ADO RDD xHarbour

[AHF]

Gunther,

I see it now.
The app I m using does not accept single quotes in fields because of some xml extraction.

Try with this in ado_putvalue

xValue := '"'+xValue+'"'

Instead of xValue := "'"+xValue+"'"

[byte-one]

Antonio, this is not functioning (MSSQL 2014). I use

Code: Select all

xValue :="'"+ STRTRAN( xValue, "'", "''" )+"'"

I do not use crypted strings.

[AHF]

Gunther,

Ill post a new build maybe today that solves this problem for all engines.

[AHF]

New build adordd version 1.0 at https://github.com/AHFERREIRA/adordd.git

Changes:

Should compile now with Harbour and without FiveWin but I didnt test it.
Field replaces with special chars (single quotes, apostrophes, etc ) should work now.
DbEval improve speed.
Update 01_README.PDF with adordd error codes

[AHF]

New build adordd version 1.0 at https://github.com/AHFERREIRA/adordd.git

Small bug corrections:

Corrected bug when updating a indexkey with UDF index.
Corrected bug when adding new record with UDF index and UDF index is empty.
Corrected bug when ZAP was given an error after when refreshing recordset.

[AHF]

New build adordd ver 1.0 at https://github.com/AHFERREIRA/adordd.git

Changes:

Finally it seems that we managed to increase dramatically the opening times for huge tables and much better memory usage.

1) New SET ADO PRE OPEN THRESHOLD TO nRecords

This set will define how adordd will open the tables during app run time.

All the tables with records equal or greater than nRecords and without any WHERE opening clause
will be all open during app initialization.

You will wait a little for the app to start but after all table opening it will be very fast.
This will consume much memory at start but less during app run time.

You should balance yourself the value of nRecords accordingly to the time you are willing to wait to open
a table and your memory availability.
Its very easy opening a couple of tables > 500.000 recs to use some gigas of memory.

Even without this SET adordd will only take some time opening a table if its not open already
otherwise it will be very fast.

adordd its now extremely fast and can be truly compared to any other dbfcdx kind of rdd.

Ive made some trials and it seems to be running ok. Please check it yourselves and revert.

2) Code cleaning

3) Compile improvement with Harbour / xHarbour with or without FiveWin.

4) 01_readme.pdf new SET information.

[pieter]

Antonio,

Thank you for this adordd version and adordd in general.

I tried to compile/link this newest adordd.

I got the following error:

Error: Unresolved external '_HB_FUN_HB_PS' referenced from C:\PIETER\XHARBOUR123
\LIB\ADORDD.LIB|adordd

Pieter

[cnavarro]

I have not had time to try your great work
I hope I can use it from now
Thank you for your effort

[AHF]

Pieter,

You must compile and link adordd.prg not adordd.lib.

There is a bug in INDEX ON temporary files without any UDFs or index condition otherwise its ok.
Ill post a new corrected build asap.

[AHF]

Cristobal,

Thanks, Im looking forward to your comments.

[pieter]

Antonio, oké, thanks.

Same error:
Error: Unresolved external '_HB_FUN_HB_PS' referenced from C:\PIETER\UB404-8SEPT
EMBER2015-V4\UB\ADORDD.OBJ

Am I doing something wrong with compiling/linking (or is it because of the bug)?

Pieter

[AHF]

Pieter,

Are you using Harbour or xHarbour?

Try comment #include "hbcompat.ch"

[pieter]

Antonio, I use Xharbour. I comment #include "hbcompat.ch" like you suggested. it is solved now. Thanks.

[AHF]

Pieter,

Strange I use xHarbour and it compiles ok with that line not commented.

Could you try this to see if it works?

#ifndef __XHARBOUR__
#include "hbcompat.ch"
#endif

[pieter]

Antonio,

#ifndef __XHARBOUR__
#include "hbcompat.ch"
#endif

With above code my program compiles and link.

Pieter