On Mon, Mar 08, 1999 at 03:07:09PM +0100, Reiner Dassing wrote:
> Hello!
> Thank you very much all of you who have given me some hints for solving
> my problem of the compilation of PostgreSQL under Digital UNIX V4.0D
>
> Here are the steps of the successful compilation:
>
> 1. The option "-E" of cc gave the hint that the preprocessor has deleted
> the const attribut of the function "PGconn *PQconnectdb(const char
> *conninfo)"
> in file "src/interfaces/libpq/fe-connect.c"
>
> 2. By using the options "-O4 -std" of cc this effect was gone (this is magic
> to me)
> and it was possible to compile it
There are two ways to explain this "magic":
* The -std flag might cause some preprocessor macro to be set, which
in turn may effect a macro that controls how prototypes are
declared.
* The -std flag could also change the response of the tests done with
the compiler by the configure script, and this, again, could change
the prototype declaration macro.
> The regression tests do show the following results:
> boolean .. ok
> char .. ok
> name .. ok
> varchar .. ok
> text .. ok
> strings .. failed
> int2 .. failed
> int4 .. failed
> int8 .. failed
This looks strange to me -- from replies, which I got from this list
some time ago, I know that a failing int8 test is not necessarily a
problem, but I'd be concerned if int2 and int4 don't work. What are
the differences between the actual and the expected results?
Greetinx, Jan
--
+- Jan T. Kim -------------------------------------------------------+
| email: [EMAIL PROTECTED] |
| WWW: http://www.mpiz-koeln.mpg.de/~kim/ |
*-----=< hierarchical systems are for files, not for humans >=-----*
