prints out info if bytes > 0). only set the handler if SIGINFO is defined
* hijack SIGQUIT to be the same as SIGINFO (foreign ports have this, and it's
annoying to have SIGQUIT dump core on netbsd when it prints info on other
systems)
* in {recv,send}request(), factor a lot of duplicated code out into a
`cleanup' section at the end
* rework shell() a bit
* implement xsignal_restart(), which only sets the SA_RESTART flag if
specifically requested
* xsignal() is now a wrapper to xsignal_restart(). INFO, USR1, USR2 and WINCH
are restartable, ALRM, INT, PIPE and QUIT are not.
* improve getreply()'s timeout code to take advantage of the above.
other changes:
* improve wording of how globbing works for `classic' URLs (host:path).
suggested by John Refling <johnr@imageworks.com> in relation to PRs
[bin/8519] and [bin/8520]
* always compile in the `edit' command even if NO_EDITCOMPLETE defined.
it's just a no-op in the latter case, which is more consistent to
the users.
* always compile in about: support (i.e, remove NO_ABOUT).
i'm entitled to some vanity in this program...
* clean up some whitespace
the disabled state can be overridden by toggling epsv4.
(I got sick of the errors about EPSV not being supported on almost
every server I connect to. This way we retain support for epsv4, but
it's not so whiny after the first failure...)
to "extern" if it's not set. define GLOBAL to (empty) in main.c.
this effectively moves all the globals into main.c whilst retaining
namespace access to them in other source files.
(global vars in header files confuse foreign linkers)
should be in host order. found/fixed by Matthias Pfaller <leo@dachau.marco.de>
* parse_url(): improve checking of portnum, and add an extra argument to pass
back the parsed portnum to the caller (reduces a bit of code duplication)
* Move the KAME/WIDE copyrights after the BSD/TNFi ones. Since there was
significantly less code added under the former, it's only fair on the latter.
ETA might be over and will appear to drop rapidly, rather than being under
and appear not to change. the original code makes sense when you're testing
by suspending & resuming the client. however, the unfudged number is probably
better in reality, especially for slow spurty networks.
requested by ITOH Yasufumi <itohy@netbsd.org> in [bin/7977]
lpage page local files
pdir as dir, but through your $PAGER
pls as ls, but through your $PAGER
* implement docase() (a la dotrans() et al) and use appropriately, rototilling
some duplicated code
* globulize(): modify to return a pointer to the strdup()ed result in all cases,
and hack the code that calls it to take this into account
* replace strcpy() and strncpy() with strlcpy()
* put(), getit(): use some aptly named local vars instead of argv[...]
* delint
prototype for idle() in <unistd.h> (which i thought was against namespace
and sensibility guidelines, but...)
* consistently use xsignal() instead of signal(). we get known behaviour
in all cases (SA_RESTART), which is good for some borken foreign systems.
* remove signal.h from most files; it's unnecessary now
* fetch_url(): use `long chunksize' instead of ssize_t; it's more portable, and
we're setting chunksize with strtol() anyway
* xsignal(): only use SA_RESTART if it exists. SunOS 4.x doesn't have it
but has the inverse (SA_INTERRUPT). the original function i was inspired
from had this support (lib/signal.c, W. Richard Stevens' `UNP 2nd ed Vol 1').
* remove <termios.h> from util.c; it should be unnecessary now
prototype for idle() in <unistd.h> (which i thought was against namespace
and sensibility guidelines, but...)
* consistently use xsignal() instead of signal(). we get known behaviour
in all cases (SA_RESTART), which is good for some borken foreign systems.
* remove signal.h from most files; it's unnecessary now
* fetch_url(): use `long chunksize' instead of ssize_t; it's more portable, and
we're setting chunksize with strtol() anyway
* xsignal(): only use SA_RESTART if it exists. SunOS 4.x doesn't have it
but has the inverse (SA_INTERRUPT). the original function i was inspired
from had this support (lib/signal.c, W. Richard Stevens' `UNP 2nd ed Vol 1').
* remove <termios.h> from util.c; it should be unnecessary now
it's more portable and more obvious
* remove the mkgmtime() && HAVE_TIMEGM stuff:
a) why should netbsd have to define HAVE_TIMEGM to compile cleanly?
b) foreign compiles of ftp should just be linked with working
timegm function
a more portable version of this ftp client will be released as a 3rdparty
product; no use polluting our code with half-baked attempts...
* abort_remote(): replace borken MIN(4,BUFSIZ) with just BUFSIZ; it
should have been MAX(4,BUFSIZ), but it's probably safe to assume that
BUFSIZ is at least 3... (fix from simonb)
* auth_url(): use the correct variable when calculating a buffer size.
add lots of comments about how to size up the buffers, and add extra
checks to hopefully ensure that there won't be an overflow (unless
someone modifies the length of the sprintf()s).
* as part of the above, slightly rework the way the `*' bar is calculated.
also fixes a display bug when > 160 stars were needed to be printed.
the maximum progress bar width at this time is 256.
* remove some code that checks the port that was #if 0-ed out as part of the
ipv6 migration; it's not going to be used again.
* document the above three commands
* rototill the way the sndbuf and rcvbuf work. remove resetsockbufsize()
* use the appropriate socket buffer size as the size of the buffer that
the read()/write() loops use. speeds up things in some cases.
list of target nodes returned by Make_ExpandUse(). We have to search the
cohorts explicitly while iterating through the tree. So, tweak CompatMake()
to do this.
can (i.e., everything except environment variables, which aren't
stored in hash tables).
While we're here, inline the body of VarDelete into Var_Delete since
it's the only caller and it's just simpler that way when v->name can
share storage with the hash entry and may not need to be freed
separately.
Speeds up the infamous libc build benchhmark maybe 1% on PIII, 4% on
alpha pc164
Suggested by Perry Metzger.
Build a list of `cohorts' as before, but do *not* link each one into all the
parent nodes; instead, copy the `cohort' lists into the stream of targets to
be built inside Make_ExpandUse(). Also do the attribute propagation as a
separate pass after parsing.
This eliminates several O(n^2) algorithms.
passing in the LstNode of the child being inspected. Shaves off another few %,
particularly when there are long child lists containing $ expansions (e.g. in
libc).
* Treat search patterns as BREs only, per 1003.2 and XCU5.
* Do not perform any special treatment when an unqoted * or @ is given as
the first character of the search pattern.
Addresses PR bin/8269 by Chris Demetriou.
connection destination, hoping this to help ftpd's behavior with
scoped IPv6 addresses.
I'm not sure if it is the right way, but it is the best way available to us.
LPRT or EPRT command gives no information about which interface (or scope)
to be used for new data connection.
ftp(1): On data connection establishment, warn if scoped address is used.
If peer (ftp daemon) does not handle scoped address, data connection
may not work right.
This seems to be sort of protocol spec hole, not implementation issue.
a malloc'ed pointer and it tries to realloc(3) it if it had to grow it
before. su(1) gave it a pointer from the stack which caused realloc to
core dump.
- just display the hostname:port of the proxy url, rather than the full url.
this prevents someone `shoulder surfing' a proxy username/password
in $http_proxy. [suggested by perry]
- compact verbose notes for http fetchs; now displays
(via host:port, with authorization, with proxy authorization)
with each component being optional.
(and a couple introduced with the ipv6 mods...)
- don't override host with the canonical name; this prevented fetches from
http/1.1 virtual hosts from working if the virtual host was a CNAME.
[noted by bernd]
- call freeaddrinfo() if res was built with getaddrinfo()
some of them are caused by the kvm_getfiles() implementation being
TOTALLY BROKEN for dead kernels. Print out the KVM descriptor's
error so that error causes are more easily diagnosed.
runpath > built-in default; this is the behaviour of the SVR4 shared loader,
and gives users the opportunity to override the runpath. (Addresses a report
on current-users by John Kohl.)
chown/chgrp stages which would otherwise blow out if not running as
root. A future version may log the owner/group to allow correction by
a later privileged cleanup pass.
Just warn about madvise failure; it's not something to blow up about.
Document new option, and clean up option grouping for -d (install
directory) option.
I hereby place this into the public domain, the script contains
no copyright header to make it small. Replacing the binary
implementation saves 8k on /usr.
See also PRs 6614 and 361.
- fix command name in SYNOPSIS section of man page.
- link ktrace man page to ktruss man page.
- add executable and man page for ktruss in distrib sets.
not SDYING, so s/SDYING/SDEAD/.
And add the new element "dying" to the state_abbrev array.
XXX Is is correct to have two elements with the number 3 in sorted_state?
1. If fork()ing, the program is executed twice.
2. If the ktruss(1) output is bufferd, the final output is not flushed
(because it is executing).
% ktrace -f - rm >/dev/null
usage: rm [-dfiPRrW] file ...
usage: rm [-dfiPRrW] file ...
% ktruss date | cat
Fri Jul 23 12:09:45 JST 1999
Fri Jul 23 12:09:46 JST 1999
%
same name, match files' entire paths against regular expressions.
-regex is case sensitive, -iregex is case-insensitive. Note that these
primaries are _not_ entirely compatible with the GNU find primaries,
because their BREs appear to support alternation with \| whereas our BREs
do not. Also note there are no primaries which provide extended regular
expressions matching, though if they are desired they would be trivial
to implement.
the an empty directory (e.g, between `some' and `path' in
`ftp://host/some//path'), then execute `CWD ' (without a path).
This command will probably fail on rfc 959 compliant servers, so
issue a warning in this case and bail. [noted by cgd].
(i wonder if the people who wrote rfc 1738 actually realised that this
requirement appears to contravene the spec for `cwd' in rfc 959 ?)
* replace isurl() with isipv6addr(), and use appropriately. fixes
auto-login with `classic ftp URLs' (e.g, `ftp somehost:')
* cleanup and rework some of the ipv6 stuff in parse_url()
* prevent potential coredump in fetch_ftp() when parsing `;type=X'
* KNF a few lines
* fix a couple of comments
* cleanup the man page a bit
this is mainly for (hypothetical) ftp server which disconnect clients
that use EPSV/EPRT. I've never seen any ftp server like this, but
epsv4 command may be of use when such an ftp server is found.
in the presence of .PATH directives by specifying:
.PATH: .DOTLAST
This will be used to fixup the build system to work with both crypto-us
and crypto-intl sub-trees.
Make(1) changes by Christos Zoulas, after much badgering by me :-)
xfer rate stuff, but i never completed the changes that didn't need it
set).
fixes a coredump noticed on current-users@ by Chan Yiu Wah <c5666305@hkstar.com>
features:
---------
* transfer rate throttling with the new `rate' command. syntax:
rate direction [max [incr]]
where direction is `all', `get' or `put'.
if max is not supplied, the current settings are displayed.
if max is supplied, then transfers in the given direction will
be throttled to this value.
if incr is supplied, the increment for the `on-the-fly' scaling
will be set to that, otherwise `1024' is used.
currently implemented for binary get, binary put, and url fetches.
not yet supported for ascii get or put, or local file copies.
* on-the-fly scaling of the throttle based on signals:
- SIGUSR1 raises the throttle rate by the increment for that direction
- SIGUSR2 lowers the throttle rate by the increment for that direction
* -T dir,max[,incr] option to set rate from the command line
* `k', `m', `g' suffix support for bytecounts in the `hash', `rate',
`rcvbuf' and `sndbuf' commands)
bug fixes and code mods:
------------------------
* fix up ftp_login() so that ruserpass() is always called, even for
command-line url fetches.
* implement strsuftoi(), which parses a given number into a int with
suffix support. replaces getsockbufsize()
* implement parserate(), which does the argv parsing for -T and rate
* save and restore errno in signal handlers (may not be necessary, but
it doesn't hurt)
notes:
------
the rate command has had reasonable testing, but I'd like feedback
if it doesn't do the right thing, especially from people on slower
(i.e, modem) links.
I haven't tested the rate throttle against a http server which does
`transfer-encoding: chunked' because I couldn't find a server to
test against.
transfer stats are printed when verbose is enabled even if ftp is compiled
with -DNO_PROGRESS/-DSMALL. this adds back some feedback that existed
in traditional ftp that was lost when i added the progress meter. (noticed
when ftp-ing from the -DSMALL version of ftp that's on the 1.4 boot floppy)
to 0,0 was unnecessary because wclear() is documented to do that. also,
the immediate wrefresh() done by msg_clear() means that if we can take
advantage of full screen clear sequence, we will.
is _bogus_ in the face of printf-like message formatting! Instead,
calcuate the max size to format when the message window is set. We know
that we'll never want to format more characters than can fit in the window.
NO_PROGRESS. -DSMALL still implies all of those. progress meter support
isn't necessary for the smallest possible ftp client, but it adds very
little space and makes users' lives much better. Therefore, it should
be enabled for installation media if at all possible.
(say width or height).
* work harder to make menu fit width-wise. (if necessary, move the menu
further left, and if that's not good enough exit with an error.)
* move menu system initialization out of process_menu() so that it can
be called earlier (e.g. by the sysinst main()), since it initializes
curses and other code might want to access the curses data structures
before process_menu() is invoked.
which is one too many). Also, msgc manual page says that 'def' and 'val'
can be the same string, but the way input was done (characters typed
went directly into val) meant that the contents of 'def' would be corrupted.
If the user backspaced to the beginning of the line and hit return (to accept
the default), they'd get a combination of the old default string and the
new characters they typed. alloca() a buffer an put new input there to
avoid this problem.
they check at that level whether or not to build binaries.
A reference to telnet was also here (protected by .if make(obj)...,
but since A) it's already referenced in the main SUBDIR list and b)
the telnet Makefile does this checking as well, it's gone too.
the latter only seemed to work for TIS Gauntlet and not TIS fwtk.
thanks to simonb@netbsd.org for testing this. fixes [bin/5556].
* if EOF (e.g, ^D) is entered at a username/password/account prompt which
happens to use fgets(), exit the login rather than treating EOF as CR.
* don't use the comma operator where separate statements are valid
* always use snprintf to copy stuff into malloced buffers, just in case
typos creep in and mean that the buffer ends up being overflowed
lukem