Re: Sprintf and decimal point

I think it's premature to recommend a solution. My concern is that "someone" has set up this job with two locales -- neither of which use the period as a decimal point. IT_IT_E (Italian with Euro support) specifies that a space be used for the decimal point while NO_NO (Norwegian) specifies that a comma be used for the decimal point.

This leads me to ask "who" is complaining about the comma being used?

If the end user, then most end users I've met would be very upset if some of their applications used a comma and some used a period (they usually want a consistent representation). Which do they want? If they want "everything" Norwegian EXCEPT the decimal point then I would create a new (customized) locale for them and associate that locale with the job. The source for all of the IBM provided locales can be found in QSYSLOCALE/QLOCALESRC with the member name matching the locale name. To create your own locale I would copy the IBM source to another source file (and another member name), modify the decimal point definition, create a new locale using the CRTLOCALE command, and then associate that locale with the user.

If it's the application programmer complaining because they are parsing the buffer returned by sprintf() for a hardcoded period, then I would say "too bad". They are trying to run a non-internationalized application in an international environment -- fix the program and don't modify the user environment to accomodate the developer. The current value for the decimal point (and many, many, many other national language settings) can be found easily with the QlgRetrieveLocaleInformation API. Set the precedent now to retrieve the user preferred decimal format and use that when scanning the buffer. It's much better to do it right from the start :)

I hope this helps,
Bruce Vining
Bruce Vining Services

beppecosta <beppecosta@xxxxxxxxxxx> wrote:
Hi Bruce,

At present, WRKENVVAR shows two variables:

lang = '/QSYS.LIB/IT_IT_E.LOCALE'
LANG = '/QSYS.LIB/NO_NO.LOCALE'

I think I should change the uppercase one.

Is there a way to code it in the program, without
impacting the other programs that maybe run in the job
?

Thanks-
Giuseppe.

--- Bruce Vining ha scritto:

The sprintf() API is locale sensitive so what you
need to do is specify a locale that uses the period
as a decimal point. One quick way to do that is to
WRKENVVAR and change the job level environment
variable LANG from whatever it currently is to
"/QSYS.LIB/EN_US.LOCALE". Assuming you haven't
modified this IBM supplied locale, running your
program should now show the period being used for
the formatting of doubles.

There are many, many other ways to specify a
locale. This is simply one quick and easy way.
Note that this change may also impact the running of
several other C library functions that are locale
sensitive.

Hope this helps,
Bruce
Bruce Vining Services

beppecosta wrote:
I have a C program that prints a float with
sprintf.

The decimal point is comma. How can I have the point
separator instead ?

I have tried to compile with TGTCCSID(37), to
compile
from a job with CCSID(37) but it still prints with a
comma.

Thanks.




___________________________________
L'email della prossima generazione? Puoi averla con
la nuova Yahoo! Mail:
http://it.docs.yahoo.com/nowyoucan.html
--
This is the C programming iSeries / AS400 (C400-L)
mailing list
To post a message email: C400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit:
http://lists.midrange.com/mailman/listinfo/c400-l
or email: C400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the
archives
at http://archive.midrange.com/c400-l.




Bruce
Bruce Vining Services
507-206-4178
--
This is the C programming iSeries / AS400 (C400-L)
mailing list
To post a message email: C400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit:
http://lists.midrange.com/mailman/listinfo/c400-l
or email: C400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the
archives
at http://archive.midrange.com/c400-l.