>>>>

wMessBody = wMessBody + CHR(13) + ALLTRIM(STR(ScanGrL.OrdNmb))

>>>>
>>>>When I get time I might compare interpreted chr(13) versus a variable containing chr(13), also transform() versus alltrim(str()) ...
>>>
>>>Do not forget comparing with literal containing chr(13) -- 0h0D. Finally there is no need to call a function or store something to a variable what is a clear literal.
>>>;)
>>
>>Learned NOT to use certain characters in string literals (in particular anything with ASCII value of 128~255). Such characters are used for encoding characters in double-byte character systems used in some East Asian locales (in particular Chinee, Japanese, and Korean). A program with these would work OK in English and various other Western European locales, but will trigger all sorts of runtime errors when running in East Asian locales (due to the double-byte character encodings). The problem would *not* occur if you'd constructed the string by concatenating with CHR() with values in the 128~255 range.
>
>Thanks for the info - I will consider this if I go into this market. For now I'm in a niche with a German customer - there is no danger. Anyway 0h0D is below 0h80 by a wide margin. :)

FYI -- this occurred with a few of our customers in the USA -- where they had some of their computers set up with an East Asian locale.