Nabble - GNADE - Dev

Re: GNADE

2008-07-01T14:47:01Z

I prefer to write some code or do some design. Is there anything that needs to be coded or designed?

Andrew Carroll
OSU Software Services
405-744-7559
andrew.carroll@...

-----Original Message-----
From: Michael Erdmann [mailto:michael.erdmann@...]
Sent: Tuesday, July 01, 2008 12:48 PM
To: Carroll, Andrew
Cc: gnade-develop@...
Subject: Re: [Gnade-develop] GNADE

Carroll, Andrew wrote:
> I am interested in contributing. Are you hiring?
>
Contribution is fine, but i am not hiring any people. There is absolute
no money in here!

Is there any specific contribution you are thingking about?

Regards
M.Erdmann

> Andrew Carroll
> OSU Software Services
> 405-744-7559
> andrew.carroll@...
>
>
>
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://sourceforge.net/services/buy/index.php
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>

-------------------------------------------------------------------------
Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW!
Studies have shown that voting for your favorite open source project,
along with a healthy diet, reduces your potential for chronic lameness
and boredom. Vote Now at http://www.sourceforge.net/community/cca08
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: GNADE

2008-07-01T10:48:24Z

Carroll, Andrew wrote:
> I am interested in contributing. Are you hiring?
>
Contribution is fine, but i am not hiring any people. There is absolute
no money in here!

Is there any specific contribution you are thingking about?

Regards
M.Erdmann

GNADE

2008-07-01T06:22:07Z

I am interested in contributing. Are you hiring?

Andrew Carroll
OSU Software Services
405-744-7559
andrew.carroll@...

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: Why AUnit 1.xx Series ?

2008-02-27T21:51:44Z

Done; but i have not yet tested it.

Michael Erdmann wrote:

> Ups,
>
> I will get the latest vrsion from there and merge it in!
>
> i have an account there, no Problem :-)
>
> Michael
>
>
> Per Sandberg wrote:
>> Michael
>> That page is rather old but her is a full link
>> https://libre.adacore.com/
>> GNAT GPL Edition
>> Download GNAT GPL Edition
>> aunit
>> aunit-gpl-2.01-1-src.tgz
>>
>> Note you need to login first
>> /Per
>>
>>
>> Michael Erdmann wrote:
>>> Strange,
>>>
>>> i googled a bit and a found the home location of aunit in the net
>>>
>>> https://libre.adacore.com/aunit/main.html
>>>
>>> there is only version 1.03?!
>>>
>>> Where is the 2.xx version in the net?
>>>
>>>
>>> Michael
>>>
>>>
>>> Per Sandberg wrote:
>>>
>>>> Michael,
>>>> Why AUnit 1. series ?? Since AUnit 2.xx is out and way not just save
>>>> the original .tar.gz file and say that AUNit shall be installed.
>>>> Then have the makefile contain something like
>>>>
>>>> install:
>>>> tar -xzf aunit-xxx.tgz
>>>> If necessary apply patches
>>>> Make -C aunit-xxx
>>>> Make -C aunit-xxx install
>>>>
>>>> And have the installation of Aunit as a very optional task that could be
>>>> performed if AUnit isent installed (and AUnit 2.x is part of GNAT-GPL 2007)
>>>>
>>>> /Per
>>>>
>>>>
>>>>
>>>>
>>> -------------------------------------------------------------------------
>>> This SF.net email is sponsored by: Microsoft
>>> Defy all challenges. Microsoft(R) Visual Studio 2008.
>>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>>> _______________________________________________
>>> Gnade-develop mailing list
>>> http://gnade.sourceforge.net/
>>> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>>>
>>>
>> -------------------------------------------------------------------------
>> This SF.net email is sponsored by: Microsoft
>> Defy all challenges. Microsoft(R) Visual Studio 2008.
>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>> _______________________________________________
>> Gnade-develop mailing list
>> http://gnade.sourceforge.net/
>> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: Why AUnit 1.xx Series ?

2008-02-03T03:18:42Z

Ups,

I will get the latest vrsion from there and merge it in!

i have an account there, no Problem :-)

Michael

Per Sandberg wrote:

> Michael
> That page is rather old but her is a full link
> https://libre.adacore.com/
> GNAT GPL Edition
> Download GNAT GPL Edition
> aunit
> aunit-gpl-2.01-1-src.tgz
>
> Note you need to login first
> /Per
>
>
> Michael Erdmann wrote:
>> Strange,
>>
>> i googled a bit and a found the home location of aunit in the net
>>
>> https://libre.adacore.com/aunit/main.html
>>
>> there is only version 1.03?!
>>
>> Where is the 2.xx version in the net?
>>
>>
>> Michael
>>
>>
>> Per Sandberg wrote:
>>
>>> Michael,
>>> Why AUnit 1. series ?? Since AUnit 2.xx is out and way not just save
>>> the original .tar.gz file and say that AUNit shall be installed.
>>> Then have the makefile contain something like
>>>
>>> install:
>>> tar -xzf aunit-xxx.tgz
>>> If necessary apply patches
>>> Make -C aunit-xxx
>>> Make -C aunit-xxx install
>>>
>>> And have the installation of Aunit as a very optional task that could be
>>> performed if AUnit isent installed (and AUnit 2.x is part of GNAT-GPL 2007)
>>>
>>> /Per
>>>
>>>
>>>
>>>
>>
>> -------------------------------------------------------------------------
>> This SF.net email is sponsored by: Microsoft
>> Defy all challenges. Microsoft(R) Visual Studio 2008.
>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>> _______________________________________________
>> Gnade-develop mailing list
>> http://gnade.sourceforge.net/
>> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>>
>>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: Why AUnit 1.xx Series ?

2008-02-03T02:36:41Z

Michael
That page is rather old but her is a full link
https://libre.adacore.com/
GNAT GPL Edition
Download GNAT GPL Edition
aunit
aunit-gpl-2.01-1-src.tgz

Note you need to login first
/Per

Michael Erdmann wrote:

> Strange,
>
> i googled a bit and a found the home location of aunit in the net
>
> https://libre.adacore.com/aunit/main.html
>
> there is only version 1.03?!
>
> Where is the 2.xx version in the net?
>
>
> Michael
>
>
> Per Sandberg wrote:
>
>> Michael,
>> Why AUnit 1. series ?? Since AUnit 2.xx is out and way not just save
>> the original .tar.gz file and say that AUNit shall be installed.
>> Then have the makefile contain something like
>>
>> install:
>> tar -xzf aunit-xxx.tgz
>> If necessary apply patches
>> Make -C aunit-xxx
>> Make -C aunit-xxx install
>>
>> And have the installation of Aunit as a very optional task that could be
>> performed if AUnit isent installed (and AUnit 2.x is part of GNAT-GPL 2007)
>>
>> /Per
>>
>>
>>
>>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>
>

Re: Why AUnit 1.xx Series ?

2008-02-03T02:27:24Z

Strange,

i googled a bit and a found the home location of aunit in the net

https://libre.adacore.com/aunit/main.html

there is only version 1.03?!

Where is the 2.xx version in the net?

Michael

Per Sandberg wrote:

> Michael,
> Why AUnit 1. series ?? Since AUnit 2.xx is out and way not just save
> the original .tar.gz file and say that AUNit shall be installed.
> Then have the makefile contain something like
>
> install:
> tar -xzf aunit-xxx.tgz
> If necessary apply patches
> Make -C aunit-xxx
> Make -C aunit-xxx install
>
> And have the installation of Aunit as a very optional task that could be
> performed if AUnit isent installed (and AUnit 2.x is part of GNAT-GPL 2007)
>
> /Per
>
>
>

Why AUnit 1.xx Series ?

2008-02-02T22:55:00Z

Michael,
Why AUnit 1. series ?? Since AUnit 2.xx is out and way not just save
the original .tar.gz file and say that AUNit shall be installed.
Then have the makefile contain something like

install:
tar -xzf aunit-xxx.tgz
If necessary apply patches
Make -C aunit-xxx
Make -C aunit-xxx install

And have the installation of Aunit as a very optional task that could be
performed if AUnit isent installed (and AUnit 2.x is part of GNAT-GPL 2007)

/Per

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: GNADE 2.0

2008-02-01T07:42:59Z

Jérôme Roussel wrote:

>>> Concerning the regression tests, that's a very good thing to have them
>>> now. It should be indicated somewhere in the README file that Aunit is
>>> mandatory (and which version) or a test should be added in the Makefile
>>> structure to de/active the regression tests depending of Aunit
>> detection.
>>
>> My intention is to add a vendor brach with the latest Version in
>> the svn. The result will be that we will use always the same Aunit
>> Version. Any objections against this approach?
>>
>> The production version will not include this SW.
>
> AUnit does not seem to change very often so that may be a good idea,

The main idea is that requiered changes to products e.g. makefile
or missing features are more easy to maintain.

Michael

>
> Jérôme
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: GNADE 2.0

2008-01-30T21:55:16Z

Per Sandberg a écrit :

> Jérôme Roussel wrote:
>>>> Concerning the regression tests, that's a very good thing to have them
>>>> now. It should be indicated somewhere in the README file that Aunit is
>>>> mandatory (and which version) or a test should be added in the
>>>> Makefile
>>>> structure to de/active the regression tests depending of Aunit
>>>>
>>> detection.
>>> I think it is enough to just update the README file and then
>>> avoid using
>>> "make test" if not AUnit is in place so I just updated the file
>>> since it
>>> should be there any way.
>>>
>>> /Thanks
>>> /Per
>>>
>>>
>>
>> Ok, up to now, I have mostly used "make", and not "make compile" then
>> "make
>> test". Do you think that the regression tests should be kept as part
>> of the
>> "make" target? I don't know what the standard is but in most projects,
>> "make" is used just to compile so is synonym of "make compile", right ?
>>
>> Thanks,
>>
>> Jérôme
>>
>>
>
> Jérôme
>
> Concerning the default target all or compile I got no preferences at
> all so I am happy ether way so why not change in the following way
> Default target: compile
> and all => compile install
> Apart from that keep it as is.
> /Thanks
> /Per

Ok, I have done the following change:
- Default target => compile
- all => compile test install

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: GNADE 2.0

2008-01-30T11:12:42Z

>> Concerning the regression tests, that's a very good thing to have them
>> now. It should be indicated somewhere in the README file that Aunit is
>> mandatory (and which version) or a test should be added in the Makefile
>> structure to de/active the regression tests depending of Aunit
> detection.
>
> My intention is to add a vendor brach with the latest Version in
> the svn. The result will be that we will use always the same Aunit
> Version. Any objections against this approach?
>
> The production version will not include this SW.

AUnit does not seem to change very often so that may be a good idea,

Jérôme

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: GNADE 2.0

2008-01-30T10:28:17Z

Jérôme Roussel wrote:

> Hi,
>
> Per Sandberg a écrit :
>> Michael Erdmann wrote:
>>
>>> Per Sandberg wrote:
>>>
>>>> The GNADE 2.0 tree is now getting to a point where it is in
>>>> presentable shape, there are still
>>>> outstanding issues but at least it builds out of the box on both on
>>>> Linux and win32
>>>>
> It happens very often that compilation out of the box of a fresh svn
> checkout does not work anymore... We should try to be very careful about
> the state of the svn repository.
>
> Today for instance, it just stopped on my computer while trying to
> compile esql. I don't know if the fix is correct/enough but it seemed
> wrong that the NAME variable was not defined in src/esql/Makefile, so I
> have just added it.
>
> I have also applied a patch to correct the detection of the value of
> IS_WORKING in Makefile.conf. It was not working and PREFIX was not set
> correctly then.
>
> Concerning the regression tests, that's a very good thing to have them
> now. It should be indicated somewhere in the README file that Aunit is
> mandatory (and which version) or a test should be added in the Makefile
> structure to de/active the regression tests depending of Aunit detection.

My intention is to add a vendor brach with the latest Version in
the svn. The result will be that we will use always the same Aunit
Version. Any objections against this approach?

The production version will not include this SW.

I will do my best to rework the odbc section for regression by end of
this weekend :-)

>
> Thanks,
>
> Jérôme
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: GNADE 2.0

2008-01-29T22:33:21Z

Jérôme Roussel wrote:

> Hi,
>
> Per Sandberg a écrit :
>
>> Michael Erdmann wrote:
>>
>>
>>> Per Sandberg wrote:
>>>
>>>
>>>
>>>> The GNADE 2.0 tree is now getting to a point where it is in presentable
>>>> shape, there are still
>>>> outstanding issues but at least it builds out of the box on both on
>>>> Linux and win32
>>>>
>>>>
> It happens very often that compilation out of the box of a fresh svn
> checkout does not work anymore... We should try to be very careful about
> the state of the svn repository.
>

Yes agree on that, and now since GNADE is getting closer to a shape
where we may think of doing a release and the infra structure and the
regression tests are in place it makes sense to do a
"make clean", "make compile", "make test" before commit.

I maybe should have made the above statement "builds out of the box most
of the times".
> Today for instance, it just stopped on my computer while trying to
> compile esql. I don't know if the fix is correct/enough but it seemed
> wrong that the NAME variable was not defined in src/esql/Makefile, so I
> have just added it.
>
>
The fix is correct, I also added a regression for esql and then realized
that the project file in esql needs a bit of refactoring, to get the
support packages to the installation.

> I have also applied a patch to correct the detection of the value of
> IS_WORKING in Makefile.conf. It was not working and PREFIX was not set
> correctly then.
>
> Concerning the regression tests, that's a very good thing to have them
> now. It should be indicated somewhere in the README file that Aunit is
> mandatory (and which version) or a test should be added in the Makefile
> structure to de/active the regression tests depending of Aunit detection.
>
I think it is enough to just update the README file and then avoid using
"make test" if not AUnit is in place so I just updated the file since it
should be there any way.
> Thanks,
>
> Jérôme
>
>
/Thanks
/Per

> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: GNADE 2.0

2008-01-29T21:06:10Z

Hi,

Per Sandberg a écrit :

> Michael Erdmann wrote:
>
>> Per Sandberg wrote:
>>
>>
>>> The GNADE 2.0 tree is now getting to a point where it is in presentable
>>> shape, there are still
>>> outstanding issues but at least it builds out of the box on both on
>>> Linux and win32
>>>

It happens very often that compilation out of the box of a fresh svn
checkout does not work anymore... We should try to be very careful about
the state of the svn repository.

Today for instance, it just stopped on my computer while trying to
compile esql. I don't know if the fix is correct/enough but it seemed
wrong that the NAME variable was not defined in src/esql/Makefile, so I
have just added it.

I have also applied a patch to correct the detection of the value of
IS_WORKING in Makefile.conf. It was not working and PREFIX was not set
correctly then.

Concerning the regression tests, that's a very good thing to have them
now. It should be indicated somewhere in the README file that Aunit is
mandatory (and which version) or a test should be added in the Makefile
structure to de/active the regression tests depending of Aunit detection.

Thanks,

Jérôme

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: GNADE 2.0

2008-01-23T08:11:14Z

Michael Erdmann wrote:

> Per Sandberg wrote:
>
>> The GNADE 2.0 tree is now getting to a point where it is in presentable
>> shape, there are still
>> outstanding issues but at least it builds out of the box on both on
>> Linux and win32
>>
>
> The configure stuff does not work any more. To my knowledge there are
> no operating depedencies in the ODBC and the native drivers and neither
> in gesql. In GANDE 1.x is basically used to guess some installtion
> paramters which do depend on the operating system.
>
> I don't think that is is needed any more?! For the time beeing i would
> propose to delete the stuff.
>

Yes all configure is taken care of i Makefile.conf and since there
shouldn't be any differences between gnat hosts and i don't think anyone
is considering a cross build I am 99% shure that the
build model will work.
But of course it has to be verified for all hosts and databases an I am
only using odbc and SQLite on win32.

>> /Per
>>
>> -------------------------------------------------------------------------
>> This SF.net email is sponsored by: Microsoft
>> Defy all challenges. Microsoft(R) Visual Studio 2008.
>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>> _______________________________________________
>> Gnade-develop mailing list
>> http://gnade.sourceforge.net/
>> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>>
>>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>
>

Re: GNADE 2.0

2008-01-20T08:15:29Z

Per Sandberg wrote:
> The GNADE 2.0 tree is now getting to a point where it is in presentable
> shape, there are still
> outstanding issues but at least it builds out of the box on both on
> Linux and win32

The configure stuff does not work any more. To my knowledge there are
no operating depedencies in the ODBC and the native drivers and neither
in gesql. In GANDE 1.x is basically used to guess some installtion
paramters which do depend on the operating system.

I don't think that is is needed any more?! For the time beeing i would
propose to delete the stuff.

>
> /Per
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>

Re: GNADE 2.0 - Samples & Regression

2008-01-20T08:11:10Z

Jérôme Roussel wrote:
> Michael,
>
> The test suite should check at least that the build infrastructure and
> project files do work as expected. It should be possible to detect when
> project files are broken by adding simple tests to build against the
> different supported databases.

Actually the bild environement for gnade2.0 seems to me still quite
unstable. I will put this on priority two.

>
> A second kind of tests could try to check the major features of the
> different supported databases. It would be a good way to detect
> implementation changes in the database libraries, leading to a needed
> change in GNADE too.

This is something i thought first of :-) I will start with this.

I got on my box for testing: MimerSQL, portges and mysql.
Which odbc managers should be used, i am using UnixODBC.

In order to implement regression tests i am using a test harenss
which consists of a test case generator and AUnit. I will make
this available in the svn. For Aunit i will open a vendor branch.

Michael

>
> Jérôme
>
>
> Michael Erdmann wrote:
>> Dear all,
>>
>> i like to create a regression test suite for the ODBC. Actually
>> this was missing, any ideas about the scope of the tests?
>>
>>
>> Michael
>>
>> -------------------------------------------------------------------------
>> This SF.net email is sponsored by: Microsoft
>> Defy all challenges. Microsoft(R) Visual Studio 2008.
>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>> _______________________________________________
>> Gnade-develop mailing list
>> http://gnade.sourceforge.net/
>> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: GNADE 2.0 - Samples & Regression

2008-01-19T19:48:00Z

Michael,

The test suite should check at least that the build infrastructure and
project files do work as expected. It should be possible to detect when
project files are broken by adding simple tests to build against the
different supported databases.

A second kind of tests could try to check the major features of the
different supported databases. It would be a good way to detect
implementation changes in the database libraries, leading to a needed
change in GNADE too.

Jérôme

Michael Erdmann wrote:

> Dear all,
>
> i like to create a regression test suite for the ODBC. Actually
> this was missing, any ideas about the scope of the tests?
>
>
> Michael
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>

Re: GNADE 2.0

2008-01-19T12:55:21Z

Per Sandberg wrote:
> The GNADE 2.0 tree is now getting to a point where it is in presentable
> shape, there are still
> outstanding issues but at least it builds out of the box on both on
> Linux and win32

Does somebody mind if we remove gsql from the release? I have looked at
this thing for about 4 years ;-/ There are a lot better programms
available in the open source world?

GNADE 2.0 - Samples & Regression

2008-01-19T11:02:13Z

Dear all,

i like to create a regression test suite for the ODBC. Actually
this was missing, any ideas about the scope of the tests?

Michael

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

project files

2008-01-17T17:37:03Z

Per,

1/ Thanks for the quick fixes. I confirm it works much better with GNAT
4.1.2 now!

I have detected a new regression however: with a simple: $ make,
gnade.gpr is not installed anymore.

$ svn update
À la révision 86.

$ ls /home/roussel/gnade-inst/lib/gnat/
gnade-base.gpr gnade-odbc.gpr gnade-sqlite3.gpr
gnade-support.gpr
gnade-mysql.gpr gnade-postgres.gpr gnade-sqlite3-impl_lib.gpr

I have to launch:

$ make _install
mkdir -p /home/roussel/gnade-inst/lib/gnat
cp -f gnade_.gpr /home/roussel/gnade-inst/lib/gnat/gnade.gpr
chmod a-w /home/roussel/gnade-inst/lib/gnat/gnade.gpr

$ ls /home/roussel/gnade-inst/lib/gnat/
gnade-base.gpr gnade-mysql.gpr gnade-postgres.gpr
gnade-sqlite3-impl_lib.gpr
gnade.gpr gnade-odbc.gpr gnade-sqlite3.gpr gnade-support.gpr

and then gnade.gpr is correctly installed. Do you have the same
behaviour on your setup ?

2/ Now, I am testing the project structure with a simple example using
sqlite3.

$ gnatmake -P test_sqlite.gpr
gnade-sqlite3.gpr:8:24: library directory
"/home/roussel/gnade-inst/lib/gnat/../GNADE/SQLite3" does not exist
gnade-sqlite3.gpr:10:25: "../../include/GNADE/SQLite3" is not a valid
directory
gnade-support.gpr:7:24: library directory
"/home/roussel/gnade-inst/lib/gnat/../GNADE/Support" does not exist
gnade-support.gpr:9:25: "../../include/GNADE/Support" is not a valid
directory
gnade-base.gpr:6:24: library directory
"/home/roussel/gnade-inst/lib/gnat/../gnade/base" does not exist
gnade-base.gpr:7:25: "../../include/gnade/base" is not a valid directory
gnade-sqlite3-impl_lib.gpr:4:24: library directory
"/home/roussel/gnade-inst/lib/gnat/../GNADE/SQLite3" does not exist
gnatmake: "test_sqlite.gpr" processing failed

There is a mix of lower case and upper case, which prevents gnatmake
from finding the files in the proper directories. The GNAT convention
should be used I guess, which means using lower case for project path
directories for both $PREFIX/lib/gnade and $PREFIX/include/gnade (GNADE
is upper case for the moment). Once I've corrected the pathes in the
different installed project files, gnatmake does not complain anymore :-)

3/ Finally, here is the result I get:

$ gnatmake -P test_sqlite.gpr
gnatbind -I- -x /home/roussel/dev/sqlite/test_sqlite.ali
gnatlink /home/roussel/dev/sqlite/test_sqlite.ali
-L/home/roussel/gnade-inst/lib/gnade/sqlite3 -lgnadesqlite3sd
-L/home/roussel/gnade-inst/lib/gnade/support -lgnadesupportsd
-L/home/roussel/gnade-inst/lib/gnade/sqlite3 -lsqlite3
-L/home/roussel/gnade-inst/lib/gnade/base -lgnadebase -o
/home/roussel/dev/sqlite/test_sqlite
/usr/bin/ld: cannot find -lgnadebase
collect2: ld returned 1 exit status
gnatlink: cannot call /usr/bin/gcc-4.1
gnatmake: *** link failed.

$ ls /home/roussel/gnade-inst/lib/gnade/base/
gnu.ali gnu-db.ali gnu-db-version.ali libgnadebasesd.a

I am not sure how the convention lib...name...sd.a works right now but
clearly the solution is in gnade-base.gpr. Once this is corrected, my
test program works perfectly.

I could make direct changes in the svn sources but since you're the one
putting in place the build infrastructure, I think it's much easier if I
just test it and point the few things to fix.

Hope it helps !

Jérôme

Per Sandberg a écrit :

> Jérôme
> Thanks for the note I have taken care of both:
> * removed the "-p" in gnatmake
> * Create the $PREFIX dir.
> All commited to the repository
>
> /Per
>
> Jérôme Roussel wrote:
>> Hi,
>>
>> Sorry for my silence too. It seems that the build process has been
>> cleaned up since my last trials so no need to spend some time on the
>> issue from my previous email. The README file should just be updated
>> to remove the reference to:
>> $ make config-<mode>
>> with mode being dynamic since this target does not exist.
>>
>> I've succeeded the compilation using GCC 4.1.2 but
>>
>> 1) I had to remove all the -p switch in every Makefile
>>
>> compile: # Compile SQLite3
>> - $(GNATMAKE) -p -Pgnade-${NAME}.gpr
>> + $(GNATMAKE) -Pgnade-${NAME}.gpr
>>
>> 2) I had to create the missing lib-obj/Debug/lib directories.
>>
>> Is-it acceptable to add the missing directories directly in the
>> repository (or to create them in the Makefiles) and to remove the -p
>> switch which is currently specific to GNAT GPL only?
>>
>> Another little suggestion: the PREFIX directory defined in
>> Makefile.conf is not created if it does not exist, which stops the
>> build process. That could be done automatically.
>>
>> Thanks for this excellent project !
>>
>> Jérôme
>>
>>
>> Michael Erdmann a écrit :
>>
>>> roussel@... wrote:
>>>
>>>> Hi,
>>>>
>>> Sorry for the long silence, but unfortunatly i had some other
>>> priorities.
>>>
>>> Currently i am using.
>>>
>>> GNAT GPL 2007 (20070405-41)
>>> Copyright 1996-2007, Free Software Foundation, Inc.
>>>
>>> But unfortunatly the target config-dynamic does not exist in the svn.
>>>
>>> gnade@hal:~/gnade2.0> make config-dynamic
>>> make: *** No rule to make target `config-dynamic'. Stop.
>>>
>>>
>>> Which revision of trunk did you check out?
>>>
>>> Michael
>>>
>>>
>>
>> -------------------------------------------------------------------------
>>
>> This SF.net email is sponsored by: Microsoft
>> Defy all challenges. Microsoft(R) Visual Studio 2008.
>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>> _______________________________________________
>> Gnade-develop mailing list
>> http://gnade.sourceforge.net/
>> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>>
>>

GNADE 2.0

2008-01-17T13:44:47Z

The GNADE 2.0 tree is now getting to a point where it is in presentable
shape, there are still
outstanding issues but at least it builds out of the box on both on
Linux and win32

/Per

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: which version of gnat needed in the compilation process for version 2.0 ?

2008-01-17T13:39:29Z

Jérôme
Thanks for the note I have taken care of both:
* removed the "-p" in gnatmake
* Create the $PREFIX dir.
All commited to the repository

/Per

Jérôme Roussel wrote:

> Hi,
>
> Sorry for my silence too. It seems that the build process has been
> cleaned up since my last trials so no need to spend some time on the
> issue from my previous email. The README file should just be updated to
> remove the reference to:
> $ make config-<mode>
> with mode being dynamic since this target does not exist.
>
> I've succeeded the compilation using GCC 4.1.2 but
>
> 1) I had to remove all the -p switch in every Makefile
>
> compile: # Compile SQLite3
> - $(GNATMAKE) -p -Pgnade-${NAME}.gpr
> + $(GNATMAKE) -Pgnade-${NAME}.gpr
>
> 2) I had to create the missing lib-obj/Debug/lib directories.
>
> Is-it acceptable to add the missing directories directly in the
> repository (or to create them in the Makefiles) and to remove the -p
> switch which is currently specific to GNAT GPL only?
>
> Another little suggestion: the PREFIX directory defined in Makefile.conf
> is not created if it does not exist, which stops the build process. That
> could be done automatically.
>
> Thanks for this excellent project !
>
> Jérôme
>
>
> Michael Erdmann a écrit :
>
>> roussel@... wrote:
>>
>>> Hi,
>>>
>> Sorry for the long silence, but unfortunatly i had some other
>> priorities.
>>
>> Currently i am using.
>>
>> GNAT GPL 2007 (20070405-41)
>> Copyright 1996-2007, Free Software Foundation, Inc.
>>
>> But unfortunatly the target config-dynamic does not exist in the svn.
>>
>> gnade@hal:~/gnade2.0> make config-dynamic
>> make: *** No rule to make target `config-dynamic'. Stop.
>>
>>
>> Which revision of trunk did you check out?
>>
>> Michael
>>
>>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>
>

Re: which version of gnat needed in the compilation process for version 2.0 ?

2008-01-16T19:51:25Z

Hi,

Sorry for my silence too. It seems that the build process has been
cleaned up since my last trials so no need to spend some time on the
issue from my previous email. The README file should just be updated to
remove the reference to:
$ make config-<mode>
with mode being dynamic since this target does not exist.

I've succeeded the compilation using GCC 4.1.2 but

1) I had to remove all the -p switch in every Makefile

compile: # Compile SQLite3
- $(GNATMAKE) -p -Pgnade-${NAME}.gpr
+ $(GNATMAKE) -Pgnade-${NAME}.gpr

2) I had to create the missing lib-obj/Debug/lib directories.

Is-it acceptable to add the missing directories directly in the
repository (or to create them in the Makefiles) and to remove the -p
switch which is currently specific to GNAT GPL only?

Another little suggestion: the PREFIX directory defined in Makefile.conf
is not created if it does not exist, which stops the build process. That
could be done automatically.

Thanks for this excellent project !

Jérôme

Michael Erdmann a écrit :

> roussel@... wrote:
>> Hi,
>
> Sorry for the long silence, but unfortunatly i had some other
> priorities.
>
> Currently i am using.
>
> GNAT GPL 2007 (20070405-41)
> Copyright 1996-2007, Free Software Foundation, Inc.
>
> But unfortunatly the target config-dynamic does not exist in the svn.
>
> gnade@hal:~/gnade2.0> make config-dynamic
> make: *** No rule to make target `config-dynamic'. Stop.
>
>
> Which revision of trunk did you check out?
>
> Michael
>

Re: A Simple Persistance Mechanism

2008-01-05T06:23:52Z

Graham Stark wrote:

> On Thu, 2008-01-03 at 09:25 +0100, Michael Erdmann wrote:
>.... snip .. snip ..
>
> Thank you.
>
>> If you like we could put the source code in the SVN or CSV
>> of the gnade project.
>>
>
> If anyone else out there felt that this was a worthwhile approach, then
> by all means.

At least it provides a good platform and it enforces later
contributions by other people, if there anny.

- Michael

> Graham
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: A Simple Persistance Mechanism

2008-01-04T01:51:58Z

On Thu, 2008-01-03 at 09:25 +0100, Michael Erdmann wrote:

> Graham Stark wrote:
> > Hi,
> > I've written a very simple Ada persistence mechanism. It's a port to
> > Ada of Propel (http://propel.phpdb.org/), a very popular PHP database
> > system for the PHP scripting language. Like Propel, it's essentially a
> > code generating system which builds Ada source and an SQL schema from a
> > common XML definition file.
>
> Thanks for the contribution. I have put a reference on the GNADE home
> page.
>

Thank you.

> If you like we could put the source code in the SVN or CSV
> of the gnade project.
>

If anyone else out there felt that this was a worthwhile approach, then
by all means.

Graham

--
Graham Stark, Virtual Worlds
Phone (+44) 01908 618239 Mobile (+44) 07952633185 Skype graham_k_stark

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: which version of gnat needed in the compilation process for version 2.0 ?

2008-01-03T00:36:53Z

roussel@... wrote:
> Hi,

Sorry for the long silence, but unfortunatly i had some other
priorities.

Currently i am using.

GNAT GPL 2007 (20070405-41)
Copyright 1996-2007, Free Software Foundation, Inc.

But unfortunatly the target config-dynamic does not exist in the svn.

gnade@hal:~/gnade2.0> make config-dynamic
make: *** No rule to make target `config-dynamic'. Stop.

Which revision of trunk did you check out?

Michael

> I have been looking at GNADE for quite some time now and I've decided to
> look at the latest version 2.0 from svn.
>
> From a clean svn checkout, here what happens on my side
>
> $ cd gnade2.0/trunk
> $ make config-dynamic
> ...
> gnatmake -p -P ./gnade-base.gpr
> gnatmake: invalid switch: p
> make[2]: *** [all] Erreur 4
> ...
>
> I got the above error, which is easy to correct since it's only a
> problem of gnat version. I am using gnat FSF version 4.1.2. You should
> indicate somewhere in the documentation on which version of gnat the
> compilation process relies on.
>
> Concerning the FSF version of GCC, the -p flag is only available in the
> head version, the future 4.3.X. Is it really mandatory to use this
> feature ? Is it not possible to just create the needed directory in the
> svn tree for the time being ?
>
> There also must be something wrong in the build process because if I
> relaunch the last command used:
> $ make config-dynamic
> I don't get the same error message. It means that the build process does
> not go throw the same path each time it is launched, which is really
> surprising... I expect to get the same error message, when I just
> relaunch make in a project.
>
> Thanks for your great work !
>
> Jérôme
>
> -------------------------------------------------------------------------
> SF.Net email is sponsored by: The Future of Linux Business White Paper
> from Novell. From the desktop to the data center, Linux is going
> mainstream. Let it simplify your IT future.
> http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: A Simple Persistance Mechanism

2008-01-03T00:25:07Z

Graham Stark wrote:
> Hi,
> I've written a very simple Ada persistence mechanism. It's a port to
> Ada of Propel (http://propel.phpdb.org/), a very popular PHP database
> system for the PHP scripting language. Like Propel, it's essentially a
> code generating system which builds Ada source and an SQL schema from a
> common XML definition file.

Thanks for the contribution. I have put a reference on the GNADE home
page.

If you like we could put the source code in the SVN or CSV
of the gnade project.

Michael

>
> It's at a fairly early stage, with a limited feature set, and has been
> tested only on my computers, but I'm using it successfully on a fairly
> large simulation project I'm working on.
>
> The generator is written in Python, and uses Gnade ODBC bindings as the
> back-end.
>
> You can get it here:
>
> http://virtual-worlds.biz/downloads/mill.tgz
>
> There are some examples in the xml/ directory, and some instructions in
> doc/Brief_Guide.pdf
>
> I was thinking of putting it on SourceForge if it generated any
> interest.
>
> I'm calling it 'Mill', for no very compelling reason.
>
> regards, and a happy new year,
>
> Graham Stark
>

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

A Simple Persistance Mechanism

2008-01-02T13:02:11Z

Hi,
I've written a very simple Ada persistence mechanism. It's a port to
Ada of Propel (http://propel.phpdb.org/), a very popular PHP database
system for the PHP scripting language. Like Propel, it's essentially a
code generating system which builds Ada source and an SQL schema from a
common XML definition file.

It's at a fairly early stage, with a limited feature set, and has been
tested only on my computers, but I'm using it successfully on a fairly
large simulation project I'm working on.

The generator is written in Python, and uses Gnade ODBC bindings as the
back-end.

You can get it here:

http://virtual-worlds.biz/downloads/mill.tgz

There are some examples in the xml/ directory, and some instructions in
doc/Brief_Guide.pdf

I was thinking of putting it on SourceForge if it generated any
interest.

I'm calling it 'Mill', for no very compelling reason.

regards, and a happy new year,

Graham Stark

--
Graham Stark, Virtual Worlds
Phone (+44) 01908 618238 Mobile (+44) 07952633185 Skype graham_k_stark

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: which version of gnat needed in the compilation process for version 2.0 ?

2007-12-16T21:50:53Z

Michael Erdmann wrote:
> roussel@... wrote:
>> Hi,
> Actually i am not even able to check out the stuff!

checkout works now; ...

>
> $ svn co https://gnade.svn.sourceforge.net/svnroot/gnade/gnade2.0/trunk
> A trunk/regression
> A trunk/regression/SQLite3
> A trunk/regression/Makefile
> A trunk/regression/README
> ....
> A trunk/src/sqlite3/gnu-db-sqlite3.adb
>
> get some how blocked at gnu-dg-sqlite3.adb. Per, did you put some kind
> of notification trigger on this file?
>
>
> Michael
>
>
>
>> I have been looking at GNADE for quite some time now and I've decided to
>> look at the latest version 2.0 from svn.
>>
>> From a clean svn checkout, here what happens on my side
>>
>> $ cd gnade2.0/trunk
>> $ make config-dynamic
>> ...
>> gnatmake -p -P ./gnade-base.gpr
>> gnatmake: invalid switch: p
>> make[2]: *** [all] Erreur 4
>> ...
>>
>> I got the above error, which is easy to correct since it's only a
>> problem of gnat version. I am using gnat FSF version 4.1.2. You should
>> indicate somewhere in the documentation on which version of gnat the
>> compilation process relies on.
>>
>> Concerning the FSF version of GCC, the -p flag is only available in the
>> head version, the future 4.3.X. Is it really mandatory to use this
>> feature ? Is it not possible to just create the needed directory in the
>> svn tree for the time being ?
>>
>> There also must be something wrong in the build process because if I
>> relaunch the last command used:
>> $ make config-dynamic
>> I don't get the same error message. It means that the build process does
>> not go throw the same path each time it is launched, which is really
>> surprising... I expect to get the same error message, when I just
>> relaunch make in a project.
>>
>> Thanks for your great work !
>>
>> Jérôme
>>
>> -------------------------------------------------------------------------
>> SF.Net email is sponsored by: The Future of Linux Business White Paper
>> from Novell. From the desktop to the data center, Linux is going
>> mainstream. Let it simplify your IT future.
>> http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
>> _______________________________________________
>> Gnade-develop mailing list
>> http://gnade.sourceforge.net/
>> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>>
>
>
> -------------------------------------------------------------------------
> SF.Net email is sponsored by:
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services
> for just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>

-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
Gnade-develop mailing list
http://gnade.sourceforge.net/
https://lists.sourceforge.net/lists/listinfo/gnade-develop

Re: which version of gnat needed in the compilation process for version 2.0 ?

2007-12-16T08:43:49Z

roussel@... wrote:
> Hi,
Actually i am not even able to check out the stuff!

$ svn co https://gnade.svn.sourceforge.net/svnroot/gnade/gnade2.0/trunk
A trunk/regression
A trunk/regression/SQLite3
A trunk/regression/Makefile
A trunk/regression/README
....
A trunk/src/sqlite3/gnu-db-sqlite3.adb

get some how blocked at gnu-dg-sqlite3.adb. Per, did you put some kind
of notification trigger on this file?

Michael

>
> I have been looking at GNADE for quite some time now and I've decided to
> look at the latest version 2.0 from svn.
>
> From a clean svn checkout, here what happens on my side
>
> $ cd gnade2.0/trunk
> $ make config-dynamic
> ...
> gnatmake -p -P ./gnade-base.gpr
> gnatmake: invalid switch: p
> make[2]: *** [all] Erreur 4
> ...
>
> I got the above error, which is easy to correct since it's only a
> problem of gnat version. I am using gnat FSF version 4.1.2. You should
> indicate somewhere in the documentation on which version of gnat the
> compilation process relies on.
>
> Concerning the FSF version of GCC, the -p flag is only available in the
> head version, the future 4.3.X. Is it really mandatory to use this
> feature ? Is it not possible to just create the needed directory in the
> svn tree for the time being ?
>
> There also must be something wrong in the build process because if I
> relaunch the last command used:
> $ make config-dynamic
> I don't get the same error message. It means that the build process does
> not go throw the same path each time it is launched, which is really
> surprising... I expect to get the same error message, when I just
> relaunch make in a project.
>
> Thanks for your great work !
>
> Jérôme
>
> -------------------------------------------------------------------------
> SF.Net email is sponsored by: The Future of Linux Business White Paper
> from Novell. From the desktop to the data center, Linux is going
> mainstream. Let it simplify your IT future.
> http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
> _______________________________________________
> Gnade-develop mailing list
> http://gnade.sourceforge.net/
> https://lists.sourceforge.net/lists/listinfo/gnade-develop
>