`normalize_identifier` is called quite often in the generator, once for
every variable name or path element that is written.
This PR aims to speed up the function by
* using a per-length input string length replacement map
* binary searching the replacement map instead of a linear search
Diffent, but functionally equivalent implementations were compared:
```text
* linear search in one big map: 348.44 µs
* binary search in one big map: 334.46 µs
* linear search in a per-length map: 178.84 µs
* binary search in a per-length map: 154.54 µs
* perfect hashing: 170.87 µs
```
The winner of this competition is "binary search in a per-length map".
It does not introduce new dependencies, but has the slight disadvantage
that it uses one instance of `unsafe` code. I deem this disadvantage
acceptable, though.
Nb. It was also tested if a variant that only stores the replaced string
would be faster. This "optimization" proved to be slower for all
implementations except "binary search in a per-length map", for which it
has the same runtime. Without a clear advantage to use the "optimized
version", I chose to use the more easy to read slice of tuples variant.
Obviously, for all measurements: YMMV.
This commit solves issue #719.
This is done by making the markdown filter borrow the string and
simplifying the filter to accept `&str` instead of `AsRef<str>`
Add test for the markdown filter using as input
Revert markdown filter changes
Revert unnecessary changes
Improve test_markdown_owned_string test
Use cargo fmt
This change
* allows using empty arrays `[]` in expessions,
* adds a cut when the leading `[` was encountered, and
* fixes the interaction between arrays and boolean OR.
IMO the restriction that you couldn't use empty arrays is not needed.
The missing cut made error messages slictly worse if you forget to add
the closing `]`.
Filter expressions must not have white spaces before the pipe `|`. The
white space is used to tell a filter expressions, and `std::ops::Or`
apart.
This commit introduces a shorthand for defining and calling macros when
using them as a reusable substitute for variables assigned complex values
(e.g. string literals with or without newline escapes). The use-case is
formatting - from my experience it's easier to visually parse a `macro`
`endmacro` block than a multiline variable assignment.
Signed-off-by: mataha <mataha@users.noreply.github.com>
In <https://github.com/dtolnay/trybuild/pull/219> the output of error
messages was subtly changed, because they introduced a subdirectoy in
their temp path.
This PR fixes the mismatch between the expected and the actual output.
I was going through the book when I saw the version was still 0.8.
This commit ensures all references to Askama version are to 0.11.2.
Most of them already were.
Merge whenever appropriate for release, #722 related
The support for the magic `_parent` field is deprecated since v0.8.0
or issue #180. It's bothersome to keep this feature alive, when no-one
should be using it for 3 years.
Just migrated a repo from tera to askama and this was one of the only
things that was different. This is also coherent with `{% block %}` for
which I added the same feature years ago.
