Revert "Revert "feat: add geometry to database args"" (#12165)

crates/node/core/src/args/database.rs

@@ -1,6 +1,6 @@
 //! clap [Args](clap::Args) for database configuration
 
-use std::time::Duration;
+use std::{fmt, str::FromStr, time::Duration};
 
 use crate::version::default_client_version;
 use clap::{
@@ -22,6 +22,12 @@ pub struct DatabaseArgs {
     /// NFS volume.
     #[arg(long = "db.exclusive")]
     pub exclusive: Option<bool>,
+    /// Maximum database size (e.g., 4TB, 8MB)
+    #[arg(long = "db.max-size", value_parser = parse_byte_size)]
+    pub max_size: Option<usize>,
+    /// Database growth step (e.g., 4GB, 4KB)
+    #[arg(long = "db.growth-step", value_parser = parse_byte_size)]
+    pub growth_step: Option<usize>,
     /// Read transaction timeout in seconds, 0 means no timeout.
     #[arg(long = "db.read-transaction-timeout")]
     pub read_transaction_timeout: Option<u64>,
@@ -33,8 +39,9 @@ impl DatabaseArgs {
         self.get_database_args(default_client_version())
     }
 
-    /// Returns the database arguments with configured log level and given client version.
-    pub const fn get_database_args(
+    /// Returns the database arguments with configured log level, client version,
+    /// max read transaction duration, and geometry.
+    pub fn get_database_args(
         &self,
         client_version: ClientVersion,
     ) -> reth_db::mdbx::DatabaseArguments {
@@ -48,6 +55,8 @@ impl DatabaseArgs {
             .with_log_level(self.log_level)
             .with_exclusive(self.exclusive)
             .with_max_read_transaction_duration(max_read_transaction_duration)
+            .with_geometry_max_size(self.max_size)
+            .with_growth_step(self.growth_step)
     }
 }
 
@@ -89,10 +98,84 @@ impl TypedValueParser for LogLevelValueParser {
         Some(Box::new(values))
     }
 }
+
+/// Size in bytes.
+#[derive(Debug, Default, Clone, Copy, PartialEq, Eq)]
+pub struct ByteSize(pub usize);
+
+impl From<ByteSize> for usize {
+    fn from(s: ByteSize) -> Self {
+        s.0
+    }
+}
+
+impl FromStr for ByteSize {
+    type Err = String;
+
+    fn from_str(s: &str) -> Result<Self, Self::Err> {
+        let s = s.trim().to_uppercase();
+        let parts: Vec<&str> = s.split_whitespace().collect();
+
+        let (num_str, unit) = match parts.len() {
+            1 => {
+                let (num, unit) =
+                    s.split_at(s.find(|c: char| c.is_alphabetic()).unwrap_or(s.len()));
+                (num, unit)
+            }
+            2 => (parts[0], parts[1]),
+            _ => {
+                return Err("Invalid format. Use '<number><unit>' or '<number> <unit>'.".to_string())
+            }
+        };
+
+        let num: usize = num_str.parse().map_err(|_| "Invalid number".to_string())?;
+
+        let multiplier = match unit {
+            "B" | "" => 1, // Assume bytes if no unit is specified
+            "KB" => 1024,
+            "MB" => 1024 * 1024,
+            "GB" => 1024 * 1024 * 1024,
+            "TB" => 1024 * 1024 * 1024 * 1024,
+            _ => return Err(format!("Invalid unit: {}. Use B, KB, MB, GB, or TB.", unit)),
+        };
+
+        Ok(Self(num * multiplier))
+    }
+}
+
+impl fmt::Display for ByteSize {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        const KB: usize = 1024;
+        const MB: usize = KB * 1024;
+        const GB: usize = MB * 1024;
+        const TB: usize = GB * 1024;
+
+        let (size, unit) = if self.0 >= TB {
+            (self.0 as f64 / TB as f64, "TB")
+        } else if self.0 >= GB {
+            (self.0 as f64 / GB as f64, "GB")
+        } else if self.0 >= MB {
+            (self.0 as f64 / MB as f64, "MB")
+        } else if self.0 >= KB {
+            (self.0 as f64 / KB as f64, "KB")
+        } else {
+            (self.0 as f64, "B")
+        };
+
+        write!(f, "{:.2}{}", size, unit)
+    }
+}
+
+/// Value parser function that supports various formats.
+fn parse_byte_size(s: &str) -> Result<usize, String> {
+    s.parse::<ByteSize>().map(Into::into)
+}
+
 #[cfg(test)]
 mod tests {
     use super::*;
     use clap::Parser;
+    use reth_db::mdbx::{GIGABYTE, KILOBYTE, MEGABYTE, TERABYTE};
 
     /// A helper type to parse Args more easily
     #[derive(Parser)]
@@ -108,6 +191,101 @@ mod tests {
         assert_eq!(args, default_args);
     }
 
+    #[test]
+    fn test_command_parser_with_valid_max_size() {
+        let cmd = CommandParser::<DatabaseArgs>::try_parse_from([
+            "reth",
+            "--db.max-size",
+            "4398046511104",
+        ])
+        .unwrap();
+        assert_eq!(cmd.args.max_size, Some(TERABYTE * 4));
+    }
+
+    #[test]
+    fn test_command_parser_with_invalid_max_size() {
+        let result =
+            CommandParser::<DatabaseArgs>::try_parse_from(["reth", "--db.max-size", "invalid"]);
+        assert!(result.is_err());
+    }
+
+    #[test]
+    fn test_command_parser_with_valid_growth_step() {
+        let cmd = CommandParser::<DatabaseArgs>::try_parse_from([
+            "reth",
+            "--db.growth-step",
+            "4294967296",
+        ])
+        .unwrap();
+        assert_eq!(cmd.args.growth_step, Some(GIGABYTE * 4));
+    }
+
+    #[test]
+    fn test_command_parser_with_invalid_growth_step() {
+        let result =
+            CommandParser::<DatabaseArgs>::try_parse_from(["reth", "--db.growth-step", "invalid"]);
+        assert!(result.is_err());
+    }
+
+    #[test]
+    fn test_command_parser_with_valid_max_size_and_growth_step_from_str() {
+        let cmd = CommandParser::<DatabaseArgs>::try_parse_from([
+            "reth",
+            "--db.max-size",
+            "2TB",
+            "--db.growth-step",
+            "1GB",
+        ])
+        .unwrap();
+        assert_eq!(cmd.args.max_size, Some(TERABYTE * 2));
+        assert_eq!(cmd.args.growth_step, Some(GIGABYTE));
+
+        let cmd = CommandParser::<DatabaseArgs>::try_parse_from([
+            "reth",
+            "--db.max-size",
+            "12MB",
+            "--db.growth-step",
+            "2KB",
+        ])
+        .unwrap();
+        assert_eq!(cmd.args.max_size, Some(MEGABYTE * 12));
+        assert_eq!(cmd.args.growth_step, Some(KILOBYTE * 2));
+
+        // with spaces
+        let cmd = CommandParser::<DatabaseArgs>::try_parse_from([
+            "reth",
+            "--db.max-size",
+            "12 MB",
+            "--db.growth-step",
+            "2 KB",
+        ])
+        .unwrap();
+        assert_eq!(cmd.args.max_size, Some(MEGABYTE * 12));
+        assert_eq!(cmd.args.growth_step, Some(KILOBYTE * 2));
+
+        let cmd = CommandParser::<DatabaseArgs>::try_parse_from([
+            "reth",
+            "--db.max-size",
+            "1073741824",
+            "--db.growth-step",
+            "1048576",
+        ])
+        .unwrap();
+        assert_eq!(cmd.args.max_size, Some(GIGABYTE));
+        assert_eq!(cmd.args.growth_step, Some(MEGABYTE));
+    }
+
+    #[test]
+    fn test_command_parser_max_size_and_growth_step_from_str_invalid_unit() {
+        let result =
+            CommandParser::<DatabaseArgs>::try_parse_from(["reth", "--db.growth-step", "1 PB"]);
+        assert!(result.is_err());
+
+        let result =
+            CommandParser::<DatabaseArgs>::try_parse_from(["reth", "--db.max-size", "2PB"]);
+        assert!(result.is_err());
+    }
+
     #[test]
     fn test_possible_values() {
         // Initialize the LogLevelValueParser