Re: comp.databases.informix.4gl: Tool Update and New Tools

In article <3rjqd3$egi@cssun.mathcs.emory.edu>, Nils.Myklebust@CCMAIL.telemax.no : > >Some of your tools are realy interesting, but: Your make files doesn't work with >4GL ver. 6.xx. We use OnLine 7.1/4GL 6.xx for development, but also compile for >SE 5/4GL 4.xx. Yes, I do the same thing. And you're right. I submitted a 6+ makefile generator yesterday. And today Johnathan Leffler sent me his recommendations for making it better. I like what he sent me. YOU could also make it better with the vi editor and a little time. >Your make files are extremly counter productive to us. They aren't my makefiles when YOU create them! :-) >We would appreciate simple make files that use c4gl and work for all versions. Simply replace the 4-step dance with Ic4gl1 thru 4 and put the c4gl program in their place. You also won't need most of the ENV variables. You'll notice the 6+ make generator was derived directly from c4gl. I must emphasize that the c4gl for 4.whatever doesn't have half the action that the c4gl for 6+ contains. These are different animals. You don't notice the difference because you're making the call to c4gl based on your $INFORMIXDIR when you switch from the 5+ engine to 7+ engine. This is probably your preference in your make files, and it's easy enough to change. >I am at a loss to understand what the problem is with c4gl. > There's no problem with it. I just prefer seeing what is happening during the compilation process, and having a little more understanding of what is going on. It doesn't buy me any more control, other than how I add modules together, and include/exclude the RDS make. c4gl in its' previous incantations has been so simple, that building makefiles was relatively easy. I also prefer staying consistent with the rest of the UNIX programming world, and not getting into "Phase 1...", "Phase 2...", etc. 6+ has a lot more going on, especially with the shared libraries. I am also the kind of person who hates shell scripts that are not easy to understand. c4gl was written in a style that hopes for the best of situations, without a lot of logic going on. The only alternative to c4gl is that interactive programming environment where you can do a "planned compile". If you want to talk about productivity going down the toilet, this is it. UNIX programming has the tools already available for programming that this fine product really isn't necessary, and thus the reason why I prefer make. And as long as I'm going to create a makefile, I may as well break out the rules to create an executable from the 4GL. Yes this is a non- portable approach when you consider jumping from 4.12 to 6.0, but there's no reason why you can't create two different makefile cranks to help you out. And that's what I give you today. I get the impression that c4gl is designed to protect folks from the compilation process. This is a contradictory position in light of INFORMIX endorsing PVCS source control, and the use of nmake for New Era. 4GL should benefit from as much product development as New Era, even in its' sundown years. If it has to come from the fringe, like developers like me :-), then so be it. BTW, I like the new docs for 6+ 4gl. They're not bad, and it looks like the training dept gave up some of their stuff to the docs. By the way, CC: me when you post so I am aware of your questions or comments. Thanks and best wishes with MAKE. Tim -- \\\\|// (O|O) ==============================---o00--(_)--00o---============================ Tim Schaefer tschaefe@gate.net http://www.gate.net/~tschaefe The Computer Business Company, Inc. INXUTIL 2.0 is now absolutely FREE =============================================================================