Jay Oliveri (ilnero at gmx.net) wrote:
> Patch applied.
In IRC, mjr said he already did it.
> btw, what was it failing on exactly? ..because I
> haven't been able to break fcpget, probably because I've been focused
> on fcpputsite for some time.
It failed on everything except the help message. I couldn't fetch any
keys without segfaults. It was dereferencing pointers that had not been
initialized (they contained random garbage). I cleaned up the realloc()
stuff and added some initializations (to NULL).
And by the way, fcpputsite was segfaulting intermittently for me also;
but I haven't tried it for several days.
fcpget still doesn't actually fetch any keys, even if I know they're
in my data store -- it just says "Failed to open 'KSK at gpl.txt'". But
that's a lot farther than it got before.
> > I also removed 9 Ctrl-M (CR) characters from the header file. Fred
> > only knows what that does to cpp/gcc.
> Dunno exatly what u mean from this. Those carriage returns don't
> affect anything other than readability.
Carriage returns are not supposed to be present in Unix text files.
CR-LF pairs are how MS-DOS terminates lines. LF is what Unix uses.
Nine of the lines in the header file contained MS-DOS CR-LF pairs,
and all the other lines contained Unix LFs.
They probably weren't hurting anything in this particular case. But
imagine you have something like this:
#define a_macro(x,y) (function_foo(x) + function_bar(y) + \
function_blah(x,y))
In this case, if the first line has a CR at the end of it, the backslash
won't work as a line-continuation character, and the preprocessor
will puke. So I removed the CRs on general principle.
--
Greg Wooledge | "Truth belongs to everybody."
greg at wooledge.org | - The Red Hot Chili Peppers
http://wooledge.org/~greg/ |
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL:
<https://emu.freenetproject.org/pipermail/devl/attachments/20020223/0b2f59fb/attachment.pgp>
