Zeoslib Portal

Well, I doubt the endianness of the current platform could affect GUID to string conversion. Otherwise it is bug. Actually I see no real cases where difference of GUID endianness could play role. Could you imagine a real, believable example? By real I mean those which are useful not theoretical case...

Hi,
not much people use events they are not even being checked in test suite. So if you provide a simple test case, it likely will be included to catch such cases

Hello Jan, If you have to do a manual search, which value would you put into your search? a) AC76BA861033F40077601000003D0004 b) 86BA76AC331000F477601000003D0004 c) something else? What do you mean by "manual search" here? If SELECT GuidField FROM ... then there will be no problems as FB d...

maybe we should do a conversion here between big endian and little endian? Otherwise the ASCII representation of GUIDs will not be the same? Well, we could. But is this difference really important? Server does no byte swapping, it's all about how it interprets the value. And this only affects strin...

Mostly done here https://github.com/Fr0sT-Brutal/Zeos_dev/tree/branches/testing-7.3 Additionally done: - Set Guid type by field names (DS-level option) - Set Guid type to SP result fields - by type or by field names Changes will be committed to SVN as soon as SF ends their long-lasting disfunction p...

Michael, sure, done

Michael, Let's rollback my stub. Yep I think it would be better. There's no sense in bloating the code if it won't help. Question do we talk about RangeChecks or OverFlow checks here? Range checks. Check this option in compiler settings of Test suite - and get C1118. Uncheck it - and app will be com...

Nope this is hopeless. Adding some typecasts won't help because then appears another errors: [Error] ZDbcResultSet.pas(358): Declaration of 'GetULong' differs from declaration in interface 'IZResultSet' which could be cured only by disabling range checking for the whole unit. Here are my changes to ...

I can't check now, but doubt something changed. This error is hardly related to Zeos code...

Try r4100. I did bugfix there regarding GUIDs

Michael, nevermind :) with keywords array totally hosted by Metadata all discussion regarding creation/caching objects loses its sense. The solution is brilliant! I use Zeos in mutithreading env. too, but plain dbc only. Using in MT is OK as long as you keep each component in its own thread. Driver ...

Michael, I also added some words to the previous post Thanks for that new solution is pretty nice! Cool :) which means the converter is assigned somwhere before the Connection is opened? I'm not sure I got what you mean... Metadata is created with an opened connection, AFAIU. But DBInfo is created a...

Hi Michael, could you please rollback the simplification Yep, did it just now. Having paths declared in IDE settings makes this kind of bugs hard to detect... Does it meen you prebuild a TStringDynArray oslt? Thought Pos() schould be faster? Is it slower if the String is precomposed like in my examp...

note the IdentiferConv (short IC ) makes loads of roundtrips to dermine Lower/Upper/Mixed cases. This could be optimes to one rountrip if we retrieve an enum. You've guessed right my thoughts :) I even implemented this change (just a test class I extracted from identconv, and I haven't tested it wi...

Do it! I've noticed you did a nice simplification on my ugly code Okay, I just didn't want to break your commit combo :lol: Will it? Needs to be tested. IDK, this quoting is used in ASA and DbLib only which I do not use... And this job can't be made in atomic commits. Either all or nothing.. I'll r...