Thread: reference problem of manifest.(win32.mak of libpq.dll)

reference problem of manifest.(win32.mak of libpq.dll)

From
"Hiroshi Saito"
Date:
Hi Magnus, and Dave.

It seems that it is different in nmake although the initial value of
VisualStudio is embedding. Then,  It sees a reference problem by
the dll independent. Therefore, embedding considers like an ideal.

Please take this into consideration.

Regards,
Hiroshi Saito

Attachment

Re: reference problem of manifest.(win32.mak of libpq.dll)

From
Magnus Hagander
Date:
On Wed, Jan 09, 2008 at 04:01:20PM +0900, Hiroshi Saito wrote:
> Hi Magnus, and Dave.
>
> It seems that it is different in nmake although the initial value of
> VisualStudio is embedding. Then,  It sees a reference problem by
> the dll independent. Therefore, embedding considers like an ideal.
>
> Please take this into consideration.

+ !IF "$(_NMAKE_VER)" != "6.00.9782.0"

I don't think that's safe. We need to do a range check. Perhaps check if
vesion is >= 7.0 or something?

There can be other 6.00.<something> versions, no? Say with different
servicepacks or something?

//Magnus

Re: reference problem of manifest.(win32.mak of libpq.dll)

From
"Hiroshi Saito"
Date:
Hi.

----- Original Message -----
From: "Magnus Hagander" <magnus@hagander.net>


> On Wed, Jan 09, 2008 at 04:01:20PM +0900, Hiroshi Saito wrote:
>> Hi Magnus, and Dave.
>>
>> It seems that it is different in nmake although the initial value of
>> VisualStudio is embedding. Then,  It sees a reference problem by
>> the dll independent. Therefore, embedding considers like an ideal.
>>
>> Please take this into consideration.
>
> + !IF "$(_NMAKE_VER)" != "6.00.9782.0"
>
> I don't think that's safe. We need to do a range check. Perhaps check if
> vesion is >= 7.0 or something?
>
> There can be other 6.00.<something> versions, no? Say with different
> servicepacks or something?

Condition understanding of '>=' is not made as for namke of VC6 to a regrettable
thing, but it causes an error to it....:-(
So, except all  thought that it was good.

Regards,
Hiroshi Saito




Re: reference problem of manifest.(win32.mak of libpq.dll)

From
Magnus Hagander
Date:
On Thu, Jan 10, 2008 at 11:52:15PM +0900, Hiroshi Saito wrote:
> Hi.
>
> ----- Original Message -----
> From: "Magnus Hagander" <magnus@hagander.net>
>
>
> >On Wed, Jan 09, 2008 at 04:01:20PM +0900, Hiroshi Saito wrote:
> >>Hi Magnus, and Dave.
> >>
> >>It seems that it is different in nmake although the initial value of
> >>VisualStudio is embedding. Then,  It sees a reference problem by
> >>the dll independent. Therefore, embedding considers like an ideal.
> >>
> >>Please take this into consideration.
> >
> >+ !IF "$(_NMAKE_VER)" != "6.00.9782.0"
> >
> >I don't think that's safe. We need to do a range check. Perhaps check if
> >vesion is >= 7.0 or something?
> >
> >There can be other 6.00.<something> versions, no? Say with different
> >servicepacks or something?
>
> Condition understanding of '>=' is not made as for namke of VC6 to a
> regrettable thing, but it causes an error to it....:-(
> So, except all  thought that it was good.

Hmm. Crap.
Perhaps there's something else we can check on? Like a feature or
environment variable only present in newer versions or something?

/Magnus


Re: reference problem of manifest.(win32.mak of libpq.dll)

From
"Hiroshi Saito"
Date:
From: "Magnus Hagander" <magnus@hagander.net>

>> Condition understanding of '>=' is not made as for namke of VC6 to a
>> regrettable thing, but it causes an error to it....:-(
>> So, except all  thought that it was good.
>
> Hmm. Crap.
> Perhaps there's something else we can check on? Like a feature or
> environment variable only present in newer versions or something?

I don't think of a good idea. However, since our document has announced
officially, saying >=VC7.1. Therefore, I think that it is satisfactory.

Regards,
Hiroshi Saito

Re: reference problem of manifest.(win32.mak of libpq.dll)

From
Magnus Hagander
Date:
On Fri, Jan 11, 2008 at 12:15:53AM +0900, Hiroshi Saito wrote:
> From: "Magnus Hagander" <magnus@hagander.net>
>
> >>Condition understanding of '>=' is not made as for namke of VC6 to a
> >>regrettable thing, but it causes an error to it....:-(
> >>So, except all  thought that it was good.
> >
> >Hmm. Crap.
> >Perhaps there's something else we can check on? Like a feature or
> >environment variable only present in newer versions or something?
>
> I don't think of a good idea. However, since our document has announced
> officially, saying >=VC7.1. Therefore, I think that it is satisfactory.

Ah, good point, I forgot about that.

But - if we do that, why do we need that IF check *at all*?

//Magnus

Re: reference problem of manifest.(win32.mak of libpq.dll)

From
"Hiroshi Saito"
Date:
From: "Magnus Hagander" <magnus@hagander.net>

>> I don't think of a good idea. However, since our document has announced
>> officially, saying >=VC7.1. Therefore, I think that it is satisfactory.
>
> Ah, good point, I forgot about that.
>
> But - if we do that, why do we need that IF check *at all*?

It is because it is saved by it although VC6 is private. probably, present all will
be possible by it. I feel that it is better than exclusion for it.

Regards,
Hiroshi Saito

Re: reference problem of manifest.(win32.mak of libpq.dll)

From
Magnus Hagander
Date:
On Fri, Jan 11, 2008 at 12:29:53AM +0900, Hiroshi Saito wrote:
> From: "Magnus Hagander" <magnus@hagander.net>
>
> >>I don't think of a good idea. However, since our document has announced
> >>officially, saying >=VC7.1. Therefore, I think that it is satisfactory.
> >
> >Ah, good point, I forgot about that.
> >
> >But - if we do that, why do we need that IF check *at all*?
>
> It is because it is saved by it although VC6 is private. probably, present
> all will be possible by it. I feel that it is better than exclusion for it.

Ok. Applied.

//Magnus