Firebird

Is there a cross plaform way of getting the UDF directory, or even the 
firebird directory?

Regards
Ian Newby

On Tuesday 27 March 2007 15:44, Ian Newby wrote:
> Is there a cross plaform way of getting the UDF directory, or even the
> firebird directory?

There is UNDOCUMENTED (read - may be changed/removed in the future) function, 
which does the trick. It adds firebird directory before file name 'file' and 
puts result into resultString, which must be at least MAX_PATH_LEN size. 

void API_ROUTINE gds__prefix(TEXT* resultString, const TEXT* file)

I.e.
gds__prefix(result, "UDF/your.file);
will give in result something like:
"/opt/firebird/UDF/your.file"

> the existance of environment variables, namely HOME, TEMP and TMP
TMPDIR is the standard for *nix.  Does Ubuntu set TMPDIR?

-----Original Message-----
From: firebird-devel-bounces@...
[mailto:firebird-devel-bounces@...] On Behalf Of Ian
Newby
Sent: Tuesday, March 27, 2007 7:37 AM
To: firebird-devel@...
Subject: [Firebird-devel] UDF Temp Files

Hi Folks,
I have a udf library which generates guids and firebird "uuids"=20
(compressed reversed guids 22 characters long) which runs on both
Windows and Linux.

I have a small problem. The library needs to store a state file on the
local file system. Currently, I find a location for this by checking for
  the existance of environment variables, namely HOME, TEMP and TMP.

I've just upgraded one of our servers to Ubuntu 6.10 (from Mandrake
2005) and the library has stopped working. It seems that the Ubuntu
doesn't set any of the environment variables.

Whats the best cross platform way of finding somewhere to put this file?

Regards
Ian Newby

Disclaimer: This message (including attachments) is confidential and may =
be privileged. If you have received it by mistake please notify the =
sender by return e-mail and delete this message from your system. Any =
unauthorized use or dissemination of this message in whole or in part is =
strictly prohibited. Please note that e-mails are susceptible to change. =
RxStrategies, Inc. shall not be liable for the improper or incomplete =
transmission of the information contained in this communication or for =
any delay in its receipt or damage to your system. RxStrategies, Inc. =
does not guarantee that the integrity of this communication has been =
maintained nor that this communication is free from viruses, =
interceptions or interference.=20

-=| Dimitry Sibiryakov, 27.03.2007 11:20 |=-
> On 27 Mar 2007 at 17:05, Argus Connect wrote:
> 
>> I also followed this guide when I tried to build it on an Intel Mac 
>> but I get the following error(s) when running make.
> 
>   Too new gcc. FB1.5 can't be built with gcc4.

Hi, Argus,

Debian uses three patches for 1.5 to make it compile with gcc4.1. See
[1], [2] and [3].


HTH,
	dam
[1]
http://svn.debian.org/wsvn/pkg-firebird/1.5/trunk/debian/patches/gcc4.patch?op=file

[2]
http://svn.debian.org/wsvn/pkg-firebird/1.5/trunk/debian/patches/gcc4.1.patch?op=file

[3]
http://svn.debian.org/wsvn/pkg-firebird/1.5/trunk/debian/patches/gcc4+amd64.patch?op=file
-- 
Damyan Ivanov                           Modular Software Systems
phone +359(2)928-2611, 929-3993              fax +359(2)920-0994
mobile +359(88)856-6067              JID dam@...

Jorge Andrés Brugger wrote:
> Helen:
>
> I´ve tried to store data using es_es collation and selecting it using 
> es_es_ci_ai and doesn't work.
>
> If  I´ve the following records in a single varchar field defined with 
> es_es collation:
>
> ABC
> Abcd
> ab
>
> and I do
>
> select * from table where field starting with 'A' collate es_es_ci_ai
>
> or
>
> select * from table where field like '%A%' collate es_es_ci_ai
>
> result set is
>
> ABC
> Abcd
>
> (it misses 'ab')
>
> Seems to be a bug when using collations in expressions ... should I 
> report it?
AFAIR, the standard says that the collation from the field is preferred, 
but I don't know if this is valid when a explicit COLLATE clause is 
specified.
But we have a very ugly logic for this, sometimes the max collation id 
dictate what is preferred.

You could put the COLLATE clause after the field.


Adriano

Yes, it works fine putting collate just after field.

My new question: Will it use the index? (STARTING WITH clause)

I´m sure that select * from table where field starting with 'a' will use 
the index by 'field', but select * from table where field collate 
es_es_ci_ai starting with 'a' will?

Regards

Adriano dos Santos Fernandes escribió:
> You could put the COLLATE clause after the field.
>
>
> Adriano
>
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
>
>
>   

-- 
Jorge Andrés Brugger
Informática
DASU - Obra Social del Personal de la Universidad Nacional de la Patagonia
Comodoro Rivadavia, Chubut, Argentina
Teléfono (0297) 446-4444 int. 103
Correo electrónico: jbrugger@...
Website: http://www.dasu.com.ar

Jorge Andrés Brugger wrote:
> Yes, it works fine putting collate just after field.
>
> My new question: Will it use the index? (STARTING WITH clause)
>
> I´m sure that select * from table where field starting with 'a' will use 
> the index by 'field', but select * from table where field collate 
> es_es_ci_ai starting with 'a' will?
>   
No.


Adriano

So I need what I've asked before and by private e-mail ... a es_es_ci_ci 
standardization ...

Any chance to get some help about?

Thanks!

Adriano dos Santos Fernandes escribió:
> Jorge Andrés Brugger wrote:
>   
>> Yes, it works fine putting collate just after field.
>>
>> My new question: Will it use the index? (STARTING WITH clause)
>>
>> I´m sure that select * from table where field starting with 'a' will use 
>> the index by 'field', but select * from table where field collate 
>> es_es_ci_ai starting with 'a' will?
>>   
>>     
> No.
>
>
> Adriano
>
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
>
>
>   

-- 
Jorge Andrés Brugger
Informática
DASU - Obra Social del Personal de la Universidad Nacional de la Patagonia
Comodoro Rivadavia, Chubut, Argentina
Teléfono (0297) 446-4444 int. 103
Correo electrónico: jbrugger@...
Website: http://www.dasu.com.ar

2.1 new features (specials-first, etc)... will use indexes if available?

Adriano dos Santos Fernandes escribió:
> Jorge Andrés Brugger wrote:
>   
>> Yes, it works fine putting collate just after field.
>>
>> My new question: Will it use the index? (STARTING WITH clause)
>>
>> I´m sure that select * from table where field starting with 'a' will use 
>> the index by 'field', but select * from table where field collate 
>> es_es_ci_ai starting with 'a' will?
>>   
>>     
> No.
>
>
> Adriano
>
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> Firebird-Devel mailing list, web interface at https://lists.sourceforge.net/lists/listinfo/firebird-devel
>
>
>   

-- 
Jorge Andrés Brugger
Informática
DASU - Obra Social del Personal de la Universidad Nacional de la Patagonia
Comodoro Rivadavia, Chubut, Argentina
Teléfono (0297) 446-4444 int. 103
Correo electrónico: jbrugger@...
Website: http://www.dasu.com.ar

> Union returns inconsistent field names
> --------------------------------------
> 
>                  Key: CORE-1181
>                  URL: http://tracker.firebirdsql.org/browse/CORE-1181
>              Project: Firebird Core
>           Issue Type: Bug
>           Components: Engine
>     Affects Versions: 2.0.1, 2.0.0
>          Environment: Windows XP
>             Reporter: michalk1
>             Priority: Minor
> 
> CREATE TABLE TESTTAB
> (ID NUMERIC(15, 2));
> 
> SELECT ID FROM TESTTAB GROUP BY ID
> UNION
> SELECT CAST(0 AS NUMERIC(15,2)) AS ID FROM RDB$DATABASE

The problem is in this code (dsql/make.cpp):


	// Dialect 1 NUMERIC and DECIMAL are stored as sub-types
	// 1 and 2 from float types dtype_real, dtype_double
	if (!any_float) {
		any_float = (desc1.dsc_sub_type == 0);
	}

Unlike the comment says, in Dialect 1 there's no guaranteed indication 
that an approximate numeric descriptor has been defined as 
NUMERIC/DECIMAL. Although the parser really always sets dsc_sub_type for 
NUMERIC/DECIMAL constants, this value is stored in RDB$FIELDS for 
Dialect 3 descriptors only. And Language Reference doesn't define 
meaning of RDB$FIELD_SUB_TYPE for real/double data types either. So the 
column ID is described as double with dsc_sub_type == 0, thus causing 
the above code snippet to consider it a plain double (non-NUMERIC) and 
hence the resulting descriptor also has scale == 0. As a result, we have 
wrong output data type.

Any smart ideas how to fix that?


Dmitry

On Tuesday 27 March 2007 12:15, phil.c@... wrote:
> Hi,
>
>  Opps ! Sorry I used the 1.5.4.4910 build. I downloaded it from
> http://www.firebirdsql.org/index.php?op=files&id=engine_154
>
>  Also I need to use 1.5.x as we also deploy to interbase clients which
> still use the old security database. It would take a while to write the
> code to figure out which version they are using when creating users for
> my app, I have not got time right now to do this.

BTW, the only security-database related problem you will get with old 
interbase client (reasonably old), is isc_(add/modify/delete)_user calls not 
working. Login/password validation and even management using services API 
should work.

As Dimitry says XCode 2.4.1 is the latest and uses gcc 4.01
You need a version of Xcode that uses gcc 3.x

Paul

Dear Paul,

 Thanks. I will try to get a version of Xcode which is supported on Intel
Mac which has gcc 3.x.

From,

Phillip

>
> As Dimitry says XCode 2.4.1 is the latest and uses gcc 4.01
> You need a version of Xcode that uses gcc 3.x
>
> Paul
>
>
>
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share
> your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> Firebird-Devel mailing list, web interface at
> https://lists.sourceforge.net/lists/listinfo/firebird-devel
>

>  Thanks. I will try to get a version of Xcode which is supported on Intel
> Mac which has gcc 3.x.

Problem is - I don't think you will have much luck, I seem to remember that
there isn't such a beast...

Paul

Hi Paul,

 I have found this link

http://developer.apple.com/releasenotes/DeveloperTools/RN-GCC3/

Which talks about gcc 3 in Xcode, you can also set the version of gcc to
use with something like

sudo /usr/sbin/gcc_select 3

 However it also says "In Xcode 2.1 and higher, the gcc 3.3 compiler does
not support generating code for x86"

I will keep looking for the highest Version 3 they released, which may
support x86.

Someone has compiled it before on x86 as I have got the binaries to work :)
Unless the are universal binaries....
http://developer.apple.com/documentation/MacOSX/Conceptual/universal_binary/index.html

 I can program in C and a bit in C++, can someone explain the reason it
will not work with gcc 4.x. Maybe I could change the makefile / code ?

From,

Phillip

>
>>  Thanks. I will try to get a version of Xcode which is supported on
>> Intel
>> Mac which has gcc 3.x.
>
> Problem is - I don't think you will have much luck, I seem to remember
> that
> there isn't such a beast...
>
> Paul
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share
> your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> Firebird-Devel mailing list, web interface at
> https://lists.sourceforge.net/lists/listinfo/firebird-devel
>