summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* | | More tests for tor_netdir, plus a bugfix and a little refactoring.Nick Mathewson2021-08-171-34/+84
| | | | | | | | | | | | | | | | | | (While messing around with test coverage on `in_same_subnet`, I found that there was a copy-paste error where we were using v4_bits when v6_bits would have been appropriate.)
* | | Make construct_netdir() return a PartialNetDir.Nick Mathewson2021-08-175-11/+13
| | | | | | | | | | | | | | | This way we can still use it even if we've removed too many microdescriptors to make a real network.
* | | Test Netdir behavior with BadExit flag.Nick Mathewson2021-08-171-2/+60
| | |
* | | Make MicrodescBuilder implement Debug and CloneNick Mathewson2021-08-171-0/+1
| | |
* | | Re-work the test network generator to take a customization function.Nick Mathewson2021-08-171-17/+104
|/ /
* | X_m tiebreaker rules are now specifiedNick Mathewson2021-08-171-2/+0
| |
* | Update tor-netdir path fractions to avoid NaN.Nick Mathewson2021-08-161-15/+34
| |
* | Remove SPEC note in tor-consdiff: dir-spec has been updated.Nick Mathewson2021-08-161-2/+1
| |
* | Note SPEC statuses in tor-netdocNick Mathewson2021-08-162-2/+1
| |
* | Remove some notes about SPEC issues that are now resolved.Nick Mathewson2021-08-162-7/+3
| |
* | Make PaddingNegotiate parsing match C tor.Nick Mathewson2021-08-162-8/+0
| |
* | More channel-cell coverage.Nick Mathewson2021-08-163-19/+95
|/ | | | "Every little helps": This gets tor-cell coverage over 90%.
* Use add_warnings.py for clippy::cast_losslessNick Mathewson2021-08-1326-25/+27
|
* Merge remote-tracking branch 'origin/mr/59'Nick Mathewson2021-08-1330-16/+44
|\
| * Add the "cast_lossless" lint.S0AndS02021-08-1230-16/+44
| | | | | | | | | | | | | | | | | | | | | | | | > Check `cast_lossless` section of Clippy documentation for details; > > https://rust-lang.github.io/rust-clippy/master/index.html#cast_lossless This adds the following Clippy configuration to crates; #![deny(clippy::cast_lossless)] And applies suggested Clippy and `cargo fmt` fixes.
* | More test cases for tor_cell::relaycell::msgNick Mathewson2021-08-132-11/+109
| |
* | Comment fixes in tor_cell::errNick Mathewson2021-08-131-3/+3
| |
* | Misc fake coverage improvementsNick Mathewson2021-08-132-0/+20
| | | | | | | | | | | | These aren't true coverage improvements; they're just either improving the coverage of test code, or telling the coverage tool that it shouldn't worry about code in the fuzzers.
* | Improve test coverage on tor-unitsNick Mathewson2021-08-131-3/+111
| |
* | Restore success and failure counts with timeout state.Nick Mathewson2021-08-131-2/+18
|/
* Merge commit 'origin/mr/46^'Nick Mathewson2021-08-123-9/+7
|\
| * Change output type in DirResponse to Vec<u8>rls2021-07-303-9/+7
| | | | | | | | | | | | | | | | | | | | Previously the DirResponse contained a String for the output. By changing it to a Vec<u8> the caller has to parse the Vec<u8> and deal with potential failures. With this change `fetch_multiple`'s `useful_responses` should also contain non-UTF-8 responses. This will case an Err to be returned in the `download_attempt` function if the DirResponse does not contain valid UTF-8.
* | Fix a few rustdoc link warnings.Nick Mathewson2021-08-122-3/+6
| |
* | Merge remote-tracking branch 'origin/mr/58'Nick Mathewson2021-08-1220-38/+42
|\ \
| * | Fix code snippet in tor-persist docsLennart Kloock2021-08-122-2/+2
| | |
| * | Fix code snippet in arti docsLennart Kloock2021-08-122-2/+2
| | |
| * | Fix warning sign emoji in tor-client docLennart Kloock2021-08-122-2/+2
| | |
| * | Reorder doc headlines in tor-circmgrLennart Kloock2021-08-122-2/+6
| | |
| * | Fix typos in tor-chanmgrLennart Kloock2021-08-122-4/+4
| | |
| * | Add brackets and remove colon for better consistencyLennart Kloock2021-08-124-12/+12
| | |
| * | Rerun readmes.shLennart Kloock2021-08-113-1/+3
| | |
| * | Sync docs with readmesLennart Kloock2021-08-112-3/+3
| | |
| * | Fix code snippet in tor-socksproto README.md and docsLennart Kloock2021-08-112-2/+2
| | |
| * | Fix typos in tor-protover README.md and docsLennart Kloock2021-08-112-4/+4
| | |
| * | Fix previously fixed typos in docs tooLennart Kloock2021-08-101-2/+2
| | |
| * | Fix a typo in tor-rtmock README.mdLennart Kloock2021-08-101-1/+0
| | | | | | | | | | | | Remove a '#'
| * | Fix a few typos in tor-rtcompat README.mdLennart Kloock2021-08-101-3/+3
| | |
| * | Fix typo in tor-rtcompat README.mdLennart Kloock2021-08-101-1/+1
| | | | | | | | | | | | Change 'if' to 'is'
* | | Fix documentation link issues.Nick Mathewson2021-08-123-4/+5
| | |
* | | Use real network parameter values to tune test circuits.Nick Mathewson2021-08-123-6/+30
| | | | | | | | | | | | Finishes #145.
* | | Periodically launch timeout-testing circuits.Nick Mathewson2021-08-122-35/+78
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Our path-spec says that, if we don't have enough build-time data to make a good estimate for our timeouts, we should periodically try launching testing circuits just to see how long they take to build. Here we implement that. The implementation is a bit wonky because of how our crates are factored; see arti#161 for more info. Closes #145. Closes #57.
* | | WIP: Launch testing circuits as needed to find timeoutsNick Mathewson2021-08-112-0/+64
| | | | | | | | | | | | This won't work yet, since it needs directories and doesn't have them.
* | | For timeout testing, support for "useless" circuits.Nick Mathewson2021-08-112-0/+32
| | | | | | | | | | | | | | | We need the idea of launching a circuit for timeout testing, and getting something that either exits someplace or exits noplace at all.
* | | tor-client: remove workaround now that #155 is in.Nick Mathewson2021-08-111-10/+0
| | |
* | | Add ability to launch a circuit to _some_ exit, without giving ports.Nick Mathewson2021-08-113-3/+67
| | | | | | | | | | | | | | | | | | We'll want this for testing circuits and for RESOLVE requests. Closes #155.
* | | Add functions to check whether port policies are nonempty.Nick Mathewson2021-08-112-1/+33
| | |
* | | fallbackdir: Regenerate listDavid Goulet2021-08-111-869/+879
| | | | | | | | | | | | | | | | | | Closes #160 Signed-off-by: David Goulet <[email protected]>
* | | Bump to latest arti-corpora.Nick Mathewson2021-08-111-0/+0
| | |
* | | Upgrade to latest fslock and use its per-fd exclusion feature.Nick Mathewson2021-08-116-23/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The latest version of the fslock crate has an optional feature that allows file-locks to exclude one another even if they are opened within the same process. On Windows this behavior comes for free; on Unix it depends on an internal map of (dev,inode) pairs. Here we upgrade to the latest version of the fslock crate, and use this feature.
* | | Use the correct unused-circuit timeout when we're learning timeouts.Nick Mathewson2021-08-105-5/+32
|/ /