This makes copying such types much cheaper and will allow us to
intern common struct types in the future too. This further
drops the sample cost for langtype.rs from ~6.6% down to 4.0%.
We are now also able to share/intern common struct types.
Before:
```
Time (mean ± σ): 1.073 s ± 0.021 s [User: 0.759 s, System: 0.215 s]
Range (min … max): 1.034 s … 1.105 s 10 runs
allocations: 3074261
```
After:
```
Time (mean ± σ): 1.034 s ± 0.026 s [User: 0.733 s, System: 0.201 s]
Range (min … max): 1.000 s … 1.078 s 10 runs
allocations: 2917476
```
This allows us to cheaply copy the langtype::Type values which
contain such a type. The runtime impact is small and barely noticable
but a sampling profiler shows a clear reduction in samples pointing
at langtype.rs, roughly reducing that from ~8.6% inclusive cost
down to 6.6% inclusive cost.
Furthermore, this allows us to share/intern common types.
Before:
```
Benchmark 1: ./target/release/slint-viewer ../slint-perf/app.slint
Time (mean ± σ): 1.089 s ± 0.026 s [User: 0.771 s, System: 0.216 s]
Range (min … max): 1.046 s … 1.130 s 10 runs
allocations: 3152149
```
After:
```
Time (mean ± σ): 1.073 s ± 0.021 s [User: 0.759 s, System: 0.215 s]
Range (min … max): 1.034 s … 1.105 s 10 runs
allocations: 3074261
```
- include the ' ' char that doesn't have a bounding box but need to be
included for the advance_x
- Fix an off by one in rendering where the last pixel was missing
Initially the character map is ordered by glyph index. Maintain that order for glyph embedding, so that we can safely collect() the bitmaps. Just before creating the final BitmapFont data structure, sort by code point for fast lookups at run-time.
Fix#6588
Note for the test: parse error at the end of the file are actually
reported at offset 0. Use a trick to add more `^` at the beginning to
mean that.
(Offset 0 is actually offset 1 as the column starts at 1)
Slintpad uses URLs to images. Do not fail when we "embed"
those so that we find the list of resources on the Documents
later.
This fixes image loading in slintpad again.
Do not extract the file loading into a fuinction: That keeps
`self` around which is part of a borrowed state. That is a
problem as during the load we end up in `ensure_document_loaded`
again, which will try to borrow_mut that same state.
This removes a lot of allocations and speeds up the compiler step
a bit. Sadly, this patch is very invasive as it touches a lot of
files. That said, each individual hunk is pretty trivial.
For a non-trivial real-world example, the impact is significant,
we get rid of ~29% of all allocations and improve the runtime by
about 4.8% (measured until the viewer loop would start).
Before:
```
Benchmark 1: ./target/release/slint-viewer ../slint-perf/app.slint
Time (mean ± σ): 664.2 ms ± 6.7 ms [User: 589.2 ms, System: 74.0 ms]
Range (min … max): 659.0 ms … 682.4 ms 10 runs
allocations: 4886888
temporary allocations: 857508
```
After:
```
Benchmark 1: ./target/release/slint-viewer ../slint-perf/app.slint
Time (mean ± σ): 639.5 ms ± 17.8 ms [User: 556.9 ms, System: 76.2 ms]
Range (min … max): 621.4 ms … 666.5 ms 10 runs
allocations: 3544318
temporary allocations: 495685
```
Heaptrack showed a lot of memory allocations when the strings
in the NamedReference got copied around. Instead, we now use
SmolStr to benefit from SSO. This already removes ~320000
allocations for the benchmark app I am using here. Runtime
also improves by ca 1.8%.
Before:
```
Benchmark 1: ./target/release/slint-viewer ../slint-perf/app.slint
Time (mean ± σ): 676.6 ms ± 15.6 ms [User: 594.6 ms, System: 77.4 ms]
Range (min … max): 662.9 ms … 703.4 ms 10 runs
allocations: 5202354
temporary allocations: 857511
```
After:
```
Benchmark 1: ./target/release/slint-viewer ../slint-perf/app.slint
Time (mean ± σ): 664.2 ms ± 6.7 ms [User: 589.2 ms, System: 74.0 ms]
Range (min … max): 659.0 ms … 682.4 ms 10 runs
allocations: 4886888
temporary allocations: 857508
```
Newer clang may produce warnings like this when using == with floating point numbers:
```
error: floating-point comparison is always false; constant cannot be represented exactly in type 'float' [-Werror,-Wliteral-range]
```
Use the C++ equivalent of approx_eq to avoid the warning and provide more reliable comparison.
... which will list all resources that are not going to get embedded
as `None` in the Document's `embedded_file_reosurces`.
The idea is to use that field to find all used resources in the
live preview so that we know what we can watch.
The struct held provides access to the design metrics of the font scaled
to the font pixel size used by the element.
ChangeLog: Slint Language: Added font-metrics property to `Text` and `TextInput`.
Closes#6047
