Compare ASCII converted string in C/AL

grouic

Hello,

I'm importing data from a csv file that potentially contains french or spanish characters (éàñ ... etc).
To have the correct character string in Navision, I use the well known ansi2ascii fonction, that works just fine.

However, it seems that there is a problem when comparing a converted string with a string written in C/AL !

for eg :

var_description = 'hello world' (value of the import)

description := ansi2ascii(var_description);
if description = 'hello world" then ... etc

For Navision description and 'hello world' are not the same strings ....

NB : it works if I write this :

if description = ansi2ascii('hello world') then ... etc

Anyone got this problem ?
Do I really need to convert ALL the strings I need to compare in C/AL ?????

Thx for the help

Dave_Cintron

You may need to compare the upper range of character strings you are converting (128-255) because the Navision code page is not exactly the same as the Windows code page. You would compare what's being imported with the result.

ajhvdb

I found using ansi2ascii created new problems for me. It looks a bit like yours and haven't found a solution for this yet

Sometimes (Depending on the programs you use later, f.i. Excel) it converts a "space" to a lower value which in Excel is interpreted as an "à" character. Try your code again but now with "HelloWorld" (without the space).

grouic

Hey ajhvdb, you're right. I tested a comparison on strings not containing spaces, and it works.
I'm continuing my investigation

Sog

you could replace spaces with the expression "[space]" (or another expression of your choice), convert the string with ansi2ascii and replace your expression again with regular spaces

grouic

OK I solved my problem.

In the MakeVars function (CUnit ANSI <-> ASCII converter), I just removed the last char of the strings AsciiStr and AnsiStr.
The both were visually spaces but only the char in the AnsiStr was a "real" space.

I did not find which ascii char was used as "space" in the AsciiStr (in a unix file it's displayed as "| ").

I dont know if it's a bug in the CU.

ajhvdb

Glad you solved it.

Could you test the following:
> Open csv in Excel
> Saveas new csv
> Compare the two csv files or import in NAV

In the new csv file, I still have the "à" chars but maybe it is only with my Office version