We have some CSV files that are written according to the OEM character encoding.
We're facing some problems with the "ã/é/ê/etc" characters while loading those files on a W2K server. They are not properly recognized.
This ONLY happens when the RELOAD is done on a W2K machine. When we test it on a XP one it works fine. Also when we reload on XP and futher open the QVW on W2K it works fine. Thus the problem is with the RELOAD process on W2K.
Version is 9.00.7320.7 SR2.
We need it to be OK on the W2K machine cause this is the official one (server).
As usual , I have provided a tiny QVW to test. Inside the package (ZIP) you'll have :
- The QVW file itself : TEST161P.QVW that has just been reloaded on a W2K machine. You can open it there and RELOAD on XP to see how it becomes OK.
- The input file, TEST161P.CSV written using OEM char encoding. Just in case you want to "see" what's really written in there you can open TEST161P_ANSI.CSV on Notepad and you'll have what I've written.
- The QVD file generated by the script. Notice that it is very important that not only characters are loaded OK but also are EXPORTED OK to the QVD file. Cause we use A LOT of QVD files here. We have a kind of a database of QVD files. The development stage only uses QVDs. We dont read CSVs daily. CSVs are transformed into QVD just as they are generated. The the CSVs are discarded and we keep only the QVDs (faster/smaller).
Any ideas ?
Anyone else has a W2K machine to test ?
Would there be something OUTSIDE QV itself that is disturbing the RELOAD process ?
Thanks !
BTW great forum this one !
See ya !
