Commit Graph

139 Commits (57058e5c0f70787e4ea682e557a9341b1a023934)

Author SHA1 Message Date
_ 9ec89a17e2 Merge branch 'fix-windows' 2024-01-21 19:11:14 -06:00
_ 23e4835933 🏁 fix Windows build 2023-10-30 21:13:07 -05:00
_ 3639862c04 🔊 fix deprecation warnings from new deps 2023-08-31 20:30:23 -05:00
_ 843fddb568 🚨 cargo check pass 2023-08-31 19:15:04 -05:00
(on company time) 43fe373847 Merge branch 'main' of https://six-five-six-four.com/git/reactor/ptth into main 2023-08-29 08:53:37 -05:00
(on company time) 0d6290ed60 🐛 bug: fix a config bug
In ptth_multi_call_server, the file server root was "" instead of ".".
This happened in 134035f due to a bad merge.

I'm not sure how to add a test for it.
2023-01-26 15:52:44 -06:00
_ 82cef9cfd6 📝 docs: add the new PTTH repo on self-hosted Gitea to all Cargo.toml files 2023-01-07 00:10:11 +00:00
_ a838d80d38 🐛 bug: also hack in content-type for HTML pages 2022-12-31 03:06:23 +00:00
_ 68932e00de Merge remote-tracking branch 'external/main' 2022-05-26 01:00:02 +00:00
_ ed660f3432 add `index_directories` 2022-04-29 01:07:00 +00:00
_ 820c5aeef4 bring tests up to date
I definitely remember doing this already. I must have left the changes on
my home PC.
2022-04-22 11:37:20 -05:00
_ da581f452d remove that stupid bad passwords file 2022-04-20 12:31:35 +00:00
_ c2299ab39a fix up the tests 2022-04-20 12:25:44 +00:00
_ 734602244d Merge /run/media/user/Elements/file-jar/ptth 2022-04-12 01:29:46 +00:00
_ b69770870c 🚧 wip: chasing the 30-second download timeout issue 2022-04-11 17:25:18 -05:00
_ 8c731c4cef 🐛 bug: looks like the memory leak for large files might have been in Tokio or reqwest. 2022-04-11 16:56:59 -05:00
_ 4c8521c50a 🐛 bug: fix Content-Length header 2022-04-09 20:47:46 +00:00
Trisha 800dbcb019 add hit counter 2022-03-25 16:39:26 -05:00
Trisha fb9b0c67f5 add multi-root 2022-03-25 16:14:22 -05:00
Trisha 436adb98ef 🚧 swich to the new routing func 2022-03-25 16:04:51 -05:00
Trisha de5338f4f2 ♻️ refactor: extract routing func 2022-03-25 15:47:34 -05:00
Trisha e6273209f9 🚧 wip: threading the multi-root stuff into the file server module 2022-03-25 15:40:36 -05:00
Trisha 81141e2faf add `file_server_roots` config 2022-03-25 15:30:38 -05:00
_ c852efb343 add subcommand to generate a random ULID 2022-03-24 23:24:23 +00:00
_ 931ad6d81b 🔇 change logging a little 2022-03-24 22:45:19 +00:00
_ 80bd4abad3 🔊 change logs a bit 2022-02-06 20:03:27 +00:00
_ 6ea8b5c30f 🐛 bug: fix Cache-Control header so Nginx can cache my blog 2022-02-05 17:28:30 +00:00
_ e5d157e1af 🐛 bug: don't send Content-Length to relay if status code is 304 Not Modified
Thanks to Nginx for noticing this
2022-02-05 17:00:17 +00:00
_ 2656d16264 ♻️ refactor: extract `http_listen` fn 2022-02-05 16:52:56 +00:00
_ d96bf801c6 🔇 make logs quieter 2022-02-05 15:57:40 +00:00
_ c47ab6d282 🚧 wip: messing with cache parameters 2022-01-09 00:22:16 +00:00
_ b642142332 🐛 bug: fix content length when byte serving 2021-11-13 21:31:30 -06:00
_ 16aaac0965 🐛 bug: used the wrong PathBuf function in resolving the merge 2021-10-14 18:51:12 -05:00
_ 134035f198 Merge remote-tracking branch 'origin/main' 2021-10-14 18:35:31 -05:00
_ 98b43d1ba2 guess MIME for JavaScript files so that `ptth_file_server` can serve Jet Racing 4 locally.
Firfox has a bunch of odd security features that I only sometimes understand.
Among them is that it won't load JS files without the `content-type` header,
and it also disables a lot of features for the `file://` scheme. Which I think
is a shame. There's probably a good reason they do this, but I'm not aware of one.

So now you can use PTTH's file server to host Jet Racing 4.
2021-10-02 20:03:03 +00:00
_ 2972e85671 use rust_embed and default configs so you can call `ptth_file_server` from any working dir.
This is like Python's old `-m SimpleHTTPServer` but better because I wrote it.
2021-10-02 19:49:19 +00:00
_ f7b78b8a12 ⬆️ update to blake3 1.0.0 to de-dupe the cfg-if depend 2021-10-02 19:13:46 +00:00
_ f44613540e add the multi-call server
This exe can act as both a PTTH and PTTH_QUIC end server.
It's only 17 MB, which is a big improvement over the 15 + 15 for shipping both servers as their own exes
2021-10-02 18:13:14 +00:00
_ 30f8bbb0aa ♻️ refactor inject CLI args instead of reading them in main 2021-10-02 17:31:06 +00:00
_ 0b10737403 ♻️ refactor: move ptth_server's main into its library so I can make a busybox-style multi-use binary 2021-10-02 17:26:56 +00:00
_ 1abf1d194c 🚧 maybe HTTP Basic Auth? 2021-08-29 20:56:32 -05:00
_ 5376e8bba0 🚧 wip: add placeholder for client tokens 2021-08-29 20:17:25 -05:00
_ 324c1f7cd6 ♻️ refactor: remove pointless `Option <>` for file server root 2021-08-29 19:49:32 -05:00
_ 9b60acf0e0 📝 remove Github repo from Cargo.tomls. PTTH is no longer on my Github 2021-07-18 17:44:07 +00:00
_ 3a2147d50f use `MissedTickBehavior::Skip` on all intervals 2021-07-18 17:43:58 +00:00
_ d8c6e1bd14 ⬆️ update to Tokio 1.8.x to get a feature I want 2021-07-18 17:33:34 +00:00
_ d292693856 ⬆️ update rmp-serde to 0.15.5 and add it to ptth_quic 2021-07-18 17:19:05 +00:00
_ 02b6955a50 🐛 bug: allow `auto-gen-key` to also recreate the file and fix permissions 2021-07-13 16:02:49 -05:00
_ 2f2e4ef64d 📝 document how to run `ptth_server`
This commit also works well on Termux on Android.
2021-07-11 13:12:32 +00:00
_ 6f45f5105e remove `heim` dependency
It doesn't work on Android, and I can't fault it for that (Platform difference)
but it also drags in `uom` which has a bunch of template magic that takes
forever to compile, and I'm pretty sure it's dragging down my compile times.

I tried to build `ptth_server` on nightly to get compile timings, and heim
depends on `socket2` which failed to compile.

And I never ended up using the CPU monitoring. So heim is gone now.
2021-07-11 12:17:36 +00:00