Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Merge branch 'main' into 2013-int | Jeff Forcier | 2023-03-10 | 1 | -8/+13 |
|\ | |||||
| * | Massage SSHClient.connect exception-raising docstring | Jeff Forcier | 2023-01-20 | 1 | -8/+13 |
| | | | | | | | | | | | | | | | | | | - Sphinx Python domain can link to non-stdlib exception classes inside the `:raises:` fieldlist now! yay!! - Add explicit mention of 'can raise NoValidConnectionsError', which was never added when the latter was created. Was still technically correct (it is a subclass of socket.error) but that is not the best kind of correct, after all. | ||||
* | | Add timeout for opening an SSH channel | vakarisz | 2023-03-02 | 1 | -0/+5 |
|/ | |||||
* | Nuke retry_on_signal, pointless on modern Py3 | Jeff Forcier | 2023-01-16 | 1 | -2/+2 |
| | | | | | Additionally, other bits of code that retry EINTR can similarly just go away. | ||||
* | No more (object)ions | Jeff Forcier | 2023-01-16 | 1 | -1/+1 |
| | |||||
* | string_types -> str | Jeff Forcier | 2023-01-09 | 1 | -2/+1 |
| | |||||
* | Docstring tweaks, changelog entry, re #2125 | Jeff Forcier | 2022-11-04 | 1 | -2/+6 |
| | |||||
* | Write basic test re #2125, including kwarg default value tweak | Jeff Forcier | 2022-11-04 | 1 | -1/+3 |
| | |||||
* | Allow Transport factory to SSHClient.connect | Noah Pederson | 2022-10-07 | 1 | -1/+6 |
| | | | | | | | | | | Adds a transport_factory argument to `SSHClient.connect` that allows you to dynamically generate a Transport instance without, and therefore modify inner connection parameters before a connection gets established. This should address some of the issues in #2054 with minial changes to the API and no changes to Transport while allowing for arbitrary control over Transports API. | ||||
* | Merge branch '2.10' into 2.11 | Jeff Forcier | 2022-06-03 | 1 | -0/+4 |
|\ | |||||
| * | split comment to lines | yoav | 2022-06-03 | 1 | -1/+2 |
| | | |||||
| * | Close socket on failure | yoav | 2022-06-03 | 1 | -0/+3 |
| | | | | | | | | | | | | | | As mentioned in socket docs: "Sockets are automatically closed when they are garbage-collected, but it is recommended to close() them explicitly, or to use a with statement around them." Resolve #1126 | ||||
* | | Fix Free Software Foundation address | Paul Howarth | 2022-03-15 | 1 | -1/+1 |
|/ | | | | They moved from Temple Place to Franklin Street in 2005. | ||||
* | Merge branch '2.5' into 2.62.6 | Jeff Forcier | 2021-10-15 | 1 | -4/+6 |
|\ | |||||
| * | Enhance language around SSHClient.close()2.5 | Jeff Forcier | 2021-10-15 | 1 | -4/+6 |
| | | |||||
* | | Rename to disableD_algorithms everywhere for consistency | Jeff Forcier | 2019-06-21 | 1 | -4/+4 |
| | | | | | | | | | | Having it feel verb-y isn't that much better than the alternative, and consistency is a nice hobgoblin | ||||
* | | Client->Transport for disable_algorithms | Jeff Forcier | 2019-06-21 | 1 | -1/+4 |
| | | |||||
* | | Add docstring and signature support for disable_algorithms | Jeff Forcier | 2019-06-21 | 1 | -0/+6 |
| | | |||||
* | | Implement #322 - new ChannelFile subclass for stdin that calls ↵ | Jeff Forcier | 2019-06-14 | 1 | -1/+1 |
|/ | | | | shutdown_write on close() | ||||
* | Merge branch '2.4' | Jeff Forcier | 2018-05-29 | 1 | -51/+95 |
|\ | |||||
| * | Blacken under black 18.5b0 | Jeff Forcier | 2018-05-29 | 1 | -5/+5 |
| | | |||||
| * | Blacken Paramiko on 2.4 | Chris Rose | 2018-05-17 | 1 | -46/+90 |
| | | |||||
* | | Years later... | Jeff Forcier | 2017-11-29 | 1 | -0/+3 |
| | | |||||
* | | wrong link in doctring comment | Marwan Rabbâa | 2017-11-28 | 1 | -1/+1 |
| | | |||||
* | | add documentation for get_pty param in SSHClient.exec_command() | Marwan Rabbâa | 2017-11-28 | 1 | -0/+3 |
|/ | |||||
* | Implement new passphrase kwarg for SSHClient.connect() | Jeff Forcier | 2017-11-09 | 1 | -7/+20 |
| | |||||
* | flake8 for previous changeset | Jeff Forcier | 2017-10-11 | 1 | -7/+7 |
| | |||||
* | String format modernization, part 11070-remove-python26-and-33 | Jeff Forcier | 2017-10-10 | 1 | -15/+19 |
| | | | | | | | Choosing to skip it in some edge/corner cases where it's more harmful than helpful. Also choosing to replace many non-%s specifiers with regular old {} since I don't see why one would normally care. Again, eschewing that in spots where it seems like it might matter. | ||||
* | Use set literals | Jeff Forcier | 2017-10-10 | 1 | -1/+1 |
| | |||||
* | Use new(er)-style string formatting, {} instead of {0} | Jeff Forcier | 2017-10-10 | 1 | -2/+2 |
| | |||||
* | Fix gss_host setting and cleanup its logic | Sebastian Deiss | 2017-09-18 | 1 | -2/+4 |
| | | | | | The parameter 'kex_requested' is misleading, since setting 'gss_host' is also required for gssapi-with-mic. | ||||
* | Attempt to refactor gss_host setting and clean up/document its logic | Jeff Forcier | 2017-09-13 | 1 | -6/+5 |
| | |||||
* | Formatting yet again | Jeff Forcier | 2017-09-13 | 1 | -1/+2 |
| | |||||
* | versionchanged for gss_trust_dns | Jeff Forcier | 2017-09-13 | 1 | -0/+2 |
| | |||||
* | More formatting | Jeff Forcier | 2017-09-13 | 1 | -5/+7 |
| | |||||
* | Feels like we do still want set_gss_host gated by whether GSS is in play (as ↵ | Jeff Forcier | 2017-09-13 | 1 | -5/+6 |
| | | | | is in master) | ||||
* | Formatting | Jeff Forcier | 2017-09-13 | 1 | -3/+4 |
| | |||||
* | Trailing comma | Jeff Forcier | 2017-09-13 | 1 | -1/+1 |
| | |||||
* | Merge branch 'master' into 1063-int | Jeff Forcier | 2017-09-13 | 1 | -15/+58 |
|\ | |||||
| * | Merge branch '2.2' | Jeff Forcier | 2017-09-12 | 1 | -13/+12 |
| |\ | |||||
| * | | Overhaul PublicBlob and use it better within RSAKey. | Jeff Forcier | 2017-08-28 | 1 | -1/+1 |
| | | | | | | | | | | | | | | | | | | | | | | | | This allows server-side Paramiko code to correctly create cert-bearing RSAKey objects and thus verify client signatures, and now the test suite passes again, barring the stub tests. Re #1042 | ||||
| * | | Stub tests and partly-working implementation of 'load certs found alongside ↵ | Jeff Forcier | 2017-08-28 | 1 | -9/+21 |
| | | | | | | | | | | | | | | | | | | | | | key_filenames' behavior re #1042 This actually breaks existing tests due to test server not supporting certs...bah | ||||
| * | | Docstring/TODO tweaks | Jeff Forcier | 2017-08-28 | 1 | -1/+2 |
| | | | |||||
| * | | Refactor and clean up recently tweaked key loading bits in SSHClient | Jeff Forcier | 2017-08-28 | 1 | -20/+26 |
| | | | |||||
| * | | Changelog and docs re #1042 | Jeff Forcier | 2017-08-28 | 1 | -2/+16 |
| | | | |||||
| * | | string slice instead of rstrip, thanks ploxiln | Paul Kapp | 2017-08-22 | 1 | -1/+1 |
| | | | |||||
| * | | Add certificate filenames to look_for_keys | Paul Kapp | 2017-08-22 | 1 | -5/+15 |
| | | | |||||
* | | | Invent the parameter 'gss_trust_dns' for Kerberos support | Anselm Kruis | 2017-09-13 | 1 | -9/+11 |
| |/ |/| | | | | | | | | | | | | | | | | | | | | | In response to Paramiko issue #915 the parameter 'gss_trust_dns' was added for Kerberos support. Set by default to 'True' the parameter indicates whether or not the DNS is trusted to securely canonicalize the hostname of the target host. If set to 'False' the hostname entered will be passed to GSSAPI. This option behaves like GSSAPITrustDNS from OpenSSH. Also, the parameter 'gss_host' is now always set, regardless if GSSAPI is used or not. Further, a minor fix was required to make the SFTP test work again. | ||||
* | | Merge branch '2.1-gsskex-hostkeycheck-fix' into 2.2-gsskex-hostkeycheck-fix | Anselm Kruis | 2017-08-04 | 1 | -13/+12 |
|\ \ | |/ |/| | |||||
| * | Merge branch '2.0-gsskex-hostkeycheck-fix' into 2.1-gsskex-hostkeycheck-fix | Anselm Kruis | 2017-08-04 | 1 | -13/+12 |
| |\ |