YASI -> Yet Another Strange Idea. This idea is the illegitimate offspring from learning the R language and thinking about z/OS UNIX. And, like many others from this source, may well be really stupid. But I don't know the meaning of the word fear! And many other words too for that matter.
A legacy program can read or write a z/OS UNIX file via QSAM via a normal DD statement by using the PATH= instead of DSN=. Now, think of a browser. Most browsers can read a web site by using a URL prefixed with http:// or https://. But most can also access other protocols by using, for example, ftp:// or file:// . The R language, and likely others, can also use these prefixes where a file name is normally found. RFC 1738 (and likely others) talks about URLs: https://www.ietf.org/rfc/rfc1738.txt I really wonder if this would be of any use in a z/OS environment. //INPUT DD URL='file:///etc/resolv.conf' or //INPUT DD URL='http://some.web.site/download/filedata.txt' or //INPUT DD URL='ftp://user:password@host/download/filedata.txt' and so on. It might even be interesting if someone would create a "CURL" subsystem which could do the above. This could be modeled after the cURL ported command. Well, back into my hole for a time. -- There is nothing more pleasant than traveling and meeting new people! Genghis Khan Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
