On Wed, 16 Dec 2015 23:21:27 +0100, Jonas Maebe
wrote:
>
> As far as I know, Delphi only has errors and warnings. It does not have
> separate categories such as notes and hints, unlike FPC.
>
Delphi does have hints. But no notes.
--
Virgo Pärna
virgo.pa...@mail.ee
__
On 12/16/15, Jonas Maebe wrote:
> As far as I know, Delphi only has errors and warnings. It does not have
> separate categories such as notes and hints, unlike FPC.
I hardly ever use Delphi anymore for many years now, so you might be right.
Let's stop this discussion, it's getting a bit OT and a
On 16/12/15 23:16, Bart wrote:
On 12/16/15, Jonas Maebe wrote:
But D7 does not give me a warning on FillChar().
We don't give a warning either. The category of hints was created
exactly for those messages that have a high chance of being a false
positive.
That should have been:
My D7 does
On 12/16/15, Jonas Maebe wrote:
>> But D7 does not give me a warning on FillChar().
>
> We don't give a warning either. The category of hints was created
> exactly for those messages that have a high chance of being a false
> positive.
That should have been:
My D7 does not give me a warning or h
On 16/12/15 18:14, Bart wrote:
On 12/16/15, Jonas Maebe wrote:
We could, but that would be an extremely ugly hack.
Yes I know it's very ugly.
But D7 does not give me a warning on FillChar().
We don't give a warning either. The category of hints was created
exactly for those messages that
On 12/16/15, Jonas Maebe wrote:
> We could, but that would be an extremely ugly hack.
Yes I know it's very ugly.
But D7 does not give me a warning on FillChar().
But agreed, we should avoid being ugly ;-)
Bart
___
fpc-pascal maillist - fpc-pascal@l
2015-12-16 16:19 GMT+01:00 Jonas Maebe :
>
> Santiago A. wrote on Wed, 16 Dec 2015:
>
> El 16/12/2015 a las 11:55, Jonas Maebe escribió:
>>
>>> while with "out", they are first finalised
>>>
>> Sorry, what does this mean?
>>
>
> http://www.freepascal.org/docs-html/rtl/system/finalize.html
>
Not i
Bart wrote on Wed, 16 Dec 2015:
Can't we hardcode the compiler to not give a hint on FillChar and related?
Especially since these are used to initialize (null) variables.
We could, but that would be an extremely ugly hack. You can always
write and use your own wrapper if you know that you w
On 12/16/15, Jonas Maebe wrote:
> The result parameters can be changed. The BlockRead buffer parameter
> cannot be changed for the same reason that the parameters of fillchar
> and move cannot be changed:
Can't we hardcode the compiler to not give a hint on FillChar and related?
Especially
Santiago A. wrote on Wed, 16 Dec 2015:
El 16/12/2015 a las 11:55, Jonas Maebe escribió:
while with "out", they are first finalised
Sorry, what does this mean?
http://www.freepascal.org/docs-html/rtl/system/finalize.html
Jonas
___
fpc-pascal mail
El 16/12/2015 a las 11:55, Jonas Maebe escribió:
> while with "out", they are first finalised
Sorry, what does this mean?
--
Saludos
Santi
s...@ciberpiula.net
___
fpc-pascal maillist - fpc-pascal@lists.freepascal.org
http://lists.freepascal.org/cgi-
2015-12-16 11:55 GMT+01:00 Jonas Maebe :
>
> Santiago A. wrote on Wed, 16 Dec 2015:
>
> In functions blockread and blockwrite, the result parameter is "var".
>> And in blockread buffer is also "var".
>>
>> Couldn't they be marked as "out" instead of "var"?.
>>
>
> The result parameters can be chan
Santiago A. wrote on Wed, 16 Dec 2015:
In functions blockread and blockwrite, the result parameter is "var".
And in blockread buffer is also "var".
Couldn't they be marked as "out" instead of "var"?.
The result parameters can be changed. The BlockRead buffer parameter
cannot be changed for
In functions blockread and blockwrite, the result parameter is "var".
And in blockread buffer is also "var".
Couldn't they be marked as "out" instead of "var"?.
It's not an aesthetic or academic change, I would save a lot of "not
initialized" warnings that flood the message window.
Of course I
14 matches
Mail list logo
