On 2024-04-18 01:36, Bruno Haible wrote:
> Thanks for the report. In fact, many .h files in Gnulib still are not
> C++-safe.
> Instead of fixing it one by one, let me do it for all (except the two
> modules 'glibc-internal/scratch_buffer' and 'glibc-internal/dynarray', for
> which the priority is
In the specification of totalorder* [1] we read:
"in the following order:
- negative quiet NaNs, in order of decreasing payload;
- negative signaling NaNs, in order of decreasing payload;
- negative infinity;
- finite numbers, in ascending order, with negative zero before positi
Hi Bruno,
On 4/17/24 3:18 PM, Bruno Haible wrote:
> In this example, the property 'attribute' is used outside of the class,
> that is, it is public. (In the OO world, the concept of "class" and of
> "private"/"public" are tightly related, because one of the main purposes
> of a class is to hide im
The module 'nan' (previously tests/nan.h, now lib/nan.h) defines functions
for getting a NaN value reliably. This module is, so far, under GPL,
because historically only the tests use it.
I would like to use it under LGPLv2+, for the modules 'setpayload*'
and 'getpayloadl' (functions defined by IS
These two commands differ in the style of warning output:
$ ./gnulib-tool.sh --create-testdir --dir=../testdir2 --single-configure
getpayloadl
gnulib-tool: warning: module getpayloadl depends on a module with an
incompatible license: nan
gnulib-tool: warning: module getpayloadl depends on a modu
Hi,
Miro Palmu wrote:
> * lib/full-read.h: Add extern "C" for C++.
Thanks for the report. In fact, many .h files in Gnulib still are not C++-safe.
Instead of fixing it one by one, let me do it for all (except the two
modules 'glibc-internal/scratch_buffer' and 'glibc-internal/dynarray', for
which
Hi Collin,
Both patches applied.
> The reason I dislike defining instance variables outside of __init__
> is that you can run into situations like this, which I find hard to
> follow:
>
> var = GLImport(...) # self.assistant does not exist.
> var.assistant.do_something() # Thi
Hi Bruno,
On 4/17/24 11:02 AM, Bruno Haible wrote:
>> $ pylint *.py | grep W0201
>> GLError.py:119:12: W0201: Attribute 'message' defined outside __init__
>> (attribute-defined-outside-init)
>
> This warning is bogus. In a dynamic language like Python, it is
> perfectly fine to have a property u
Hi Collin,
> $ pylint *.py | grep W0201
> GLError.py:119:12: W0201: Attribute 'message' defined outside __init__
> (attribute-defined-outside-init)
This warning is bogus. In a dynamic language like Python, it is
perfectly fine to have a property used only by one method. Only if
it is used by mor
Same what commit a0b977a430af88ff58d9596a60670936eb7f1d31
does for module full-write, but for full-read.
* lib/full-read.h: Add extern "C" for C++.
Copyright-paperwork-exempt: Yes
---
ChangeLog | 4
lib/full-read.h | 11 +++
2 files changed, 15 insertions(+)
diff --git a/C
Hi Collin,
> This will require some code reordering first though. The
> GLFileAssistant used by GLImport is not defined until
> GLImport.execute(). The file name transformations are done in
> GLImport.prepare().
Good point.
> I think the correct solution there is to define the GLFileAssistant in
ISO C 23 specifies in the (optional, but normative) Annex F also functions
for extracting the payload of a quiet or signalling NaN. This patch series
implements them.
2024-04-17 Bruno Haible
getpayloadl: Add tests.
* tests/test-getpayloadl.c: New file.
* modules/getpay
12 matches
Mail list logo
