dnelson@umbio.MIAMI.EDU (Dru Nelson) (12/18/88)
Could someone please post a make for ORCA/M or APW shell if a PD one
exists?
--
Dru Nelson UUCP: ....!uunet!gould!umbio!dnelson
Miami, Florida MCI: dnelson
Internet: dnelson%umbio@umigw.miami.eduscott@claris.com (Scott Lindsey) (12/18/88)
From article <1130@umbio.MIAMI.EDU>, by dnelson@umbio.MIAMI.EDU (Dru Nelson): > Could someone please post a make for ORCA/M or APW shell if a PD one > exists? > Good luck. I played around with writing a make for a while, but gave up in frustration. I think the closest you could come is something like MPW's make, which just produces the commands to make, but doesn't execute them. There seem to be some inherent problems (bugs) with making shell calls to APW or ORCA/M. I was unable to get anything consistent or usable. The other problem is that the granularity of timestamps is one minute. That means that it's quite possible to edit a file and not change its modification time. I'm looking into converting to MPW-IIgs, but things don't look real hopeful for the conversion process.
scott@claris.com (Scott Lindsey) (12/20/88)
From article <7449@claris.com>, by scott@claris.com (Scott Lindsey): > From article <1130@umbio.MIAMI.EDU>, by dnelson@umbio.MIAMI.EDU (Dru Nelson): >> Could someone please post a make for ORCA/M or APW shell if a PD one >> exists? >> > > There seem to be some inherent problems (bugs) with making shell calls to > APW or ORCA/M. I was unable to get anything consistent or usable. OK, I'll stand corrected. I think the last time I'd tried was with a beta version of APW. After seeing MicroEMACS's success in executing shell commands, I went back & played with it myself & got usable execution of commands. So a Make that executes commands is quite possible, though the timestamp granularity is still a practical problem. -- Scott Lindsey, wombat | UUCP: {ames,apple,portal,sun,voder}!claris!scott Product Development | Internet: scott@claris.com | AppleLink: LINDSEY1 Claris Corp. | These are not the opinions of Claris or anyone else.
scott@claris.com (Scott Lindsey) (12/20/88)
From article <7484@claris.com>, by scott@claris.com (Scott Lindsey): > .... After seeing MicroEMACS's success in executing shell commands, > I went back & played with it myself & got usable execution of commands. > So a Make that executes commands is quite possible, though the timestamp > granularity is still a practical problem. OK, once more for the hard of thinking. Sorry, I spoke too soon (again). There *is* an inherent problem with running programs from shell calls. There is no such thing as a subshell, merely calls to the one sleeping shell. Executing *commands* in a shell call is no problem, but when you actually cause a program to load and run, (such as compile or assemble would), then the shell gets confused by having more than one program running from it. E.g., I ran emacs, then from emacs I ran an assembly, successfully returning. When I try to exit emacs, I usually get a crash or a system-hang. It is still possible to have a program to create an exec file that does your build for you, then a shell script that runs the make, then its output script (exec) file...