Re: varargs calling C from 4gl

In article <1993May26.041938.5727@informix.com> aland@informix.com (Colonel Panic) writes: >In article <C7Loq6.AI2@wetware.com> cse@wetware.com (Scott Ellard) writes: > >Geez, Scott, how long have you been lurking in this newsgroup? Acutally for some time now. :) > >>In article <1ttklqINNmmv@emory.mathcs.emory.edu> roderick@ibcinc.com (Roderick Schertler) writes: >>>Is there a way to defeat 4gl's `Function has been called with a >>>different number of arguments' checking? I'd really like to write a C >> >>You will need to modify the c4gl shell script to accept the '-z' flag. >>This flag is used during the first pass (the fglc pass) of the compile. >> >>This flag is documented for fglpc but not for c4gl. > >Also, it doesn't work completely right in RDS in releases below 4.10.UE1 Yes. Sorry I forgot to mention this. The vararg pcode is interpreted properly using fglgo and friends, however the pcode compiler emits bogus argument information in earlier versions. Since pcode is machine independent, I usually compile on my Sun machine (which has a fixed version) and then move it to the appropriate target machine. > >>I have it on good authority that this feature will not disappear >>as it is an important part of the Informix-TP/ToolKit product. >> >>FYI: Informix-TP/ToolKit is a product which allows 4GL programs >> to interoperate within the Tuxedo and Encina TP environments. >>Scott Ellard > >But only Tuxedo is released so far (right?) Yes. Informix-TP/ToolKit is currently only released for Tuxedo. The version for Encina is complete but we are waiting for a new release from IBM/Transarc. Also I believe the TopEnd version is due any day now. >-- >Alan Denney aland@informix.com {pyramid|uunet}!infmx!aland Scott Ellard