iLibCall fails on 64-bt Python versions

[kadler]

Original report by Kevin Adler (Bitbucket: kadler, GitHub: kadler).

---

All calls to XMLSERVICE through iLibCall fail with *NODATA. Other transports are unaffected.

Example:

#!python
from itoolkit import *
from itoolkit.lib.ilibcall import *

itransport = iLibCall()
itool = iToolKit()

itool.add(iCmd5250('wrkactjob', 'WRKACTJOB'))
print(itool.xml_in())
itool.call(itransport)
wrkactjob = itool.dict_out('wrkactjob')
print(wrkactjob)

Output:

#!text
<?xml version='1.0'?>
<xmlservice><sh error="fast" var="wrkactjob"><![CDATA[/QOpenSys/usr/bin/system WRKACTJOB]]></sh></xmlservice>

{'error': {'error': '*NODATA', 'wrkactjob': {...`

[kadler]

Original comment by Kevin Adler (Bitbucket: kadler, GitHub: kadler).

---

This is an upstream XMLSERVICE bug: https://bitbucket.org/inext/xmlservice-rpg/issues/17/calling-xmlservice-from-a-64-bit-pase-job

For now, the workaround is to use the database transport:

#!python
# Instead of this
from itoolkit.lib.ilibcall import *
itransport = iLibCall()

# Do this
from itoolkit.db2.idb2call import *
itransport = iDB2Call()

[kadler]

New XMLSERVICE bug is IBM/xmlservice#17

[dmabupt]

Original comment by Kevin Adler (Bitbucket: kadler, GitHub: kadler).

This is an upstream XMLSERVICE bug: https://bitbucket.org/inext/xmlservice-rpg/issues/17/calling-xmlservice-from-a-64-bit-pase-job

For now, the workaround is to use the database transport:

#!python
# Instead of this
from itoolkit.lib.ilibcall import *
itransport = iLibCall()

# Do this
from itoolkit.db2.idb2call import *
itransport = iDb2Call()

@kadler
It should be uppercased B ->itransport = iDB2Call() , not iDb2Call()

[jkyeung]

@kadler
It should be uppercased B ->itransport = iDB2Call() , not iDb2Call()

@dmabupt - Ah. Actually, it should be Db2, since that is consistent with IBM's current branding. However, it does look like the actual code in itoolkit.db2.idb2call still has a class named iDB2Call. It's probably reasonable to make an alias there, like

iDb2Call = iDB2Call

or vice versa.

[kadler]

@jkyeung That would make sense to do, if all the i*Call transports weren't going to be deprecated in the next release anyway.

[kadler]

@ramstein74 Seems like a different issue than this one - please open a separate issue.

[richardschoen]

Whew, good thing this was documented. I was banging my head trying to get iLibCall to work today :-)

[kadler]

This is fixed in XMLSERVICE 2.0.1. The necessary PTFs are now documented in here an the code will now attempt to detect this condition and raise an exception with the PTF numbers in the exception message.