commit by bde.
Fix bugs in floating point formatting. The 4.4lite version is similar
to revision 1.3 in old-cvs and is missing all of jtc's fixes in revision
1.4 in old-cvs. Revision 1.2 in ncvs fixed one of the old bugs but
introduced at least one new one (for %.0e).
old-cvs log:
revision 1.4
date: 1993/11/04 19:38:22; author: jtc; state: Exp; lines: +33 -20
My work from NetBSD to make printf() & friends ANSI C compliant.
Fixes several bugs in floating point formatting:
1. Trailing zeros were being stripped with %e format.
2. %g/%G formats incorrect.
3. Lots of other nits.
the copy built from here was overwritten by the other copy and the other
copy was put in library-building command lines twice. ld now objects to
duplicated modules.
from the code in strftime.c . This affects both the library code
and all the commands using it (e.g. date +%s).
Note that %s is not required by ANSI, but we've already got it in 1.1.5.1.
Suggested by: luigi@labinfo.iet.unipi.it (Luigi Rizzo)
than at compile time.
Should have same functionality as old libforms but with new mechanism.
Lots of new features that use the new mechanism are still to be added.
Since functions will come and go from libcompat as they are deprecated
it makes no sense to build a shared library out of it as it will change.
Based on freedback from Terry and Jonas on the mailing lists.
in an (unlikely) border case (maxgroups==1 and the user is on
an /etc/group line for the same group and that group only ...).
Now this case is dealt with as before ...
Add a missing apostrophe that suggests inverting the frequency to get
tick size. It read better before because `CLK_TCK' suggests a tick
size although it is actually a frequency.
as an NIS client. The pw_breakout_yp routines that are used to populate the
_pw_passwd structire only do anything if the bits in the pw_fields member
_pw_passwd are cleared. Unfortunately, we can get into a state where
pw_fields has garbage in it right before the YP lookup functions are
called, which causes the breakout functions to screw up in a big way.
Here's how to duplicate the problem:
- Configure FreeBSD as an NIS client
- Log in as a user who's password database records reside only in
the NIS passwd maps.
- Type ps -aux
Result: your processes appear to be owned by 'root' or 'deamon.'
/bin/ls can exhibit the same problem.
The reason this happens:
- When ps(1) needs to match a username to a UID, it calls getpwuid().
- root is in the local password file, so getpwuid() calls __hashpw()
and __hashpw() populates the _pw_passwd struct, including the pw_fields
member. This happens before NIS lookups take place because, by coincidence,
ps(1) tends to display processes owned by root before it happens upon
a proccess owned by you.
- When your UID comes up, __hashpw() fails to find your entry in the
local password database, so it bails out, BUT THE BITS IN THE pw_fields
STRUCTURE OF _pw_passwd ARE NEVER CLEARED AND STILL CONTAIN INFORMATION
FROM THE PREVIOUS CALL TO __hash_pw()!!
- If we have NIS enabled, the NIS lookup functions are called.
- The pw_breakout_yp routines see that the pw_fields bits are set and
decline to place the data retrieved from the NIS passwd maps into the
_pw_passwd structure.
- getpwuid() returns the results of the last __hashpw() lookup instead
of the valid NIS data.
- Hijinxs ensue when user_from_uid() caches this bogus information and
starts handing out the wrong usernames.
AAAARRRRRRRRRGGGGGGHHHHHHHHHH!!!
*Please* don't tell me I'm the only person to have noticed this.
Fixed by having __hashpw() check the state of pw_fields just before
bailing out on a failed lookup and clearing away any leftover garbage.
What a fun way to spend an afternoon.
one line long.
Fixed a bug in the input field with cursor positioning at the end of
the field.
Make the print_status function available to apps so they can print
status messages.
Updated the example for the new fib parser.
- FreeBSD's NIS server can supply a master.passwd map, which has
more fields in it than a standard passwd map, so we need a
_master_pw_breakout() fuction.
- When doing passwd map lookups, look for master.passwd.* by attempting
a _yp_first() on master.passwd.byname. If it exists, we're being served
by a FreeBSD NIS server and we should use this map.
- If we aren't the superuser, retrieve only the standard passwd maps.
If we're being served by a FreeBSD system, then the passwd map has
no passwords in it, and it won't serve us the master.passwd map unless
we're superuser anyway.
There's a small speed hit for the superuser inherent in the check for
the master.passwd map, but this lets us dynamically decide what to do
rather than rely on a non-standard config file somewhere. Since all
of this is bypassed for normal users, they shouldn't notice the
difference.
You can now specify separate attributes for selected/not selected
cases individually for each field and also an attr for the form as
a whole so you can now have colored backgrounds for the form and
different coloured fields etc.
Update the example.
Change the copyright to a BSD style one.
1) Link against object directory version of libcom_err.so.
2) Don't try to install ss_err.h if we haven't made it yet. It's not
on the critical path for `make world' at this point.
1) Changed LIB_SCCS and SYSLIB_SCCS to LIB_RCS and SYSLIB_RCS.
2) Changed sccsid[] variables to rcsid[]
3) Moved all RCSID strings into .text
4) Converted all SCCSID's to RCS $Id$'s
5) Added missing $Id$'s after copyright.
YP by disallowing `+' entries as logins in all cases. (This handles the
case of a `+' entry in the password file but YP not running, which should
never happen but is easy enough to check for so we'll apply some
prophylaxis.)
1) Don't spit out an error message if Kerberos is installed but not yet
set up.
2) Don't attempt to verify the ticket you got back, as workstations
are not intended to have srvtab files of their own.
Both behaviors can be re-enabled with KLOGIN_PARANOID.
a number of (ex-)Athena programs. Breaking my own rules for importing
somewhat, as this code does not appear to be actively maintained by anyone
(not that it really needs it).
input forms. It has the following simple fields:
Text fields: Just titles, labels etc.
Input fields: An editable text field that may or may not have an
initial default value.
Labelled input field: This is an input field that has an initial
informative entry in it but it vanishes when you start editing the
field.
Toggle fields: These are fields with a pre-defined list of options
which you cycle through using the space bar.
Action fields: These are button type fields that call functions when
they are selected.
A simple demo is included in examples.
Embalm. Rewrite to do things much the same as gcc-2: use fistpq for speed
and elegance, and mishandle overflow consistently. __fixunsdfsi() is no
longer called by gcc.
getcwd() has two off-by-one bugs in FreeBSD-2.0:
1. getcwd(buf, size) fails when the size is just large enough.
2. getcwd(buf + 1, 1) incorrectly succeeds when the current directory
is "/". buf[0] and buf[2] are clobbered.
(I modified Bruce's original patch to return the proper error code
[ERANGE] in the case of #2, but otherwise... -DG)
This program demonstrates the bug:
---
#include <stdlib.h>
#include <string.h>
#include <stdio.h>
#include <unistd.h>
int main(void)
{
char buf[5];
int errors;
errors = 0;
if (chdir("/tmp") != 0) {
perror("chdir");
abort();
}
if (getcwd(buf, 5) == NULL) {
perror("oops, getcwd failed for buffer size = size required");
++errors;
}
if (chdir("/") != 0) {
perror("chdir");
abort();
}
buf[0] = 0;
buf[2] = 1;
if (getcwd(buf + 1, 1) != NULL) {
fprintf(stderr,
"oops, getcwd succeeded for buffer size = one too small\n");
++errors;
}
if (buf[0] != 0) {
fprintf(stderr,
"oops, getcwd scribbled on memory before start of buffer\n");
++errors;
}
if (buf[2] != 1) {
fprintf(stderr,
"oops, getcwd scribbled on memory after end of buffer\n");
++errors;
}
exit(errors == 0 ? 0 : 1);
}
pointer returned by realloc(). All callers free the pointer if the
execve fails. Nuke the caching. This essentially restores buildargv()
to the 1.1.5 version. Also fix a memory leak if realloc() fails. Also
nuke similar but non-broken caching in execvp(). malloc() should be
efficient enough.
command available yet.
Changed an entry in getprcent.3 from rpcinfo(8C) to rpcinfo(8).
Changed an entry in getrpcport.3 from 3R to 3.
Changed two entries in rpc.3 from 3N to 3.
incredibly obnoxious, but also makes inverse mappings work when the local
resolver is in a cache-only configuration. (Maybe this is actually
a bug in BIND?)
pointer if len is 0. I should have looked at the revision history - I would
have found that Bruce already fixed the bug with len=0 over a month ago.
Whoever said that the bug was in 2.0 was wrong.
Change the reference for the libtermcap libtermlib link from SHLIBDIR
to LIBDIR. SHLIBDIR is undefined in the standard case.
termcap.c:
Initialize a local variable to zero. Otherwise an erroneous free call
can happen and clobber the calling program.
Seen with vi and gdb. If you have TERMCAP set with a terminal entry and
set TERM with something like huhu, vi and gdb core dumps.
for Wine support. The current snapshot of wine works fine with this.
This should go into the beta as the code which it calls in the kernel is
already there, and works fine.
$(DESTDIR)/$(LIBDIR) (I need SHLIBDIR. The / was a bug and the
$(...) style was inconsistent.)
Install ordinary libraries in ${DESTDIR}${LIBDIR} instead of in
$(DESTDIR)/$(LIBDIR).
Change remaining $(...) to ${...}.
later be applied to a number of programs (inetd for instance) to clean
out the bogus code doing the same thing, modulus all the bugs.
If you need to read a '#'-is-a-comment-file, please use these routines.
I realize that the shlib# should be bumped (for the non-US world:
increased by something), but will defer this until something significant
happens.
input fields. It reads a template file passed to init_forms(char *)
and creates a curses based form editor. See the examples directory
for a basic demo.
This effectively changes the non-DES password algoritm.
If you have the "securedist" installed you will have no problems with this.
(Though you might want to consider using this password-encryption instead
of the DES-based if your system is likely to be hacked)
If you are running a -current system without the "securedist" installed:
YOU WILL NEED TO CHANGE ALL PASSWORDS !! There is no backwards mode.
Suggested procedure is:
Update your sources
cd /usr/src/lib/libcrypt
make clean
make all
make install
passwd root
<set roots new password>
change password for any other users on the system.
This algorithm is expected to be much better than the traditional DES-
based algorithm. It uses the MD5 algorithm at what it is best at, as
opposed to the DES algorithm at something it isn't good at at all. The
algorithm is designed such that it should very hard to shortcut the
calculations needed to build a dictionary, and to make partial knowledge
(Hmm, his password starts with a 'P'...) useless. Of course if somebody
breaks the MD5 algorithm this looses too.
The salt is 48 bits (8 char @ base64).
The encrypted password is 128 bits.
And I am positively delighted to say that it takes 34 msec to crypt() a
password on a Pentium/60Mhz, so building a dictionary is not really an
option for hackers at the moment.
Given the right circumstances, a call to kvm_open can result in a core
dump.
The diff belows fixes this (note that this change is already in the
NetBSD code). Could somebody apply this?
Gary J.
Submitted by: gj
From: Chris Torek <torek@bsdi.com>
Here is a semi-official patch (apply to /usr/src/lib/libc/stdio/fseek.c,
rebuild libc, install). The current code fails when the seek:
- is optimized, and
- is to just past the end of the block currently in the buffer, and
- is followed by another seek with no intervening read operation, and
- the destination of subsequent seek is within the block left in the
buffer (seeking to the beginning of a block does not force a read,
so the buffer still contains the previous block)
so it is indeed rather obscure.
I may have a different `final' fix, as this one `loses' the buffer
contents on a seek that goes just past the end of the current block.
[Footnote: seeks are optimized only on read-only opens of regular
files that are buffered by the file's optimal I/O size. This is
what you get with fopen(path, "r") and no call to setvbuf().]
Obtained from: [ BSDI mailing list ]
I know that many of these entries are bogus and need to be revisited,
but let's get the tree working again for now and then do a pass through
looking at all the __FreeBSD__ entries, shall we?
While trying to figure out why rlogind wasn't working right for root,
I noticed that man wouldn't come back with a man page for iruserok, but
it would for ruserok. Checking the lib/net directory's Makefile.inc
file shows that the link to the rcmd man page just isn't getting
created.
>How-To-Repeat:
Do a 'man iruserok' and notihing will come back, where a 'man ruserok'
will.
Submitted by: Brian Moore <ziff@houdini.eecs.umich.edu>
Obtained from: NetBSD-bugs mailing list
on terminals with no pad char (cons25) and quote from tputs.c says so too:
! * Too bad there are no user program accessible programmed delays.
! * Transmitting pad characters slows many
! * terminals down and also loads the system.
and don't return error, if non-terminals. This fix allows curses
to work into full duplex pipes under control of main program,
like good old curses does.
getnet* configuration. (It's highly unlikely that you'd want to do
something different, and network lookups aren't common enough to justify
their own configuration file.)
DANGER WILL ROBINSON!
_PATH_UNIX is currently defined as the literal string "don't use this".
I am of two minds about this myself, but wanted to get something into the
tree as quickly as possible.
!!!!!!!!
NB
!!!!!!!!
You MUST pwd_mkdb /etc/master.passwd before attempting to use the new
libc, or things may go wrong. (I doubt anything actually /will/ go
wrong, but the actual behavior is undefined. YOU HAVE BEEN WARNED.)
The database format is, however, backwards-compatible, so old executables
will still work.
(void) setlocale(LC_ALL, "");
It will be easiest way now to make national chars available
for all ctype-oriented programs at once by simple:
setenv LANG Your_National_Charset
Default case (without "LANG" environment
variable) will be fully ANSI compatible (got "C" locale).
If "LANG" variable present, extention becomes active.
Effect of this extention is great: in one time all ctype
oriented programs can accept/print national characters
without any touching source/binary code, it is big win, IMHO.
This method is fully compatible with ISO8859-* and russian koi8-r
too (in general -- with all 8-bit character sets). I think
it is very useful.
I got this idea from Xenix locale implementation.
This extention is even never compiled in, unless you set
setenv STARTUP_LOCALE
before rebuilding crt0.c or corresponding variable in /etc/make.conf
>From: jtk@atria.com (John T. Kohl)
in rcmd:
It calls select() with a hardcoded "number of file descriptors" argument
of 32, rather than computing it based on the sockets about which it
cares.
- Now we work out the nfds arg, and do some error checking
Submitted by: Geoff.
and tiny*tiny at compile time. The evaluations are supposed to be done
at run time to set the IEEE exception flags. Many other source files
in libm and msun are missing this fix. Fixing them is not urgent since
the default IEEE exception masks don't allow use of the overflow
exception flag.
Don't add to POBJS or SOBJS. bsd.lib.mk does it. Some objects were
duplicated.
Don't add to CLEANFILES. bsd.lib.mk does it. Some objects were
quadruplicated.
Define variables that are only used once close to where they are
used.
The ifdefs for avoiding building of profiled/shared objects when
NOPROFILE/NOPIC are set were not actually committed. The ifdefs
belong in bsd.lib.mk anyway.
o __FULLINE added for AL/DL/CS optimization with __noqch.
refresh.c
o Attributes does not turned off before clearing screen, cause
highlighted screen.
o Proper usage of 'affcnt' tputs parameter, affects terminals with
padding.
o make AL/DL/CS optimize not only for __FULLWIN but for __FULLLINE.
ATTENTION: original code works _only_for_ FULLWIN, i.e. if you
use two FULLLINE windows like in 'talk', you have full slow repaint with
original code, I enhance this thing. All other fixes marked
with phrase 'wrong for non-full windows' or WFNFW is continue of this fix.
I rewrite scroll code too for proper working (see below and tty.c
changes).
o DEBUG code always use 'i' index from 0 to curscr->maxy instead of
'i - win->begy', fixed
o check added into DEBUG to be shure that index inside current window.
o ->hash assigment code is WFNFW (forget win->begy).
o when CE usage required, and last spaces number counted, code don't check
attributes, so last standouted space will be incorrectly cleared.
o cep (start pointer) forget to add win->begy/win->begx, code WFNFW.
o clsp (last space) wrong in two places at once: forget to add win->begy
(WFNFW) and incorrectly use 'win->begx * __LDATASIZE' in pointer
arithmetics.
o clsp check incorrect: was 'clsp < win->maxx * __LDATASIZE', need to
be 'clsp < win->maxx
o Attributes does not turned off before clearing end of line, cause
highlighted end of line.
o When find how many lines from the top/bottom of the screen are unchanged,
code always forget '- win->begy', WFNFW.
o NO_JERKINESS code forgets to add win->begy, WFNFW.
o Curw & Curs changed in comment description
o In search for the largest block of text not changed forget to add
'- win->begy' (several places), WFNFW.
o Forget to add '- win->begy' for non-dirty lines, WFNFW.
o touchline forget to add '- win->begy', WFNFW.
o rewrite scrolln():
* remove win parameter, we deal with whole screen (curscr) now;
* use NL or '\n' instead of sf, it is faster in any case;
(imagine: cat written on curses now use '\n' for scroll
like standard cat, no ugly escapes)
* use dl (if present) instead of DL, if abs(n) == 1, the same
about al/sr, it is faster;
* change win->maxy to 'curscr->maxy - 1', we deal with whole screen
here, WFNFW.
* SF can be correctly issued only if cursor at bottom of scroll
region (whole screen region included too), fix this;
* sr/SR can be correctly issued only if cursor at top of scroll
region (whole screen region included too), fix this;
* use pre-calculaded (in setterm.c) __usecs variable to determine
usage of CS or AL/DL;
* completely rewrite scroll region stuff using __set_scroll_region
from tty.c (see below);
tty.c
o Added __set_scroll_region function which set CS region and stays
back in old position. Use SC/RC (save/restore cursor) if possible,
else use HO and __mvcur.
o __startwin: added __set_scroll_region(whole screen) at program
startup, if __usecs;
o endwin: added __set_scroll_region(whole screen) at program
exit, if __usecs;
o Fix all tc{set/get}attrs to works properly, when stdin redirected,
use /dev/tty in this case (needed for some applications).
setterm.c
o Add new variable __usecs, if (!AL/al || !Dl/dl) && CS && (SC && RC || HO)
(save/restore cursor used in __set_scroll_region in tty.c).
o Set __noqch, if !__usecs && (!AL/al || !DL/dl).
o Proper ospeed initialization for tputs, i.e. if speed == B9600,
ospeed = 13
curses.c
o Add __usecs variable that indicates usage of CS (if AL/DL absent).
curses.h
o Allow translation with applications which includes <sgtty.h>,
undef BXXX manually to avoid redefinition and include termios
to define proper ones.
o Define old-style names curx/begx/maxx/etc. for old applications.
Define _tty like __baset too.
o Typedef SGTTY type for old applications (SGTTY == struct termios).
o wstandout/wstandend should be int and not char*, some old
applications relay on this fact. See standout.c too.
o __FULLINE added indicated line width == terminal width, needed
for refresh using AL/DL/CS with __noqch, see refresh.c changes.
o Add extern __usecs variable that indicates usage of CS (if AL/DL absent).
o Add __set_scroll_region() prototype, see tty.c and refresh.c changes
for details.
o Change winch() character mask from 0177 to 0377, we don't need to
strip high bit on national characters.
o Allow translate on systems with _BSD_VA_LIST_ undefined, such as
FreeBSD 1.1.5.1
o __tty_fileno added to allows work with stdin redirected, see tty.c
o Privately declare tputs (..., void) and externally tputs(..., int),
many applications require this. Maybe not nice thing, but needed.
o Remove _putchar definition and replace it to proper _putchar
prototype, some old apps declares: 'extern int _putchar()'
and don't even include curses.h in such modules. See putchar.c
cr_put.c
o __mvcur: if destline == destcol && outline == outcol do nothing,
i.e. don't issue any escapes.
o Proper usage of 'affcnt' tputs parameter, affects terminals with
padding.
cur_hash.c
o Change char->unsigned char for proper sum 8-bit national characters.
getch.c
o check for inp == EOF added, don't add EOF to window.
getstr.c
o check for EOF added, don't add EOF to str.
insertln.c
o add cast to (int) in comparation of y and win->cury, this produce
big number (cast to (unsigned)) if y < 0
tstp.c
o Fix all tc{set/get}attrs to works properly, when stdin redirected,
use /dev/tty in this case (needed for some applications).
o add tstp() function for compatibility, some applications wants it.
standout.c
o Some old applications relay in fact that wstandout/wstandend
returns int instead of char*, change return type to OK/ERR.
putchar.c
o Add _putchar function (which calls __cputchar),
some old apps declares: 'extern int _putchar()'
and don't even include curses.h in such modules.
automagically. -lfoo has to be right to work, but ${LIBFO0} is too
easy to forget or misspell; nothing checks it and it should be
different for shared libraries.
because libmd builds a test program before installation and if
you've used CLOBBER there's no crt.0 to link with. This ensures
that in a make world the csu objects will get installed before
reaching the libmd directory.
Reviewed by:
Submitted by:
