daveyc / pyzfile Goto Github PK

Writing records to a z/OS PDS member the square bracket characters are stored as X'BA' and X'BB' rather than X'AD' and X'BD'.

I haven't checked the other "problematic" characters yet (cent sign, not sign, caret, unbroken and broken vertical bars).

It is kind of difficult to write json output to a z/OS file without the correct square bracket characters. Is there any other encoding that will give correct square brackets in a z/OS file?

Hi

I trying to read a file and writing it to another file.

My program:

from pyzfile import *

def main():
print("Read MVS-file")
outfile = ZFile("//'BC6608.TEST.QSAMOUT'","w,type=record,seek",encoding='cp1047')
print(outfile.info())
try:
with ZFile("//'BC6608.TEST.QSAM'","rb,type=record",encoding='cp1047') as infile:
for rec in infile:
print(rec)
except ZFileError as e:
print(e)
print(outfile.info())
print(infile.info())
print(file.info())
print("All records read")

if name == "main":
main()

The first print(outfile.info()) returns:

æ'access_method': 'UNSPEC', 'blksize': 4294967295, 'device': '', 'dsname': '', 'dsorgConcat': False, 'dsorgHFS': False, 'dsorgHiper': False, 'dsorgMem': False,'dsorgPDE': False, 'dsorgPDSdir': False, 'dsorgPDSmem': False, 'dsorgPO': False, 'dsorgPS': False, 'dsorgTemp': False, 'dsorgVSAM': False, 'maxreclen': 343758143136, 'mode': æ'append': False, 'read': False, 'update': False, 'write': Falseå, 'noseek_to_seek': 'NOSWITCH', 'openmode': 'TEXT', 'recfmASA': False, 'recfmB':False, 'recfmBlk': False, 'recfmF': False, 'recfmM': False, 'recfmS': False, 'recfmU': False, 'recfmV': False, 'vsamRKP': 343740737584, 'vsamRLS': 'NORLS', 'vsamkeylen': 471371242, 'vsamtype': 'NOTVSAM'å

So no dsname.

And if I add a outfile.write(rec) after the 'print(rec)' I will get an error:
raise Exception(f"Error writing to file 'æself.filenameå': æself.lib.zfile_strerror().decode('utf-8')å")
Exception: Error writing to file '//'BC6608.TEST.QSAMOUT'': EDC5113I Bad file descriptor.

The files are nothing special:
Organization . . . : PS Current Utilization
Record format . . . : FB Used tracks . . . . : 2
Record length . . . : 80 Used extents . . . : 1
Block size . . . . : 27920
1st extent tracks . : 15
Secondary tracks . : 15 Dates
Data set name type : Creation date . . . : 2023/12/20
Data set encryption : NO Referenced date . . : 2023/12/20
Expiration date . . : None

SMS Compressible . : NO

Am I doing anything wrong?

BR
Frank Allan Rasmussen

This is of great importance especially for opening PDS data sets for write.

I was asked how someone installs this if they cannot use pip on z/OS.

They managed it by

I managed to install it by downloading the zip file, uploading it to the zOS LPAR, unzipping it and then running the pip program against the pyzfile-1.0.0.post2-py3-none-any.whl in the pyzfile/dist folder.

It would be worth adding some words like this to this documentation
Colin