>From: [EMAIL PROTECTED] (Bill Davidsen)
>> Mkisofs:
>>
>> - mkisofs now also compiles without -DUSE_LIBSCG
>>
>> - -DNON_UNIXFS for VMS removed (need to check if it still compiles)
>>
>> - Include File structure completely rewritten.
>> The old portability code from Eric now has been completely removed
>> and replaced by reusable code snipplets in suportable incude files.
>> This makes portability finally clean and is needed to add largefile
>> support in the near future.
>>
>> Please test on all possible OS if I introduced problems.
> Could you let us know what you have decided to do with the source
>naming conventions? We had a lot of good discussion, and I have the
>impression that you took some time to study the suggestions, have you
>decided what changes, if any, you will make.
I am working on better portability for a long time. The result is the
makefile system which uses re-usable object oriented makefile snipplets
and also provides include files that help to make portability a lot easier.
I believe it's a pain if you read the autoconf manual and see that you
should put this peece of code into your sources for this reason and so on.
A year later you find that this advice was wrong and does not work on OS xxx
or because of constraint yyy so you have to grab through all your sources
and try to find the buggy parts.
With the include files that are part of my makefile system, you
include e.g. timedefs.h and get a ready to use abstract view of what you
like to use. If this is wrong, you simply modify timedefs.h and don't
need to look through all your sources. The emulated interface usually
is the POSIX interface if this mya be done Well timedefs is an easy job,
inttypes.h is harder because e.g. Linux violates the POSIX/C-99 specs for
the types that should be in innttypes.h although Linux is missing this
file. So I needed introduce new types that are not already used.
If you like to use large files, you need to be much more acurate
in following progamming conventions and as mkisofs still looked like a
junk yard at some places, I edited all files in mkisofs to make sure
that e.g. mconfig.h always is included before stdio.h and any other
system include file to allow to activate the transitional LF32 interface.
When doing this, I removed that last artefacts of Erics old portability
attempts. I am now sure that everything is implemented the same way.
Note that a year ago, there have been three different ways of implementing
major()/minor() and IS_IFREG & friends in mkisofs. This caused my headaches
when I got bug-reports from e.g. FreeBSD people.
> I realize that you may not have an implementation schedule at this
>time, with the holidays coming and all the other items you noted.
My implementation schedule for the holidays is:
- make my editor ved ready for a source distibution.
- build all the parts of my beetle motor to get a new working motor ;-)
- Look a bit for Large File support with all my programs.
J�rg
EMail:[EMAIL PROTECTED] (home) J�rg Schilling D-13353 Berlin
[EMAIL PROTECTED] (uni) If you don't have iso-8859-1
[EMAIL PROTECTED] (work) chars I am J"org Schilling
URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
