| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
| |
Now everyone will know where to get my software if they see it.
|
| |
|
|
|
|
|
|
|
| |
These features share some code since they both require fetching reply
contexts, so it makes sense to implement them together.
TODO cover webmention sending with integration tests
|
| |
|
| |
|
| |
|
|
|
|
| |
Now I will know if something breaks horribly again.
|
|
|
|
|
|
| |
Iterator::skip_while() returns the last item. Reimplement the
combinator that I need using a loop over Iterator::by_ref()
instead. This will terminate after the end is reached.
|
| |
|
|
|
|
| |
Closes #4.
|
|
|
|
|
|
|
|
| |
This will ease future extraction of the media endpoint to a separate
crate. This is highly desirable since it will allow Kittybox's media
endpoint to be used separately in instances where a standalone media
endpoint is desirable (e.g. custom solutions using my code to polyfill
for desired functionality that is undesirable to implement by oneself)
|
| |
|
|
|
|
|
| |
Match blocks and ifs are actually perfectly usable as expressions. I
forgot about that when writing that code.
|
|
|
|
|
|
|
|
|
| |
Templates and utility types are now separate crates to speed up
compilation, linting and potential reuse/replacement.
Potentially more crates could be split out/modularized, resulting in
speedups, smaller binaries (whenever features are excluded) and even
more reuse capabilities.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Reads don't lock anymore. At all.
- Writes create a temporary file and use `rename(2)` to atomically
replace it
- since OpenOptions::create_new(true) is used, tempfile creation is
atomic (and since tempfile names are per-post, a post can only be
edited by one request at a time)
- Since written files get atomically replaced, readers can't read a
corrupted file
Potential pitfalls:
1. This approach is not covered by unit tests (yet)
2. Stale tempfiles can prevent editing posts (can be solved by
throwing out tempfiles that are older than, say, a day)
3. Crashed edits can leave stale tempfiles (honestly that sounds
better than corrupting the whole database, doesn't sound like a bug to
me at all!)
|
| |
|
|
|
|
| |
It's an advanced HTTP client that can do more than vanilla Hyper does.
|
|
|
|
|
| |
Sometimes it's cumbersome to include them, and I plan to make them
embedded in the app anyway. This is my reminder to do it ASAP.
|
|
|
|
|
|
| |
Now Kittybox can gracefully shutdown on SIGTERM. Nice!
TODO: consider shutting down on multiple signals
|
| |
|
|
|
|
|
|
| |
This stubs the neccesary code with enough stuff that it will work and
be accepted by most compliant Micropub implementations. Later, this
can be extended when the neccesary amendments and refactors are done.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
Warp is using hyperium/http instead of http-types, so I replaced all
of the http-types usage (mostly status codes) by Warp's
http::StatusCode.
Additionally some of the struct fields were made public to allow
initialization from public code.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
This adds the ability to use mocks that don't actually touch the
network and alternative transports such as using OpenSSL instead of
rustls (but rustls is still superior).
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
require_token() uses a token endpoint URI and an HTTP client to query
the token endpoint and return a User object if the user was
authorized, or rejecting with IndieAuthError if not. It is recommended
to use recover() and catch the IndieAuthError at the application level
to show a "not authorized" error message to the user.
This function is more intended for API consumption, but is general
enough to permit using in other scenarios.
TODO: make a variant that returns Option<User> instead of rejecting
|
|
|
|
|
|
|
|
|
|
|
| |
Warp allows requests to be applied as "filters", allowing to flexibly
split up logic and have it work in a functional style, similar to
pipes.
Tokio is just an alternative runtime. I thought that maybe switching
runtimes and refactoring the code might allow me to fish out that
pesky bug with the whole application hanging after a certain amount of
requests...
|
|
|
|
|
|
| |
This is to prevent spinning in a loop waiting for a lock. This hangs
often, though I suspect this should have been fixed in the previous
commit.
|
|
|
|
|
|
|
| |
This may or may not be the cause for the app hanging while waiting for
a lock. Now the operations with locks are never performed over an
async boundary, excluding any shenanigans that can happen when
accidentally leaving a file locked over async boundaries.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|