how to decode base64 encoded zpl data back to ASCII/EBCDIC

[Scott Klement]

data written to ifs file i do see as zpl data but when i check value of variable ZplOut using debug, after base64_Decode is called it still shows as binary or hexadecimal.

The data is not EBCDIC. If you view it as ASCII or UTF-8 instead of EBCDIC, you'd be able to see it. Alternately, if you need it in EBCDIC, you can translate it.

IFS file shows data in UTF-8 because file is created with CCSID 1208?

In my example, I created the file as CCSID 1208, which is the CCSID number for UTF-8. I don't know if that's what it should be -- but, it's certainly closer than the EBCDIC you seem to be using. Can you please find out what it's supposed to be? I've asked this a couple of times now.

Notice that when I created the IFS file, I specified O_CCSID and 1208 to specify that the file is 1208. However, I did NOT specify O_TEXTDATA or O_TEXT_CREAT, which means that the IFS APIs will not translate it from the program's EBCDIC to the file's UTF-8. I did it that way because the data is already in UTF-8, so translating would cause problems.

I need to print this zpl data to spool file to send to printer. How to do that? Here what i am doing (after zpl data written to ifs) to achieve what i need:
- Created a physical file DMYPRTFL with as400 default CCSID 65535: CRTPF FILE(TEST1/DMYPRTFL) RCDLEN(132)

Depends on how your printer is set up, et al. One easy way to preserve the values would be to use a *USERASCII spooled file -- then you can just write the data to it without converting it to EBCDIC first. The disadvantage to this approach, however, is that you can't see the contents in the spool.

cmd = 'CPY OBJ(' + #Quote + stmf + #Quote + ') ' +
'TOOBJ(' + #Quote + '/qsys.lib/test1.lib/' +
'dmyprtfl.file/dmyprtfl.mbr' +
#Quote + ') FROMCCSID(*OBJ) ' +
'TOCCSID(*JOBCCSID) DTAFMT(*TEXT) REPLACE(*YES)';

You are telling it to convert from the CCSID of the stream file (i.e. 1208, or UTF-8) to the job's CCSID. That means you are converting it to EBCDIC. If that's what you want, why not do it in the RPG program?

Are you on a recent release of RPG? (One where UTF-8 fields are allowed?) Or would you have to use APIs to translate it?

[sbehera00]

Hello Scott,
followingup on this base64 thread for another need. I got to do some change to ZPL after decoded before writing to IFS. As far as writing to IFS is all working fine but %scan does not find the string ^POI what i am looking for to replace with '^PON because ZplOut is in not in ASCII format. Trying to avoid reading IFS file after it is written and manipulate content. If this change of data can be done before writing to IFS, i will save time and better performance. Please help..

Code:
zplLen = base64_decode( %addr(B64In:*data)
: %len(B64In)
: %addr(ZplOut)
: %size(ZplOut) );
#strpos = %Scan('^POI':ZplOut);
If #strpos > 0;
ZplOut = %SubSt(ZplOut:1:#strpos-1) + '^PON' +
%SubSt(ZplOut:#strpos+4:(%Len(ZplOut)- #strp
EndIf;
there are file open etc, not putting here but just the write part
callp write(fd: %addr(ZplOut): zplLen);

[sbehera00]

Data looks like this:

EVAL ZplOut
ZPLOUT =
....5...10...15...20...25...30...35...40...45...50...55...60
1 ';ãì!&<ÅÁ>ÁÊ/ÈÁÀÂ`î;ãë;ì ;&ï;&|ñ;&ê'
61 ';ã|;åâ;ãë;ãè; +;ãàå'
121 ';ãë;ã|;åã ÃÃÃÃÃÄÄÄÃÄÃÄÃ'
181 'ÄÄÃÃÄÃÄÃÄÄÃÃÄÃÃÄÃÃÄÃÃÄÃÄ'
241 'ÄÄÃÄÃÃÃÄÃÃÃÃÃÄÄÄÃÄÃÄÃÄÄÃÃ'
301 'ÄÃÄÃÄÄÃÃÄÃÃÄÃÃÄÃÃÄÃÄÄÄÃÄÃÃÃÄ'
361 'ÃÃÃÃÃÄÄÄÃÄÃÄÃÄÄÃÃÄÃÄÃÄÄÃÃÄ'
421 'ÃÃÄÃÃÄÃÃÄÃÄÄÄÃÄÃÃÃÄÃÃÃÃÃÄÄÄÃ'
481 'ÄÃÄÃÄÄÃÃÄÃÄÃÄÄÃÃÄÃÃÄÃÃÄÃÃ'
541 'ÄÃÄÄÄÃÄÃÃÃÄÃÃÃÃÃÄÄÄÃÄÃÄÃ'
601 'ÄÄÃÃÄÃÄÃÄÄÃÃÄÃÃÄÃÃÄÃÃÄÃÄÄÄÃ'

[Scott Klement]

Why are you posting all the mistranslated characters? Am I supposed to be able to make something of that?

Seems to me that if your data is in UTF-8 and you want to work with it, you can't be scanning for EBCDIC data and expecting that to work. Instead, scan for UTF-8 data...

Use the ccsid(*utf8) character on the strings that will be in UTf-8.

Code: Select all

     D zplOut          s          20000a   ccsid(*utf8)
     D CPOI            s              4a   ccsid(*utf8) inz('^POI')
     D CPON            s              4a   ccsid(*utf8) inz('^PON')

Then use the UTF-8 strings in your work

Code: Select all

       
       #strpos = %Scan(CPOI: ZplOut);
       If #strpos > 0;
         ZplOut = %SubSt(ZplOut:1:#strpos-1) + CPON +
                  %SubSt(ZplOut:#strpos+4:(%Len(ZplOut) - #strpos) - 3);
       EndIf;

Alternately, just use the %SCANRPL or %SCAN/%REPLACE BIFs.

[sbehera00]

Hello Scott,
when i do
D ZplOutU s 65535A ccsid(*utf8)
SEU gives error "The parameter for keyword CCSID is not valid" then i changed to
D ZplOutU s 65535A ccsid(1208) i get error
The CCSID keyword is not valid with the specified definition.

I have made other way to work. Ideal way i was thinking is to do change when i get the data but at this time i am not sent to printer but just store in folder. When i am sending to printe i am reading the file and able to change this piece of data before writing to spool.

I thought there will be quick/easy way but does not look like...

[Scott Klement]

Support for ccsid(*utf8) (or ccsid(1208)) was released with V7R2 back in 2014. If you don't have that support on your system, you're VERY out of date.

That will certainly make things more difficult. You'll have to code constants with the hex values of ^POI and ^PON and scan/replace those.