Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

SY576

53 views
Skip to first unread message

Peter Zolja

unread,
Sep 16, 2003, 11:43:16 AM9/16/03
to
I have a project + steps to reproduce "Internal error: SY576" in Delphi 7. I
did some googling and it looks like nobody was able to reproduce this
internal error. I was wondering if you guys know if this was reproduced
before. If not please let me know about a QC# I could post the steps and
project (assuming that after I remove some sensitive code -- not much -- it
would still fire).

Thanks.

beny

unread,
Sep 16, 2003, 11:53:57 AM9/16/03
to
I've been getting the same error sporadically in D7 too, try closing the
project and deleting all the DCUs.
works for me
HTH

"Peter Zolja" <csu10711 [at] mail.claytonstate.net> wrote in message
news:3f67...@newsgroups.borland.com...

Peter Zolja

unread,
Sep 16, 2003, 12:45:11 PM9/16/03
to
> I've been getting the same error sporadically in D7 too, try closing the
> project and deleting all the DCUs.
> works for me

There may be more than one problem that triggers SY576. In my case it
doesn't matter if I delete the dcu or if I close and reopen the project. It
only fires on this project (i.e. that I can reproduce) and if I do certain
steps (i.e. I restart the machine, I load the project, I do the steps and I
get the error). If I don't do those steps it doesn't fire, and everything
seems to be working fine. BTW: I have to restart the IDE to get rid of the
error; once it fires you can't compile anything.

I would like to know if anybody was able to reproduce the problem, and if
not I could go ahead and post a version so that more people could test it
and see if it works for them (and eventually put it on QC).

Thank you all.


Robert Love

unread,
Sep 16, 2003, 1:27:22 PM9/16/03
to
>I have a project + steps to reproduce "Internal error: SY576" in Delphi 7. I

There is a report for the Internel Error but it need some majore work
specifically reproducible steps. QC #5951

Robert Love


Peter Zolja

unread,
Sep 16, 2003, 2:10:10 PM9/16/03
to
> There is a report for the Internel Error but it need some majore work
> specifically reproducible steps. QC #5951

Cool, but I couldn't find a way to upload an attachment... I can list the
steps required, but I need to upload the project I'm working on (it doesn't
fire on just any project).

Thanks.

Peter Zolja

unread,
Sep 16, 2003, 2:12:36 PM9/16/03
to
> Cool, but I couldn't find a way to upload an attachment...

What I mean by this is that if I click on "Attachment" I get "Internal
Application Error" -- is there a different way to attach something?

Dave Nottage (TeamB)

unread,
Sep 16, 2003, 8:15:01 PM9/16/03
to

Can you please send the attachment to borland.public.attachments, and I (or
someone else who can if they get there before me) will add it to the report.

Thanks.

--
Dave Nottage (TeamB)


Peter Zolja

unread,
Sep 16, 2003, 10:54:27 PM9/16/03
to
> Can you please send the attachment to borland.public.attachments, and I
(or
> someone else who can if they get there before me) will add it to the
report.

Sure, look for "Delphi SY576" in b.p.a.
Further instructions (i.e. steps) in UMain.pas in the archive.

P.S. FWIW In some cases when I received SY576 (not with this project) when I
was closing the project I was getting that infamous repeating AV (where you
need to end the task to get rid of the errors); so I think these two may be
related somehow...


Unknown

unread,
Sep 16, 2003, 11:04:39 PM9/16/03
to
Peter Zolja wrote:

> Sure, look for "Delphi SY576" in b.p.a.
> Further instructions (i.e. steps) in UMain.pas in the archive.

I was able to reproduce it using the attachment. I already rated
QC#5951 higher expecting someone to update it with the attachment and
the steps.

--
Leonel
"Reality is merely an illusion, albeit a very persistent one." - Albert
Einstein

Robert Love

unread,
Sep 17, 2003, 1:48:29 AM9/17/03
to
>Sure, look for "Delphi SY576" in b.p.a.

I have attached the file to the QC item. I have also updated the
steps. Thanks for taking the time to put this together.

Robert Love

Arthuro

unread,
Sep 17, 2003, 5:32:51 AM9/17/03
to
In borland.public.delphi.non-technical, Leonel <> wrote in message
<3f67...@newsgroups.borland.com>...

>Peter Zolja wrote:
>
>> Sure, look for "Delphi SY576" in b.p.a.
>> Further instructions (i.e. steps) in UMain.pas in the archive.
>
>I was able to reproduce it using the attachment. I already rated
>QC#5951 higher expecting someone to update it with the attachment and
>the steps.

Rated it, and gave it a vote. Real bugs deserve votes..
--
***Posted by Jake's Custom Newsgroup Reader***

Posted using Jake's Super Newsreader 0.9.2.953

Unknown

unread,
Sep 17, 2003, 7:04:37 AM9/17/03
to
Arthuro wrote:

> Rated it, and gave it a vote. Real bugs deserve votes..

I can't give a vote to it since all my other votes are in real bugs,
that are affecting me.

dk_sz

unread,
Sep 17, 2003, 10:25:24 AM9/17/03
to
> I can't give a vote to it since all my other votes are in real bugs,
> that are affecting me.


Hehel, I don't think Borland will fix your bugs
unless it makes it to Top10 stupid requests :-).
I have all my votes burried in real bugs as well.


best regards
Thomas


Robert Love

unread,
Sep 17, 2003, 10:57:21 AM9/17/03
to
>Hehel, I don't think Borland will fix your bugs
>unless it makes it to Top10 stupid requests :-).
>I have all my votes burried in real bugs as well.

Do you really think (If/When) Borland releases a patch
for D7 it will only include 10 bug fixes. In the past far more than
10 bugs have been fixed in major patches.

Robert Love

dk_sz

unread,
Sep 17, 2003, 12:08:59 PM9/17/03
to
> Do you really think (If/When) Borland releases a patch
> for D7 it will only include 10 bug fixes. In the past far more than
> 10 bugs have been fixed in major patches.

Of course the Top10 refers to Anders' Top10 list
of voted "bugs" (actually most were requests).
Maybe the way I put it was on the edge, but..
I think you get the idea ;-)


best regards
Thomas


Colin Wilson

unread,
Sep 19, 2003, 9:40:13 AM9/19/03
to
Peter Zolja wrote:

> I have a project + steps to reproduce "Internal error: SY576" in
> Delphi 7. I did some googling and it looks like nobody was able to
> reproduce this internal error.

See if you can track down Dennis Passmore in one of the groups. He
seems to know more about 'Internal Errors' in Delphi than anyone else,
and has helped me figure them out several times in the past.

--
Colin - using XanaNews HTTP Transport
e-mail :co...@wilsonc.demon.co.uk
web: http://www.wilsonc.demon.co.uk/delphi.htm

Posted with XanaNews 1.15.7.1

johnnie

unread,
Sep 19, 2003, 9:59:29 AM9/19/03
to

Dennis had an accident recently and will not attent at this groups for
a while for more info take a look on B.P.D.Language.Basm for the
anounsement.

regards
johnnie.

Colin Wilson

unread,
Sep 19, 2003, 10:02:34 AM9/19/03
to
johnnie wrote:

> Dennis had an accident recently

Sorry to hear that :(

John Kaster (Borland)

unread,
Sep 19, 2003, 12:47:37 PM9/19/03
to
johnnie wrote:

> Dennis had an accident recently

Ouch. Hope he recovers oompletely and quickly!


--
John Kaster, Borland Developer Relations, http://bdn.borland.com
$1280/$50K: http://homepages.borland.com/jkaster/tnt/thanks.html
Make a wish: http://qc.borland.com * Get source
http://codecentral.borland.com


dennis_passmore

unread,
Sep 21, 2003, 4:37:17 PM9/21/03
to
Accident??

Not this Dennis!

I am still lurking around here (;>).


Dennis Passmore
Ultimate Software, Inc.

Danny Thorpe

unread,
Sep 22, 2003, 2:47:57 AM9/22/03
to

"Peter Zolja" <csu10711 [at] mail.claytonstate.net> wrote in message
news:3f67...@newsgroups.borland.com...
> I have a project + steps to reproduce "Internal error: SY576" in Delphi 7.

SY576 indicates that a cross-unit symbol reference has gone bad. The
compiler set out to import (reference) a symbol from another unit, but the
symbol that came back doesn't look right.

Look for references to symbols from other units in your code. Elaborate
fully qualified references might contribute, or references to types that are
defined in units that are not explicitly named in your uses clause (this is
sometimes possible with enums).

If you can identify the specific types that are causing the problem, moving
them into the same unit would work around it.

-Danny


dennis_passmore

unread,
Sep 22, 2003, 11:55:44 AM9/22/03
to
I could not reproduce the error but I would be suspect of the following declarations:

------------------------------
type TArrayOfString = array of string;

function SplitString(source:string;delimiter:string):TArrayOfString;overload;
procedure SplitString(source:string;delimiter:string;var result:TArrayOfString);overload;
------------------------------

It is one thing to overload a function or procedure but to change the function/procedure
type in the overload declaration is something I would never attempt. No wonder the
compiler is getting confused.

Rudy Velthuis (TeamB)

unread,
Sep 22, 2003, 2:03:54 PM9/22/03
to
johnnie wrote:

>
> Dennis had an accident recently and will not attent at this groups for
> a while for more info take a look on B.P.D.Language.Basm for the
> anounsement.

That is another Dennis. I have never seen his last name, but he is not
Dennis Passmore, and not Dennis Landi. <g>

--
Rudy Velthuis (TeamB)

"Why yes-- a bulletproof vest." - James Rodges, murderer, on his final
request before the firing squad.

Danny Thorpe

unread,
Sep 23, 2003, 2:01:59 AM9/23/03
to

<Dennis Passmore> wrote in message
news:mi6umvglf6celsgn7...@4ax.com...

> I could not reproduce the error but I would be suspect of the following
declarations:
>
> ------------------------------
> type TArrayOfString = array of string;
>

That's certainly a strong candidate.


> function
SplitString(source:string;delimiter:string):TArrayOfString;overload;
> procedure SplitString(source:string;delimiter:string;var
result:TArrayOfString);overload;
> ------------------------------
>
> It is one thing to overload a function or procedure but to change the
function/procedure
> type in the overload declaration is something I would never attempt. No
wonder the
> compiler is getting confused.

These overloads shouldn't be a problem. They have different parameter
counts, so the types are pretty much irrelevant for overload resolution.
Function results are not part of overload resolution.

-Danny

0 new messages