This PR introduces a global `--offline` flag.
## Doctor
This replaces the `--offline` flag on the doctor command.
## Juvix package builds
The flag applies to juvix build commands like `juvix compile`, `juvix
repl`. This is so that users can continue to build packages offline that
have external dependencies when there's no network connection (as long
as they built the same package online previously).
Specifically, when the `--offline` flag is used in a package that has
external git dependencies.
* No `git clone` or `git fetch` commands are used
* `git checkout` will continue to be used
* Clones from previous builds are reused
This means that you can update the `ref` field in a git dependency, as
long as the ref existed the last time that the project was built without
the `--offline` flag.
* Closes https://github.com/anoma/juvix/issues/2333
This PR adds external git dependency support to the Juvix package
format.
## New dependency Git item
You can now add a `git` block to the dependencies list:
```yaml
name: HelloWorld
main: HelloWorld.juvix
dependencies:
- .juvix-build/stdlib
- git:
url: https://my.git.repo
name: myGitRepo
ref: main
version: 0.1.0
```
Git block required fields:
* `url`: The URL of the git repository
* `ref`: The git reference that should be checked out
* `name`: The name for the dependency. This is used to name the
directory of the clone, it is required. Perhaps we could come up with a
way to automatically name the clone directory. Current ideas are to
somehow encode the URL / ref combination or use a UUID. However there's
some value in having the clone directory named in a friendly way.
NB:
* The values of the `name` fields must be unique among the git blocks in
the dependencies list.
## Behaviour
When dependencies for a package are registered, at the beginning of the
compiler pipeline, all remote dependencies are processed:
1. If it doesn't already exist, the remote dependency is cloned to
`.juvix-build/deps/$name`
2. `git fetch` is run in the clone
3. `git checkout` at the specified `ref` is run in the clone
The clone is then processed by the PathResolver in the same way as path
dependencies.
NB:
* Remote dependencies of transitive dependencies are also processed.
* The `git fetch` step is required for the case where the remote is
updated. In this case we want the user to be able to update the `ref`
field.
## Errors
1. Missing fields in the Git dependency block are YAML parse errors
2. Duplicate `name` values in the dependencies list is an error thrown
when the package file is processed
3. The `ref` doesn't exist in the clone or the clone directory is
otherwise corrupt. An error with a suggestion to `juvix clean` is given.
The package file path is used as the location in the error message.
4. Other `git` command errors (command not found, etc.), a more verbose
error is given with the arguments that were passed to the git command.
## Future work
1. Add an offline mode
2. Add a lock file mechanism that resolves branch/tag git refs to commit
hashes
* closes https://github.com/anoma/juvix/issues/2083
---------
Co-authored-by: Jan Mas Rovira <janmasrovira@gmail.com>
- Closes#2293.
- Closes#2319
I've added an effect for termination. It keeps track of which functions
failed the termination checker, which is run just after translating to
Internal. During typechecking, non-terminating functions are not
normalized. After typechecking, if there is at least one function which
failed the termination checker, an error is reported.
Additionally, we now properly check for termination of functions defined
in a let expression in the repl.
- Closes#2188.
This pr introduces a new syntactical statement for defining aliases:
```
syntax alias newName := oldName;
```
where `oldName` can be any name in the expression namespace. Fixity and
module aliases are not supported at the moment.
- The `newName` does not inherit the fixity of `oldName`. We have agreed
that the goal is to inherit the fixity of `oldName` except if `newName`
has a fixity statement, but this will be done in a separate pr as it
requires #2310.
Stack LTS 21.6 uses GHC 9.4.5, binaries for HLS are available via ghcup.
Changes required:
1. Fix warnings about type level `:` and `[]` used without backticks.
2. Fix warnings about deprecation of builtin `~` - replaced with `import
Data.Type.Equality ( type (~) )` in the Prelude
3. SemVer is no longer a monoid
4. `path-io` now contains the `AnyPath` instances we were defining
(thanks to Jan) so they can be removed.
5. Added `aeson-better-errors-0.9.1.1` as an extra-dep. The reason it is
not part of the resolver is only because it has a strict bound on base
which is not compatible with ghc 9.4.5. To work around this I've set:
```
allow-newer: true
allow-newer-deps:
- aeson-better-errors
```
which relaxed the upper constraint bounds for `aeson-better-errors`
only. When the base constraints have been updated we can remove this
workaround.
6. Use stack2cabal to generate the cabal.project file and to freeze
dependency versions.
https://www.stackage.org/lts-21.6/cabal.config now contains the
constraint `haskeline installed`, which means that the version of
haskeline that is globally installed with GHC 9.4.5 will be used, see:
* https://github.com/commercialhaskell/stackage/issues/7002
GHC 9.4.5 comes with haskeline 0.8.2 preinstalled but our configuration
contains the source-repository-package for haskeline 0.8.2.1 (required
because we're using a fork) so if you try to run` cabal build` you get a
conflict.
Constraints from cabal imports cannot yet be overridden so it's not
possible to get rid of this conflict using the import method. So we need
to use stack2cabal with an explicit freeze file instead.
7. Remove `runTempFilePure` as this is unused and depends on
`Polysemy.Fresh` in `polysemy-zoo` which is not available in the
resolver. It turns out that it's not possible to use the `Fresh` effect
in a pure context anyway, so it was not possible to use
`runTempFilePure` for its original purpose.
8. We now use https://github.com/benz0li/ghc-musl as the base container
for static linux builds, this means we don't need to maintain our own
Docker container for this purpose.
9. The PR for the nightly builds is ready
https://github.com/anoma/juvix-nightly-builds/pull/2, it should be
merged as soon as this PR is merged.
Thanks to @benz0li for maintaining https://github.com/benz0li/ghc-musl
and (along with @TravisCardwell) for help with building the static
binary.
* Closes https://github.com/anoma/juvix/issues/2166
- Closes#2258
# Overview
When we define a type with a single constructor and one ore more fields,
a local module is generated with the same name as the inductive type.
This module contains a projection for every field. Projections can be
used as any other function.
E.g. If we have
```
type Pair (A B : Type) := mkPair {
fst : A;
snd : B;
};
```
Then we generate
```
module Pair;
fst {A B : Type} : Pair A B -> A
| (mkPair a b) := a;
snd : {A B : Type} : Pair A B -> B
| (mkPair a b) := b;
end;
```
GEB 0.3.2 introduces the following changes.
* The STLC frontend no longer requires full type information in terms.
The syntax of the terms changed.
* An error node has been introduced which allows to compile Juvix `fail`
nodes.
The following features required for compilation from Juvix are still
missing in GEB.
* Modular arithmetic types ([GEB issue
#61](https://github.com/anoma/geb/issues/61)).
* Functor/algebra iteration to implement bounded inductive types ([GEB
issue #62](https://github.com/anoma/geb/issues/62)).
- Closes#2060
- Closes#2189
- This pr adds support for the syntax described in #2189. It does not
drop support for the old syntax.
It is possible to automatically translate juvix files to the new syntax
by using the formatter with the `--new-function-syntax` flag. E.g.
```
juvix format --in-place --new-function-syntax
```
# Syntax changes
Type signatures follow this pattern:
```
f (a1 : Expr) .. (an : Expr) : Expr
```
where each `ai` is a non-empty list of symbols. Braces are used instead
of parentheses when the argument is implicit.
Then, we have these variants:
1. Simple body. After the signature we have `:= Expr;`.
2. Clauses. The function signature is followed by a non-empty sequence
of clauses. Each clause has the form:
```
| atomPat .. atomPat := Expr
```
# Mutual recursion
Now identifiers **do not need to be defined before they are used**,
making it possible to define mutually recursive functions/types without
any special syntax.
There are some exceptions to this. We cannot forward reference a symbol
`f` in some statement `s` if between `s` and the definition of `f` there
is one of the following statements:
1. Local module
2. Import statement
3. Open statement
I think it should be possible to drop the restriction for local modules
and import statements
When `juvix format` is invoked from some directory within a juvix
project then the formatter is run on all the files contained in the
project.
If `juvix format` is run from some directory outside of a Juvix project
then an error is reported. The user gets the same error as they would
get if
`juvix format` was run with a directory argument that is not within a
Juvix project.
* Closes https://github.com/anoma/juvix/issues/2087
This PR was already merged in https://github.com/anoma/juvix/pull/2173,
but main was subsequently forced pushed as part of the 0.4.0 release and
these changes were erased by mistake.
This PR changes the behaviour of the formatter when run on files that
are already formatted. Previously the source of a file that was already
formatted was not output by the formatter.
After this PR, the formatter always outputs the contents of a formatted
file (when used on a single file, and if the --check option is not
specified).
If the `format: false` pragma is set then the source is echoed verbatim,
without highlighting (because it's not possible to get the highlighting
without the formatting).
This probably helps implementing the formatter in the vscode extension,
see https://github.com/anoma/vscode-juvix/issues/98
`format` command now returns code `0` most of the time.
It will return `1` when:
* some error occur, so can not format
* file is unformatted and `--check` option is used
* One or more files are not formatted in a Juvix project.
- Fixes#2171
Say we have a module that import/open the Prelude:
Test.juvix
```
module Test;
import Stdlib.Prelude open;
```
When the module is compiled, we have a step in the compiler pipeline
which filters out unreachable symbols. For this module all symbols are
filtered because the module contains no definitions.
So if the module is loaded in the REPL, no symbols will be available to
process through the evaluator. The REPL is a place to explore the
symbols in the module so (like with Haskell's GHCi) it would be useful
if all symbols were available in the REPL session. That's what this PR
implements.
* Closes https://github.com/anoma/juvix/issues/2159
---------
Co-authored-by: Jan Mas Rovira <janmasrovira@gmail.com>
- Closes#2128
- Closes#2161
This pr fully implements the monadic pretty printer based on
`ExactPrint`, which respects comments. Until now, comments inside
expressions were printed after the current statement. Now they are
printed in the correct place, except when a comment occurs before
something that we don't store its location. E.g. parentheses,
semicolons, braces, colons, etc. I proposed that we irone out this issue
in a separate pr.
Since the old non-monadic algorithm is no longer necessary, I removed
it.
* Closes#2134
Adds the `argnames` pragma which specifies function argument names.
These will be the names used in Core and subsequently in VampIR for the
`main` function.
```
{-# argnames: [x, y] -#}
main : Nat -> Nat -> Nat;
```
If set, `JUVIX_LLVM_DIST_PATH` should point to the root of a LLVM
installation, i.e clang should be present
in`$JUVIX_LLVM_DIST_PATH`/bin/clang.
If `JUVIX_LLVM_DIST_PATH` is not set, or `clang` is not available there
then the system PATH is used instead, (this is the current behaviour).
The `juvix doctor` clang checks use the same logic as `juvix compile` to
find and check the `clang` executable.
To help with debugging the clang location, this PR also adds `juvix
doctor --verbose` which prints the location of the `clang` executable
and whether it was found using the system PATH or the
JUVIX_LLVM_DIST_PATH environment variable:
```
juvix doctor --verbose
> Checking for clang...
| Found clang at "/Users/paul/.local/share/juvix/llvmbox/bin/clang" using JUVIX_LLVM_DIST_PATH environment variable
```
or
```
juvix doctor --verbose
> Checking for clang...
| Found clang at "/Users/paul/.local/bin/clang" using system PATH
```
* Closes https://github.com/anoma/juvix/issues/2133
`AnsiText` is a type that represents some text that can be printed with
`Ansi` formatting annotations, or as plain text. It is expected that it
should have a `Semigroup` instance. This pr adds that.
---------
Co-authored-by: Paul Cadman <git@paulcadman.dev>
This pr adds a new command, `:def` to the repl. This command expects a
single identifier that must be in scope and then prints its definition.
For constructors, the whole type definition is printed.
It also applies some refactors to the code for repl command.
1. Before there was a mega `where` block of definitions. I have hoisted
most of the definitions there to the top level. I feel like now it is
easier to navigate and read.
2. Use `ExceptT` instead of local `case` expressions for errors.
3. Use forks of `haskeline` and `repline`. These forks are necessary
because these libraries do not export the constructors `HaskelineT` and
`InputT` respectively, thus, making it impossible to catch errors in
their underlying monad.
- Closes#2067
This pr adds the field `main` to `juvix.yaml`. This field is optional
and should contain a path to a juvix file that is meant to be used for
the `compile` (and `dev compile`) command when no file is given as an
argument in the CLI. This makes it possible to simply run `juvix
compile` if the `main` is specified in the `jvuix.yaml`.
I have updated the `juvix.yaml` of the milestone examples.
---------
Co-authored-by: Paul Cadman <git@paulcadman.dev>
Co-authored-by: Jonathan Cubides <jonathan.cubides@uib.no>
Currently generated links for fixing errors in the `juvix doctor`
command are broken.
Fixing that by updating the base Url link.
Also fixed the link to the installation of juvix in the contributing
guide.
Instead, always act as `--no-format` option is set to `False` as
previous default.
The change seem to not affect any current formatting, so I assume it
passes the checks on testing.
Fixes#2084
# Checklist:
- [x] My code follows the style guidelines of this project
- [ ] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
In this PR we pass through the `juvix compile` optimization flag to the
C compiler in the native compilation.
NB: Clang supports -On for any positive n. -O4 and higher is equivalent
to -O3
Also we disable optimizations when the `-g` / `--debug` option is
specified.
* Closes https://github.com/anoma/juvix/issues/2104
---------
Co-authored-by: Lukasz Czajka <lukasz@heliax.dev>
* Closes#2034.
* Adds the `vampir` target to the `compile` command.
* Adds two tests which are not yet enabled because `vamp-ir` is not
available in the CI (these and more tests will be enabled in #2103).
The prettyprinter library takes care avoid adding whitespace to empty
lines when it is rendering indented text.
See:
7e32c010ec/prettyprinter/src/Prettyprinter/Internal.hs (L1999)
However it only does this for unannotated text.
In our code we were stripping annotations from renderings within
`toTextStream` but we must remove the annotations before calling
`layoutPretty` to get the proper handling of whitespace with
indentations. That's what this PR does.
* Closes#2032.
* Adds the `juvix dev core normalize` command.
* Adds the `:n` command in JuvixCore REPL.
* Adds the `--normalize` flag to `juvix dev core read` and `juvix dev
core from-concrete`.
* Adds `pipeline-normalize` which denotes pipeline steps necessary
before normalization.
* Adds normalization tests in `tests/VampIR/positive/Core`.
* Closes#1989
* Adds optimization phases to the pipline (specified by
`opt-phase-eval`, `opt-phase-exec` and `opt-phase-geb` transformations).
* Adds the `-O` option to the `compile` command to specify the
optimization level.
* Functions can be declared for inlining with the `inline` pragma:
```
{-# inline: true #-}
const : {A B : Type} -> A -> B -> A;
const x _ := x;
```
By default, the function is inlined only if it's fully applied. One can
specify that a function (partially) applied to at least `n` explicit
arguments should be inlined.
```
{-# inline: 2 #-}
compose : {A B C : Type} -> (B -> C) -> (A -> B) -> A -> C;
compose f g x := f (g x);
```
Then `compose f g` will be inlined, even though it's not fully applied.
But `compose f` won't be inlined.
* Non-recursive fully applied functions are automatically inlined if the
height of the body term does not exceed the inlining depth limit, which
can be specified with the `--inline` option to the `compile` command.
* The pragma `inline: false` disables automatic inlining on a
per-function basis.
This PR adds support for importing modules from within a Juvix project
in the Juvix REPL.
The imported module is checked (parsed, arity-checked, type-checked etc)
as normal and added to the REPL session scope. Any errors during the
checking phase is reported to the user.
### Notes:
* You must load a file before using `import`. This is because the REPL
needs to know which Juvix project is active.
* You may only import modules from within the same Juvix project.
### Examples
After launching `juvix repl`:
#### `open import`
```
Stdlib.Prelude> open import Stdlib.Data.Int.Ord
Stdlib.Prelude> 1 == 1
true
```
#### `import as`
```
Stdlib.Prelude> import Stdlib.Data.Int.Ord as Int
Stdlib.Prelude> 1 Int.== 1
true
```
#### `import`then `open`
```
Stdlib.Prelude> import Stdlib.Data.Int.Ord as Int
Stdlib.Prelude> open Int
Stdlib.Prelude> 1 == 1
true
```
#### Line-terminating semicolons are ignored:
```
Stdlib.Prelude> import Stdlib.Data.Int.Ord as Int;;;;;
Stdlib.Prelude> 1 Int.== 1
true
```
* Closes https://github.com/anoma/juvix/issues/1951
---------
Co-authored-by: Jonathan Cubides <jonathan.cubides@uib.no>
This pr fixes a bug where the repl would crash if it had the implicit
stdlib dependency and the .juvix-build/stdlib directory did not yet
exist. This bug was not exposed in the smoke tests because the
.juvix-build was never cleared.
---------
Co-authored-by: Paul Cadman <git@paulcadman.dev>
# Description
No the filepath in the `juvix forma` command is n=made optional.
However, in that case, the `--stdin` command is required.
### Implementation details
~For now, as a quick solution, I have introduce the "fake" path that is
used for `fomat` command with stdin option.~
I needed to do a couple of big changes:
* `format` command FILE is now optional, howvere, I check that in case
of `Nothing` `--stdin` option should be present, otherwise it will fail
* `entryPointModulePaths` is now `[]` instead of `NonEmpty`
* `ScopeEff` now has `ScopeStdin` constructor as well, which would take
the input from stdin instead of having path passed around
* `RunPipelineNoFileEither` is added to the `App` with the bunch of
`*Stdin` functions that doesn't require filepath argument to be passed
Fixes#2008
## Type of change
- [x] New feature (non-breaking change which adds functionality)
# Checklist:
- [x] My code follows the style guidelines of this project
- [x] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [x] I have added tests that prove my fix is effective or that my
feature works:
- [x] smoke tests
---------
Co-authored-by: Paul Cadman <git@paulcadman.dev>
Co-authored-by: janmasrovira <janmasrovira@gmail.com>
* Closes#1965
* Implements the `unroll` pragma to control the unrolling depth on a
per-function basis.
* Implements parsing of the `inline` pragma.
---------
Co-authored-by: janmasrovira <janmasrovira@gmail.com>
This PR adds the `juvix clean` command to the CLI that removes the Juvix
project build directory.
It respects the `--internal-build-dir` global option:
```
$ juvix compile Foo.juvix --internal-build-dir /tmp/build
$ juvix clean --internal-build-dir /tmp/build
```
In addition this PR fixes the `juvix format` program brief description
string. This was too long for the `juvix --help` display. The longer
description is now only displayed when `juvix format --help` is run.
* Closes https://github.com/anoma/juvix/issues/2017
- Closes#1993
This pr makes it possible to use `~`, `..` and environment variables in
the `juvix.yaml` and all flags / input of the cli.
In the CLI, the shell will be responsible for replacing environment
variables with their value, so the usual syntax can be used. For the
`dependencies` field, I have implemented a parser that has some
restrictions:
1. Environment variables are given with the makefile-like syntax
`$(VAR)`
2. The three characters `$` `(` `)` are reserved for the environment
variables syntax.
They cannot be part of the path.
3. `~` is reserved for `$(HOME)`. I.e. the prepath `~~` will expand to
`$HOME$HOME`.
4. Nested environment variables are not allowed.
Thanks @paulcadman for the feedback. I think we are ready to merge this
nightmarish pr 👻
---------
Co-authored-by: Paul Cadman <git@paulcadman.dev>
This PR implements pretty printing of evaluation results consistently
with Juvix syntax. The printed values do not necessarily originate
directly from the source code. All functions/lambdas are printed as
`<fun>`.
The same mechanism is used to implement pretty printing of unmatched
pattern examples.
Juvix REPL now uses the new value printing mechanism to display
evaluation results. Typing `nil` in the REPL will now just display
`nil`. The command `juvix dev repl` still prints raw JuvixCore terms.
* Closes#1957
* Closes#1985
