On Thu, May 22, 2014 at 12:04 AM, Paul Eggert wrote:
> Jim Meyering wrote:
>>
>> What about the conflict with modules/streq?
>> There are over 300 uses of STREQ in gnulib and coreutils.
>> Are you suggesting to change all of those?
>
>
> Yes. But if this is too much all at once, let's just call t
Jim Meyering wrote:
What about the conflict with modules/streq?
There are over 300 uses of STREQ in gnulib and coreutils.
Are you suggesting to change all of those?
Yes. But if this is too much all at once, let's just call the inline
function 'STREQ'. We can rename it to 'streq' later, if an
On Wed, May 21, 2014 at 9:54 PM, Paul Eggert wrote:
> Kieran Colford wrote:
>>
>> I deliberately chose to use it as a macro and call it this because that
>> is how it is done everywhere else in gnulib.
>
>
> Sure, but let's take this opportunity to do a better API, while we're making
> a module fo
On 14-05-22 12:54 AM, Paul Eggert wrote:
> Kieran Colford wrote:
>> I deliberately chose to use it as a macro and call it this because that
>> is how it is done everywhere else in gnulib.
>
> Sure, but let's take this opportunity to do a better API, while we're
> making a module for it. So my s
Kieran Colford wrote:
I deliberately chose to use it as a macro and call it this because that
is how it is done everywhere else in gnulib.
Sure, but let's take this opportunity to do a better API, while we're
making a module for it. So my suggestion would be to replace STREQ with
streq unifo
On 05/21/2014 10:31 PM, Eric Blake wrote:
> On 05/21/2014 09:33 PM, Kieran Colford wrote:
>> ---
>> lib/stringops.h | 31 +++
>> modules/stringops | 22 ++
>> 2 files changed, 53 insertions(+)
>> create mode 100644 lib/stringops.h
>> create m
On 05/21/2014 09:33 PM, Kieran Colford wrote:
> ---
> lib/stringops.h | 31 +++
> modules/stringops | 22 ++
> 2 files changed, 53 insertions(+)
> create mode 100644 lib/stringops.h
> create mode 100644 modules/stringops
Your commit message
On 14-05-22 12:28 AM, Eric Blake wrote:
> On 05/21/2014 10:14 PM, Paul Eggert wrote:
>> Kieran Colford wrote:
>>> +#define STRNEQ(X, Y) (strcmp (X, Y) != 0)
>>
>> 1. This name is poorly chosen (it looks too much like "strncmp", which
>> means something quite different) and it's not needed (people
On 05/21/2014 10:14 PM, Paul Eggert wrote:
> Kieran Colford wrote:
>> +#define STRNEQ(X, Y) (strcmp (X, Y) != 0)
>
> 1. This name is poorly chosen (it looks too much like "strncmp", which
> means something quite different) and it's not needed (people can just
> use "!STREQ").
>
> 2. STREQ shoul
Kieran Colford wrote:
+#define STRNEQ(X, Y) (strcmp (X, Y) != 0)
1. This name is poorly chosen (it looks too much like "strncmp", which
means something quite different) and it's not needed (people can just
use "!STREQ").
2. STREQ should be an inline function, not a macro. There's little
---
lib/stringops.h | 31 +++
modules/stringops | 22 ++
2 files changed, 53 insertions(+)
create mode 100644 lib/stringops.h
create mode 100644 modules/stringops
diff --git a/lib/stringops.h b/lib/stringops.h
new file mode 100644
index
11 matches
Mail list logo
