[Pan-devel] Re: Pan 0.14.2 Won't Compile

[Duncan]

Greg Lee <address@hidden> posted
address@hidden, excerpted below, on  Mon, 30 Jul 2007 06:01:56
+0000:

> On Sun, 29 Jul 2007 13:33:11 -0500, Michael Satterwhite wrote:
> 
>> msort.c: In function ‘msort_with_tmp’: msort.c:68: error: invalid
>> lvalue in increment msort.c:69: error: invalid lvalue in increment
>> msort.c:74: error: invalid lvalue in increment msort.c:75: error:
>> invalid lvalue in increment
>> 
>> Looking at the source to msort.c, this appears to be occurring in the
>> use of an incremented pointer to a pointer.
> 
> I suspect this is due to some change in gcc 4.1.2, since I've seen the
> compiler complaint several times recently.  You could just recode the
> increment as a separate statement.  But better to get a more recent
> version of Pan (which no longer has an msort.c, since it's written in
> c++).

I'll second that.  pan 0.14.x is what, something like four years old 
now?  (Just checked, almost, 0.14.2 was released on August 31, 2003, so 
it's just about a month short of four years.  Time does fly...)  It's 
unsurprising newer and stricter gccs don't like the old code.

pan 0.90 was an all-new rewrite in C++, released the beginning of April, 
last year (2006).  Charles released nearly weekly updates for awhile, tho 
he seems to be taking a well earned breather ATM.  Quite a number of 
serious bugs have been found and squashed since 0.90, however, and a lot 
of rounding out of the functionality has occurred as well.  The old 0.14 
code has been officially unsupported for about a year now (and hadn't 
been updated much for a couple years before that), so unless you want to 
be doing all your own hacking, and/or fork the old pan and run with it 
yourself, the strong recommendation is to grab new-pan, now 0.131, and go 
with it.  Quite a number of folks are compiling it on their own, across 
Linux and the BSDs, OSX, and even for MSWormOS, so unlike that ancient 
code you're currently dealing with, any problems you have with the new 
code are likely to likely to have fixes pretty quickly, and aren't 
actually likely to occur at all as long as you stick to mainline Linux, 
and not too new or old a gcc and binutils.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman