[hylafax-devel] Re: Updating 4.1beta2

[Steve Williams <steve@genie96.com>]

Hi All,

I am curious about the state of the current CVS.

I have been playing in my few spare minutes trying to get it to compile on
AIX.  

I have tried with the following compilers ( all on the same system )
under AIX 4.2.1.??

  gcc 2.7.2.3
  gcc 2.95.2
  xlC 3.1.4.10

Both of the gcc's will compile emacs 20.5, so I am pretty confidant in the 
compiler compiling ( I boot strapped them both locally ).

The all die compiling  BoolArray.c++

I have tried gmake 3.8.71 as well as the standard AIX make as well ( yes, it 
made a difference when I bootstrapped the compilers ).

The errors from gcc 2.95.2 are:

/usr/local/bin/g++      -I                        -D__ANSI_CPP__ -I. -I.. -I.././regex -I.././util  -I.././util -I/usr/local/include -g -O  -c BoolArray.c++

In file included from BoolArray.c++:26:
BoolArray.h:31: syntax error before `{'
BoolArray.h:31: parse error before `public'
BoolArray.h:31: parse error before `&'
BoolArray.h: In function `class fxBoolArray Ptr(...)':
BoolArray.h:31: `other' undeclared (first use this function)
BoolArray.h:31: (Each undeclared identifier is reported only once
BoolArray.h:31: for each function it appears in.)
BoolArray.h: At top level:
.. and it goes on for quite a way..

How were the latest checkins to the CVS compiled?  

Am I the only one having problems, or are other people as well, but no-one
( like me ) is saying anything?

Thanks,
-- 
	Steve Williams, Calgary, Alberta, Canada
	Genie Computer Systems Inc.
	steve@genie96.com

"A man doesn't begin to attain wisdom until he recognizes that he is 
 no longer indispensable."
- Admiral Richard E. Byrd ( 1888-1957 )


> 
> Nico Kadel-Garcia <raoul@adsl-151-203-22-73.bellatlantic.net> wrote:
> > 
> > Hi, guys! I've got a bit of time now that I've switched jobs, and there
> > are a few minor updates I'd like to suggest.
> >
> 
> Hope the new job is going well.
> 
> > First, we should bring the 4.1beta2 release up to 4.1beta3  with all
> > the changes in the CVS release, or finally ship it out the door as
> > 4.2.
> >
> 
> Agreed. Probably a 4.1beta3 as a release candidate is the prudent option.
> 
> > Third, I've got a small set of patches at
> > http://cirl.meei.harvard.edu/hylafax/patches/, some of which are
> > already incorporated into the CVS version.
> > 
> 
> Ok; Here is my list of patches still to be applied to the CVS :
> 
> 1. AIX patches by Steve Williams - untested and uncomplete???
> 
> 2. faxcron-failures.patch by Seth Chaiklin 20/03/98 
>                  - I think this was just accidently missed 
> 
> 3. RAND_MAX.patch by Nico Kadel-Garcia 22/10/99
>                  - patch to the configure script for SunOS
> 
> 4. posix-rename.patch by John Williams  23/10/99
>                  - patch to FaxRequest.c++
> 
> 5. mkdepend-v4.patch by James Lee 01/01/00
>                  - SunOS patch by our self-confessed sed expert
>                    with obviously nothing better to do on the
>                    first day of a new Millenium.
> 
> 6. faxaddmodem-stty.patch (untested) suggested by Phil Watkinson
>                  - eliminate warning message '/bin/stty:standard
>                 input: unable to perform all requested operations'
>                  by removing line 773 of script.
> 
> 7. configure-tiff.patch suggested by Eric Wood
>                  - allow tiff library 3.5 as well as 3.4
> 
> 8. Modem Configuration files : 
>                  - performance-rts by Nico Kadel-Garcia
>                  - rockwell-k56
>                       xonxoff amendment by Glenn Burkhardt
>                       add SupraExpress 56e PRO by Phil Watkinson
> 
> 9. sys.h.patch by Mark Burton 09/01/00
>                 - missing :: operator in line 128 of Sys.h
> 
> 10. faxcover.patch by Mark Burton 09/01/00
>                 - variable not declared in faxCover::tildeExpand
> 
> 11. cvtDateTime.patch by Dmitry Bely 10/02/00
>                 - patch to both recvstats and xferfaxstats re Y2K
>                   rollover bug resulting in faxcron reporting all
>                   faxes ever sent.
> 
> To which needs to be added the patches on Nico's web site not included
> in the above :
> 
> 12. bzero.patch
> 13. munmap.patch
> 14. zlib.patch
> 15. hylafax-client.patch
> 16. fixhtml.patch
> 
> I think that patches 14 and 15 have been applied to the CVS but this
> needs checking.
> 
> We need to decide if there is any reason *not* to apply these patches
> to the source code. I am willing to do the work if given the okay by
> Robert (well, the easy ones anyway).
> 
> We also need a list of showstopping bugs still to be fixed. The major
> one is, of course, faxq maxing out when working with tiff-3.5.x. Are
> there any others???
> 
> Sorry for the long post,
> Phil.
> 
> 
> 
> 
>