🚀 go-pugleaf

I love those rules that dont apply to TeamB members :D

I'm going to download the new Indy anyway.

"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
news:46bd01a2$1@newsgroups.borland.com...
>I don't mean to cross-post this, but I know people using Indy are here, and
> we need much help and quickly.
> 

"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
news:46bd01a2$1@newsgroups.borland.com...
>I don't mean to cross-post this

Yes you do, or you wouldn't have.  :^)

> Due to deadlines we're working under, we need a release ASAP....
> A LOT of changes have been made to the inner code, and to some of the
> interfaces....
> Unfortunately, the current Indy 10 snapshot is not ready for production 
> use
> just yet, in our opinion....
> We need to make sure
> things even still run at all.

Do you have any idea what kind of nightmare ensues from this scenario?  Far, 
far better IMHO to miss a deadline than to dump alpha code on an 
unsuspecting public, and that's what this is.  Wouldn't touch it with a 10 
foot pole.

Best of luck,
Van Swofford
Tybee Jet Corp. 

Will download & install in a vmware Delphi tomorrow. Will give it a 
bashing...

JD

"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
news:46bd01a2$1@newsgroups.borland.com...
>I don't mean to cross-post this, but I know people using Indy are here, and
> we need much help and quickly.
>
> Indy is gearing up for a new major release.  One other person and I have
> been working around the clock for several weeks now trying to make that
> happen.  Due to deadlines we're working under, we need a release ASAP.
>
> A LOT of changes have been made to the inner code, and to some of the
> interfaces.  For instance, there is UTF-8 support at the IOHandler and
> IdBuffer levels now, more C++Builder and FreePascal support has been
> implemented, IOHandler and IdStack optimizations have been made, the IdSys
> and IdObjs units have been removed, etc.
>
> Unfortunately, the current Indy 10 snapshot is not ready for production 
> use
> just yet, in our opinion.  Since we seem to be the only two people working
> on Indy 10 at this point, we're not going to make the deadlines we have at
> the rate we're going.  So we need more people to get involved on this, and
> quickly.  We need Indy 10 to be thoroughly re-tested.  We need to make 
> sure
> things even still run at all.  Make sure any bugs that have crept in are
> fixed.  And so on.
>
> So PLEASE PLEASE PLEASE, anyone and everyone that can, please download the
> current snapshot from http://indy.fulgan.com and start hammering away on 
> it
> over the next few days, and report any problems that arise.
>
>
> Gambit
>
> 

Hi,

I don't use many components, but I'll try to install it on BCB6 and test the 
components I do use.

Paul

"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
news:46bd01a2$1@newsgroups.borland.com...
>
> So PLEASE PLEASE PLEASE, anyone and everyone that can, please download the
> current snapshot from http://indy.fulgan.com and start hammering away on 
> it
> over the next few days, and report any problems that arise.
>
> Gambit

Remy,

| Indy is gearing up for a new major release.  One other person and I
| have been working around the clock for several weeks now trying to
| make that happen.  Due to deadlines we're working under, we need a
| release ASAP.

The "rush to release deadline" symptom is the cause of most horrible
software releases!

Take a deep breath, say "to heck with the 'deadline'," and take the
necessary time to do things RIGHT!!!

You will not be sorry in the long run and your customers will be
happier and much better served by a solid and more reliable release
product.

-- 
Q

08/12/2007 12:10:20

XanaNews Version 1.17.5.7  [Q's salutation mod]

I can't build the Indy 10 download, are there instructions for BCB6 Pro. 
I've got instructions for Indy 9 and done it many times, but I can't see 
anything in the Indy10 zip file for building this version?



"Paul at NCF" <notplmacca@clara.co.uk> wrote in message 
news:46bec059$1@newsgroups.borland.com...
> Hi,
>
> I don't use many components, but I'll try to install it on BCB6 and test 
> the components I do use.
>
> Paul
>
> "Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
> news:46bd01a2$1@newsgroups.borland.com...
>>
>> So PLEASE PLEASE PLEASE, anyone and everyone that can, please download 
>> the
>> current snapshot from http://indy.fulgan.com and start hammering away on 
>> it
>> over the next few days, and report any problems that arise.
>>
>> Gambit
>
> 

"Paul at NCF" <notplmacca@clara.co.uk> wrote in message
news:46c0089b$1@newsgroups.borland.com...

> I can't build the Indy 10 download, are there instructions
> for BCB6 Pro.

The current snapshot has a working FULLC6.BAT script that I have been using
without problem to compile BPI/BPL files that can then be installed into the
IDE.

Supposedly, the code is able to be compiled in the IDE directly now in BDS
2006+ versions. But BPKs have not been made for BCB 5/6 yet.


Gambit

"Q Correll" <qcorrell@pacNObell.net> wrote in message
news:46bf5be2$1@newsgroups.borland.com...

> The "rush to release deadline" symptom is the cause of
> most horrible software releases!
>
> Take a deep breath, say "to heck with the 'deadline'," and
> take the necessary time to do things RIGHT!!!
>
> You will not be sorry in the long run and your customers will
> be happier and much better served by a solid and more reliable
> release product.

Hey, I totally agree.  I wish we had more time to do that.  But the deadline
is not of our own creation.  It is set by a partner of ours how wants a new
release of Indy to bundle with the next release of their own product.  We
weren't given much advance notice to get it ready, and there were a lot of
pending updates that hadn't been committed into the codebase yet.  So now
we're scrambling to get it all done.  But if we don't do this, they their
stuck with the old version that was quite buggy and incomplete.  So either
way, we've been backed into a corner again.


Gambit

You are right, I'm not sure how I missed those files.


"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
news:46c08e92$1@newsgroups.borland.com...
>
> "Paul at NCF" <notplmacca@clara.co.uk> wrote in message
> news:46c0089b$1@newsgroups.borland.com...
>
>> I can't build the Indy 10 download, are there instructions
>> for BCB6 Pro.
>
> The current snapshot has a working FULLC6.BAT script that I have been 
> using
> without problem to compile BPI/BPL files that can then be installed into 
> the
> IDE.
>
> Supposedly, the code is able to be compiled in the IDE directly now in BDS
> 2006+ versions. But BPKs have not been made for BCB 5/6 yet.
>
>
> Gambit
>
> 

Hi,

I'm now running the fullc6.bat file:

I've copied my previous computil.exe from Indy9.  And the batch file fails 
at this line, any clues?

Borland Delphi Version 14.0
Copyright (c) 1983,2002 Borland Software Corporation
dclIndyProtocols60.dpk(45)
46 lines, 0.12 seconds, 16268 bytes code, 209 bytes data.
Could Not Find C:\Program Files\Indy\C6\*.rsp

"Paul at NCF" <notplmacca@clara.co.uk> wrote in message
news:46c0d350$1@newsgroups.borland.com...

> the batch file fails at this line, any clues?

That is not a failure.  Look at what the script is actually doing at the
end.  When it is deleting temp files, there is no .rsp file present, so the
'del' command issues a warning.


Gambit

Remi,

Would I be able to install and test the latest snapshot in BCB5?

I would like to test it in BCB5.

JLoz
"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
news:46c08e92$1@newsgroups.borland.com...
>
> "Paul at NCF" <notplmacca@clara.co.uk> wrote in message
> news:46c0089b$1@newsgroups.borland.com...
>
>> I can't build the Indy 10 download, are there instructions
>> for BCB6 Pro.
>
> The current snapshot has a working FULLC6.BAT script that I have been 
> using
> without problem to compile BPI/BPL files that can then be installed into 
> the
> IDE.
>
> Supposedly, the code is able to be compiled in the IDE directly now in BDS
> 2006+ versions. But BPKs have not been made for BCB 5/6 yet.
>
>
> Gambit
>
> 

Thanks.

The original Indy6 (in BCB6?) and Indy9 produced *.dcu files.  I don't see 
these in this version, are these redundant files and can be ignored?  I'm in 
the process of installing Indy10 now to try building some of my 
applications.

Also, I've added the following to my copy of batch files:

cleanup.bat:  echo action
------------

ECHO.
ECHO Cleanup - deleting any unnecessary files in %1
ECHO.

fullc6.bat: pass param to cleanup.bat and echo cleanup action at the end
----------

:start
call clean.bat ..\Lib\        << It is Lib it is tidying up? So this outputs 
the dir name in cleanup.bat
computil SetupC6

REM ***************************************************
REM Clean-up IndyProtocols60
REM ***************************************************
ECHO.
ECHO Build Tidyup - deleting any unnecessary files in dir:
cd
ECHO.


"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
news:46c0d808$1@newsgroups.borland.com...
> That is not a failure.  Look at what the script is actually doing at the
> end.  When it is deleting temp files, there is no .rsp file present, so 
> the
> 'del' command issues a warning.

I get the following compiler error when building against the Indy10:

[C++ Error] IdWinSock2.hpp(218): E2257 , expected
[C++ Error] IdWinSock2.hpp(222): E2257 , expected
[C++ Error] IdWinSock2.hpp(236): E2257 , expected
[C++ Error] IdWinSock2.hpp(252): E2257 , expected
[C++ Error] IdWinSock2.hpp(254): E2257 , expected
[C++ Error] IdWinSock2.hpp(256): E2257 , expected
[C++ Error] IdWinSock2.hpp(274): E2257 , expected
[C++ Error] IdWinSock2.hpp(280): E2257 , expected
[C++ Error] IdWinSock2.hpp(311): E2040 Declaration terminated incorrectly
[C++ Error] IdWship6.hpp(27): E2257 , expected
[C++ Error] IdWship6.hpp(29): E2303 Type name expected
[C++ Error] IdWship6.hpp(31): E2303 Type name expected

eg. Line 218 =>    typedef WSANAMESPACE_INFOEXW  TWSANameSpace_InfoExW;

eg. Line 27 =>      typedef SOCKET_SECURITY_SETTINGS_IPSEC 
*PSOCKET_SECURITY_SETTINGS_IPSEC;


"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message
news:46c0d808$1@newsgroups.borland.com...
> That is not a failure.  Look at what the script is actually doing at the
> end.  When it is deleting temp files, there is no .rsp file present, so
> the
> 'del' command issues a warning.

"Paul at NCF" <notplmacca@clara.co.uk> wrote in message
news:46c15276@newsgroups.borland.com...

> The original Indy6 (in BCB6?) and Indy9 produced *.dcu files.
> I don't see these in this version

They are deleted by the script.  You don't need them anyway.  They don't
apply in C++ at all.  They are Delphi files.

> Also, I've added the following to my copy of batch files:

I wouldn't get too comfortable with the .BAT files for much longer.  I
intend to remove them, just as soon as I can make new BPK files for BCB 5/6.
BDS 2006 and Delphi/C++ 2007 don't use .BAT files anymore, so I don't see
any reason why BCB 5/6 should anymore as well.


Gambit

"Paul at NCF" <notplmacca@clara.co.uk> wrote in message
news:46c15c33$1@newsgroups.borland.com...

> I get the following compiler error when building against the Indy10:

That is a known bug that hasn't been addressed yet.  Those structures don't
exist in the SDKs that ship with BCB 6 and earlier.


Gambit

"JLoz" <jfsierra@excite.com> wrote in message
news:46c12048@newsgroups.borland.com...

> Would I be able to install and test the latest snapshot in BCB5?

Indy has not been officially tested with BCB 5, and the FULLC5.BAT file is
not up-to-date.  Other people have tried adapting FULLC6.BAT to work with
BCB 5, but I don't know how successful they were.


Gambit

Hi

Nothing more I can do at the moment then to test Indy10?  If so, I'll return 
back to Indy9.

Paul


"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
news:46c16442@newsgroups.borland.com...
>
> "Paul at NCF" <notplmacca@clara.co.uk> wrote in message
> news:46c15c33$1@newsgroups.borland.com...
>
>> I get the following compiler error when building against the Indy10:
>
> That is a known bug that hasn't been addressed yet.  Those structures 
> don't
> exist in the SDKs that ship with BCB 6 and earlier.
>
>
> Gambit
>
> 

> Indy has not been officially tested with BCB 5, and the FULLC5.BAT file is
> not up-to-date.  Other people have tried adapting FULLC6.BAT to work with
> BCB 5, but I don't know how successful they were.

I was one of them, and up until 1 month ago I was able to generate the bpl 
and bpi files I needed.  Now it does not compile at all.

If any of you guys would have the time, how long would it take for you to 
get a working FULLC5.bat?

JLoz

>
>
> Gambit
>

> I was one of them, and up until 1 month ago I was able to generate the bpl
> and bpi files I needed.  Now it does not compile at all.
>
> If any of you guys would have the time, how long would it take for you to 
> get a working FULLC5.bat?

Never mind.  I should have read the whole thread first where you said you 
intend to create bpk files for Indy.

How soon do you think you would be able to do this?

Thanks again,

JLoz

>
> JLoz
>
>>
>>
>> Gambit
>>
>
> 

"Paul at NCF" <notplmacca@clara.co.uk> wrote in message
news:46c200c7$1@newsgroups.borland.com...

> Nothing more I can do at the moment then to test Indy10?

Just comment out the offending declarations in the .hpp file for now.


Gambit

"JLoz" <jfsierra@excite.com> wrote in message
news:46c20bbb@newsgroups.borland.com...

> I was one of them, and up until 1 month ago I was able to
> generate the bpl and bpi files I needed.  Now it does not
> compile at all.

Can you be more specific?


Gambit

>> I was one of them, and up until 1 month ago I was able to
>> generate the bpl and bpi files I needed.  Now it does not
>> compile at all.
>
> Can you be more specific?

I have downloaded the snapshot three time during the last week and a half 
and I get different compiler errors.  The latest one is below.

IdWinsock2.pas(3317) Error: ',' or ')' expected but '=' found
IdWinsock2.pas(3445) Error: ',' or ')' expected but '=' found
IdWinsock2.pas(6129)
IdWship6.pas(39) Fatal: Could not compile used unit 'IdWinsock2.pas'


>
>
> Gambit
>
> 

"JLoz" <jfsierra@excite.com> wrote in message
news:46c2f14b$1@newsgroups.borland.com...

> IdWinsock2.pas(3317) Error: ',' or ')' expected but '=' found
> IdWinsock2.pas(3445) Error: ',' or ')' expected but '=' found

That is a known issue that we haven't fixed yet.  Earlier versions of the
Pascal compiler (including the one in BCB 5) do not support ordinal values
being specified for individual items of an enum.  That feature was added in
the v6 compiler.  Enums must always start at 0 in earlier versions.


Gambit

"Q Correll" <qcorrell@pacNObell.net> wrote in message 
news:46bf5be2$1@newsgroups.borland.com...
> Remy,
>
> | Indy is gearing up for a new major release.  One other person and I
> | have been working around the clock for several weeks now trying to
> | make that happen.  Due to deadlines we're working under, we need a
> | release ASAP.
>
> The "rush to release deadline" symptom is the cause of most horrible
> software releases!

Agreed.  It's too bad that Borland/CodeGear has had the habit of saying, 
"We'll release it when it's ready," and then releasing it way too early 
anyway (and getting away with it -- which is our own fault, because we buy 
it anyway), because now everybody else thinks it's just the way you should 
do software.  Argh...sigh.

> Take a deep breath, say "to heck with the 'deadline'," and take the
> necessary time to do things RIGHT!!!

No kidding.  The Boost maintainers are a *good* example; they are very 
conscientious about stable releases.  Of course, nobody makes money on Boost 
(and there are no stockholders or a board of directors to keep happy), but 
still, when it comes to releasing software, they're a good example of what 
to do, as opposed to Borland/CodeGear's example of what NOT to do (IMO).

- Dennis 

>> IdWinsock2.pas(3317) Error: ',' or ')' expected but '=' found
>> IdWinsock2.pas(3445) Error: ',' or ')' expected but '=' found
>
> That is a known issue that we haven't fixed yet.  Earlier versions of the
> Pascal compiler (including the one in BCB 5) do not support ordinal values
> being specified for individual items of an enum.  That feature was added 
> in
> the v6 compiler.  Enums must always start at 0 in earlier versions.
>
Is that going to be fixed for this release?

What about the bpk for BCB5/6?  How soon do you think we will see them?

Thanks,

>
> Gambit
>
> 

Hi,

I've downloaded the latest Indy10, and now I am getting a bit further.

1)  I've rebuilt my Syslog server and client test harness (after updating 
the project files).  Observations so far:

a)  I had to remove include lines for AntiFreeze headers, as the linker was 
reporting it couldn't find AntiFreeze.obj.  Once these were commented out, 
it linked successfully.

b)  The server and client test harnesses are working (including receiving 
SysLog data from my wireless ADSL router), however I notice a difference in 
behaviour for two different send calls.  In Indy9 the hostname is an IP 
address, with Indy10 it is either blank or the PC name (the test was done 
using local loop on the same machine):

With SysLog Client rebuilt only (and using Indy9 Server) the data received 
is:

Indy9 SysLog Client Output:
    2007-08-18,15:09:05:0000,192.168.1.109,127.0.0.1,4099,SysLogClient,23,Debug,Mail,Test 
Message
    2007-08-18,15:09:05:0000,192.168.1.109,127.0.0.1,4097,SysLogClient,23,Debug,Mail,Test 
Message

Indy 10 SysLog Client Output:
    2007-08-18,15:15:40:0000,,127.0.0.1,4099,SysLogClient,23,Debug,Mail,Test 
Message
    2007-08-18,15:15:40:0000,MAINPC2,127.0.0.1,4097,SysLogClient,23,Debug,Mail,Test 
Message

With the SysLog Server Application rebuilt (and my Service version) the 
output is as follows; so this is defintely a difference in the client 
component.

    2007-08-18,15:26:50:0000,,127.0.0.1,4099,SysLogClient,23,Debug,Mail,Test 
Message
    2007-08-18,15:26:50:0000,MAINPC2,127.0.0.1,4097,SysLogClient,23,Debug,Mail,Test 
Message

This send leaves the host name blank:
        IdSysLog->SendLogMessage( sl, false );
    When sl is created like this:
        TIdSysLogMessage * sl = new TIdSysLogMessage(NULL);
        sl->Msg->Process = ProcessName;
        sl->Msg->PIDAvailable = true;
        sl->Msg->PID     = ProcessID;
        sl->TimeStamp    = Now();
        sl->Msg->Content = message;
        sl->Severity     = severity;
        sl->Facility     = facility;

This send fills in the PC name:
            IdSysLog->SendLogMessage( In_ProcessName, In_Message, 
GetFacility(In_Facility), GetSeverity(In_Severity), true, In_ProcessID );


2)  UDP Client & Server Components, I get the following compile errors. 
Have all these interfaces changed?

------------------------
IdUDPServer->OnUDPRead = IdUDPServerUDPRead;

[C++ Error] CUDPServerTask.Cpp(56): E2034 Cannot convert 'void (_fastcall * 
(_closure )(TObject *,TStream *,TIdSocketHandle *))(TObject *,TStream 
*,TIdSocketHandle *)' to 'void (_fastcall * (_closure )(TIdUDPListenerThread 
*,TByteDynArray,TIdSocketHandle *))(TIdUDPListenerThread 
*,TByteDynArray,TIdSocketHandle *)'
------------------------
IdUDPServer->SendBuffer(ABinding->PeerIP, ABinding->PeerPort, 
IdUDPServer->TxBuf->Udata,TxBuf->Ulength );

[C++ Error] CUDPServerTask.Cpp(258): E2285 Could not find a match for 
'TIdUDPBase::SendBuffer(AnsiString,unsigned short,unsigned char *,unsigned 
long)'
------------------------
IdUDPClient->SendBuffer( Udata, Ulength );

[C++ Error] CUDPClientTask.cpp(258): E2285 Could not find a match for 
'TIdUDPClient::SendBuffer(unsigned char *,unsigned long)'
------------------------
RxBuf->Ulength = IdUDPClient->ReceiveBuffer( Data, DataSize, Wait );

[C++ Error] CUDPClientTask.cpp(278): E2285 Could not find a match for 
'TIdUDPClient::ReceiveBuffer(unsigned char *,unsigned long,unsigned long)'
------------------------

I think thats it.

Regards,

Paul



"Paul at NCF" <notplmacca@clara.co.uk> wrote in message
news:46bec059$1@newsgroups.borland.com...
> Hi,
>
> I don't use many components, but I'll try to install it on BCB6 and test
> the components I do use.
>
> Paul
> 

"Paul at NCF" <notplmacca@clara.co.uk> wrote in message
news:46c70d07$1@newsgroups.borland.com...

> I had to remove include lines for AntiFreeze headers

Please be more specific.  What exactly did you remove, and from where
exactly?

> The server and client test harnesses are working (including
> receiving SysLog data from my wireless ADSL router),
> however I notice a difference in behaviour for two different
> send calls.

What calls are you referring to exactly?  Where are you seeing the data
appear?  Again, you need to be more specific.

> With SysLog Client rebuilt only (and using Indy9 Server) the
> data received is:

Since there is no such output being generated by Indy, I have to assume that
you are formatting the message data into your own string value.  Please show
your actual code so we can see what fields you are using.

> This send leaves the host name blank:

You are not assigning anything to the TIdSysLogMessage::Hostname property in
your code.  Looking at the source code for TIdSysLogMessage, I see that it
tries to assign a default value to that property inside of the
TIdSysLogMessage::InitComponent() method.  However, something I observed
last week was that the InitComponent() method is not always called correctly
under C++.  I have no idea why yet.

> UDP Client & Server Components, I get the following compile errors.
> Have all these interfaces changed?

Yes, which you can see if you look at the declarations in the header files.

The send/receive methods need some explanation.  You can't pass raw memory
buffers to Indy anymore, mainly due to the .NET support.  In this particular
case, you will have to wrap your data into a TIdBytes.  For sending data,
you can use RawToBytes().  For receiving data, you can use BytesToRaw().
For example:

    IdUDPClient->SendBuffer(RawToBytes(Udata, Ulength));

    TIdBytes buf;
    buf.Length = DataSize;
    DataSize = IdUDPClient->ReceiveBuffer(buf, Wait);
    BytesToRaw(buf, Data, DataSize);
    RxBuf->Ulength = DataSize;


Gambit

"Remy Lebeau (TeamB)" <no.spam@no.spam.com> wrote in message 
news:46c9d309@newsgroups.borland.com...
>
> Please be more specific.  What exactly did you remove, and from where
> exactly?

This is in my mainform header.  Not sure why or if I needed to have these 
before, but leaving them in causes the linker error with Indy10.  Commenting 
them out of my application links okay, and doesn't appear to cause any 
problems.

[Linker Fatal Error] Fatal: Unable to open file 'IDANTIFREEZE.OBJ'

#ifndef MainFormH
#define MainFormH
//---------------------------------------------------------------------------
#include <Classes.hpp>
#include <Controls.hpp>
#include <StdCtrls.hpp>
#include <Forms.hpp>
#include <SysLogServerTask.h>
#include <IdBaseComponent.hpp>
//#include <IdAntiFreeze.hpp>
//#include <IdAntiFreezeBase.hpp>
//---------------------------------------------------------------------------

> What calls are you referring to exactly?  Where are you seeing the data
> appear?  Again, you need to be more specific.

I assume my email answered this further on

> Since there is no such output being generated by Indy, I have to assume 
> that
> you are formatting the message data into your own string value.  Please 
> show
> your actual code so we can see what fields you are using.

Yes, this is my output, eg.

SysLogClientTask::ProcessSysLogMessage(TIdSysLogMessage * msg)
{
            AnsiString tstr = "";
            AnsiString seve = "";
            AnsiString faci = "";
            AnsiString mpid = msg->Msg->PID;
            AnsiString mesg = msg->Msg->Content;
            AnsiString proc = msg->Msg->Process;
            AnsiString host = msg->Hostname;
            AnsiString peer = msg->Peer;
            AnsiString prio = msg->Pri;
            TDateTime  time = msg->TimeStamp;
            .... some processing of time string and then
            tstr += host;
            tstr += Separator;
            tstr += peer;
            tstr += Separator;
            tstr += mpid;
            tstr += Separator;
            tstr += proc;
            tstr += Separator;
            tstr += prio;
            tstr += Separator;
            tstr += seve;
            tstr += Separator;
            tstr += faci;
            tstr += Separator;
            tstr += mesg;
            tstr += "\r\n";

> You are not assigning anything to the TIdSysLogMessage::Hostname property 
> in
> your code.  Looking at the source code for TIdSysLogMessage, I see that it
> tries to assign a default value to that property inside of the
> TIdSysLogMessage::InitComponent() method.  However, something I observed
> last week was that the InitComponent() method is not always called 
> correctly
> under C++.  I have no idea why yet.

This is what I saw, I presume there is a way of filling in the property, but 
since the code looked like it should be doing it (and does in one particular 
scenario), I raised it as a query on the Indy10 snapshot.

> The send/receive methods need some explanation.  You can't pass raw memory
> buffers to Indy anymore, mainly due to the .NET support.  In this 
> particular
> case, you will have to wrap your data into a TIdBytes.  For sending data,
> you can use RawToBytes().  For receiving data, you can use BytesToRaw().
> For example:
>
>    IdUDPClient->SendBuffer(RawToBytes(Udata, Ulength));
>
>    TIdBytes buf;
>    buf.Length = DataSize;
>    DataSize = IdUDPClient->ReceiveBuffer(buf, Wait);
>    BytesToRaw(buf, Data, DataSize);
>    RxBuf->Ulength = DataSize;

Thanks, just wanted to be sure before diving down a dead end.  I'll give 
this a go when I can (as I'm also trying to get a system out by next week)

Paul 

Remy,

I waited a couple of weeks and just downloaded the snapshot and tried to 
compile it for BCB5.

I am using the modified FULLC6.bat file and I get an error when executing

dcc32.exe IndyProtocols50.dpk /O..\Lib\Protocols\objs /DBCB /M /H /W /JPHN 
/N. /U. -$d-l-n+p+r-s-t-w-y- %2 %3 %4

The error is below
. .
Borland Delphi  Version 13.0  Copyright (c) 1983,99 Inprise Corporation
IdCompilerDefines.inc(337)
IdASN1Util.pas(402)
IdCompilerDefines.inc(337)
IdCompilerDefines.inc(337)
IdCompilerDefines.inc(337)
IdCompilerDefines.inc(337)
IdCompilerDefines.inc(337)
IdCompilerDefines.inc(337)
IdCharsets.pas(3607)
IdCompilerDefines.inc(337)
IdResourceStringsProtocols.pas(936)
IdGlobalProtocols.pas(1531) Error: Undeclared identifier: 'DirectoryExists'
IdGlobalProtocols.pas(3167)
IdHeaderList.pas(132) Fatal: Could not compile used unit 
'IdGlobalProtocols.pas'

Error!


I have been looking forward to testing this new release but I have not been 
able to compile for a while.  In case you want to know what I am doing I am 
copying the whole batch file below.  There it goes:

*******************************************************************

@echo off

if (%1)==() goto test_command
if (%1)==(start) goto start
goto endok

:test_command
ECHO *************  TEST COMMAND   ***********
PAUSE
if (%COMSPEC%)==() goto no_command
%COMSPEC% /E:9216 /C %0 start %1 %2 %3
goto endok

:no_command
ECHO *************  NO COMMAND   ***********
PAUSE
echo No Command Interpreter found
goto endok

:start
REM call clean.bat
ECHO *************  BEFORE CALLING COMPUTIL   ***********
PAUSE
computil SetupC5
if exist setenv.bat call setenv.bat
if not exist ..\C5\*.* md ..\C5 >nul
REM if exist ..\C5\*.* call clean.bat ..\C5\
if (%NDC5%)==() goto enderror

REM ***************************************************
REM Compile Runtime Package IndySystem60
REM ***************************************************
ECHO *************  BEFORE COPYING THE FILES   ***********
PAUSE

cd System
copy *.pas ..\..\C5
copy *.dpk ..\..\C5
copy *.obj ..\..\C5
copy *.inc ..\..\C5
copy *.res ..\..\C5
copy *.dcr ..\..\C5
copy *.rsp ..\..\C5
if not exist .\objs\*.* md .\objs >nul
cd ..\..\C5

ECHO *************  COMPILING INDYSYSTEM50.DPK   ***********
PAUSE

dcc32.exe IndySystem50.dpk /O..\Lib\System\objs /DBCB /M /H /W /JPHN 
/N. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror
dcc32.exe IndySystem50.dpk /O..\Lib\System\objs /DBCB /M /H /W 
/N. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror
ECHO *************  CREATING THE LIBRARY   ***********
PAUSE
for %%f in (*.obj) do tlib.exe IndySystem50.lib +%%f /P64 > nul

ECHO *************  CLEANING UP FILES   ***********
PAUSE

REM ***************************************************
REM Clean-up IndySystem60
REM ***************************************************
del *.dcu > nul
del *.pas > nul
del *.dpk > nul
del *.obj > nul
del *.inc > nul
del *.res > nul
del *.dcr > nul
del *.rsp > nul

REM ***************************************************
REM Compile Runtime Package IndyCore60
REM ***************************************************
cd ..\Lib\Core

ECHO *************  COPYING CORE FILES   ***********
PAUSE

copy *.pas ..\..\C5
copy *.dpk ..\..\C5
copy *.obj ..\..\C5
copy *.inc ..\..\C5
copy *.res ..\..\C5
copy *.dcr ..\..\C5
copy *.rsp ..\..\C5

if not exist .\objs\*.* md .\objs >nul
cd ..\..\C5

ECHO *************  COMPILING INDYCORE50.DPK   ***********
PAUSE

dcc32.exe IndyCore50.dpk /O..\Lib\Core\objs /DBCB /M /H /W /JPHN /N. 
/U. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror
dcc32.exe IndyCore50.dpk /O..\Lib\Core\objs /DBCB /M /H /W /N. 
/U. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror

ECHO *************  CREATING CORE LIBRARY   ***********
PAUSE

for %%f in (*.obj) do tlib.exe IndyCore50.lib +%%f /P64 > nul
del *.obj > nul

REM ***************************************************
REM Compile Designtime Package dclIndyCore60
REM ***************************************************

ECHO *************  COMPILING DCLINDYCORE50.DPK   ***********
PAUSE

dcc32.exe dclIndyCore50.dpk /O..\Lib\Core\objs /DBCB /M /H /W /Z /JPHN /N. 
/U. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror
dcc32.exe dclIndyCore50.dpk /O..\Lib\Core\objs /DBCB /M /H /W /Z /N. 
/U. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror

ECHO *************  CREATING DCLCORE LIBRARY   ***********
PAUSE
for %%f in (*.obj) do tlib.exe dclIndyCore50.lib +%%f /P64 > nul

ECHO *************  CLEANING UP CORE FILES   ***********
PAUSE

REM ***************************************************
REM Clean-up IndyCore60
REM ***************************************************
del *.dcu > nul
del *.pas > nul
del *.dpk > nul
del *.obj > nul
del *.inc > nul
del *.res > nul
del *.dcr > nul
del *.rsp > nul

REM ***************************************************
REM Compile Runtime Package IndyProtocols60
REM ***************************************************

ECHO *************  COPYING PROTOCOL FILES  ***********
PAUSE

cd ..\Lib\Protocols
copy *.pas ..\..\C5
copy *.dpk ..\..\C5
copy *.obj ..\..\C5
copy *.inc ..\..\C5
copy *.res ..\..\C5
copy *.dcr ..\..\C5
copy *.rsp ..\..\C5
if not exist .\objs\*.* md .\objs >nul
cd ..\..\C5

ECHO *************  COMPILING PROTOCOLS ***********
PAUSE
ECHO *************  FIRST ***********
PAUSE
dcc32.exe IndyProtocols50.dpk /O..\Lib\Protocols\objs /DBCB /M /H /W /JPHN 
/N. /U. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror
ECHO *************  SECOND ***********
PAUSE
dcc32.exe IndyProtocols50.dpk /O..\Lib\Protocols\objs /DBCB /M /H /W /N. 
/U. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror

ECHO *************  DELETING THIRD PARTY FILES   ***********
PAUSE

REM ***************************************************
REM Delete third-party .obj files
REM before compiling the .lib file
REM ***************************************************
del adler32.obj > nul
del compress.obj > nul
del crc32.obj > nul
del deflate.obj > nul
del example.obj > nul
del gzio.obj > nul
del infback.obj > nul
del inffast.obj > nul
del inflate.obj > nul
del inftrees.obj > nul
del minigzip.obj > nul
del trees.obj > nul
del uncompr.obj > nul
del zutil.obj > nul
ECHO *************  CREATING PROTOCOLS LIBRARY ***********
PAUSE

for %%f in (*.obj) do tlib.exe IndyProtocols50.lib +%%f /P64 > nul
del *.obj > nul

REM ***************************************************
REM Compile Designtime Package dclIndyProtocols60
REM ***************************************************

ECHO *************  COMPILING dclPROTOCOLS   ***********
PAUSE
dcc32.exe dclIndyProtocols50.dpk /O..\Lib\Protocols\objs /DBCB /M /H /W 
/JPHN /N. /U. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror
dcc32.exe dclIndyProtocols50.dpk /O..\Lib\Protocols\objs /DBCB /M /H /W /N. 
/U. -$d-l-n+p+r-s-t-w-y- %2 %3 %4
if errorlevel 1 goto enderror
for %%f in (*.obj) do tlib.exe dclIndyProtocols50.lib +%%f /P64 > nul

ECHO *************  DELETING PROTOCOLS FILES  ***********
PAUSE

REM ***************************************************
REM Clean-up IndyProtocols60
REM ***************************************************
del *.dcu > nul
del *.pas > nul
del *.dpk > nul
del *.obj > nul
del *.inc > nul
del *.res > nul
del *.dcr > nul
del *.rsp > nul
cd ..\Lib
goto endok

:enderror
REM call ..\Lib\clean.bat
cd ..\Lib
echo Error!

:endok
pause


*******************************************************************

"JLoz" <jfsierra@excite.com> wrote in message
news:46cf50c7@newsgroups.borland.com...

> IdGlobalProtocols.pas(1531) Error: Undeclared identifier:
'DirectoryExists'

Change line 1531 to call IndyDirectoryExists() instead of DirectoryExists().
In BCB 5, DirectoryExists() still existed in the FilCtrl unit, which Indy
strictly avoids.  DirectoryExists() was moved to the SysUtils unit in BCB 6.


Gambit