summaryrefslogtreecommitdiffhomepage
AgeCommit message (Collapse)Author
2023-01-09string_types -> strJeff Forcier
2023-01-09Move b, u helpers to util module for nowJeff Forcier
Later on we can likely nuke many uses of these
2023-01-09Remove or transmute all use of long()Jeff Forcier
- When wrapping literals: just go away - When wrapping variables whose values are already definitely integers (eg output of 'id()'): ditto - When wrapping variables of unknown provenance or which are definitely NOT integers: replaced with int()
2023-01-09Update BytesIOJeff Forcier
2023-01-09Remove py3compat.PY2, including related streamliningJeff Forcier
2023-01-09integer_types -> intJeff Forcier
2023-01-09Update StringIOJeff Forcier
2023-01-09Migrate some byte related helpers aroundJeff Forcier
I feel like we should be able to just nuke byte_chr and friends at this point, but it's not entirely obvious, so let's rock that boat later.
2023-01-09Require modern versions of bcrypt, pynaclJeff Forcier
We may end up using new Cryptography primitives instead of pynacl eventually, but for now, just using both of these libraries' "3.6+" releases as new baseline.
2023-01-09Move to cryptography 3.3 as baseline depJeff Forcier
and bump top end test matrix cell to 40ish
2023-01-09Tidy up requires/entrypoints/install docsJeff Forcier
- Realized at this date that _moving_ Ed25519 deps behind an entrypoint is going the wrong direction, that key type should be used _more_ now, not _less_. I no longer even remember why folks were pissy about pynacl/bcrypt exactly. - We can in fact just use ed25519 from cryptography itself now, so no more pynacl anyways (but still bcrypt) - Twiddle install language a bit re: invoke used in more places, and again removing the notes about ed25519 being special
2023-01-09Make py3compat drop its own changelog entry, and nix the file itselfJeff Forcier
NOTE: imports not fixed yet so this will break CI!
2023-01-09Classifiers, python_requires updateJeff Forcier
2023-01-09Bump invoke requirement to modern revJeff Forcier
May technically function with invoke 1.x but much easier to support if we ensure everything's working with the same transitive dep tree.
2023-01-09Changelog entry for py2 dropJeff Forcier
2023-01-09Overhaul dev-reqs again: modern stuff and less pinningJeff Forcier
2023-01-09Add 3.11 to test matrix tooJeff Forcier
2023-01-09add 3.10 to test matrixPaul Kehrer
2023-01-09update dev-requirements pinsPaul Kehrer
pin 3.6 and then 3.7+
2023-01-09CI config tweaksJeff Forcier
2023-01-09bump dev versionJeff Forcier
2022-11-04Cut 2.12.0Jeff Forcier
2022-11-04Merge branch '2.11'Jeff Forcier
2022-11-04Cut 2.11.1Jeff Forcier
2022-11-04Merge branch '2.11'Jeff Forcier
2022-11-04Merge branch '2.10' into 2.11Jeff Forcier
2022-11-04Merge branch '2.11'Jeff Forcier
2022-11-04Fix busted merge, le whoopsJeff Forcier
2022-11-04Cut 2.10.6Jeff Forcier
2022-11-04Merge branch '2.11'Jeff Forcier
2022-11-04Merge branch '2.10' into 2.11Jeff Forcier
2022-11-04Fix changelog linkJeff Forcier
2022-11-04Docstring tweaks, changelog entry, re #2125Jeff Forcier
2022-11-04Write basic test re #2125, including kwarg default value tweakJeff Forcier
2022-10-07Allow Transport factory to SSHClient.connectNoah Pederson
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.
2022-09-27Merge branch '2.11'Jeff Forcier
2022-09-27Merge branch '2.10' into 2.11Jeff Forcier
2022-09-27Mark CVE bug as 2.10-only in changelogJeff Forcier
2022-06-10TODO tweakJeff Forcier
2022-06-03Merge branch '2.10' into 2.11Jeff Forcier
2022-06-03Merge branch '2.9' into 2.10Jeff Forcier
2022-06-03Merge branch '2.8' into 2.9Jeff Forcier
2022-06-03Add test proving basic behavior of #1822Jeff Forcier
2022-06-03Changelog re #1822, closes #1822Jeff Forcier
2022-06-03split comment to linesyoav
2022-06-03Close socket on failureyoav
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
2022-06-03Merge branch '2.10' into 2.11Jeff Forcier
2022-06-03Merge branch '2.9' into 2.10Jeff Forcier
2022-06-03Merge branch '2.8' into 2.9Jeff Forcier
2022-06-03Skip tests requiring invoke if it's not installedMichał Górny
Since invoke is an optional dependency and only one group of tests require it, skip them gracefully rather than failing if it's not present.