Closes#311
Similar to how `bat` does it, we now have a global struct which holds
all the queried system directories. We prefer the XDG directories, but
use the "native" directories as a fallback. This should ensure that when
MacOs users upgrade from an older version, their existing config and
cache are still used. After deleting the old directories, the new ones
should be used by tealdeer automatically.
This is useful when the cache is managed externally, e.g. provisioned
from a Nix store path or by a package manager, where the directory's
mtime doesn't reflect the cache's real age, causing a spurious warning
on every invocation. Today the only way to silence it is `--quiet`,
which must be passed every call and hides all other output too.
Allows directories in the configuration file (cache & custom pages) to
be relative to the user's home directory by expanding the common `~`
path prefix notation. Works for the current user only (`~diferentUser/`
syntax is not supported, and will cause an error if attempted).
Works for Linux/Unix (via `HOME` env var) and Windows (via `USERPROFILE`
env var).
Examples (assuming a user called "foo" on a Linux system):
```
~/my/custom-pages # /home/foo/custom-pages
~ # /home/foo
~bar/cache # error
```
Closes#473
This regression was introduced in
593e9309b9 (#414) and leads to a panic
when trying to display pages where characters line up like in the issue
or the test. I checked other places and found that a similar panic could
occur when parsing language strings, so I added code to ignore them
instead.
- Add regression test
- Fix prefix check
- Skip non-ASCII locales
Closes#453
- Add note about changing tls_backend setting
- impl Display for TlsBackend
- Remove trailing slash in default archive source to make URL in error
look nicer
Closes#402
This adds special handling for escaped placeholders as required by the
client spec. The added tests include examples from current pages that
rely on this behavior.
The text replacements use `str::replace` which constructs a new
allocated `String`. I considered using a custom `replace_inplace` method
on `&mut str` (which works if the replacement string is at most as long
as the pattern to be replaced), but decided against it because I think
that the performance improvement is not significant enough to justify
adding `unsafe` code. It is also possible to avoid `unsafe` by
re-checking UTF-8 validity after all modifications, but the code still
felt a bit out of place for tealdeer. We can always add these
optimizations later if we want to.
For #251, I want to use the `Language<'_>` type inside `Config`. The languages will either refer to values read from the config file, or to static strings from `get_languages_from_env`, so just using `Language<'static>` is not an option.
Instead, some input for the `Config` needs to be persisted in the main function for the duration of the program so that the config can reference it. At first I was hoping that this input would be the `contents` string from `RawConfig::load`, but as it turns out you cannot (in general) deserialize strings from toml without having to alter them, for example when they contain escapes like `\n`. Thus, the toml parser seemingly doesn't even try and just throws an error when deserializing into a borrowed string (even if it could in theory just return the correct substring from the input). Given that `RawConfig` should stay static then, the raw config itself is the next best thing to keep alive and have the config reference into.
While this change might seem a bit drastic for little benefit, I am actually pretty happy with it because I want to unify the configuration anyways at some point so that the CLI arguments, environment variables, and the config file are merged at the beginning of the program and then only a single config is used for the everything (no more `enable_styles` everywhere!). At this time, the `Config` would have references into `Cli` anyways, and having the `ConfigLoader` as an entity for this merging also seems natural.
* Remove unneeded Clone bound on Dedup
* Introduce Language struct to represent language strings
* Move language directory name logic into own method on Language type
Currently, `main` uses `print_error` and `process::exit` in several
places. These could be unified by propagating errors via `Result`s.
There is one place left that manually calls `print_error`, because the
error should be followed by help instructions printed without any
highlighting.
- Adapt to `thing.paint(style)` API; was `style.paint(thing)`
- Remove `yansi::Paint::enable_windows_ascii()` in style usage decision;
removed in yansi commit b186eb5bfb, which introduced "automatic"
support for Windows: "If support is not available, styling is disabled
and no styling sequences are emitted", fitting the `Auto` option
- Respect `--color=always` even if we know it won't work
---------
Co-authored-by: Niklas Mohrin <dev@niklasmohrin.de>
* Change custom page files to use a `.md` extension.
* Add a breaking change warning to the documentation for the usage of custom pages.
* Add a temporary check for old custom pages to assist in the migration to the new naming convention.
Co-authored-by: Danilo Bargen <mail@dbrgn.ch>
Right now we only support building with Rustls. However, there are quite
a few architectures that aren't yet supported by "ring" (the crypto
library used by Rustls), for example MIPS, PowerPC or SPARC.
To offer an alternative, I added the "native-tls" feature that can be
used instead of "native-roots". When used, the native TLS stack is used
instead of Rustls (i.e. SChannel on Windows, Secure Transport on macOS
and OpenSSL otherwise).