On 28/10/2013 11:39 AM, Timothy Sipples wrote:
David Crayford writes:
I'm not aware of any previous requirement for a mainframe
COBOL program to access a data base running on a different
platform.
That's fairly common. To pick an example, Oracle offers a product called
Oracle Access Manager for CICS that allows COBOL (and other programming
language) programs running in CICS to access Oracle databases. I think
OAM4CICS also supports applications running outside CICS as it happens.
I wonder if there is a market for mainframe legacy applications to
access NoSQL data stores?
To pick another example, CICS Transaction Server and IMS Transaction
Manager support Web Services. COBOL applications are frequently Web
Services consumers (clients), accessing remote applications and databases
via Web Services protocols.
Do you know of many customers using COBOL web service consumers in
production? After all the hullabaloo
about SOA a few years ago it seems to have gone a bit quiet. And the
protocol that was being suggested, SOAP, is already
obsolete. If we take IMS as an example, it has a SOAP gateway and XML
adapters for IMS Connect. That's pretty old
school in the world of web services these days where REST/JSON is the
architecture of choice.
Does anybody know how I can call a RESTful web service from an IMS COBOL
program?
To pick yet another example, it's a very common pattern for COBOL (and
other programming language) applications running on z/VSE (inside or
outside CICS Transaction Server) to access DB2 (or other databases) running
on Linux on zEnterprise (or perhaps also on other platforms). z/VSE started
providing standard, no additional charge support for remote database access
several years ago. It's called the z/VSE Database Connector (DBCLI), and
more information is available here:
http://www.ibm.com/systems/z/os/zvse/products/connectors.html#dbcli
Ze'ev Atlas asks:
Does the current incarnation of COBOL (and perhaps PL/I) have
a native UTF-8 string type.
Yes, UTF-8 is supported in Enterprise COBOL. To get started, take a look at
the following Enterprise COBOL 5.1 functions: ULENGTH, USUBSTR,
USUPPLEMENTARY, UVALID, and UWIDTH. Those are all new in Version 5.1. More
information here:
http://pic.dhe.ibm.com/infocenter/pdthelp/v1r1/topic/com.ibm.entcobol.doc_5.1/PGandLR/tasks/tputf8in.html
The following page applies to Enterprise COBOL at least as far back as
Version 4.1:
http://pic.dhe.ibm.com/infocenter/pdthelp/v1r1/topic/com.ibm.entcobol.doc_5.1/PGandLR/tasks/tpstr31.html
And here's another code sample (applicable to Enterprise COBOL 5.1) which
takes a table of Czech composer names (fun!) represented in UTF-8,
determines their initials, and outputs the results (in EBCDIC):
http://pic.dhe.ibm.com/infocenter/pdthelp/v1r1/topic/com.ibm.entcobol.doc_5.1/PGandLR/ref/rputf8e.html
Thus interacting with MongoDB in UTF-8 from your COBOL applications should
be no problem whatsoever.
--------------------------------------------------------------------------------------------------------
Timothy Sipples
GMU VCT Architect Executive (Based in Singapore)
E-Mail: [email protected]
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
