-response-line"): Add test. --- module/web/http.scm| 6 -- test-suite/tests/web-http.test | 2 ++ 2 files

bug#51129: Check for gperf if running from git: this avoids breaking at runtime.

From 0b70a54dd7df0075c392067a7ce17d0f05a59adf Mon Sep 17 00:00:00 2001 From: Arne Babenhauserheide Date: Mon, 11 Oct 2021 07:33:23 +0200 Subject: [PATCH] autoconf: Check for gperf if running from git this avoids breaking at runtime. --- configure.ac | 9 + 1 file changed, 9

bug#45214: guile segfaults on 32-bit big-endian targets

On 10/4/21 18:08, John David Anglin wrote: > On 2021-10-03 5:38 a.m., John Paul Adrian Glaubitz wrote: >> I can confirm that adding "-Oresolve-primitives -Ocps" to GUILE_OPTIMIZATIONS >> in bootstrap/Makefile.am fixes the problem for me on 32-bit PowerPC on >> Debian. > > It also fixes the build

bug#45214: guile segfaults on 32-bit big-endian targets

On 2021-10-03 5:38 a.m., John Paul Adrian Glaubitz wrote: > I can confirm that adding "-Oresolve-primitives -Ocps" to GUILE_OPTIMIZATIONS > in bootstrap/Makefile.am fixes the problem for me on 32-bit PowerPC on Debian. It also fixes the build on Debian hppa. Dave -- John David Anglin

bug#45214: guile segfaults on 32-bit big-endian targets

Hi Efraim! On 4/14/21 16:18, Efraim Flashner wrote: > I spent some more time bisecting guile so I could create a diff against > 3.0.5 and then spent another two weeks or so bootstrapping Guix from my > custom bootstrap binaries out to mesa. > > I've attached the patch that's likely to go into

bug#50730: suspendable-ports.test: test case fails

This bug was originally reported to bug-g...@gnu.org. Copying here. Guix bug report: https://debbugs.gnu.org/cgi/bugreport.cgi?bug=50729 I suspect it may be a bug in my kernel if not a bug in guile itself. The test case fails on both master and v3.0.7 (git). ./build-aux/config.guess

bug#50719: Cannot add a no-applicable-method implementation for a generic with no method

Dear guile, The following program should just return 'foo, but it enters an infinite loop: (use-modules (oop goops)) (define-class ()) (define-method (no-method (generic ) args) 'foo) (define-method (no-applicable-method (generic ) args) 'foo) (define hello (make #:name 'hello))

bug#50609: number overflow

This does not compile on guile 3.0.7, (define (f . l) (let lp2 ((i 0) (s 0) (l l)) (if (and (pair? l) (< i 64)) (lp2 (+ i 1) (if (car l) (logior (ash 1 i) s) s) (cdr l)) s))) While compiling expression: Throw to key `numerical-overflow' with args `("ash" "Numerical ove

bug#50608: Fix for ‘, trace (method ()) --> no applicable method for ...’

Hi guile, Attached is a fix for and a similar issue for 'procedure-name'. Greetings, Maxime. From fe518ed4fb2c7e55f69a229349e3183ccfdcfc97 Mon Sep 17 00:00:00 2001 From: Maxime Devos Date: Wed, 15 Sep 2021 19:57:20 +0200 Subject: [PATCH 1/2]

bug#50608: ,trace (method ()) --> no applicable method for ...

Hi guile, Try running the following from a repl: (use-modules (oop goops)) (method ()) ;; --> $1 = #< () 7fbb1bcb47c0> ,trace (method ()) ---> [...] trace: | | (_ #< 7fbb1bc67c80> (#:specializers () #:formals () #:body ((if #f …)) …)) trace: | | | (struct-ref/unboxed #< 7fbb1bc67c80> 5)

<    2   3   4   5   6   7   8   9   10   11   >