Is it possible to extend an enumeration, while preserving strong type
checking? In other words, given existing code like
typeTUpstreamServerCapabilitiesBase= (uscConnected, uscCanIssueGUID);
TLocalServerCapabilitiesBase= (lscConnected, lscCanIssueGUID,
lscIsFirebird);
TUps
It is not possible to extend an enumeration type. If you are not
willing to use separate sets to hold common and extended elements, you
can use a generic (not parametric) set type, say SET OF 0..31, that
can hold all possible elements, and define the possible elements as
constants.
On Fri, Dec 28
On 27.12.2012 17:11, Patrick wrote:
>
>> This is exactly what I thought. You are trying to use the GTK1 unit.
>> The unit for GTK+ is called "gtk2". Take a look into "packages/gtk"
>> and "packages/gtk2" of a Free Pascal source folder to see the
>> differences in the units (just look at the *.pas/
On 28.12.2012 05:47, Patrick wrote:
Free Pascal Compiler version 2.6.0 [2011/12/23] for x86_64
Copyright (c) 1993-2011 by Florian Klaempfl and others
Target OS: Linux for x86-64
Compiling helloworld.pas
helloworld.pas(28,11) Error: Incompatible types: got "Boolean32"
expected "LongInt"
helloworld
On 28/12/2012 11:42, Mark Morgan Lloyd wrote:
Is it possible to extend an enumeration, while preserving strong type
checking? In other words, given existing code like
typeTUpstreamServerCapabilitiesBase= (uscConnected, uscCanIssueGUID);
TLocalServerCapabilitiesBase= (lscConnected, l
JC Chu wrote:
It is not possible to extend an enumeration type. If you are not
willing to use separate sets to hold common and extended elements, you
can use a generic (not parametric) set type, say SET OF 0..31, that
can hold all possible elements, and define the possible elements as
constants.
Ludo Brands wrote:
On 28/12/2012 11:42, Mark Morgan Lloyd wrote:
Is it possible to extend an enumeration, while preserving strong type
checking? In other words, given existing code like
typeTUpstreamServerCapabilitiesBase= (uscConnected, uscCanIssueGUID);
TLocalServerCapabilitiesBa
On 28/12/2012 12:48, Mark Morgan Lloyd wrote:
Ludo Brands wrote:
type
TAllServerCapabilitiesBase=(uscConnected,
uscCanIssueGUID,lscConnected, lscCanIssueGUID, lscIsFirebird);
TUpstreamServerCapabilitiesBase= uscConnected .. uscCanIssueGUID;
TLocalServerCapabilitiesBase= lscConnected..
On 28/12/2012 12:59, Ludo Brands wrote:
On 28/12/2012 12:48, Mark Morgan Lloyd wrote:
Ludo Brands wrote:
type
TAllServerCapabilitiesBase=(uscConnected,
uscCanIssueGUID,lscConnected, lscCanIssueGUID, lscIsFirebird);
TUpstreamServerCapabilitiesBase= uscConnected .. uscCanIssueGUID;
TLo
In our previous episode, Sven Barth said:
> > Fatal: Compilation aborted
> > Error: /usr/bin/ppcx64 returned an error exitcode (normal if you did not
> > specify a source file to be compiled)
>
> How does your helloworld.pas look like? It's hard to guess what's wrong
> without seeing the code...
If you do not allow clients of your unit to define new elements or
change the set values, and what you want to do is prevent, say,
accidentally adding an extended element to the set in the base class,
you can declare extended elements as class-local constants. Otherwise
you will need to add checks
Sven Barth wrote:
On 28.12.2012 05:47, Patrick wrote:
Free Pascal Compiler version 2.6.0 [2011/12/23] for x86_64
Copyright (c) 1993-2011 by Florian Klaempfl and others
Target OS: Linux for x86-64
Compiling helloworld.pas
helloworld.pas(28,11) Error: Incompatible types: got "Boolean32"
expected "
On Fri, Dec 28, 2012 at 6:47 AM, Patrick wrote:
> I am also finding the site confusing. If I click on AMD64/x86_64 on the
> binary download page and then choose sourceforge, I end up being offered a
> package that says fpc-2.6.0.i386-linux.tar.
>
> I did find another link and installed the compi
Ludo Brands wrote:
On 28/12/2012 12:59, Ludo Brands wrote:
On 28/12/2012 12:48, Mark Morgan Lloyd wrote:
Ludo Brands wrote:
type
TAllServerCapabilitiesBase=(uscConnected,
uscCanIssueGUID,lscConnected, lscCanIssueGUID, lscIsFirebird);
TUpstreamServerCapabilitiesBase= uscConnected .. usc
On 12/28/2012 08:39 AM, Juha Manninen wrote:
On Fri, Dec 28, 2012 at 6:47 AM, Patrick wrote:
I am also finding the site confusing. If I click on AMD64/x86_64 on the
binary download page and then choose sourceforge, I end up being offered a
package that says fpc-2.6.0.i386-linux.tar.
I di
Hope I am not wearing out the warm welcome I have received
I installed the 64 bit .deb package from the website. I am still getting
a lot of errors. I am starting to wonder if I have borked my install. I
am on Debian stable right now. I can move to Mint Mate desktop...
At the end of the e
On 28.12.2012 19:04, Patrick wrote:
However I am delighted to see a lua package in there. I have more
experience with lua. Make is running without errors but I expected to
see a liblua.a archive file after and did not see it, there were more
warnings. (I have them at the end of the email)
You w
Could you try to compile the example using "-st" and then run the
generated ppas.sh script and report it's output here? I hope to get
more error information from the linker this way...
Regards,
Sven
___
fpc-pascal maillist - fpc-pascal@lists.freep
On 28.12.2012 22:53, Patrick wrote:
Ah ! the liblua.a file was there at the start and then after make clean,
make it was not. Thanks !
The liblua.a file is not necessarily what you think it is. On *nix
platforms the .a files are used for smartlinking, not for linking
against something.
I
19 matches
Mail list logo
