| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
|
|
| |
Nothing in libftpio uses _file, and the only consumer in the tree
(sysinstall) doesn't invoke fileno() on the FILE.
MFC after: 2 months
|
|
|
|
| |
(Sorry, forgot to commit this earlier.)
|
|
|
|
|
|
|
|
| |
accessor functions for its benefit now thaat FILE is opaque.
I'm sure there's a better way. I leave that for people to work
on in a src tree that isn't broken.
Pointy hat: jhb
|
|
|
|
| |
Submitted by: Antoine Brodin
|
| |
|
|
|
|
|
|
| |
scope identifier).
MFC after: 3 weeks
|
| |
|
| |
|
|
|
|
| |
- The ftpPassive()
|
|
|
|
| |
Don't assume 227 and 228 replies enclose remote address with parentheses.
|
|
|
|
|
|
| |
This is required for forthcoming IPv6 ready installer.
Obtained from: KAME
|
|
|
|
|
| |
Reviewed by: des
Obtained from: FreeBSD's src/usr.bin/ftp/main.c
|
|
|
|
| |
Suggested by: Eric D. Futch <efutch@nyct.net>
|
|
|
|
|
|
| |
initialization of sa_flags added so that the 'struct sigaction'
can be declared local in both functions that use the global
(static) declaration. Remove the global declaration.
|
| |
|
|
|
|
|
|
|
| |
(and its friends) more firewall friendly.
PR: 10580
Submitted by: nsayer
|
|
|
|
|
|
| |
PR: 6259
Reviewed & slightly modified by: phk
Submitted by: Archie Cobbs <archie@whistle.com>
|
|
|
|
|
| |
function from 'errno' to 'error' so that there is no conflict with the
thread-safe definition of errno in errno.h.
|
|
|
|
|
|
|
|
|
|
|
| |
can return UNIX errnos. When UNIX errnos catch up with FTP status
codes (e.g. at 100) a new way will have to be found to tell which
is which.
This allows fetch to print errors like
fetch: ftp.fu-berlin.de: No route to host
instead of
fetch: ftp.fu-berlin.de: Unknown error
|
|
|
|
|
|
| |
nuked file descriptor. This is probably why sysinstall's ftp xfer
occasionally SEGV'd if you left things alone for a long time and
the timeout code got called. Whoops!
|
|
|
|
|
|
|
|
|
| |
plain 0 should be used. This happens to work because we #define
NULL to 0, but is stylistically wrong and can cause problems
for people trying to port bits of code to other environments.
PR: 2752
Submitted by: Arne Henrik Juul <arnej@imf.unit.no>
|
| |
|
|
|
|
| |
just going to hang forever on the close).
|
|
|
|
|
|
|
|
|
|
| |
the handler with SA_RESTART set, so the system calls I wanted to have the
timeout effect will just restart instead (which is NOT what I wanted).
Sheepishly use sigaction() like a good boy and make timeouts actually do
something.
Also pass errors out more effectively so that fetch(1) actually understands
what went wrong.
|
|
|
|
|
|
|
|
| |
This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.
Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.
|
| |
|
|
|
|
| |
(lack of sleep)
|
|
|
|
|
| |
connection at the end of operation, so it not leave opened
file without a reason.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
FTP error return code because
1) They return NULL, it means that ftpErrno can't be used because
it takes file pointer
2) They don't have FILE-type argument as f.e. ftpGet/ftpPut to use
it for ftpErrno instead.
For that functions I add yet one int* type argument to store
FTP error return code. It is impossible to add some global variable
for that reason, because user can have multiply FTP connections
opened at the same time.
So, interface changed, major number bumped.
Userland changes will follows.
Minor bugfixes, the code:
Forget to close file in few places, when failure occurse
Forget to NULL cached host name, multiply free is possible
|
|
|
|
|
|
| |
at the same time, so add new con_state to avoid QUIT recursion
Still should go to 2.2
|
|
|
|
|
|
|
|
| |
send QUIT to closed connection.
2) Preserve login failure code, don't overwrite it with ftp_close
code
Should go to 2.2
|
|
|
|
| |
Submitted-By: Archie Cobbs <archie@whistle.com>
|
|
|
|
|
|
|
| |
Major version bumped (by me) since the ftpGet() public interface has
changed (an "int *" becomes and "off_t *")
Submitted by: Jason Thorpe <thorpej@nas.nasa.gov>, PR#1640
|
|
|
|
|
|
|
| |
use it.
Correct a typo bogon that had REST mistyped as RETR. No wonder fetch's
restart command didn't work! :-(
|
| |
|
|
|
|
|
| |
be zero'd only if the operation *fails*, indicating that the file
starting offset is effectively zero. This makes more sense.
|
|
|
|
|
| |
codes.
Submitted-By: Archie Cobbs <archie@whistle.com>
|
|
|
|
|
|
| |
2. Remove pkg_* support - tcl7.5's channel interface has rendered this
almost entirely unsupportable (at least in the way it currently stands).
Submitted-By: jmz & jkh
|
|
|
|
| |
Submitted by: jmz
|
|
|
|
| |
ftpAscii() call sets connection to ascii as counterpart to ftpBinary().
|
|
|
|
|
|
| |
This will do as a stop-gap until I figure out a more fault-tolerant
way of having deferred closes against the control connection work
without blocking.
|
| |
|
|
|
|
| |
Make ftpChdir return the server status again - it was more convenient.
|
|
|
|
|
|
| |
from jmz was a hopeless kludge (sorry Jean-Marc :) and handled the problem
in the wrong way. ftpRestart() has now gone away and ftpGet() has grown a
new parameter.
|
|
|
|
|
| |
(the convention as established by pkg_install(1)), select passive mode
FTP automatically.
|
|
packages and also sort of give the (correct) impression that this basically
sits on top of stdio and deals with stream pointers (FILE*).
|