Compare commits

...

4 Commits

Author SHA1 Message Date
Omar Abdulla 6b8fec2a12 Flush the buffer after writing the entire file output 2025-10-09 18:01:19 +03:00
Omar Abdulla 73fce7a250 Make output format deserializable 2025-10-09 17:42:08 +03:00
Omar fde07b7c0d Allow for succeeding tests to be ignored (#186) 2025-10-09 14:35:09 +00:00
Omar ebc24a588b Add different output formats (#185)
* Add different output formats

* Add the mode to the output
2025-10-09 14:24:14 +00:00
17 changed files with 587 additions and 388 deletions
Generated
+13 -3
View File
@@ -782,6 +782,15 @@ dependencies = [
"libc",
]
[[package]]
name = "ansi_term"
version = "0.12.1"
source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "d52a9bb7ec0cf484c551830a7ce27bd20d67eac647e1befb56b0be4ee39a55d2"
dependencies = [
"winapi",
]
[[package]]
name = "anstream"
version = "0.6.18"
@@ -2328,7 +2337,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "8d162beedaa69905488a8da94f5ac3edb4dd4788b732fadb7bd120b2625c1976"
dependencies = [
"data-encoding",
"syn 1.0.109",
"syn 2.0.101",
]
[[package]]
@@ -5564,6 +5573,7 @@ dependencies = [
"clap",
"moka",
"once_cell",
"regex",
"schemars 1.0.4",
"semver 1.0.26",
"serde",
@@ -5611,6 +5621,7 @@ name = "revive-dt-core"
version = "0.1.0"
dependencies = [
"alloy",
"ansi_term",
"anyhow",
"bson",
"cacache",
@@ -5641,7 +5652,6 @@ dependencies = [
"alloy",
"anyhow",
"futures",
"regex",
"revive-common",
"revive-dt-common",
"schemars 1.0.4",
@@ -5943,7 +5953,7 @@ dependencies = [
"security-framework 3.3.0",
"security-framework-sys",
"webpki-root-certs 0.26.11",
"windows-sys 0.52.0",
"windows-sys 0.59.0",
]
[[package]]
+1
View File
@@ -22,6 +22,7 @@ revive-dt-node-pool = { version = "0.1.0", path = "crates/node-pool" }
revive-dt-report = { version = "0.1.0", path = "crates/report" }
revive-dt-solc-binaries = { version = "0.1.0", path = "crates/solc-binaries" }
ansi_term = "0.12.1"
anyhow = "1.0"
async-stream = { version = "0.3.6" }
bson = { version = "2.15.0" }
+1
View File
@@ -14,6 +14,7 @@ anyhow = { workspace = true }
clap = { workspace = true }
moka = { workspace = true, features = ["sync"] }
once_cell = { workspace = true }
regex = { workspace = true }
semver = { workspace = true }
serde = { workspace = true }
schemars = { workspace = true }
+265
View File
@@ -1,6 +1,11 @@
use crate::iterators::EitherIter;
use crate::types::VersionOrRequirement;
use anyhow::{Context as _, bail};
use regex::Regex;
use schemars::JsonSchema;
use semver::Version;
use serde::{Deserialize, Serialize};
use std::collections::HashSet;
use std::fmt::Display;
use std::str::FromStr;
use std::sync::LazyLock;
@@ -33,6 +38,19 @@ impl Display for Mode {
}
}
impl FromStr for Mode {
type Err = anyhow::Error;
fn from_str(s: &str) -> Result<Self, Self::Err> {
let parsed_mode = ParsedMode::from_str(s)?;
let mut iter = parsed_mode.to_modes();
let (Some(mode), None) = (iter.next(), iter.next()) else {
bail!("Failed to parse the mode")
};
Ok(mode)
}
}
impl Mode {
/// Return all of the available mode combinations.
pub fn all() -> impl Iterator<Item = &'static Mode> {
@@ -171,3 +189,250 @@ impl ModeOptimizerSetting {
!matches!(self, ModeOptimizerSetting::M0)
}
}
/// This represents a mode that has been parsed from test metadata.
///
/// Mode strings can take the following form (in pseudo-regex):
///
/// ```text
/// [YEILV][+-]? (M[0123sz])? <semver>?
/// ```
///
/// We can parse valid mode strings into [`ParsedMode`] using [`ParsedMode::from_str`].
#[derive(Clone, Debug, PartialEq, Eq, Hash, Deserialize, Serialize, JsonSchema)]
#[serde(try_from = "String", into = "String")]
pub struct ParsedMode {
pub pipeline: Option<ModePipeline>,
pub optimize_flag: Option<bool>,
pub optimize_setting: Option<ModeOptimizerSetting>,
pub version: Option<semver::VersionReq>,
}
impl FromStr for ParsedMode {
type Err = anyhow::Error;
fn from_str(s: &str) -> Result<Self, Self::Err> {
static REGEX: LazyLock<Regex> = LazyLock::new(|| {
Regex::new(r"(?x)
^
(?:(?P<pipeline>[YEILV])(?P<optimize_flag>[+-])?)? # Pipeline to use eg Y, E+, E-
\s*
(?P<optimize_setting>M[a-zA-Z0-9])? # Optimize setting eg M0, Ms, Mz
\s*
(?P<version>[>=<^]*\d+(?:\.\d+)*)? # Optional semver version eg >=0.8.0, 0.7, <0.8
$
").unwrap()
});
let Some(caps) = REGEX.captures(s) else {
anyhow::bail!("Cannot parse mode '{s}' from string");
};
let pipeline = match caps.name("pipeline") {
Some(m) => Some(
ModePipeline::from_str(m.as_str())
.context("Failed to parse mode pipeline from string")?,
),
None => None,
};
let optimize_flag = caps.name("optimize_flag").map(|m| m.as_str() == "+");
let optimize_setting = match caps.name("optimize_setting") {
Some(m) => Some(
ModeOptimizerSetting::from_str(m.as_str())
.context("Failed to parse optimizer setting from string")?,
),
None => None,
};
let version = match caps.name("version") {
Some(m) => Some(
semver::VersionReq::parse(m.as_str())
.map_err(|e| {
anyhow::anyhow!(
"Cannot parse the version requirement '{}': {e}",
m.as_str()
)
})
.context("Failed to parse semver requirement from mode string")?,
),
None => None,
};
Ok(ParsedMode {
pipeline,
optimize_flag,
optimize_setting,
version,
})
}
}
impl Display for ParsedMode {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
let mut has_written = false;
if let Some(pipeline) = self.pipeline {
pipeline.fmt(f)?;
if let Some(optimize_flag) = self.optimize_flag {
f.write_str(if optimize_flag { "+" } else { "-" })?;
}
has_written = true;
}
if let Some(optimize_setting) = self.optimize_setting {
if has_written {
f.write_str(" ")?;
}
optimize_setting.fmt(f)?;
has_written = true;
}
if let Some(version) = &self.version {
if has_written {
f.write_str(" ")?;
}
version.fmt(f)?;
}
Ok(())
}
}
impl From<ParsedMode> for String {
fn from(parsed_mode: ParsedMode) -> Self {
parsed_mode.to_string()
}
}
impl TryFrom<String> for ParsedMode {
type Error = anyhow::Error;
fn try_from(value: String) -> Result<Self, Self::Error> {
ParsedMode::from_str(&value)
}
}
impl ParsedMode {
/// This takes a [`ParsedMode`] and expands it into a list of [`Mode`]s that we should try.
pub fn to_modes(&self) -> impl Iterator<Item = Mode> {
let pipeline_iter = self.pipeline.as_ref().map_or_else(
|| EitherIter::A(ModePipeline::test_cases()),
|p| EitherIter::B(std::iter::once(*p)),
);
let optimize_flag_setting = self.optimize_flag.map(|flag| {
if flag {
ModeOptimizerSetting::M3
} else {
ModeOptimizerSetting::M0
}
});
let optimize_flag_iter = match optimize_flag_setting {
Some(setting) => EitherIter::A(std::iter::once(setting)),
None => EitherIter::B(ModeOptimizerSetting::test_cases()),
};
let optimize_settings_iter = self.optimize_setting.as_ref().map_or_else(
|| EitherIter::A(optimize_flag_iter),
|s| EitherIter::B(std::iter::once(*s)),
);
pipeline_iter.flat_map(move |pipeline| {
optimize_settings_iter
.clone()
.map(move |optimize_setting| Mode {
pipeline,
optimize_setting,
version: self.version.clone(),
})
})
}
/// Return a set of [`Mode`]s that correspond to the given [`ParsedMode`]s.
/// This avoids any duplicate entries.
pub fn many_to_modes<'a>(
parsed: impl Iterator<Item = &'a ParsedMode>,
) -> impl Iterator<Item = Mode> {
let modes: HashSet<_> = parsed.flat_map(|p| p.to_modes()).collect();
modes.into_iter()
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_parsed_mode_from_str() {
let strings = vec![
("Mz", "Mz"),
("Y", "Y"),
("Y+", "Y+"),
("Y-", "Y-"),
("E", "E"),
("E+", "E+"),
("E-", "E-"),
("Y M0", "Y M0"),
("Y M1", "Y M1"),
("Y M2", "Y M2"),
("Y M3", "Y M3"),
("Y Ms", "Y Ms"),
("Y Mz", "Y Mz"),
("E M0", "E M0"),
("E M1", "E M1"),
("E M2", "E M2"),
("E M3", "E M3"),
("E Ms", "E Ms"),
("E Mz", "E Mz"),
// When stringifying semver again, 0.8.0 becomes ^0.8.0 (same meaning)
("Y 0.8.0", "Y ^0.8.0"),
("E+ 0.8.0", "E+ ^0.8.0"),
("Y M3 >=0.8.0", "Y M3 >=0.8.0"),
("E Mz <0.7.0", "E Mz <0.7.0"),
// We can parse +- _and_ M1/M2 but the latter takes priority.
("Y+ M1 0.8.0", "Y+ M1 ^0.8.0"),
("E- M2 0.7.0", "E- M2 ^0.7.0"),
// We don't see this in the wild but it is parsed.
("<=0.8", "<=0.8"),
];
for (actual, expected) in strings {
let parsed = ParsedMode::from_str(actual)
.unwrap_or_else(|_| panic!("Failed to parse mode string '{actual}'"));
assert_eq!(
expected,
parsed.to_string(),
"Mode string '{actual}' did not parse to '{expected}': got '{parsed}'"
);
}
}
#[test]
fn test_parsed_mode_to_test_modes() {
let strings = vec![
("Mz", vec!["Y Mz", "E Mz"]),
("Y", vec!["Y M0", "Y M3"]),
("E", vec!["E M0", "E M3"]),
("Y+", vec!["Y M3"]),
("Y-", vec!["Y M0"]),
("Y <=0.8", vec!["Y M0 <=0.8", "Y M3 <=0.8"]),
(
"<=0.8",
vec!["Y M0 <=0.8", "Y M3 <=0.8", "E M0 <=0.8", "E M3 <=0.8"],
),
];
for (actual, expected) in strings {
let parsed = ParsedMode::from_str(actual)
.unwrap_or_else(|_| panic!("Failed to parse mode string '{actual}'"));
let expected_set: HashSet<_> = expected.into_iter().map(|s| s.to_owned()).collect();
let actual_set: HashSet<_> = parsed.to_modes().map(|m| m.to_string()).collect();
assert_eq!(
expected_set, actual_set,
"Mode string '{actual}' did not expand to '{expected_set:?}': got '{actual_set:?}'"
);
}
}
}
@@ -7,7 +7,10 @@ pragma solidity >=0.6.9;
import "./callable.sol";
contract Main {
function main(uint[1] calldata p1, Callable callable) public returns(uint) {
function main(
uint[1] calldata p1,
Callable callable
) public pure returns (uint) {
return callable.f(p1);
}
}
+106 -47
View File
@@ -12,19 +12,18 @@ use std::{
use alloy::{
genesis::Genesis,
hex::ToHexExt,
network::EthereumWallet,
primitives::{FixedBytes, U256},
primitives::{B256, FixedBytes, U256},
signers::local::PrivateKeySigner,
};
use clap::{Parser, ValueEnum, ValueHint};
use revive_dt_common::types::PlatformIdentifier;
use semver::Version;
use serde::{Serialize, Serializer};
use serde::{Deserialize, Serialize, Serializer};
use strum::{AsRefStr, Display, EnumString, IntoStaticStr};
use temp_dir::TempDir;
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
#[command(name = "retester")]
pub enum Context {
/// Executes tests in the MatterLabs format differentially on multiple targets concurrently.
@@ -200,8 +199,30 @@ impl AsRef<ReportConfiguration> for Context {
}
}
#[derive(Clone, Debug, Parser, Serialize)]
impl AsRef<IgnoreSuccessConfiguration> for Context {
fn as_ref(&self) -> &IgnoreSuccessConfiguration {
match self {
Self::Test(context) => context.as_ref().as_ref(),
Self::Benchmark(..) => unreachable!(),
Self::ExportJsonSchema => unreachable!(),
}
}
}
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct TestExecutionContext {
/// The set of platforms that the differential tests should run on.
#[arg(
short = 'p',
long = "platform",
default_values = ["geth-evm-solc", "revive-dev-node-polkavm-resolc"]
)]
pub platforms: Vec<PlatformIdentifier>,
/// The output format to use for the tool's output.
#[arg(short, long, default_value_t = OutputFormat::CargoTestLike)]
pub output_format: OutputFormat,
/// The working directory that the program will use for all of the temporary artifacts needed at
/// runtime.
///
@@ -215,14 +236,6 @@ pub struct TestExecutionContext {
)]
pub working_directory: WorkingDirectoryConfiguration,
/// The set of platforms that the differential tests should run on.
#[arg(
short = 'p',
long = "platform",
default_values = ["geth-evm-solc", "revive-dev-node-polkavm-resolc"]
)]
pub platforms: Vec<PlatformIdentifier>,
/// Configuration parameters for the corpus files to use.
#[clap(flatten, next_help_heading = "Corpus Configuration")]
pub corpus_configuration: CorpusConfiguration,
@@ -278,9 +291,13 @@ pub struct TestExecutionContext {
/// Configuration parameters for the report.
#[clap(flatten, next_help_heading = "Report Configuration")]
pub report_configuration: ReportConfiguration,
/// Configuration parameters for ignoring certain test cases based on the report
#[clap(flatten, next_help_heading = "Ignore Success Configuration")]
pub ignore_success_configuration: IgnoreSuccessConfiguration,
}
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct BenchmarkingContext {
/// The working directory that the program will use for all of the temporary artifacts needed at
/// runtime.
@@ -457,6 +474,12 @@ impl AsRef<ReportConfiguration> for TestExecutionContext {
}
}
impl AsRef<IgnoreSuccessConfiguration> for TestExecutionContext {
fn as_ref(&self) -> &IgnoreSuccessConfiguration {
&self.ignore_success_configuration
}
}
impl Default for BenchmarkingContext {
fn default() -> Self {
Self::parse_from(["execution-context"])
@@ -548,7 +571,7 @@ impl AsRef<ReportConfiguration> for BenchmarkingContext {
}
/// A set of configuration parameters for the corpus files to use for the execution.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct CorpusConfiguration {
/// A list of test corpus JSON files to be tested.
#[arg(short = 'c', long = "corpus")]
@@ -556,7 +579,7 @@ pub struct CorpusConfiguration {
}
/// A set of configuration parameters for Solc.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct SolcConfiguration {
/// Specifies the default version of the Solc compiler that should be used if there is no
/// override specified by one of the test cases.
@@ -565,7 +588,7 @@ pub struct SolcConfiguration {
}
/// A set of configuration parameters for Resolc.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct ResolcConfiguration {
/// Specifies the path of the resolc compiler to be used by the tool.
///
@@ -576,7 +599,7 @@ pub struct ResolcConfiguration {
}
/// A set of configuration parameters for Polkadot Parachain.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct PolkadotParachainConfiguration {
/// Specifies the path of the polkadot-parachain node to be used by the tool.
///
@@ -600,7 +623,7 @@ pub struct PolkadotParachainConfiguration {
}
/// A set of configuration parameters for Geth.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct GethConfiguration {
/// Specifies the path of the geth node to be used by the tool.
///
@@ -620,7 +643,7 @@ pub struct GethConfiguration {
}
/// A set of configuration parameters for kurtosis.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct KurtosisConfiguration {
/// Specifies the path of the kurtosis node to be used by the tool.
///
@@ -635,7 +658,7 @@ pub struct KurtosisConfiguration {
}
/// A set of configuration parameters for Kitchensink.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct KitchensinkConfiguration {
/// Specifies the path of the kitchensink node to be used by the tool.
///
@@ -659,7 +682,7 @@ pub struct KitchensinkConfiguration {
}
/// A set of configuration parameters for the revive dev node.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct ReviveDevNodeConfiguration {
/// Specifies the path of the revive dev node to be used by the tool.
///
@@ -691,7 +714,7 @@ pub struct ReviveDevNodeConfiguration {
}
/// A set of configuration parameters for the ETH RPC.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct EthRpcConfiguration {
/// Specifies the path of the ETH RPC to be used by the tool.
///
@@ -711,7 +734,7 @@ pub struct EthRpcConfiguration {
}
/// A set of configuration parameters for the genesis.
#[derive(Clone, Debug, Default, Parser, Serialize)]
#[derive(Clone, Debug, Default, Parser, Serialize, Deserialize)]
pub struct GenesisConfiguration {
/// Specifies the path of the genesis file to use for the nodes that are started.
///
@@ -749,15 +772,14 @@ impl GenesisConfiguration {
}
/// A set of configuration parameters for the wallet.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct WalletConfiguration {
/// The private key of the default signer.
#[clap(
long = "wallet.default-private-key",
default_value = "0x4f3edf983ac636a65a842ce7c78d9aa706d3b113bce9c46f30d7d21715b23b1d"
)]
#[serde(serialize_with = "serialize_private_key")]
default_key: PrivateKeySigner,
default_key: B256,
/// This argument controls which private keys the nodes should have access to and be added to
/// its wallet signers. With a value of N, private keys (0, N] will be added to the signer set
@@ -775,7 +797,8 @@ impl WalletConfiguration {
pub fn wallet(&self) -> Arc<EthereumWallet> {
self.wallet
.get_or_init(|| {
let mut wallet = EthereumWallet::new(self.default_key.clone());
let mut wallet =
EthereumWallet::new(PrivateKeySigner::from_bytes(&self.default_key).unwrap());
for signer in (1..=self.additional_keys)
.map(|id| U256::from(id))
.map(|id| id.to_be_bytes::<32>())
@@ -793,15 +816,8 @@ impl WalletConfiguration {
}
}
fn serialize_private_key<S>(value: &PrivateKeySigner, serializer: S) -> Result<S::Ok, S::Error>
where
S: Serializer,
{
value.to_bytes().encode_hex().serialize(serializer)
}
/// A set of configuration for concurrency.
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct ConcurrencyConfiguration {
/// Determines the amount of nodes that will be spawned for each chain.
#[clap(long = "concurrency.number-of-nodes", default_value_t = 5)]
@@ -839,14 +855,14 @@ impl ConcurrencyConfiguration {
}
}
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct CompilationConfiguration {
/// Controls if the compilation cache should be invalidated or not.
#[arg(long = "compilation.invalidate-cache")]
pub invalidate_compilation_cache: bool,
}
#[derive(Clone, Debug, Parser, Serialize)]
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct ReportConfiguration {
/// Controls if the compiler input is included in the final report.
#[clap(long = "report.include-compiler-input")]
@@ -857,6 +873,13 @@ pub struct ReportConfiguration {
pub include_compiler_output: bool,
}
#[derive(Clone, Debug, Parser, Serialize, Deserialize)]
pub struct IgnoreSuccessConfiguration {
/// The path of the report generated by the tool to use to ignore the cases that succeeded.
#[clap(long = "ignore-success.report-path")]
pub path: Option<PathBuf>,
}
/// Represents the working directory that the program uses.
#[derive(Debug, Clone)]
pub enum WorkingDirectoryConfiguration {
@@ -866,6 +889,24 @@ pub enum WorkingDirectoryConfiguration {
Path(PathBuf),
}
impl Serialize for WorkingDirectoryConfiguration {
fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
where
S: Serializer,
{
self.as_path().serialize(serializer)
}
}
impl<'a> Deserialize<'a> for WorkingDirectoryConfiguration {
fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>
where
D: serde::Deserializer<'a>,
{
PathBuf::deserialize(deserializer).map(Self::Path)
}
}
impl WorkingDirectoryConfiguration {
pub fn as_path(&self) -> &Path {
self.as_ref()
@@ -915,15 +956,6 @@ impl Display for WorkingDirectoryConfiguration {
}
}
impl Serialize for WorkingDirectoryConfiguration {
fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
where
S: serde::Serializer,
{
self.as_path().serialize(serializer)
}
}
fn parse_duration(s: &str) -> anyhow::Result<Duration> {
u64::from_str(s)
.map(Duration::from_millis)
@@ -958,3 +990,30 @@ pub enum TestingPlatform {
/// A polkadot/Substrate based network
Zombienet,
}
/// The output format to use for the test execution output.
#[derive(
Clone,
Copy,
Debug,
PartialEq,
Eq,
PartialOrd,
Ord,
Hash,
Serialize,
Deserialize,
ValueEnum,
EnumString,
Display,
AsRefStr,
IntoStaticStr,
)]
#[strum(serialize_all = "kebab-case")]
pub enum OutputFormat {
/// The legacy format that was used in the past for the output.
Legacy,
/// An output format that looks heavily resembles the output from `cargo test`.
CargoTestLike,
}
+1
View File
@@ -21,6 +21,7 @@ revive-dt-node = { workspace = true }
revive-dt-node-interaction = { workspace = true }
revive-dt-report = { workspace = true }
ansi_term = { workspace = true }
alloy = { workspace = true }
anyhow = { workspace = true }
bson = { workspace = true }
@@ -86,6 +86,7 @@ pub async fn handle_differential_benchmarks(
&full_context,
metadata_files.iter(),
&platforms_and_nodes,
None,
reporter.clone(),
)
.await
+139 -58
View File
@@ -7,14 +7,15 @@ use std::{
time::{Duration, Instant},
};
use ansi_term::{ANSIStrings, Color};
use anyhow::Context as _;
use futures::{FutureExt, StreamExt};
use revive_dt_common::types::PrivateKeyAllocator;
use revive_dt_common::{cached_fs::read_to_string, types::PrivateKeyAllocator};
use revive_dt_core::Platform;
use tokio::sync::{Mutex, RwLock, Semaphore};
use tracing::{Instrument, error, info, info_span, instrument};
use revive_dt_config::{Context, TestExecutionContext};
use revive_dt_config::{Context, OutputFormat, TestExecutionContext};
use revive_dt_report::{Reporter, ReporterEvent, TestCaseStatus};
use crate::{
@@ -71,11 +72,20 @@ pub async fn handle_differential_tests(
info!("Spawned the platform nodes");
// Preparing test definitions.
let only_execute_failed_tests = match context.ignore_success_configuration.path.as_ref() {
Some(path) => {
let report = read_to_string(path)
.context("Failed to read the report file to ignore the succeeding test cases")?;
Some(serde_json::from_str(&report).context("Failed to deserialize report")?)
}
None => None,
};
let full_context = Context::Test(Box::new(context.clone()));
let test_definitions = create_test_definitions_stream(
&full_context,
metadata_files.iter(),
&platforms_and_nodes,
only_execute_failed_tests.as_ref(),
reporter.clone(),
)
.await
@@ -176,7 +186,7 @@ pub async fn handle_differential_tests(
.report_completion_event()
.expect("Can't fail")
});
let cli_reporting_task = start_cli_reporting_task(reporter);
let cli_reporting_task = start_cli_reporting_task(context.output_format, reporter);
tokio::task::spawn(async move {
loop {
@@ -196,21 +206,15 @@ pub async fn handle_differential_tests(
}
#[allow(irrefutable_let_patterns, clippy::uninlined_format_args)]
async fn start_cli_reporting_task(reporter: Reporter) {
async fn start_cli_reporting_task(output_format: OutputFormat, reporter: Reporter) {
let mut aggregator_events_rx = reporter.subscribe().await.expect("Can't fail");
drop(reporter);
let start = Instant::now();
const GREEN: &str = "\x1B[32m";
const RED: &str = "\x1B[31m";
const GREY: &str = "\x1B[90m";
const COLOR_RESET: &str = "\x1B[0m";
const BOLD: &str = "\x1B[1m";
const BOLD_RESET: &str = "\x1B[22m";
let mut number_of_successes = 0;
let mut number_of_failures = 0;
let mut global_success_count = 0;
let mut global_failure_count = 0;
let mut global_ignore_count = 0;
let mut buf = BufWriter::new(stderr());
while let Ok(event) = aggregator_events_rx.recv().await {
@@ -223,55 +227,132 @@ async fn start_cli_reporting_task(reporter: Reporter) {
continue;
};
let _ = writeln!(buf, "{} - {}", mode, metadata_file_path.display());
for (case_idx, case_status) in case_status.into_iter() {
let _ = write!(buf, "\tCase Index {case_idx:>3}: ");
let _ = match case_status {
TestCaseStatus::Succeeded { steps_executed } => {
number_of_successes += 1;
writeln!(
buf,
"{}{}Case Succeeded{} - Steps Executed: {}{}",
GREEN, BOLD, BOLD_RESET, steps_executed, COLOR_RESET
)
match output_format {
OutputFormat::Legacy => {
let _ = writeln!(buf, "{} - {}", mode, metadata_file_path.display());
for (case_idx, case_status) in case_status.into_iter() {
let _ = write!(buf, "\tCase Index {case_idx:>3}: ");
let _ = match case_status {
TestCaseStatus::Succeeded { steps_executed } => {
global_success_count += 1;
writeln!(
buf,
"{}",
ANSIStrings(&[
Color::Green.bold().paint("Case Succeeded"),
Color::Green
.paint(format!(" - Steps Executed: {steps_executed}")),
])
)
}
TestCaseStatus::Failed { reason } => {
global_failure_count += 1;
writeln!(
buf,
"{}",
ANSIStrings(&[
Color::Red.bold().paint("Case Failed"),
Color::Red.paint(format!(" - Reason: {}", reason.trim())),
])
)
}
TestCaseStatus::Ignored { reason, .. } => {
global_ignore_count += 1;
writeln!(
buf,
"{}",
ANSIStrings(&[
Color::Yellow.bold().paint("Case Ignored"),
Color::Yellow.paint(format!(" - Reason: {}", reason.trim())),
])
)
}
};
}
TestCaseStatus::Failed { reason } => {
number_of_failures += 1;
writeln!(
buf,
"{}{}Case Failed{} - Reason: {}{}",
RED,
BOLD,
BOLD_RESET,
reason.trim(),
COLOR_RESET,
)
}
TestCaseStatus::Ignored { reason, .. } => writeln!(
let _ = writeln!(buf);
}
OutputFormat::CargoTestLike => {
writeln!(
buf,
"{}{}Case Ignored{} - Reason: {}{}",
GREY,
BOLD,
BOLD_RESET,
reason.trim(),
COLOR_RESET,
),
};
"\t{} {} - {}\n",
Color::Green.paint("Running"),
metadata_file_path.display(),
mode
)
.unwrap();
let mut success_count = 0;
let mut failure_count = 0;
let mut ignored_count = 0;
writeln!(buf, "running {} tests", case_status.len()).unwrap();
for (case_idx, case_result) in case_status.iter() {
let status = match case_result {
TestCaseStatus::Succeeded { .. } => {
success_count += 1;
global_success_count += 1;
Color::Green.paint("ok")
}
TestCaseStatus::Failed { reason } => {
failure_count += 1;
global_failure_count += 1;
Color::Red.paint(format!("FAILED, {reason}"))
}
TestCaseStatus::Ignored { reason, .. } => {
ignored_count += 1;
global_ignore_count += 1;
Color::Yellow.paint(format!("ignored, {reason:?}"))
}
};
writeln!(buf, "test case_idx_{} ... {}", case_idx, status).unwrap();
}
writeln!(buf).unwrap();
let status = if failure_count > 0 {
Color::Red.paint("FAILED")
} else {
Color::Green.paint("ok")
};
writeln!(
buf,
"test result: {}. {} passed; {} failed; {} ignored",
status, success_count, failure_count, ignored_count,
)
.unwrap();
writeln!(buf).unwrap();
buf = tokio::task::spawn_blocking(move || {
buf.flush().unwrap();
buf
})
.await
.unwrap();
}
}
let _ = writeln!(buf);
}
// Summary at the end.
let _ = writeln!(
buf,
"{} cases: {}{}{} cases succeeded, {}{}{} cases failed in {} seconds",
number_of_successes + number_of_failures,
GREEN,
number_of_successes,
COLOR_RESET,
RED,
number_of_failures,
COLOR_RESET,
start.elapsed().as_secs()
);
match output_format {
OutputFormat::Legacy => {
writeln!(
buf,
"{} cases: {} cases succeeded, {} cases failed in {} seconds",
global_success_count + global_failure_count + global_ignore_count,
Color::Green.paint(global_success_count.to_string()),
Color::Red.paint(global_failure_count.to_string()),
start.elapsed().as_secs()
)
.unwrap();
}
OutputFormat::CargoTestLike => {
writeln!(
buf,
"run finished. {} passed; {} failed; {} ignored; finished in {}s",
global_success_count,
global_failure_count,
global_ignore_count,
start.elapsed().as_secs()
)
.unwrap();
}
}
}
+39 -5
View File
@@ -5,9 +5,8 @@ use std::{borrow::Cow, path::Path};
use futures::{Stream, StreamExt, stream};
use indexmap::{IndexMap, indexmap};
use revive_dt_common::iterators::EitherIter;
use revive_dt_common::types::PlatformIdentifier;
use revive_dt_common::types::{ParsedMode, PlatformIdentifier};
use revive_dt_config::Context;
use revive_dt_format::mode::ParsedMode;
use serde_json::{Value, json};
use revive_dt_compiler::Mode;
@@ -17,7 +16,7 @@ use revive_dt_format::{
metadata::MetadataFile,
};
use revive_dt_node_interaction::EthereumNode;
use revive_dt_report::{ExecutionSpecificReporter, Reporter};
use revive_dt_report::{ExecutionSpecificReporter, Report, Reporter, TestCaseStatus};
use revive_dt_report::{TestSpecificReporter, TestSpecifier};
use tracing::{debug, error, info};
@@ -30,6 +29,7 @@ pub async fn create_test_definitions_stream<'a>(
context: &Context,
metadata_files: impl IntoIterator<Item = &'a MetadataFile>,
platforms_and_nodes: &'a BTreeMap<PlatformIdentifier, (&dyn Platform, NodePool)>,
only_execute_failed_tests: Option<&Report>,
reporter: Reporter,
) -> impl Stream<Item = TestDefinition<'a>> {
stream::iter(
@@ -140,7 +140,7 @@ pub async fn create_test_definitions_stream<'a>(
)
// Filter out the test cases which are incompatible or that can't run in the current setup.
.filter_map(move |test| async move {
match test.check_compatibility() {
match test.check_compatibility(only_execute_failed_tests) {
Ok(()) => Some(test),
Err((reason, additional_information)) => {
debug!(
@@ -200,12 +200,16 @@ pub struct TestDefinition<'a> {
impl<'a> TestDefinition<'a> {
/// Checks if this test can be ran with the current configuration.
pub fn check_compatibility(&self) -> TestCheckFunctionResult {
pub fn check_compatibility(
&self,
only_execute_failed_tests: Option<&Report>,
) -> TestCheckFunctionResult {
self.check_metadata_file_ignored()?;
self.check_case_file_ignored()?;
self.check_target_compatibility()?;
self.check_evm_version_compatibility()?;
self.check_compiler_compatibility()?;
self.check_ignore_succeeded(only_execute_failed_tests)?;
Ok(())
}
@@ -313,6 +317,36 @@ impl<'a> TestDefinition<'a> {
))
}
}
/// Checks if the test case should be executed or not based on the passed report and whether the
/// user has instructed the tool to ignore the already succeeding test cases.
fn check_ignore_succeeded(
&self,
only_execute_failed_tests: Option<&Report>,
) -> TestCheckFunctionResult {
let Some(report) = only_execute_failed_tests else {
return Ok(());
};
let test_case_status = report
.test_case_information
.get(&(self.metadata_file_path.to_path_buf().into()))
.and_then(|obj| obj.get(&self.mode))
.and_then(|obj| obj.get(&self.case_idx))
.and_then(|obj| obj.status.as_ref());
match test_case_status {
Some(TestCaseStatus::Failed { .. }) => Ok(()),
Some(TestCaseStatus::Ignored { .. }) => Err((
"Ignored since it was ignored in a previous run",
indexmap! {},
)),
Some(TestCaseStatus::Succeeded { .. }) => {
Err(("Ignored since it succeeded in a prior run", indexmap! {}))
}
None => Ok(()),
}
}
}
pub struct TestPlatformInformation<'a> {
-1
View File
@@ -16,7 +16,6 @@ revive-common = { workspace = true }
alloy = { workspace = true }
anyhow = { workspace = true }
futures = { workspace = true }
regex = { workspace = true }
tracing = { workspace = true }
schemars = { workspace = true }
semver = { workspace = true }
+5 -2
View File
@@ -1,9 +1,12 @@
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use revive_dt_common::{macros::define_wrapper_type, types::Mode};
use revive_dt_common::{
macros::define_wrapper_type,
types::{Mode, ParsedMode},
};
use crate::{mode::ParsedMode, steps::*};
use crate::steps::*;
#[derive(Debug, Default, Serialize, Deserialize, Clone, Eq, PartialEq, JsonSchema)]
pub struct Case {
-1
View File
@@ -3,6 +3,5 @@
pub mod case;
pub mod corpus;
pub mod metadata;
pub mod mode;
pub mod steps;
pub mod traits;
+2 -2
View File
@@ -16,11 +16,11 @@ use revive_dt_common::{
cached_fs::read_to_string,
iterators::FilesWithExtensionIterator,
macros::define_wrapper_type,
types::{Mode, VmIdentifier},
types::{Mode, ParsedMode, VmIdentifier},
};
use tracing::error;
use crate::{case::Case, mode::ParsedMode};
use crate::case::Case;
pub const METADATA_FILE_EXTENSION: &str = "json";
pub const SOLIDITY_CASE_FILE_EXTENSION: &str = "sol";
-257
View File
@@ -1,257 +0,0 @@
use anyhow::Context as _;
use regex::Regex;
use revive_dt_common::iterators::EitherIter;
use revive_dt_common::types::{Mode, ModeOptimizerSetting, ModePipeline};
use schemars::JsonSchema;
use serde::{Deserialize, Serialize};
use std::collections::HashSet;
use std::fmt::Display;
use std::str::FromStr;
use std::sync::LazyLock;
/// This represents a mode that has been parsed from test metadata.
///
/// Mode strings can take the following form (in pseudo-regex):
///
/// ```text
/// [YEILV][+-]? (M[0123sz])? <semver>?
/// ```
///
/// We can parse valid mode strings into [`ParsedMode`] using [`ParsedMode::from_str`].
#[derive(Clone, Debug, PartialEq, Eq, Hash, Deserialize, Serialize, JsonSchema)]
#[serde(try_from = "String", into = "String")]
pub struct ParsedMode {
pub pipeline: Option<ModePipeline>,
pub optimize_flag: Option<bool>,
pub optimize_setting: Option<ModeOptimizerSetting>,
pub version: Option<semver::VersionReq>,
}
impl FromStr for ParsedMode {
type Err = anyhow::Error;
fn from_str(s: &str) -> Result<Self, Self::Err> {
static REGEX: LazyLock<Regex> = LazyLock::new(|| {
Regex::new(r"(?x)
^
(?:(?P<pipeline>[YEILV])(?P<optimize_flag>[+-])?)? # Pipeline to use eg Y, E+, E-
\s*
(?P<optimize_setting>M[a-zA-Z0-9])? # Optimize setting eg M0, Ms, Mz
\s*
(?P<version>[>=<]*\d+(?:\.\d+)*)? # Optional semver version eg >=0.8.0, 0.7, <0.8
$
").unwrap()
});
let Some(caps) = REGEX.captures(s) else {
anyhow::bail!("Cannot parse mode '{s}' from string");
};
let pipeline = match caps.name("pipeline") {
Some(m) => Some(
ModePipeline::from_str(m.as_str())
.context("Failed to parse mode pipeline from string")?,
),
None => None,
};
let optimize_flag = caps.name("optimize_flag").map(|m| m.as_str() == "+");
let optimize_setting = match caps.name("optimize_setting") {
Some(m) => Some(
ModeOptimizerSetting::from_str(m.as_str())
.context("Failed to parse optimizer setting from string")?,
),
None => None,
};
let version = match caps.name("version") {
Some(m) => Some(
semver::VersionReq::parse(m.as_str())
.map_err(|e| {
anyhow::anyhow!(
"Cannot parse the version requirement '{}': {e}",
m.as_str()
)
})
.context("Failed to parse semver requirement from mode string")?,
),
None => None,
};
Ok(ParsedMode {
pipeline,
optimize_flag,
optimize_setting,
version,
})
}
}
impl Display for ParsedMode {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
let mut has_written = false;
if let Some(pipeline) = self.pipeline {
pipeline.fmt(f)?;
if let Some(optimize_flag) = self.optimize_flag {
f.write_str(if optimize_flag { "+" } else { "-" })?;
}
has_written = true;
}
if let Some(optimize_setting) = self.optimize_setting {
if has_written {
f.write_str(" ")?;
}
optimize_setting.fmt(f)?;
has_written = true;
}
if let Some(version) = &self.version {
if has_written {
f.write_str(" ")?;
}
version.fmt(f)?;
}
Ok(())
}
}
impl From<ParsedMode> for String {
fn from(parsed_mode: ParsedMode) -> Self {
parsed_mode.to_string()
}
}
impl TryFrom<String> for ParsedMode {
type Error = anyhow::Error;
fn try_from(value: String) -> Result<Self, Self::Error> {
ParsedMode::from_str(&value)
}
}
impl ParsedMode {
/// This takes a [`ParsedMode`] and expands it into a list of [`Mode`]s that we should try.
pub fn to_modes(&self) -> impl Iterator<Item = Mode> {
let pipeline_iter = self.pipeline.as_ref().map_or_else(
|| EitherIter::A(ModePipeline::test_cases()),
|p| EitherIter::B(std::iter::once(*p)),
);
let optimize_flag_setting = self.optimize_flag.map(|flag| {
if flag {
ModeOptimizerSetting::M3
} else {
ModeOptimizerSetting::M0
}
});
let optimize_flag_iter = match optimize_flag_setting {
Some(setting) => EitherIter::A(std::iter::once(setting)),
None => EitherIter::B(ModeOptimizerSetting::test_cases()),
};
let optimize_settings_iter = self.optimize_setting.as_ref().map_or_else(
|| EitherIter::A(optimize_flag_iter),
|s| EitherIter::B(std::iter::once(*s)),
);
pipeline_iter.flat_map(move |pipeline| {
optimize_settings_iter
.clone()
.map(move |optimize_setting| Mode {
pipeline,
optimize_setting,
version: self.version.clone(),
})
})
}
/// Return a set of [`Mode`]s that correspond to the given [`ParsedMode`]s.
/// This avoids any duplicate entries.
pub fn many_to_modes<'a>(
parsed: impl Iterator<Item = &'a ParsedMode>,
) -> impl Iterator<Item = Mode> {
let modes: HashSet<_> = parsed.flat_map(|p| p.to_modes()).collect();
modes.into_iter()
}
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_parsed_mode_from_str() {
let strings = vec![
("Mz", "Mz"),
("Y", "Y"),
("Y+", "Y+"),
("Y-", "Y-"),
("E", "E"),
("E+", "E+"),
("E-", "E-"),
("Y M0", "Y M0"),
("Y M1", "Y M1"),
("Y M2", "Y M2"),
("Y M3", "Y M3"),
("Y Ms", "Y Ms"),
("Y Mz", "Y Mz"),
("E M0", "E M0"),
("E M1", "E M1"),
("E M2", "E M2"),
("E M3", "E M3"),
("E Ms", "E Ms"),
("E Mz", "E Mz"),
// When stringifying semver again, 0.8.0 becomes ^0.8.0 (same meaning)
("Y 0.8.0", "Y ^0.8.0"),
("E+ 0.8.0", "E+ ^0.8.0"),
("Y M3 >=0.8.0", "Y M3 >=0.8.0"),
("E Mz <0.7.0", "E Mz <0.7.0"),
// We can parse +- _and_ M1/M2 but the latter takes priority.
("Y+ M1 0.8.0", "Y+ M1 ^0.8.0"),
("E- M2 0.7.0", "E- M2 ^0.7.0"),
// We don't see this in the wild but it is parsed.
("<=0.8", "<=0.8"),
];
for (actual, expected) in strings {
let parsed = ParsedMode::from_str(actual)
.unwrap_or_else(|_| panic!("Failed to parse mode string '{actual}'"));
assert_eq!(
expected,
parsed.to_string(),
"Mode string '{actual}' did not parse to '{expected}': got '{parsed}'"
);
}
}
#[test]
fn test_parsed_mode_to_test_modes() {
let strings = vec![
("Mz", vec!["Y Mz", "E Mz"]),
("Y", vec!["Y M0", "Y M3"]),
("E", vec!["E M0", "E M3"]),
("Y+", vec!["Y M3"]),
("Y-", vec!["Y M0"]),
("Y <=0.8", vec!["Y M0 <=0.8", "Y M3 <=0.8"]),
(
"<=0.8",
vec!["Y M0 <=0.8", "Y M3 <=0.8", "E M0 <=0.8", "E M3 <=0.8"],
),
];
for (actual, expected) in strings {
let parsed = ParsedMode::from_str(actual)
.unwrap_or_else(|_| panic!("Failed to parse mode string '{actual}'"));
let expected_set: HashSet<_> = expected.into_iter().map(|s| s.to_owned()).collect();
let actual_set: HashSet<_> = parsed.to_modes().map(|m| m.to_string()).collect();
assert_eq!(
expected_set, actual_set,
"Mode string '{actual}' did not expand to '{expected_set:?}': got '{actual_set:?}'"
);
}
}
}
+7 -7
View File
@@ -16,7 +16,7 @@ use revive_dt_compiler::{CompilerInput, CompilerOutput, Mode};
use revive_dt_config::Context;
use revive_dt_format::{case::CaseIdx, corpus::Corpus, metadata::ContractInstance};
use semver::Version;
use serde::Serialize;
use serde::{Deserialize, Serialize};
use serde_with::{DisplayFromStr, serde_as};
use tokio::sync::{
broadcast::{Sender, channel},
@@ -415,7 +415,7 @@ impl ReportAggregator {
}
#[serde_as]
#[derive(Clone, Debug, Serialize)]
#[derive(Clone, Debug, Serialize, Deserialize)]
pub struct Report {
/// The context that the tool was started up with.
pub context: Context,
@@ -440,7 +440,7 @@ impl Report {
}
}
#[derive(Clone, Debug, Serialize, Default)]
#[derive(Clone, Debug, Serialize, Deserialize, Default)]
pub struct TestCaseReport {
/// Information on the status of the test case and whether it succeeded, failed, or was ignored.
#[serde(skip_serializing_if = "Option::is_none")]
@@ -451,7 +451,7 @@ pub struct TestCaseReport {
/// Information related to the status of the test. Could be that the test succeeded, failed, or that
/// it was ignored.
#[derive(Clone, Debug, Serialize)]
#[derive(Clone, Debug, Serialize, Deserialize)]
#[serde(tag = "status")]
pub enum TestCaseStatus {
/// The test case succeeded.
@@ -475,7 +475,7 @@ pub enum TestCaseStatus {
}
/// Information related to the platform node that's being used to execute the step.
#[derive(Clone, Debug, Serialize)]
#[derive(Clone, Debug, Serialize, Deserialize)]
pub struct TestCaseNodeInformation {
/// The ID of the node that this case is being executed on.
pub id: usize,
@@ -486,7 +486,7 @@ pub struct TestCaseNodeInformation {
}
/// Execution information tied to the platform.
#[derive(Clone, Debug, Default, Serialize)]
#[derive(Clone, Debug, Default, Serialize, Deserialize)]
pub struct ExecutionInformation {
/// Information related to the node assigned to this test case.
#[serde(skip_serializing_if = "Option::is_none")]
@@ -506,7 +506,7 @@ pub struct ExecutionInformation {
}
/// Information related to compilation
#[derive(Clone, Debug, Serialize)]
#[derive(Clone, Debug, Serialize, Deserialize)]
#[serde(tag = "status")]
pub enum CompilationStatus {
/// The compilation was successful.
+3 -4
View File
@@ -93,17 +93,16 @@ echo ""
# Run the tool
cargo build --release;
RUST_LOG="info,alloy_pubsub::service=error" ./target/release/retester test \
--platform revive-dev-node-revm-solc \
--platform geth-evm-solc \
--corpus "$CORPUS_FILE" \
--working-directory "$WORKDIR" \
--concurrency.number-of-nodes 10 \
--concurrency.number-of-threads 5 \
--concurrency.number-of-concurrent-tasks 1000 \
--concurrency.ignore-concurrency-limit \
--wallet.additional-keys 100000 \
--kitchensink.path "$SUBSTRATE_NODE_BIN" \
--revive-dev-node.path "$REVIVE_DEV_NODE_BIN" \
--eth-rpc.path "$ETH_RPC_BIN" \
> logs.log \
2> output.log
> logs.log
echo -e "${GREEN}=== Test run completed! ===${NC}"