home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!elroy.jpl.nasa.gov!usc!wupost!uwm.edu!zazen!doug.cae.wisc.edu!umn.edu!cybrspc!roy
- From: roy%cybrspc@cs.umn.edu (Roy M. Silvernail)
- Newsgroups: comp.os.msdos.programmer
- Subject: Re: Summary: P.D. MAKE for msdos
- Message-ID: <XwNkPB3w165w@cybrspc.UUCP>
- Date: Sat, 15 Aug 92 18:46:56 CDT
- References: <1992Aug14.152501.17351@exu.ericsson.se>
- Organization: Villa CyberSpace, Minneapolis, MN
- Lines: 23
-
- exuhag@exu.ericsson.se (James Hague) writes:
-
- > The downside to Borland MAKE, as has already been mentioned, is
- > that the documentation really stinks.
-
- Now that you mention it.....
-
- I've got a library in progress, and I'm adding modules occasionally.
- Since each module has to be tested before addition, the .obj is
- natuarally not out of date against its .c file.
-
- So... how do I write rules that specify the .lib depend on the component
- .obj's and force the recompilation of said .c files and their
- addition/update into the lib? (they need to be recompiled to remove the
- main() that lies inside an #ifdef TEST) The Borland MAKE documentation
- says nothing helpful. In fact, I can't seem to find anything at all
- that would help here. Default rules fail, because the name of the .c
- file differs from the name of the .lib. I don't particulary want to
- have to write a unique rule for each and every module.
- --
- Due to failure of my mail connection, please use the following emergency
- addresses to reply:
- roy%cybrspc@tfsquad.mn.org cybrspc!roy@tfsquad.mn.org roy@tfsquad.mn.org
-