| Age | Commit message (Collapse) | Author |
|---|
|
|
|
argument number.
|
|
reducing some complex fats from ci.c.
OK joris@
|
|
CALL or AMPER or NAME node to a function or pointer to a function. This
will come in handy later.
|
|
which corresponded to the argument # (1-based, with 0 meaning it is not
an argument). This sucked because it is impossible to include the function
name or argument name in warnings when all you have is an argument number.
Introduce a new type farg_t which represents a passed function argument.
struct farg contains argument number, function name, and argument symbol
information from the prototype).
|
|
OK joris@
|
|
is the easy way out. The code should be able to handle having
pb.author, pb.description, and pb.symbol as const char * strings
pointing to rcs_optarg values without using xstrdup at all. The
code is a bit hairy so it's difficult to verify that it's done
correctly, so that'll be a plan for another day. In the meantime,
this works.
OK joris@
|
|
OK xsa@
|
|
code detection and later on, other things.
|
|
|
|
calling rcs_close() so we don't try to rcs_close() the same file
twice if the working file does not exist, resulting in a core dump.
The current code also returns the status of the last file, so if
we do:
$ touch file
$ ci nonexistent file
that will return 0. GNU returns 1. Fix that.
Additionally, it returns -1 on error, which turns into 255. It
should return 1.
OK niallo@
|
|
Remove their remaining code, since they were not working before.
Initial diff from Steffen Wendzel.
ok millert@
|
|
From Steffen Wendzel. ok marius@
|
|
Found by lint.
OK joris@
|
|
OK joris@
|
|
ok joris@ xsa@
|
|
iov array; well spotted canacar@!
|
|
OK by otto and millert.
Offered back to bell-labs.
|
|
commands, which would result in a separate tiny packet on the wire by
using atomiciov(writev, ...) to write the length and the command in one
pass; ok deraadt@
|
|
interrupted transfers like atomicio() does for read/write;
feedback deraadt@ dtucker@ stevesk@ ok deraadt@
|
|
buffer became too large and the remote end had advertised a big window.
The problem was a mismatch in the backoff math between the channels code
and the buffer code, so make a buffer_check_alloc() function that the
channels code can use to propsectivly check whether an incremental
allocation will succeed. bz #1131, debugged with the assistance of
cove AT wildpackets.com; ok dtucker@ deraadt@
|
|
nuke rlog_rev_select(). OK niallo@.
|
|
"looks gut" mickey@
|
|
|
|
|
|
- show magic branches in rlog (thanks for that nicer hack niall).
- correct output in rcsdiff, so we perfectly match gnu's
all these bugs were found by sturm@ while he was using cvsweb
(which uses the RCS tools).
"it is in my view that you should put it in" niallo@
|
|
adding a really nasty hack to make some upcoming rlog fixes from joris work.
ok joris@
|
|
``crazy'' joris@
|
|
The comparison function was previously changed by me to use strcmp,
which is probably wrong since a buffer can theoretically have NULs
in it. This fixes it. (In my defense, the original code converted
the buffer text to char * values and compared until there was either
a difference or until a NUL was reached, so my previous change was
no worse.)
OK joris@
|
|
OK joris@
|
|
OK joris@
|
|
OK joris@
|
|
|
|
|
|
no binary change; ok ray
|
|
prodded by deraadt@
|
|
fudges the output to stay clean when intr time wasnt shown. not needed
since we show it now.
|
|
blind, too. seek help. no binary change, ok joris
|
|
tweaks by deraadt@ ok by many
|
|
|
|
Remove unused variables found by lint.
Some parts OK xsa@, ``throw it in :)'' joris@
|
|
-o allows users to delete revisions in the specified RCS files.
example:
rcs -o1.4:1.6 foo deletes revisions 1.4 - 1.6
rcs -o1.3 foo deletes revision 1.3
joint work with niallo@
okay niallo@
|
|
when he yanked the code out of rcs.c and placed it in rcstime.c.
struct tm ltb, *tb;
tb = <b;
return (tb);
isn't really the most perfect idiom to return a value to the caller, now is it?
as a bonus -z for rlog now works too.
okay niallo@
|
|
getting our buffer reset to 0 (empty) all the time.
i noticed this when i was trying to fix -z for keyword expansion. sigh.
okay niallo@
|
|
OK xsa@
|
|
Found by lint.
OK xsa@
|
|
`ci -m file'. This changes it so if -m is given without an argument
a customized error message is printed.
While I was there I sorted the optstring according to style(9).
``Oh, I see it now.'' niallo@
|
|
"cool" niallo@
|
|
|
|
original diff from ray@, crafted into shape by myself.
okay ray@
|
