I imagine it is quite a few steps ahead of z/OS Unix System Services keeping up with Unix / Linux commands.
On Fri, Sep 27, 2019 at 12:48 PM Charles Mills <[email protected]> wrote: > > Alan, thanks. We've spoken at the TDM. I hope I was not too hard on you > guys.. I'm aware of the product of which you speak and why IBM supports it. > (Can I say its name and the reason here or is that NDA?) > > I've semi-retired from full-time hard-core C++ development and product > architecting, so my opinion no longer matters a whole lot. But if I were > setting product direction I would be wanting to know what is the strategic > IBM C++ compiler? The "real" IBM product or the open source download? I would > not want to commit to a new and quite different C++ compiler only to find out > that no, IBM was now heading off in some other directions. > > I think -- and as I say, what I think does not really matter -- I think that > IBM needs to stand up and say unambiguously "this is our strategic compiler, > it supports C and C++, it supports 31- and 64-bit, it is our intent that it > will lag only a year or three behind the standards, and it will be our > strategic compiler for the foreseeable future." Otherwise frankly I wonder if > a developer today would be making a mistake to make the commitment I made 9 > years ago to C++ on Z. IBM has a strategic COBOL compiler; why not a > strategic C/C++ compiler? > > FWIW, POSIX(ON) is not a problem. "My" code all runs POSIX(ON). > > Charles > > > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On > Behalf Of Allan Kielstra > Sent: Friday, September 27, 2019 4:41 PM > To: [email protected] > Subject: Re: casting with XL C\C++ compiler > > Hi Charles > > We are listening! > https://www-01.ibm.com/servers/resourcelink/svc00100.nsf/pages/xlCC++V231ForZOsV23?OpenDocument > > This is a web deliverable that co-exists (does not overwrite) the existing > xlC compiler. I have personally used it to develop C++ code with initializer > lists, emplacements in collections, lambdas, variadic template classes and > other more modern features. > > One thing I should point out. For any non-trivial program, the resulting > executable will require POSIX(ON). Also, you should compile an entire > application with the new compiler or the old one. (The object models are not > the same.) > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
