[DUG]: MDI

2001-01-10 Thread Matthew Comb
Are MDI forms still being used or are they fairly redundant these days. What is the easiest way to make sure that your whole app only gets one icon in the taskbar and when you minimise second and third etc forms that the whole application minimises? Cheers, Matt.

[DUG]: ADO

2001-01-10 Thread Stephen Bertram
I'm new to ADO and have a quick question. When I open an ADOQuery and navigate through the result set till eof is true I cannot close the ADOQuery - it returns the error: "Either BOF or EOF is True, or the current record has been deleted. Requested operation requires a current record" If not

[DUG]: Multi language development

2001-01-10 Thread Stephen Bertram
I'm about to start a project with multilanguage support. I presume that there is somenone on the list who will be doing this already. I am interested in their thoughts. Is the Delphi language handling comprehensive? Is it preferable to roll your own conversion methods? What rules must be

[DUG]: What MIGHT be going on

2001-01-10 Thread vss
Hi all. Last week 1/2 I have been writting a mail server for work. Its al working fine etc, but I have observed something that puzzles me. If I goto the task manager, then to the processes tab, I can locat my exe in the list. It runs, when open at roughly 4500K, which is fine.(mainly due to me

Re: [DUG]: ADO

2001-01-10 Thread Neven MacEwan
Stephen Is this using MDAC 2.6 'cause I think its a known bug (there is a fix out for it) or you could go back to 2.5 Neven - Original Message - From: Stephen Bertram [EMAIL PROTECTED] To: Multiple recipients of list delphi [EMAIL PROTECTED] Sent: Thursday, 11 January 2001 09:56

Re: [DUG]: What MIGHT be going on

2001-01-10 Thread Nic Wise
AFAIK its a windows thing - windows minimizes its working set when you minimize the app, 'cos you are not interacting with it, so it doesn't have to care about things like back buffers, keyboard interaction etc. JBuilder and Delphi are the same. On an initial boot with about 15 files open, its

Re: [DUG]: What MIGHT be going on

2001-01-10 Thread Greg Amer
AFAIK the memory usage in the tack/process manager is a (extremely) vague guideline and not that close to real memory usage. Everything I have used to check real memory usage has always reported wildly different amounts to what the windows task manager has reported. Nic Wise wrote: AFAIK its a

Re: [DUG]: What MIGHT be going on

2001-01-10 Thread Peter Hyde
Nic wrote: AFAIK its a windows thing - windows minimizes its working set when you minimize the app, 'cos you are not interacting with it, so it doesn't have Not purely a Windows things. It's underdocumented and I don't recall the "why" details, but a Delphi-related "fixer" unit was

Re: [DUG]: What MIGHT be going on

2001-01-10 Thread jnorth
Do you use this Peter? TrimMem infomation removed but acknowledged Thanks for the other reply as well. I might put it into the bug list anyway and see what happens... --- New Zealand Delphi Users group - Delphi

Re: [DUG]: What MIGHT be going on

2001-01-10 Thread Peter Hyde
Jed wrote: Do you use this Peter? TrimMem infomation removed but acknowledged Lots at first, but no longer need to in most cases as the same logic was incorporated into WebHub's initialisation handling, so all our W/H projects, at least, are covered. cheers, peter

[DUG]: ignore...just a test

2001-01-10 Thread vss
please ignore --- New Zealand Delphi Users group - Delphi List - [EMAIL PROTECTED] Website: http://www.delphi.org.nz To UnSub, send email to: [EMAIL PROTECTED] with body of "unsubscribe delphi"

Re: [DUG]: ADO

2001-01-10 Thread Neven MacEwan
Stephen Last time i lloked it was an 'interim' fix but now its official http://www.borland.com/devsupport/delphi/mdac26.html Neven - Original Message - From: Stephen Bertram [EMAIL PROTECTED] To: Multiple recipients of list delphi [EMAIL PROTECTED] Sent: Thursday, 11 January 2001

[DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Stephen Bertram
I need a bulletproof routine to create unique integer IDs for inserting rows into multiple tables in a MSSQL 7 database. The catches are that I want the ID's to be unique in the database, not just each table, and the generator must be available to stored procedures. Also the generator needs to

[DUG]: XML DOM/Parser in Delphi

2001-01-10 Thread David Brennan
We are about to start a small EDI project which will involve passing around XML documents. Initially our end of the bargain is mainly producing XML but eventually we will probably be receiving (and hence parsing) it as well. We would like to do this in an official manner (ie stick to standards on

RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Edge, Martin (SSABSA)
use the NewID() function IE: Create table fred (Globally_Unique_Id uniqueidentifier default NewID() not null) will give you a GUID that is unique across everything. The NewID() function will return a GUID and it will never return the smae one again. I need a bulletproof routine to create

Re: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread David Brennan
The NewID() function will return a GUID and it will never return the smae one again. ... or not for a couple of quadrillion years anyway! :-) David Brennan. DB Solutions Ltd. --- New Zealand Delphi Users group -

RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Gary T. Benner
[Reply] Your requirement needs to have a process that is not available in most databases. You need some logic to not only give you a sequential number, but to also remember "roll backed" ID's and issue those before generating new ID's. This usually requires the creation of a pool of available

Re: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Neven MacEwan
Stephen Code your own 'generator' as a stored proc using a UniqueIdentifier (GUID) datatype and the NEWID() function Neven - Original Message - From: Stephen Bertram [EMAIL PROTECTED] To: Multiple recipients of list delphi [EMAIL PROTECTED] Sent: Thursday, 11 January 2001 15:31

RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Paul Ritchie
I need a bulletproof routine to create unique integer IDs for inserting rows into multiple tables in a MSSQL 7 database. If you are happy to use 2 integers to uniquely identify such a row, then you could use the ID of the table in sysobjects as well as an identity column in the table itself.

RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Stephen Bertram
The discussion so far has brought up 2 issues for me - Why not use GUIDs Size Difficulty in manual SQL manipulation Lack of a directly mapable datatype in Delphi Unavailability in some databases - we interface to Ingres Why use integers? Habit

Re: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Nic Wise
The suggestion that Gary has put forward is exactly what I was thinking of, but I still don't see how to "remember" rolled back ID's. do you really need to? does 1 2 3 6 7 8 9 10 14 16 17 make a difference, as long as they are all in order whats it for? N

RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Stephen Bertram
The problem is : Process 1 generates 11 Process 2 generates 12 Process 3 generates 13 Process 1 issues a rollback - ID 11 is now unused Process 4 generates 11 - based on rolledback seed value. Process 5 generates 12 !! Duplicate Process 6 generates 13 !! Duplicate Or am I missing something

RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Paul Ritchie
but I would be more tempted to add a seperate timestamp or datetime column containing the time the row was inserted. Whoops. Upon reflection I don't think a timestamp column would add any more information as to the order the rows were added than the original identity column already provides.

Re: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Neven MacEwan
Stephen I assume that one of the advantages of NewID() is that it is fast (or should be) So you could use NewID() to generate the ID and then just use the top 4 bytes (these are sequential) in your stored proc (in SQL2000 can a user defined procedure be called as a default?) ie Create Table test

Re: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Neven MacEwan
Re The problem is : Process 1 generates 11 Process 2 generates 12 Process 3 generates 13 Process 1 issues a rollback - ID 11 is now unused Process 4 generates 11 - based on rolledback seed value. Process 5 generates 12 !! Duplicate Process 6 generates 13 !! Duplicate more like

Re: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Nic Wise
Process 1 generates 11 Process 2 generates 12 Process 3 generates 13 Process 1 issues a rollback - ID 11 is now unused Process 4 generates 11 - based on rolledback seed value. Process 5 generates 12 !! Duplicate Process 6 generates 13 !! Duplicate Or am I missing something here? I can't

RE: RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Gary T. Benner
[Reply] Stephen, I think that from what you say, gaps are not a problem, it is uniqueness that you are looking for. Am I right? Assuming I am (for once this year), then all you need to do is to generate the unqiue ID outside the transaction that will use it. eg Process 1: Start

RE: RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Paul Ritchie
(Interbase is so easy in this sense because generators are produced outside any transactions). So are SQL Server identity columns, and I've got to say they're even easier. ducks for cover Paul Ritchie. --- New

Re: RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Neven MacEwan
Paul The two probs with "identity columns" are a/ They are only unique to a connection and b/ They are only held for the last action (ie if your insert fires a trigger and the trigger insert and idenity) you are stuffed PS Did I note that Gary had posted this discussion to DUG tut tut

[DUG]: testing again....ignore

2001-01-10 Thread Jeremy Coulter
Please just ignore this

RE: RE: [DUG]: [DUG-DB]: Unique IDs

2001-01-10 Thread Paul Ritchie
Neven, Well you learn something new every day eh? a/ They are only unique to a connection and I didn't think so. Aren't they generated from within the database? I use them in stored procedures and I don't see how they could be only unique to a connection. Can you point me to documentation

RE: [DUG]: testing again....ignore

2001-01-10 Thread Patrick Dunford
Who knows, there is probably a test group out there somewhere that you could test your mail server on. -Original Message-From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Jeremy CoulterSent: Thursday, 11 January 2001 18:04To: Multiple recipients of list

[DUG]: last test....sorry

2001-01-10 Thread Jeremy Coulter