029b8989db
[project @ robey@lag.net-20061215222108-7pu0151970w1e1lp]
...
add a ResourceManager to replace __del__ methods, and use it in SSHClient
to automatically close any open transport when the SSHClient is collected.
this won't work on Transport itself (to close the attached packetizer)
because Transport starts up its own thread, and the threading library
keeps a Transport object alive to run that thread. i think that's okay;
the SSHClient interface is meant to be the easier one, so that's the one
where it's important that some auto-cleanup is attempted.
2006-12-15 14:21:08 -08:00
7058f5ead2
[project @ robey@lag.net-20061215211936-rgc3uzy5ai9h6qho]
...
bug 75370: notice garbage sftp packets
since sftp packets shouldn't be larger than about 32k, if the first length
byte is non-zero (ie, the packet size > 16M), raise an exception.
2006-12-15 13:19:36 -08:00
76285309cf
[project @ robey@lag.net-20061203023334-72zo500ys0l1le30]
...
potential port forwarding support; probably buggly cuz i haven't added unit tests yet :)
2006-12-02 18:33:34 -08:00
aaa97d72c0
[project @ robey@lag.net-20061127192023-ns5lhww0jc84f21t]
...
absorb socket.errors caused by a closed socket if we knew it was closed
2006-11-27 11:20:23 -08:00
fec76c51b1
[project @ robey@lag.net-20061121001117-8mf8zzltvfvzzrv7]
...
add support for opening x11 channels, and a unit test
2006-11-20 16:11:17 -08:00
48bb10694b
[project @ robey@lag.net-20061120192137-1rcpiiq9mcd58m76]
...
reorder the closing of the pipe in Channel.close() to make sure it happens even when the channel is closed by the remote host first
2006-11-20 11:21:37 -08:00
b956cc4831
[project @ robey@lag.net-20061119205558-p5p2tmqx0xt7hq9r]
...
bump version to 1.6.4 (yanma) and trim the README changes-history a bit
2006-11-19 12:55:58 -08:00
a9c51b23ce
[project @ robey@lag.net-20061113041742-e24468a63d31b8bd]
...
sometimes the sftp module is used with raw sockets, not just paramiko
Channels. in this case, calling recv() will never return. so notice
this and use select() to give python a chance to notice a closed socket.
this kind of thing is especially useful for unit tests.
2006-11-12 20:17:42 -08:00
48afc3082a
[project @ robey@lag.net-20061112063054-bf71926ff8c98c68]
...
a bit of cleanup in close() to try harder to avoid having lingering threads
2006-11-11 22:30:54 -08:00
e2d8357622
[project @ robey@lag.net-20061112055617-098a150cf051bffa]
...
try a trick that should let 'hostkeys[hostname][keytype] = key' work for
HostKeys objects again.
2006-11-11 21:56:17 -08:00
bee3535484
[project @ robey@lag.net-20061111005330-08c517e2576693bd]
...
use python from env for unit tests
2006-11-10 16:53:30 -08:00
e736341e20
[project @ robey@lag.net-20061111004413-7bab08f1bad7f96f]
...
bug 70398:
allow constructions like:
hostkeys['hostname'] = {}
to create an empty host entry object, so that future attempts to set
keys will at least not throw an exception. (they'll still silently do
nothing, though.)
2006-11-10 16:44:13 -08:00
482d0bcef2
[project @ robey@lag.net-20061031223747-9fda3f2c1112c5dc]
...
(bug 69222)
on some recent linux kernels, a socket can return "readable" from select,
but a subsequent read() will return EAGAIN. this is against the contract
of select(), so python's socketmodule doesn't catch it or handle it.
therefore, we need to. EAGAIN should now be treated the same as a
socket timeout.
2006-10-31 14:37:47 -08:00
4628a53acf
[project @ robey@lag.net-20061031190617-1113fb493ff5b690]
...
(bug 69330)
in SFTPClient._auth, check for the existence of the rsa/dsa keys before
trying to open them, so that an I/O exception doesn't mask an earlier
one.
2006-10-31 11:06:17 -08:00
f076c51d4e
[project @ robey@lag.net-20061031185035-b57ad8184c39063a]
...
include setup_helper.py in the manifest
2006-10-31 10:50:35 -08:00
7e27adacf1
[project @ robey@lag.net-20061015015628-371a037b5165d634]
...
bump version to 1.6.3
2006-10-14 18:56:28 -07:00
f4199b5a28
[project @ robey@lag.net-20061015012126-36370734592aff17]
...
might as well assert this
2006-10-14 18:21:26 -07:00
4d090c71a6
[project @ robey@lag.net-20061008002032-d6803b57bac4f62c]
...
i think the compression needs to be inside this lock in order to preserve order
2006-10-07 17:20:32 -07:00
93ab477408
[project @ robey@lag.net-20061001034525-8f78a989ecde720c]
...
bug fix from mike looijmans: notify any thread waiting in accept() when the transport dies
2006-09-30 20:45:25 -07:00
64b89b6efd
[project @ robey@lag.net-20061001034455-bf6880fdbd4dd29e]
...
trivial doc fixes
2006-09-30 20:44:55 -07:00
6e49ddbc44
[project @ robey@lag.net-20060912073153-4d8c5623d4c2aa7b]
...
merge patches from wouter van heyst and john arbash-meinel for fixing tarballs on os x, and fixing md5 on linux
2006-09-12 00:31:53 -07:00
a4f09c31ed
[project @ robey@lag.net-20060912072641-dc427855d7569dda]
...
clean docs too
2006-09-12 00:26:41 -07:00
31c26eafb9
[project @ robey@lag.net-20060911175604-bf2798fcebd62780]
...
trap EOF errors when reading a response, and translate into SSHException
2006-09-11 10:56:04 -07:00
4acb9589ce
[project @ robey@lag.net-20060908023959-de70f0675f25fa64]
...
fixes from mike barber for chdir and sftp attr __str__
2006-09-07 19:39:59 -07:00
305a4ae417
[project @ robey@lag.net-20060908022816-ea2856cefb9d83e0]
...
fix from john arbash-meinel for the stub sftp server's default open mode
2006-09-07 19:28:16 -07:00
682185bac1
[project @ robey@lag.net-20060908015516-1b8ac2af031586fe]
...
better way to change the module name
2006-09-07 18:55:16 -07:00
df20443b29
[project @ robey@lag.net-20060901205930-021bcc34e3d5d8c8]
...
sadly, revert the append optimization -- it breaks for openssh
2006-09-01 13:59:30 -07:00
1f4a3f1976
[project @ robey@lag.net-20060829181836-1cd17df7729c4db8]
...
remove usage of has_key, which is dangerous and deprecated
2006-08-29 11:18:36 -07:00
b23079b135
[project @ robey@lag.net-20060829012231-328fee7de77036be]
...
fix a couple of pcheck warnings
2006-08-28 18:22:31 -07:00
2b8738d3ce
[project @ robey@lag.net-20060828234834-51542dc36057b361]
...
fix __setitem__ to do the right thing
2006-08-28 16:48:34 -07:00
c2b38a1fb8
[project @ robey@lag.net-20060823025610-adee7f6d9bd92725]
...
ok, ok, switch back to open()
2006-08-22 19:56:10 -07:00
145ceab54c
[project @ robey@lag.net-20060823025538-3f8a4d761d7d4118]
...
when a file is open for append, don't stat to get the file position unless the user asks for it explicitly
2006-08-22 19:55:38 -07:00
738e81033a
[project @ robey@lag.net-20060822182819-26cf01113495c9a3]
...
don't automatically make epydocs for the private (internal) methods and classes
2006-08-22 11:28:19 -07:00
cc3e383baf
[project @ robey@lag.net-20060822025601-aab3b723b95d3a9c]
...
add 'flags' param to SFTPHandle and make the default impl avoid calling tell() when in append mode; add proper append-mode support to stub_sftp
2006-08-21 19:56:01 -07:00
49418d1145
[project @ robey@lag.net-20060816213132-2e094c077d22dc50]
...
bump version to 1.6.2
2006-08-16 14:31:32 -07:00
bfe8fac5bf
[project @ robey@lag.net-20060816213106-99cffe9de5d555c5]
...
don't need odict anymore
2006-08-16 14:31:06 -07:00
929ce8df7a
[project @ robey@lag.net-20060813195045-ca4148e233794d4b]
...
add a unit test to verify that all the paramiko symbols are exported correctly
2006-08-13 12:50:45 -07:00
6d37b6b810
[project @ robey@lag.net-20060813193546-c92b58af1c27d156]
...
missing commas in exported class list (bug 55946)
2006-08-13 12:35:46 -07:00
4707b2d8d1
[project @ robey@lag.net-20060813193233-10db2cfe4c53a1df]
...
create a gzip file also, when building a release
2006-08-13 12:32:33 -07:00
4bedcfbf39
[project @ robey@lag.net-20060801182715-f0977a00fbcce456]
...
fix logic in checking prefetch buffers: don't stop looking for prefetched data just because all the buffers have arrived
2006-08-01 11:27:15 -07:00
05de66db6d
[project @ robey@lag.net-20060801182357-e030af9570314145]
...
use hexlify instead of home-grown hexify
2006-08-01 11:23:57 -07:00
074dc36e6b
[project @ robey@lag.net-20060801182143-088ff687e0945937]
...
use hexlify in tests
2006-08-01 11:21:43 -07:00
7ac433a6ac
[project @ robey@lag.net-20060801182053-a74d120228b8009a]
...
use hexlify instead of custom-made hexify in demos
2006-08-01 11:20:53 -07:00
157484f5eb
[project @ robey@lag.net-20060731070137-571efc087c20347a]
...
clean up test.py a bit and allow filtering from the command line
2006-07-31 00:01:37 -07:00
5bbe1a8f36
[project @ robey@lag.net-20060731065604-ec418957546f7ec9]
...
explain about the new launchpad site
2006-07-30 23:56:04 -07:00
4086a5cd35
[project @ robey@lag.net-20060729182801-e5e501844712bd85]
...
oops, forgot to import SSHException
2006-07-29 11:28:01 -07:00
6821b6e8e8
[project @ robey@lag.net-20060727025519-bffe5bcc0556d779]
...
variant of a patch from warren young to preserve the order of host entries from the 'known_hosts' file and preserve knowlege of which lines had multiple hostnames on them
2006-07-26 19:55:19 -07:00
17a93bce4c
[project @ robey@lag.net-20060726191515-6ff627b4d7216073]
...
if a chunk has been requested in prefetch, don't bother doing an overlapping prefetch during readv
2006-07-26 12:15:15 -07:00
c731a077fb
[project @ robey@lag.net-20060726002947-e60cb0a3d7b86919]
...
don't fetch readv chunks that we have reason to believe are already in prefetch buffers. no longer need to order the prefetch requests either.
2006-07-25 17:29:47 -07:00
c24db3e38c
[project @ robey@lag.net-20060726000907-b9a2d46eecc64cec]
...
allow prefetch + readv to occur at the same time (even though it will be really inefficient). instead of a moving pointer, use the prefetched buffers as an indication of what we've downloaded so far. break up large readv requests into the max packet size. add 2 more unit tests to test this stuff.
2006-07-25 17:09:07 -07:00
Robey Pointer