Skip to content
This repository was archived by the owner on Oct 21, 2026. It is now read-only.
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions .github/ISSUE_TEMPLATE/alpha-feedback.md
Original file line number Diff line number Diff line change
@@ -1,20 +1,20 @@
---
name: Alpha feedback
about: Report a wrong result, a crash or a surprise from an alrunner alpha or edge build
about: Report a wrong result, a crash or a surprise from an al-runner alpha or edge build
title: ""
labels: alpha-feedback
---

A wrong result is the most useful report: a test that passes or fails in
alrunner but behaves differently on real Business Central.
al-runner but behaves differently on real Business Central.

**alrunner version** (output of `alrunner --version`)
**al-runner version** (output of `al-runner --version`)

**Operating system** (for example Windows 11 x64, macOS 15 arm64, Ubuntu 24.04)

**Command** (the full command line)

**What alrunner did** (outcome: passed, failed, not run with its reason, crash, or wrong output; paste the lines that matter)
**What al-runner did** (outcome: passed, failed, not run with its reason, crash, or wrong output; paste the lines that matter)

**What you expected**

Expand Down
12 changes: 6 additions & 6 deletions .github/workflows/build-packages.yml
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
name: Build packages

# Builds alrunner for the three alpha targets, smoke tests each binary, packages it and attests it.
# Builds al-runner for the three alpha targets, smoke tests each binary, packages it and attests it.
# Used by release-tag.yml (numbered alphas, and the dry run) and edge.yml (the rolling edge build).
# Output: the artifact "packages" holding the archives and one SHA256SUMS. Nothing of Microsoft's
# is built in or downloaded here: the smoke test runs a fixture project of this repository.
Expand Down Expand Up @@ -85,18 +85,18 @@ jobs:
- name: Smoke test
shell: bash
run: |
exe=target/${{ matrix.target }}/release/alrunner
exe=target/${{ matrix.target }}/release/al-runner
[ "${{ matrix.ext }}" = zip ] && exe="$exe.exe"
release/smoke.sh "$exe"
- name: Package
shell: bash
run: |
name="alrunner-${{ inputs.label }}-${{ matrix.target }}"
name="al-runner-${{ inputs.label }}-${{ matrix.target }}"
mkdir -p stage dist
if [ "${{ matrix.ext }}" = zip ]; then
cp "target/${{ matrix.target }}/release/alrunner.exe" stage/
cp "target/${{ matrix.target }}/release/al-runner.exe" stage/
else
cp "target/${{ matrix.target }}/release/alrunner" stage/
cp "target/${{ matrix.target }}/release/al-runner" stage/
fi
cp LICENSE stage/LICENSE
cp release/README.txt stage/README.txt
Expand Down Expand Up @@ -133,7 +133,7 @@ jobs:
- name: SHA256SUMS and install scripts
run: |
cp release/install.sh release/install.ps1 dist/
(cd dist && sha256sum alrunner-*.tar.xz alrunner-*.zip > SHA256SUMS && cat SHA256SUMS)
(cd dist && sha256sum al-runner-*.tar.xz al-runner-*.zip > SHA256SUMS && cat SHA256SUMS)
- uses: actions/upload-artifact@v4
with:
name: packages
Expand Down
6 changes: 3 additions & 3 deletions .github/workflows/edge.yml
Original file line number Diff line number Diff line change
Expand Up @@ -48,7 +48,7 @@ jobs:
# caller's event (schedule or workflow_dispatch), so this input is checked first.
CALL_SHA: ${{ inputs.sha }}
run: |
# The UTC date of this build, shown in `alrunner --version` as +edge.YYYYMMDD.
# The UTC date of this build, shown in `al-runner --version` as +edge.YYYYMMDD.
echo "version_suffix=+edge.$(date -u +%Y%m%d)" >> "$GITHUB_OUTPUT"
if [ -n "$CALL_SHA" ]; then
echo "Nightly build of $CALL_SHA."
Expand Down Expand Up @@ -135,8 +135,8 @@ jobs:
done
gh release upload edge dist/* --clobber
gh release edit edge --prerelease --latest=false --draft=false \
--title "alrunner edge ($(echo "$SHA" | cut -c1-9))" --notes-file notes.md
--title "al-runner edge ($(echo "$SHA" | cut -c1-9))" --notes-file notes.md
else
gh release create edge dist/* --prerelease --latest=false --verify-tag \
--title "alrunner edge ($(echo "$SHA" | cut -c1-9))" --notes-file notes.md
--title "al-runner edge ($(echo "$SHA" | cut -c1-9))" --notes-file notes.md
fi
14 changes: 7 additions & 7 deletions .github/workflows/release-tag.yml
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@ name: Release
# The workflow checks the tag, runs the CI checks, builds and smoke tests the three targets and
# creates a normal GitHub release (not a draft, not a pre-release) marked "latest", so that
# `mise use github:aldevtools/BusinessCentral.AL.Runner` and `mise up` find it (owner decision 2026-10-20).
# Assets: alrunner-<version>-<target>.tar.xz or .zip, SHA256SUMS, install.sh, install.ps1.
# Assets: al-runner-<version>-<target>.tar.xz or .zip, SHA256SUMS, install.sh, install.ps1.
# workflow_dispatch is a dry run: it builds and packages but creates no release.

on:
Expand Down Expand Up @@ -84,12 +84,12 @@ jobs:
VERSION: ${{ needs.plan.outputs.version }}
run: |
mkdir bin
tar -C bin -xJf "dist/alrunner-$VERSION-x86_64-unknown-linux-musl.tar.xz"
printed="$(./bin/alrunner --version)"
tar -C bin -xJf "dist/al-runner-$VERSION-x86_64-unknown-linux-musl.tar.xz"
printed="$(./bin/al-runner --version)"
echo "$printed"
case "$printed" in
"alrunner $VERSION "*|"alrunner $VERSION") ;;
*) echo "::error::alrunner --version prints '$printed', not version $VERSION of tag $GITHUB_REF_NAME"; exit 1 ;;
"al-runner $VERSION "*|"al-runner $VERSION") ;;
*) echo "::error::al-runner --version prints '$printed', not version $VERSION of tag $GITHUB_REF_NAME"; exit 1 ;;
esac
- name: Create the release
env:
Expand All @@ -98,7 +98,7 @@ jobs:
VERSION: ${{ needs.plan.outputs.version }}
run: |
cat > notes.md <<NOTES
alrunner $VERSION. An alpha: no AL Runner v2 parity, some tests are refused ("not run")
al-runner $VERSION. An alpha: no AL Runner v2 parity, some tests are refused ("not run")
with a reason, and results may change.

The binaries are not signed. See "Try the alpha" in the README for the macOS and
Expand All @@ -114,4 +114,4 @@ jobs:
sed -i 's/^ //' notes.md
# A normal release, marked latest. The commit list follows the fixed text (--generate-notes).
gh release create "$GITHUB_REF_NAME" dist/* --verify-tag --latest --generate-notes \
--title "alrunner $VERSION" --notes-file notes.md
--title "al-runner $VERSION" --notes-file notes.md
2 changes: 1 addition & 1 deletion crates/al-bcdb-std/src/provider.rs
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
//! The rows of a restored company, from a SQL Server backup (`.bak`) or a BC cloud export
//! (`.bacpac`), as `alrunner test --test-data` reads them (ADR-020 point 3).
//! (`.bacpac`), as `al-runner test --test-data` reads them (ADR-020 point 3).
//!
//! The provider does no naming of its own beyond finding the database table: a table is its AL
//! name with SQL's character substitutions, selected by company and, when two apps declare the same
Expand Down
2 changes: 1 addition & 1 deletion crates/al-deps/src/remote.rs
Original file line number Diff line number Diff line change
Expand Up @@ -142,7 +142,7 @@ pub fn http_agent() -> ureq::Agent {
let config = ureq::Agent::config_builder()
.timeout_global(Some(std::time::Duration::from_secs(300)))
.timeout_connect(Some(std::time::Duration::from_secs(20)))
.user_agent(concat!("al-runtime-rs/", env!("CARGO_PKG_VERSION")))
.user_agent(concat!("al-runner/", env!("CARGO_PKG_VERSION")))
.http_status_as_error(false)
.build();
ureq::Agent::new_with_config(config)
Expand Down
2 changes: 1 addition & 1 deletion crates/al-host/src/testdata.rs
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
//! Test data from a database file (ADR-020 point 3): the capability a host gives the runtime to
//! read the rows of one table of a restored company, and the counts of what a run took from it.
//!
//! The core never sees a file. A host (`alrunner`) implements [`TableProvider`] over a BC database
//! The core never sees a file. A host (`al-runner`) implements [`TableProvider`] over a BC database
//! backup; the runtime asks it for a table the first time a session touches that table. The
//! provider answers in plain typed cells under the table's SQL column names. Mapping those names
//! to the fields of the compiled program is the runtime's job (the program is the schema, never a
Expand Down
4 changes: 2 additions & 2 deletions crates/al-test-cli/src/main.rs
Original file line number Diff line number Diff line change
Expand Up @@ -76,7 +76,7 @@ fn run() -> Result<ExitCode, String> {
}
// The corpus pins that real BC's web address in its CI is http, has a port and the path /BC, and that its host is
// not `localhost` (session/lt-geturl-shape/GetUrlShape.al:283); the host name itself is masked there and stays the
// host setting of ADR-042. So the corpus tool runs with an address of that shape instead of alrunner's default.
// host setting of ADR-042. So the corpus tool runs with an address of that shape instead of al-runner's default.
al_test_cli::host::set_server_url(Some(CORPUS_SERVER_URL.into()));
if let Some(dir) = &args.files_dir {
al_test_cli::host::set_files_dir(dir.clone());
Expand Down Expand Up @@ -516,7 +516,7 @@ fn run_version(
source_root: Some(root.to_path_buf()),
// Only the user's folders. ADR-014 section 3, step 2 (the service folder of the platform package the run
// selected) is skipped: the artifact cache holds no service folder, and BcContainerHelper's
// `~/.bcartifacts.cache/sandbox`, which `alrunner` reads, is a machine's own folder, often of another build
// `~/.bcartifacts.cache/sandbox`, which `al-runner` reads, is a machine's own folder, often of another build
// of the major. Read here, it made a corpus case pass or be refused by what HOME held (99100
// AL_JsonTextWriter_SimpleObject needs Newtonsoft.Json from it), so one build gave two results
// (corpus_harness.rs `outcomes_do_not_depend_on_threads_or_home`). A missing assembly is refused, as on CI.
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testkit/Cargo.toml
Original file line number Diff line number Diff line change
Expand Up @@ -15,5 +15,5 @@ al-ir.workspace = true
al-testrun = { workspace = true, optional = true }

[features]
# `outcomes_in`: run the tests of a codeunit the way `alrunner test` does (al-testrun).
# `outcomes_in`: run the tests of a codeunit the way `al-runner test` does (al-testrun).
runner = ["dep:al-testrun"]
2 changes: 1 addition & 1 deletion crates/al-testkit/src/lib.rs
Original file line number Diff line number Diff line change
Expand Up @@ -148,7 +148,7 @@ pub fn run_with(src: &str, object: &str, routine: &str, args: Vec<Value>) -> Res
b.call(oi, ri, args)
}

/// With the `runner` feature: compiles `src` and runs the tests of the codeunit named `codeunit` the way `alrunner test` does
/// With the `runner` feature: compiles `src` and runs the tests of the codeunit named `codeunit` the way `al-runner test` does
/// (one session per codeunit, the install code first), giving each test's name and outcome.
#[cfg(feature = "runner")]
pub fn outcomes_in(src: &str, codeunit: &str) -> Vec<(String, al_testrun::Outcome)> {
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testrun/src/baseline_cache.rs
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
//! apps and `Company-Initialize` takes longer than reading what they leave, and it is the same for the same program
//! and the same world around it. The entry lives in the compile cache's folder (`<cache>/compile/`, file extension
//! `albr`), is tied to the same build of the runtime, is pruned with the other entries, and goes with
//! `alrunner cache clear` and is replaced by `--rebuild-compile-cache`. There is no second cache.
//! `al-runner cache clear` and is replaced by `--rebuild-compile-cache`. There is no second cache.
//!
//! The file is the [`BaselineRun`]: the data the install code left, the notes about roots that did not run to their
//! end, and the instant the run began. The file is written to a name of its own and renamed, so a reader never sees
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testrun/src/cache_gc.rs
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
//! Cleanup of the compile cache folder (ADR-022, "Cleanup of old builds").
//!
//! Every entry file name starts with the tag of the alrunner build that wrote it (`al_cache::build_tag`), and a
//! Every entry file name starts with the tag of the al-runner build that wrote it (`al_cache::build_tag`), and a
//! cache hit sets the file's modification time to now. So the cleanup needs `stat` only, never a file read:
//!
//! - an entry of another build (or from before the tag existed) goes after 7 days unused: several builds run on one
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testrun/src/coverage.rs
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
//! Statement coverage for `alrunner test --coverage`: the statements of the workspace's apps, and a per-thread sink in
//! Statement coverage for `al-runner test --coverage`: the statements of the workspace's apps, and a per-thread sink in
//! which a test codeunit's run leaves what each of its tests executed.
//!
//! A statement is a `Stmt::Pos` marker of the lowered program (`ir::Program::positions`); the VM counts how often each
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testrun/src/dotnet.rs
Original file line number Diff line number Diff line change
Expand Up @@ -467,7 +467,7 @@ fn resolve(cfg: &SidecarConfig) -> Result<Resolved, String> {
Ok(Resolved { dotnet, sidecar })
}

/// The sidecar's source files, built into this binary, so a copied or installed `alrunner` can
/// The sidecar's source files, built into this binary, so a copied or installed `al-runner` can
/// find or build its helper without the repository (the cache is keyed by the same hash as the
/// repository's own build). A test keeps the list equal to the folder.
const EMBEDDED: &[(&str, &str)] = &[
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testrun/src/events.rs
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
//! What a run tells while it runs: one event per test as it finishes, and a flag that stops the run after the current
//! test (`alrunner serve`, ADR-028). Installed per thread by the code that runs a test codeunit, so the signatures of
//! test (`al-runner serve`, ADR-028). Installed per thread by the code that runs a test codeunit, so the signatures of
//! the run functions did not change.

use crate::Outcome;
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testrun/src/explain.rs
Original file line number Diff line number Diff line change
Expand Up @@ -91,7 +91,7 @@ pub fn render(head: &str, outcome: &Outcome, oracle: &str, detail: Option<&TestD
render_with(head, Some(oracle), outcome, detail)
}

/// The same for a run that has no oracle (`alrunner test --explain`).
/// The same for a run that has no oracle (`al-runner test --explain`).
pub fn render_without_oracle(head: &str, outcome: &Outcome, detail: Option<&TestDetail>) -> String {
render_with(head, None, outcome, detail)
}
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testrun/src/junit.rs
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
//! The JUnit XML of a run, written in one place (ADR-037 item 6). The two command lines write two dialects of it,
//! and both stay byte for byte what they were: `Dialect::Oracle` is the file `al-test-cli --junit` writes and the
//! oracle parser reads back; `Dialect::Runner` is the v2-shaped file `alrunner test --junit` writes. Each caller
//! oracle parser reads back; `Dialect::Runner` is the v2-shaped file `al-runner test --junit` writes. Each caller
//! builds the suites (how tests are grouped differs) and this module writes them.

use crate::Outcome;
Expand Down
14 changes: 7 additions & 7 deletions crates/al-testrun/src/runtime/compiled.rs
Original file line number Diff line number Diff line change
Expand Up @@ -21,29 +21,29 @@ pub struct RunOptions {
/// to read them (`deps::load_project_with`). The run reads what it needs; a missing entry means
/// the app has its text.
pub lazy: Vec<Option<al_sema::LazySource>>,
/// `alrunner test --isolation`: the isolation of every test codeunit, instead of the runner's property (`None`: the runner's,
/// `al-runner test --isolation`: the isolation of every test codeunit, instead of the runner's property (`None`: the runner's,
/// else Codeunit).
pub isolation: Option<TestIsolation>,
/// Count how often each statement of the workspace's apps runs, per test (`alrunner test --coverage`).
/// Count how often each statement of the workspace's apps runs, per test (`al-runner test --coverage`).
/// Off, a run pays one branch per statement.
pub coverage: bool,
/// A scratch run (`alrunner serve`'s `execute`): the `OnRun` trigger of one codeunit per `SuiteRun`'s app runs
/// A scratch run (`al-runner serve`'s `execute`): the `OnRun` trigger of one codeunit per `SuiteRun`'s app runs
/// instead of the tests, and the text of its `Message` calls is kept in `TestResult::messages`.
pub execute: bool,
/// Capture the values of the locals and the loops of the apps' code, per test (`alrunner serve`, ADR-030). Off,
/// Capture the values of the locals and the loops of the apps' code, per test (`al-runner serve`, ADR-030). Off,
/// a run pays nothing for it.
pub capture: al_vm::CaptureConfig,
/// Callbacks while the run goes on: a line per test as it finishes, and a flag that stops the run after the
/// current test (`alrunner serve`).
/// current test (`al-runner serve`).
pub hooks: crate::events::Hooks,
/// Where the run clock of a test codeunit starts (codeunit id to Unix milliseconds), instead of now: the corpus
/// oracle recorded values derived from `Today` on the day real BC ran, so the harness starts each codeunit at the
/// time of that codeunit's suite in the oracle.
pub clock_starts: Option<Arc<std::collections::HashMap<u32, i64>>>,
/// Do not print the `baseline:` detail lines to stderr (`alrunner test` prints them only with `--verbose`). The counts
/// Do not print the `baseline:` detail lines to stderr (`al-runner test` prints them only with `--verbose`). The counts
/// stay in `Phases::baseline_roots`. Default `false`: print, as the corpus job does.
pub quiet_baseline: bool,
/// `alrunner impact-index build` (ADR-043 step 2): note per test what it raised and touched
/// `al-runner impact-index build` (ADR-043 step 2): note per test what it raised and touched
/// (`TestResult::impact`), and what the install baseline raised and touched (`CorpusRuntime::impact_baseline`).
/// The baseline then runs instead of being read from the compile cache. Off, a run pays one branch per raise and
/// record operation.
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testrun/src/runtime/run.rs
Original file line number Diff line number Diff line change
Expand Up @@ -403,7 +403,7 @@ pub(crate) fn run_codeunit_core(
Err(why) => Ok(Err(Signal::Unsupported(why))),
Ok(()) => std::panic::catch_unwind(std::panic::AssertUnwindSafe(|| {
// A debugging aid: the named test panics here, to show that a panic ends one codeunit and not the
// process (tests of `alrunner serve`). Not documented for users.
// process (tests of `al-runner serve`). Not documented for users.
if std::env::var("ALRUNNER_DEBUG_PANIC_IN").is_ok_and(|n| n == routine.name) {
panic!("requested by ALRUNNER_DEBUG_PANIC_IN");
}
Expand Down
2 changes: 1 addition & 1 deletion crates/al-testrun/tests/execute.rs
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
//! A scratch run (`alrunner serve`'s `execute`): the `OnRun` trigger of one codeunit per app runs, and the
//! A scratch run (`al-runner serve`'s `execute`): the `OnRun` trigger of one codeunit per app runs, and the
//! text of its `Message` calls comes back with the result.

use al_sema::workspace::AppSource;
Expand Down
2 changes: 1 addition & 1 deletion crates/al-vm/src/callstack.rs
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,7 @@ impl Session {
}

/// Sets the frames of the test runner that stand below a test's frames (the corpus job and
/// `alrunner` set them; a session without a runner has none).
/// `al-runner` set them; a session without a runner has none).
pub fn set_runner_frames(&mut self, frames: String) {
self.runner_frames = frames;
}
Expand Down
4 changes: 2 additions & 2 deletions crates/al-vm/src/impact.rs
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
//! The per-test record of what the code raised and touched, for the impact index of `alrunner impact-index build`
//! The per-test record of what the code raised and touched, for the impact index of `al-runner impact-index build`
//! (ADR-043 section 2.2). Off unless asked for (`Session::enable_impact_trace`): then each note costs one branch.
//!
//! What is noted, and where:
Expand Down Expand Up @@ -46,7 +46,7 @@ impl Touches {
}

impl crate::Session {
/// Starts noting what the code raises and touches (`alrunner impact-index build`).
/// Starts noting what the code raises and touches (`al-runner impact-index build`).
pub fn enable_impact_trace(&mut self) {
*self.coverage.impact.borrow_mut() = Some(Touches::default());
self.coverage.impact_on = true;
Expand Down
Loading
Loading