topic Re: Conversion of char to float causing decimal wrong decimal value in Application Development and Automation Discussions

Conversion of char to float causing decimal wrong decimal value

Former Member — Tue, 11 Feb 2014 09:22:02 GMT

Hi ,

I have float value 2.6000000000000003E-03 which i converted to char using fm RS_CHECK_CONV_EX_2_IN_NO_DD

the result of it is 0.002600

later when i move the value back to float result is 2.5999999999999993E-03

same code :

data:lv_temp(128),

lv_value type zflpt. " zfltp domian is FLTP

lv_value = '2.6000000000000003E-03'.

lv_temp = lv_value.

ls_convert-type = 'P'.

ls_convert-length = 13.

ls_convert-decimals = 6.

ls_convert-ddic_leng = 13.

ls_convert-sign = c_true.

CALL FUNCTION 'RS_CHECK_CONV_EX_2_IN_NO_DD'

EXPORTING

input_external = lv_temp

descr = ls_convert

IMPORTING

input_i_format = lv_temp

lv_temp has value 0.002600

but when i move lv_temp to lv_value value is 2.5999999999999993E-03.

Please help me in solving this.

Regards,

Srikanth

Re: Conversion of char to float causing decimal wrong decimal value

Former Member — Tue, 11 Feb 2014 11:37:42 GMT

Hi Srikanth,

Can you please specify why you need to convert back from char to float ?

Re: Conversion of char to float causing decimal wrong decimal value

Former Member — Tue, 11 Feb 2014 11:44:04 GMT

Hello Rachna agarwal,

Actually we have a scenario we check for the decimals its our application specific, and i ave found solution too by using fm 'CHAR_FLTP_CONVERSION' we ot the desired value, thank you.

Re: Conversion of char to float causing decimal wrong decimal value

ThangaPrakash — Tue, 11 Feb 2014 12:08:02 GMT

Hello Srikanth,

Try using nthe below FM.

CALL FUNCTION 'CHAR_FLTP_CONVERSION' "Format conversion: Character --> Floating point

EXPORTING

* dyfld = SPACE "Screen Field name for cursor positioning

* maskn = SPACE "Masking character if input screen

* maxdec = '16' " t006-decan Maximum number of input decimal places

* maxexp = '59+' " t006-expon Maximum gross power of 10

* minexp = '60-' " t006-expon Minimum gross power of 10

string = <pass ur char value here> "CHAR input field for conversion to FLTP

* msgtyp_decim = 'W' " sref-msgtyp Message type for 'TOO_MANY_DECIM' (see domains)

IMPORTING

decim = "Number of decimal places determined

expon = "Gross power of 10 determined

flstr = "FLTP output field, from CHAR conversion

ivalu = "Value in STRING flag

EXCEPTIONS

EXPONENT_TOO_BIG = 1 " Gross power of 10 is too big

EXPONENT_TOO_SMALL = 2 " Gross power of 10 is too small

STRING_NOT_FLTP = 3 " STRING cannot be interpreted as floating point

TOO_MANY_DECIM = 4 " Too many decimal places.

Regards,

Thanga