Add default values to compiler arguments (#451)

## Description

Adds default values to the compiler arguments below when parsed by
`clap`. This does not change previous user-facing behavior as these
defaults were implemented using different logic, but improves the UX and
the visibility of the exact defaults used.

**Example output from `resolc --help`:**
(Only affected arguments are shown here, and some of their descriptions
have been omitted here.)
```
Options:

  -O, --optimization <OPTIMIZATION>
          Set the optimization parameter -O[0 | 1 | 2 | 3 | s | z]. Use `3` for best performance and `z` for minimal size
          
          [default: z]

      --metadata-hash <METADATA_HASH>
          Set the metadata hash type. Available types: `none`, `ipfs`, `keccak256`
          
          [default: keccak256]

      --heap-size <HEAP_SIZE>
          The emulated EVM linear heap memory static buffer size in bytes.
          ...
          
          [default: 131072]

      --stack-size <STACK_SIZE>
          The contracts total stack size in bytes.
          ...
          
          [default: 131072]
```

## Resolved Issues

Closes #446

crates/llvm-context/src/optimizer/settings/mod.rs

@@ -64,14 +64,7 @@ impl Settings {
     /// Creates settings from a CLI optimization parameter.
     pub fn try_from_cli(value: char) -> anyhow::Result<Self> {
         Ok(match value {
-            '0' => Self::new(
-                // The middle-end optimization level.
-                inkwell::OptimizationLevel::None,
-                // The middle-end size optimization level.
-                SizeLevel::Zero,
-                // The back-end optimization level.
-                inkwell::OptimizationLevel::None,
-            ),
+            '0' => Self::none(),
             '1' => Self::new(
                 inkwell::OptimizationLevel::Less,
                 SizeLevel::Zero,
@@ -84,23 +77,14 @@ impl Settings {
                 // The back-end does not currently distinguish between O1, O2, and O3.
                 inkwell::OptimizationLevel::Default,
             ),
-            '3' => Self::new(
-                inkwell::OptimizationLevel::Aggressive,
-                SizeLevel::Zero,
-                inkwell::OptimizationLevel::Aggressive,
-            ),
+            '3' => Self::cycles(),
             's' => Self::new(
                 // The middle-end optimization level is ignored when SizeLevel is set.
                 inkwell::OptimizationLevel::Default,
                 SizeLevel::S,
                 inkwell::OptimizationLevel::Aggressive,
             ),
-            'z' => Self::new(
-                // The middle-end optimization level is ignored when SizeLevel is set.
-                inkwell::OptimizationLevel::Default,
-                SizeLevel::Z,
-                inkwell::OptimizationLevel::Aggressive,
-            ),
+            'z' => Self::size(),
             char => anyhow::bail!("Unexpected optimization option '{}'", char),
         })
     }
@@ -126,6 +110,7 @@ impl Settings {
     /// Returns the settings for the optimal size.
     pub fn size() -> Self {
         Self::new(
+            // The middle-end optimization level is ignored when SizeLevel is set.
             inkwell::OptimizationLevel::Default,
             SizeLevel::Z,
             inkwell::OptimizationLevel::Aggressive,

crates/resolc/src/resolc/arguments.rs

@@ -4,10 +4,12 @@ use std::collections::BTreeSet;
 use std::path::Path;
 use std::path::PathBuf;
 
-use clap::Parser;
+use clap::{parser::ValueSource, ArgMatches, CommandFactory, Parser};
 use path_slash::PathExt;
 use revive_common::MetadataHash;
-use revive_solc_json_interface::SolcStandardJsonOutputError;
+use revive_solc_json_interface::{
+    PolkaVMDefaultHeapMemorySize, PolkaVMDefaultStackMemorySize, SolcStandardJsonOutputError,
+};
 
 /// Compiles the provided Solidity input files (or use the standard input if no files
 /// are given or "-" is specified as a file name). Outputs the components based on the
@@ -57,8 +59,8 @@ pub struct Arguments {
 
     /// Set the optimization parameter -O[0 | 1 | 2 | 3 | s | z].
     /// Use `3` for best performance and `z` for minimal size.
-    #[arg(short = 'O', long = "optimization")]
-    pub optimization: Option<char>,
+    #[arg(short = 'O', long = "optimization", default_value = "z")]
+    pub optimization: char,
 
     /// Disable the `solc` optimizer.
     /// Use it if your project uses the `MSIZE` instruction, or in other cases.
@@ -116,9 +118,8 @@ pub struct Arguments {
 
     /// Set the metadata hash type.
     /// Available types: `none`, `ipfs`, `keccak256`.
-    /// The default is `keccak256`.
-    #[arg(long)]
-    pub metadata_hash: Option<String>,
+    #[arg(long, default_value_t = MetadataHash::Keccak256)]
+    pub metadata_hash: MetadataHash,
 
     /// Output PolkaVM assembly of the contracts.
     #[arg(long = "asm")]
@@ -185,9 +186,9 @@ pub struct Arguments {
     /// You are incentiviced to keep this value as small as possible:
     /// 1.Increasing the heap size will increase startup costs.
     /// 2.The heap size contributes to the total memory size a contract can use,
-    ///   which includes the contracts code size
-    #[arg(long = "heap-size")]
-    pub heap_size: Option<u32>,
+    ///   which includes the contracts code size.
+    #[arg(long = "heap-size", default_value_t = PolkaVMDefaultHeapMemorySize)]
+    pub heap_size: u32,
 
     /// The contracts total stack size in bytes.
     ///
@@ -198,16 +199,17 @@ pub struct Arguments {
     /// eventually revert execution at runtime!
     ///
     /// You are incentiviced to keep this value as small as possible:
-    /// 1.Increasing the heap size will increase startup costs.
+    /// 1.Increasing the stack size will increase startup costs.
     /// 2.The stack size contributes to the total memory size a contract can use,
-    ///   which includes the contracts code size
-    #[arg(long = "stack-size")]
-    pub stack_size: Option<u32>,
+    ///   which includes the contracts code size.
+    #[arg(long = "stack-size", default_value_t = PolkaVMDefaultStackMemorySize)]
+    pub stack_size: u32,
 }
 
 impl Arguments {
     /// Validates the arguments.
     pub fn validate(&self) -> Vec<SolcStandardJsonOutputError> {
+        let argument_matches = Self::command().get_matches();
         let mut messages = Vec::new();
 
         if self.version && std::env::args().count() > 2 {
@@ -226,9 +228,13 @@ impl Arguments {
             ));
         }
 
-        if self.metadata_hash == Some(MetadataHash::IPFS.to_string()) {
+        if self.metadata_hash == MetadataHash::IPFS {
             messages.push(SolcStandardJsonOutputError::new_error(
-                "`IPFS` metadata hash type is not supported. Please use `keccak256` instead.",
+                format!(
+                    "`{}` metadata hash type is not supported. Please use `{}` instead.",
+                    MetadataHash::IPFS,
+                    MetadataHash::Keccak256,
+                ),
                 None,
                 None,
             ));
@@ -350,34 +356,34 @@ impl Arguments {
             }
             if self.disable_solc_optimizer {
                 messages.push(SolcStandardJsonOutputError::new_error(
-                    "Disabling the solc optimizer must specified in standard JSON input settings.",
+                    "Disabling the solc optimizer must be specified in standard JSON input settings.",
                     None,
                     None,
                 ));
             }
-            if self.optimization.is_some() {
+            if Self::is_argument("optimization", &argument_matches) {
                 messages.push(SolcStandardJsonOutputError::new_error(
-                    "LLVM optimizations must specified in standard JSON input settings.",
+                    "LLVM optimizations must be specified in standard JSON input settings.",
                     None,
                     None,
                 ));
             }
-            if self.metadata_hash.is_some() {
+            if Self::is_argument("metadata_hash", &argument_matches) {
                 messages.push(SolcStandardJsonOutputError::new_error(
-                    "Metadata hash mode must specified in standard JSON input settings.",
+                    "Metadata hash mode must be specified in standard JSON input settings.",
                     None,
                     None,
                 ));
             }
 
-            if self.heap_size.is_some() {
+            if Self::is_argument("heap_size", &argument_matches) {
                 messages.push(SolcStandardJsonOutputError::new_error(
                     "Heap size must be specified in standard JSON input polkavm memory settings.",
                     None,
                     None,
                 ));
             }
-            if self.stack_size.is_some() {
+            if Self::is_argument("stack_size", &argument_matches) {
                 messages.push(SolcStandardJsonOutputError::new_error(
                     "Stack size must be specified in standard JSON input polkavm memory settings.",
                     None,
@@ -403,6 +409,19 @@ impl Arguments {
         messages
     }
 
+    /// Checks whether the value from the argument with the given `argument_id`
+    /// came from the command line.
+    ///
+    /// This can be used for determining if a user has explicitly provided an
+    /// argument, even if it is the same value as the default one.
+    ///
+    /// Panics if the `id` is not a valid argument id.
+    fn is_argument(argument_id: &str, argument_matches: &ArgMatches) -> bool {
+        argument_matches
+            .value_source(argument_id)
+            .is_some_and(|source| source == ValueSource::CommandLine)
+    }
+
     /// Returns remappings from input paths.
     pub fn split_input_files_and_remappings(
         &self,

crates/resolc/src/resolc/main.rs

@@ -1,13 +1,10 @@
 //! Solidity to PolkaVM compiler binary.
 
-use std::str::FromStr;
 use std::{io::Write, path::PathBuf};
 
 use clap::error::ErrorKind;
 use resolc::Process;
-use revive_common::{
-    deserialize_from_str, EVMVersion, MetadataHash, EXIT_CODE_FAILURE, EXIT_CODE_SUCCESS,
-};
+use revive_common::{deserialize_from_str, EVMVersion, EXIT_CODE_FAILURE, EXIT_CODE_SUCCESS};
 use revive_llvm_context::{initialize_llvm, DebugConfig, OptimizerSettings, PolkaVMTarget};
 use revive_solc_json_interface::{
     ResolcWarning, SolcStandardJsonInputSettingsPolkaVMMemory,
@@ -32,7 +29,7 @@ fn main() -> anyhow::Result<()> {
 
     let is_standard_json = arguments.standard_json.is_some();
     let mut messages = arguments.validate();
-    if messages.iter().all(|error| error.severity != "error") {
+    if messages.iter().all(|error| !error.is_error()) {
         if !is_standard_json {
             std::io::stderr()
                 .write_all(
@@ -170,27 +167,21 @@ fn main_inner(
         None => None,
     };
 
-    let mut optimizer_settings = match arguments.optimization {
-        Some(mode) => OptimizerSettings::try_from_cli(mode)?,
-        None => OptimizerSettings::size(),
-    };
+    let mut optimizer_settings = OptimizerSettings::try_from_cli(arguments.optimization)?;
     optimizer_settings.is_verify_each_enabled = arguments.llvm_verify_each;
     optimizer_settings.is_debug_logging_enabled = arguments.llvm_debug_logging;
 
-    let metadata_hash = match arguments.metadata_hash {
-        Some(ref hash_type) => MetadataHash::from_str(hash_type.as_str())?,
-        None => MetadataHash::Keccak256,
-    };
-
-    let memory_config =
-        SolcStandardJsonInputSettingsPolkaVMMemory::new(arguments.heap_size, arguments.stack_size);
+    let memory_config = SolcStandardJsonInputSettingsPolkaVMMemory::new(
+        Some(arguments.heap_size),
+        Some(arguments.stack_size),
+    );
 
     let build = if arguments.yul {
         resolc::yul(
             &solc,
             input_files.as_slice(),
             arguments.libraries.as_slice(),
-            metadata_hash,
+            arguments.metadata_hash,
             messages,
             optimizer_settings,
             debug_config,
@@ -200,7 +191,7 @@ fn main_inner(
     } else if let Some(standard_json) = arguments.standard_json {
         resolc::standard_json(
             &solc,
-            metadata_hash,
+            arguments.metadata_hash,
             messages,
             standard_json.map(PathBuf::from),
             arguments.base_path,
@@ -215,7 +206,7 @@ fn main_inner(
             &solc,
             input_files.as_slice(),
             arguments.libraries.as_slice(),
-            metadata_hash,
+            arguments.metadata_hash,
             messages,
             evm_version,
             format,
@@ -240,7 +231,7 @@ fn main_inner(
             &solc,
             input_files.as_slice(),
             arguments.libraries.as_slice(),
-            metadata_hash,
+            arguments.metadata_hash,
             messages,
             evm_version,
             !arguments.disable_solc_optimizer,

crates/resolc/src/tests/cli/standard_json.rs

@@ -1,6 +1,8 @@
 //! The tests for running resolc with standard JSON option.
 
-use revive_solc_json_interface::SolcStandardJsonOutput;
+use revive_solc_json_interface::{
+    PolkaVMDefaultHeapMemorySize, PolkaVMDefaultStackMemorySize, SolcStandardJsonOutput,
+};
 
 use crate::cli_utils::{
     assert_command_success, assert_equal_exit_codes, execute_resolc_with_stdin_input,
@@ -10,12 +12,31 @@ use crate::cli_utils::{
 
 const JSON_OPTION: &str = "--standard-json";
 
+/// Asserts that the standard JSON output has at least one error with the given `error_message`.
+fn assert_standard_json_errors_contain(output: &SolcStandardJsonOutput, error_message: &str) {
+    assert!(
+        output
+            .errors
+            .iter()
+            .any(|error| error.is_error() && error.message.contains(error_message)),
+        "the standard JSON output should contain the error message `{error_message}`"
+    );
+}
+
+/// Converts valid JSON text to `SolcStandardJsonOutput`.
+fn to_solc_standard_json_output(json_text: &str) -> SolcStandardJsonOutput {
+    serde_json::from_str(json_text).unwrap()
+}
+
 #[test]
 fn runs_with_valid_input_file() {
     let arguments = &[JSON_OPTION];
     let resolc_result = execute_resolc_with_stdin_input(arguments, STANDARD_JSON_CONTRACTS_PATH);
     assert_command_success(&resolc_result, "Providing a valid input file to stdin");
 
+    let resolc_output = to_solc_standard_json_output(&resolc_result.stdout);
+    assert!(!resolc_output.errors.iter().any(|error| error.is_error()));
+
     assert!(
         resolc_result.stdout.contains("contracts"),
         "Expected the output to contain a `contracts` field."
@@ -30,7 +51,7 @@ fn no_evm_codegen_requested() {
     let result = execute_resolc_with_stdin_input(&[JSON_OPTION], STANDARD_JSON_NO_EVM_CODEGEN_PATH);
     assert_command_success(&result, "EVM codegen std json input fixture should build");
 
-    let output: SolcStandardJsonOutput = serde_json::from_str(&result.stdout).unwrap();
+    let output = to_solc_standard_json_output(&result.stdout);
     assert!(!output.errors.iter().any(|msg| msg.severity == "error"))
 }
 
@@ -49,6 +70,91 @@ fn no_evm_codegen_requested_complex() {
         "the EVM codegen std json complext input fixture should build fine",
     );
 
-    let output: SolcStandardJsonOutput = serde_json::from_str(&result.stdout).unwrap();
+    let output = to_solc_standard_json_output(&result.stdout);
     assert!(!output.errors.iter().any(|msg| msg.severity == "error"))
 }
+
+#[test]
+fn invalid_extra_arguments() {
+    struct TestCase<'a> {
+        arguments: Vec<&'a str>,
+        error_message: &'static str,
+    }
+
+    let default_heap_size_string = PolkaVMDefaultHeapMemorySize.to_string();
+    let default_stack_size_string = PolkaVMDefaultStackMemorySize.to_string();
+
+    let cases = vec![
+        TestCase {
+            arguments: vec![JSON_OPTION, "--heap-size", &default_heap_size_string],
+            error_message:
+                "Heap size must be specified in standard JSON input polkavm memory settings",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--stack-size", &default_stack_size_string],
+            error_message:
+                "Stack size must be specified in standard JSON input polkavm memory settings",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--optimization", "z"],
+            error_message: "LLVM optimizations must be specified in standard JSON input settings",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "-Oz"],
+            error_message: "LLVM optimizations must be specified in standard JSON input settings",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--metadata-hash", "ipfs"],
+            error_message:
+                "`ipfs` metadata hash type is not supported. Please use `keccak256` instead",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--metadata-hash", "keccak256"],
+            error_message: "Metadata hash mode must be specified in standard JSON input settings",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--asm"],
+            error_message: "Cannot output assembly or binary outside of JSON in standard JSON mode",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--bin"],
+            error_message: "Cannot output assembly or binary outside of JSON in standard JSON mode",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--libraries", "myLib=0x0000"],
+            error_message: "Libraries must be passed via standard JSON input",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--evm-version", "osaka"],
+            error_message: "EVM version must be passed via standard JSON input",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--output-dir", "tmp"],
+            error_message: "Output directory cannot be used in standard JSON mode",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--overwrite"],
+            error_message: "Overwriting flag cannot be used in standard JSON mode",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--disable-solc-optimizer"],
+            error_message:
+                "Disabling the solc optimizer must be specified in standard JSON input settings",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "-g"],
+            error_message: "Debug info must be requested in standard JSON input polkavm settings",
+        },
+        TestCase {
+            arguments: vec![JSON_OPTION, "--llvm-arg=-riscv"],
+            error_message:
+                "LLVM arguments must be configured in standard JSON input polkavm settings",
+        },
+    ];
+
+    for case in cases {
+        let result = execute_resolc_with_stdin_input(&case.arguments, STANDARD_JSON_CONTRACTS_PATH);
+        let output = to_solc_standard_json_output(&result.stdout);
+        assert_standard_json_errors_contain(&output, case.error_message);
+    }
+}