It is still broken since we apply attributes before we compile code.
Recall that attributes such as `@[export]` and `@[extern]` must be applied before we
compile code.
On the other hand, any attribute `attrName`
```
@[attrName] def foo := ...
```
which creates auxiliary definitions that depend on `foo` must be applied
AFTER we generate code for `foo`. Otherwise, we will fail to compile the
auxiliary definition since we don't have code for `foo` yet.
I will fix the issue above by allowing attributes to specify when they
should be applied. I will start with only two options: before and after
code compilation. In the future, we may need more options (e.g., before
elaboration), but I don't see the need yet.
cc @kha
Main differences with respect to `tests/playground/parser`
1- There is a single (parametric) parser type: `Parser k`, where `k` is
used to identify whether it is a `nud` or `led` parser.
2- It assumes parsing tables are stored in the `Environment`.
3- We check precedence mismatch, and use the value `none` to represent
"use existing precedence".
4- We have support for silent (aka epsilon) parsing actions.
Remark: the experiments at `tests/playground/parser` demonstrated that the new
parsing infrastructure is at least 10x faster than the one based on the
`Parsec` monad.
We are going to start making drastic changes in the parser,
elaborator, attributes, etc. Examples:
- No View objects. I am going to implement match_syntax.
- No RecT in the parser. I am going to implement parser extensions
using an approach similar to the one I used to implement environment
extensions.
- No Parsec. I will use an approach similar to the one used in the
experiment https://github.com/leanprover/lean4/tree/master/tests/playground/parser
It is easier to perform these changes with the new frontend disabled.
I will slowly re-active it as I apply the changes.
cc @kha
The new `partial def`s allow us to define `fix` in Lean, but the Lean
implementation is not as efficient as the native one. The native one
in C++ use weak pointers to prevent a closure allocation at every
recursive invocation.
This commit also fixes the `fixCore` helper functions that were broken
after we switched to camelCase.
We have updated the test `fix1.lean` to demonstrate the native
implementation is faster. Here are the numbers on my desktop.
```
./run.sh fix1.lean 24
721420279
Time for 'native fix': 816ms
721420279
Time for 'fix in lean': 1.34s
```
@kha: I initially planned to use the UTF8 API only in very special
cases, but I found them to be super useful. They allow us to implement
an efficient String library mostly in Lean.
However, the there was a problem: `abbrev String.Pos := USize`.
This definition is fine for a low level API, but this is not the case
anymore. By having `String.Pos := USize`, we will not be able to
prove natural theorems for the `String` API. For example,
`String.map id s = s` did not hold. We would have to include the
artificial antecedent `s.length <= usizeMax` (or something like this).
I suspect it would be very painful.
So, this commit defines `String.Pos` as `Nat`. The performance
overhead seems to be very small.
The `offset` field is problematic because it prevents us from having an
efficient way of moving back and forth between `String.Pos` and `String.Iterator`.
@kha I temporarily added `String.OldIterator` for making sure the
parser doesn't break. This is a temporary fix that will be eliminated
after we replace `parsec`.
@kha It will be awesome to automate the following idiom with a macro :)
```
def defIndent := 4
def getIndent (o : Options) : Nat := o.get `format.indent defIndent
@[init] def indentOption : IO Unit :=
registerOption `format.indent { defValue := defIndent, group := "format", descr := "indentation" }
```
