| Commit message (Collapse) | Author | Age | Files | Lines |
| ... | |
| |/
|
|
|
| |
This avoids writing "[42,\n 43,\n ...]" into the storage json,
which is (a) wasteful (b) hard to read by hand.
|
| | |
|
| |
|
|
|
|
|
|
| |
tor-log-ratelim mistakenly had a non-path dependency.
Changing this, and running `cargo fetch --offline` to update the
lockfile, causes the build to prefer in-tree for many entries in the
lockfile.
|
| |\
| |
| |
| |
| |
| |
| | |
HSS: Send DosParams extension to introduction points.
Closes #723
See merge request tpo/core/arti!1740
|
| | |
| |
| |
| | |
It's available when the HsIntro=5 subprotocol version is present.
|
| |/
|
|
|
|
|
|
| |
For the most part, this is just code movement. The implementation was
copied over to `define_derive_adhoc` from `define_key_specifier`, with
some minor adjustments that were needed to make it compile.
Closes #1093
|
| |\
| |
| |
| |
| | |
New tor-log-ratelim crate (improved varsion)
See merge request tpo/core/arti!1734
|
| | |
| |
| |
| |
| |
| |
| |
| |
| | |
This crate is meant to provide rate-limited log messages
to tell users about problems that can happen too frequently
to log individually.
This version is based off an earlier design, and off of
conversations with @diziet. It still needs tests.
|
| | |
| |
| |
| |
| |
| | |
This is the approach from appendix F.2 in rend-spec.
Part of #1053.
|
| |\ \
| | |
| | |
| | |
| | | |
Upgrade a couple of dependencies
See merge request tpo/core/arti!1736
|
| | | | |
|
| | |/ |
|
| | | |
|
| |/ |
|
| | |
|
| |
|
|
| |
See comments for design notes.
|
| |
|
|
|
|
|
| |
tor-rpcbase patchlevel only, depends on async-utils.
tor-llcrypto patchlevel only, depends on basic-utils.
tor-hspow patchlevel only, depends on hscrypto.
tor-linkspec patchlevel only, depends on basic-utils
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
```
These crates have new APIs:
tor-async-utils ADDED
tor-config ADDED
tor-hscrypto ADDED
tor-netdoc ADDED, plus BREAKING-experimental.
These crates have patch-level changes only:
tor-netdir (bugfix only). (re-exposes netdoc)
arti-rpcserver (tweaks only, uses nothing that broke.)
arti 1.1.10, no stable public APIs.
```
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
```
tor-basic-utils BREAKING
tor-rtmock BREAKING.
tor-cert BREAKING
tor-cell BREAKING
tor-proto BREAKING: re-exposes cell. (Also BREAKING-experimental)
tor-chanmgr BREAKING: Re-exposes proto.
tor-ptmgr BREAKING: re-exposes tor-chanmgr
tor-guardmgr BREAKING: re-exposes proto. (Only for ClockSkew I think?)
tor-circmgr BREAKING: re-exposes proto
tor-dirclient BREAKING
tor-hsclient BREAKING, re-exposes proto.
tor-hsservice BREAKING, re-exposes proto.
tor-hsrproxy BREAKING
arti-client BREAKING: re-exposes proto.
arti-hyper: BREAKING, re-exposes arti-client.
tor-dirmgr Let's assume BREAKING, very high-level.
tor-keymgr BREAKING, but experimental.
```
|
| |
|
|
|
|
|
|
|
|
|
| |
Representing the supported HTYPEs as `HandshakeType`s instead of `u32`s
makes it more difficult to pass in wrong/invalid values to
`HsDescBuilder::create2_formats`.
This also fixes a descriptor publisher bug spotted by @jnewsome, where
the advertised CREATE2 HTYPEs included HTYPE `1`, which is actually
supposed to be a reserved value. The publisher now only advertises the
`NTOR` HTYPE (just like C Tor).
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
| |
`ahash 0.8.3` has been [yanked], which is causing the cargo audit
job to fail.
[yanked]: https://github.com/tkaitchuck/aHash/wiki/Yanked-versions
|
| | |
|
| | |
|
| |
|
|
|
| |
(This is only used in the disabled tor-events crate, but it's
still easier to keep dependencies up-to-date than not.)
|
| | |
|
| | |
|
| |
|
|
|
| |
This doesn't test all the code paths, but it does test the main path
of execution (and detected a couple of bugs).
|
| |
|
|
|
| |
This should be smaller and faster than vec, and save some
allocations. It could also avoid a tiny sidechannel.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |\
| |
| |
| |
| | |
Non-working attempt to test onion service configuration.
See merge request tpo/core/arti!1640
|
| | |
| |
| |
| |
| |
| |
| | |
The problem here is that, for some reason, the proxy_rule
configuration doesn't actually work. I suspect a bug in
Flatten, since the same configuration works fine if you pass
it to the hsrproxy config directly.
|
| |\ \
| | |
| | |
| | |
| | | |
Provide and use a SimpleMockTimeProvider
See merge request tpo/core/arti!1639
|
| | | | |
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
For our time provider we want a priority queue that lets us look up
entries and delete them. std's Heap doesn't do that. priority-queue
has the API we need and reasonable stats etc, and is maintained.
It has some IMO-tolerable licence wrinkles.
|
| |\ \ \
| | |/
| |/|
| | |
| | | |
Initial work on top-level onion service configuration
See merge request tpo/core/arti!1638
|
| | | |
| | |
| | |
| | | |
This lets us write `[onion_services.allium_cepa]`.
|
| | |/
| |
| |
| |
| | |
These don't yet attach to the rest of the ArtiConfig, or actually
launch any onion services, but they're a start.
|
| |/ |
|
| | |
|
| |
|
|
| |
(The only change in these crates is that their dependencies changed.)
|
| | |
|