Format large numbers in a more readable way. (#2046)

- In the bottom line of the TUI, print the number of tokens to 3 sigfigs with an SI suffix, e.g. "1.23K". - Elsewhere where we print a number, I figure it's worthwhile to print the exact number, because e.g. it's a summary of your session. Here we print the numbers comma-separated.
2025-09-08 14:48:48 -07:00
parent 4c46490e53
commit 18330c2362
8 changed files with 196 additions and 11 deletions
--- a/codex-rs/Cargo.lock
+++ b/codex-rs/Cargo.lock
@@ -917,6 +917,8 @@ name = "codex-protocol"
 version = "0.0.0"
 dependencies = [
 "base64 0.22.1",
+ "icu_decimal",
+ "icu_locale_core",
 "mcp-types",
 "mime_guess",
 "pretty_assertions",
@@ -926,6 +928,7 @@ dependencies = [
 "serde_with",
 "strum 0.27.2",
 "strum_macros 0.27.2",
+ "sys-locale",
 "tracing",
 "ts-rs",
 "uuid",
@@ -1758,6 +1761,17 @@ dependencies = [
 "winapi",
 ]

+[[package]]
+name = "fixed_decimal"
+version = "0.7.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "35943d22b2f19c0cb198ecf915910a8158e94541c89dcc63300d7799d46c2c5e"
+dependencies = [
+ "displaydoc",
+ "smallvec",
+ "writeable",
+]
+
 [[package]]
 name = "fixedbitset"
 version = "0.4.2"
@@ -2257,6 +2271,45 @@ dependencies = [
 "zerovec",
 ]

+[[package]]
+name = "icu_decimal"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fec61c43fdc4e368a9f450272833123a8ef0d7083a44597660ce94d791b8a2e2"
+dependencies = [
+ "displaydoc",
+ "fixed_decimal",
+ "icu_decimal_data",
+ "icu_locale",
+ "icu_locale_core",
+ "icu_provider",
+ "tinystr",
+ "writeable",
+ "zerovec",
+]
+
+[[package]]
+name = "icu_decimal_data"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b70963bc35f9bdf1bc66a5c1f458f4991c1dc71760e00fa06016b2c76b2738d5"
+
+[[package]]
+name = "icu_locale"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6ae5921528335e91da1b6c695dbf1ec37df5ac13faa3f91e5640be93aa2fbefd"
+dependencies = [
+ "displaydoc",
+ "icu_collections",
+ "icu_locale_core",
+ "icu_locale_data",
+ "icu_provider",
+ "potential_utf",
+ "tinystr",
+ "zerovec",
+]
+
 [[package]]
 name = "icu_locale_core"
 version = "2.0.0"
@@ -2270,6 +2323,12 @@ dependencies = [
 "zerovec",
 ]

+[[package]]
+name = "icu_locale_data"
+version = "2.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4fdef0c124749d06a743c69e938350816554eb63ac979166590e2b4ee4252765"
+
 [[package]]
 name = "icu_normalizer"
 version = "2.0.0"
@@ -3514,6 +3573,7 @@ version = "0.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e5a7c30837279ca13e7c867e9e40053bc68740f988cb07f7ca6df43cc734b585"
 dependencies = [
+ "serde",
 "zerovec",
 ]

@@ -4850,6 +4910,15 @@ dependencies = [
 "yaml-rust",
 ]

+[[package]]
+name = "sys-locale"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8eab9a99a024a169fe8a903cf9d4a3b3601109bcc13bd9e3c6fff259138626c4"
+dependencies = [
+ "libc",
+]
+
 [[package]]
 name = "system-configuration"
 version = "0.6.1"
--- a/codex-rs/exec/src/event_processor_with_human_output.rs
+++ b/codex-rs/exec/src/event_processor_with_human_output.rs
@@ -26,6 +26,7 @@ use codex_core::protocol::TurnAbortReason;
 use codex_core::protocol::TurnDiffEvent;
 use codex_core::protocol::WebSearchBeginEvent;
 use codex_core::protocol::WebSearchEndEvent;
+use codex_protocol::num_format::format_with_separators;
 use owo_colors::OwoColorize;
 use owo_colors::Style;
 use shlex::try_join;
@@ -194,7 +195,7 @@ impl EventProcessor for EventProcessorWithHumanOutput {
                    ts_println!(
                        self,
                        "tokens used: {}",
-                        usage_info.total_token_usage.blended_total()
+                        format_with_separators(usage_info.total_token_usage.blended_total())
                    );
                }
            }
--- a/codex-rs/protocol/Cargo.toml
+++ b/codex-rs/protocol/Cargo.toml
@@ -12,6 +12,8 @@ workspace = true

 [dependencies]
 base64 = "0.22.1"
+icu_decimal = "2.0.0"
+icu_locale_core = "2.0.0"
 mcp-types = { path = "../mcp-types" }
 mime_guess = "2.0.5"
 serde = { version = "1", features = ["derive"] }
@@ -20,6 +22,7 @@ serde_json = "1"
 serde_with = { version = "3.14.0", features = ["macros", "base64"] }
 strum = "0.27.2"
 strum_macros = "0.27.2"
+sys-locale = "0.3.2"
 tracing = "0.1.41"
 ts-rs = { version = "11", features = ["uuid-impl", "serde-json-impl"] }
 uuid = { version = "1", features = ["serde", "v4"] }
--- a/codex-rs/protocol/src/lib.rs
+++ b/codex-rs/protocol/src/lib.rs
@@ -3,6 +3,7 @@ pub mod custom_prompts;
 pub mod mcp_protocol;
 pub mod message_history;
 pub mod models;
+pub mod num_format;
 pub mod parse_command;
 pub mod plan_tool;
 pub mod protocol;
--- a/codex-rs/protocol/src/num_format.rs
+++ b/codex-rs/protocol/src/num_format.rs
@@ -0,0 +1,98 @@
+use std::sync::OnceLock;
+
+use icu_decimal::DecimalFormatter;
+use icu_decimal::input::Decimal;
+use icu_decimal::options::DecimalFormatterOptions;
+use icu_locale_core::Locale;
+
+fn make_local_formatter() -> Option<DecimalFormatter> {
+    let loc: Locale = sys_locale::get_locale()?.parse().ok()?;
+    DecimalFormatter::try_new(loc.into(), DecimalFormatterOptions::default()).ok()
+}
+
+fn make_en_us_formatter() -> DecimalFormatter {
+    #![allow(clippy::expect_used)]
+    let loc: Locale = "en-US".parse().expect("en-US wasn't a valid locale");
+    DecimalFormatter::try_new(loc.into(), DecimalFormatterOptions::default())
+        .expect("en-US wasn't a valid locale")
+}
+
+fn formatter() -> &'static DecimalFormatter {
+    static FORMATTER: OnceLock<DecimalFormatter> = OnceLock::new();
+    FORMATTER.get_or_init(|| make_local_formatter().unwrap_or_else(make_en_us_formatter))
+}
+
+/// Format a u64 with locale-aware digit separators (e.g. "12345" -> "12,345"
+/// for en-US).
+pub fn format_with_separators(n: u64) -> String {
+    formatter().format(&Decimal::from(n)).to_string()
+}
+
+fn format_si_suffix_with_formatter(n: u64, formatter: &DecimalFormatter) -> String {
+    if n < 1000 {
+        return formatter.format(&Decimal::from(n)).to_string();
+    }
+
+    // Format `n / scale` with the requested number of fractional digits.
+    let format_scaled = |n: u64, scale: u64, frac_digits: u32| -> String {
+        let value = n as f64 / scale as f64;
+        let scaled: u64 = (value * 10f64.powi(frac_digits as i32)).round() as u64;
+        let mut dec = Decimal::from(scaled);
+        dec.multiply_pow10(-(frac_digits as i16));
+        formatter.format(&dec).to_string()
+    };
+
+    const UNITS: [(u64, &str); 3] = [(1_000, "K"), (1_000_000, "M"), (1_000_000_000, "G")];
+    let f = n as f64;
+    for &(scale, suffix) in &UNITS {
+        if (100.0 * f / scale as f64).round() < 1000.0 {
+            return format!("{}{}", format_scaled(n, scale, 2), suffix);
+        } else if (10.0 * f / scale as f64).round() < 1000.0 {
+            return format!("{}{}", format_scaled(n, scale, 1), suffix);
+        } else if (f / scale as f64).round() < 1000.0 {
+            return format!("{}{}", format_scaled(n, scale, 0), suffix);
+        }
+    }
+
+    // Above 1000G, keep whole‑G precision.
+    format!(
+        "{}G",
+        format_with_separators(((n as f64) / 1e9).round() as u64)
+    )
+}
+
+/// Format token counts to 3 significant figures, using base-10 SI suffixes.
+///
+/// Examples (en-US):
+///   - 999 -> "999"
+///   - 1200 -> "1.20K"
+///   - 123456789 -> "123M"
+pub fn format_si_suffix(n: u64) -> String {
+    format_si_suffix_with_formatter(n, formatter())
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn kmg() {
+        let formatter = make_en_us_formatter();
+        let fmt = |n: u64| format_si_suffix_with_formatter(n, &formatter);
+        assert_eq!(fmt(0), "0");
+        assert_eq!(fmt(999), "999");
+        assert_eq!(fmt(1_000), "1.00K");
+        assert_eq!(fmt(1_200), "1.20K");
+        assert_eq!(fmt(10_000), "10.0K");
+        assert_eq!(fmt(100_000), "100K");
+        assert_eq!(fmt(999_500), "1.00M");
+        assert_eq!(fmt(1_000_000), "1.00M");
+        assert_eq!(fmt(1_234_000), "1.23M");
+        assert_eq!(fmt(12_345_678), "12.3M");
+        assert_eq!(fmt(999_950_000), "1.00G");
+        assert_eq!(fmt(1_000_000_000), "1.00G");
+        assert_eq!(fmt(1_234_000_000), "1.23G");
+        // Above 1000G we keep whole‑G precision (no higher unit supported here).
+        assert_eq!(fmt(1_234_000_000_000), "1,234G");
+    }
+}
--- a/codex-rs/protocol/src/protocol.rs
+++ b/codex-rs/protocol/src/protocol.rs
@@ -16,6 +16,7 @@ use crate::custom_prompts::CustomPrompt;
 use crate::mcp_protocol::ConversationId;
 use crate::message_history::HistoryEntry;
 use crate::models::ResponseItem;
+use crate::num_format::format_with_separators;
 use crate::parse_command::ParsedCommand;
 use crate::plan_tool::UpdatePlanArgs;
 use mcp_types::CallToolResult;
@@ -645,19 +646,26 @@ impl From<TokenUsage> for FinalOutput {
 impl fmt::Display for FinalOutput {
    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
        let token_usage = &self.token_usage;
+
        write!(
            f,
            "Token usage: total={} input={}{} output={}{}",
-            token_usage.blended_total(),
-            token_usage.non_cached_input(),
+            format_with_separators(token_usage.blended_total()),
+            format_with_separators(token_usage.non_cached_input()),
            if token_usage.cached_input() > 0 {
-                format!(" (+ {} cached)", token_usage.cached_input())
+                format!(
+                    " (+ {} cached)",
+                    format_with_separators(token_usage.cached_input())
+                )
            } else {
                String::new()
            },
-            token_usage.output_tokens,
+            format_with_separators(token_usage.output_tokens),
            if token_usage.reasoning_output_tokens > 0 {
-                format!(" (reasoning {})", token_usage.reasoning_output_tokens)
+                format!(
+                    " (reasoning {})",
+                    format_with_separators(token_usage.reasoning_output_tokens)
+                )
            } else {
                String::new()
            }
--- a/codex-rs/tui/src/bottom_pane/chat_composer.rs
+++ b/codex-rs/tui/src/bottom_pane/chat_composer.rs
@@ -1,4 +1,5 @@
 use codex_core::protocol::TokenUsageInfo;
+use codex_protocol::num_format::format_si_suffix;
 use crossterm::event::KeyCode;
 use crossterm::event::KeyEvent;
 use crossterm::event::KeyEventKind;
@@ -1276,8 +1277,11 @@ impl WidgetRef for ChatComposer {
                    let token_usage = &token_usage_info.total_token_usage;
                    hint.push("   ".into());
                    hint.push(
-                        Span::from(format!("{} tokens used", token_usage.blended_total()))
-                            .style(Style::default().add_modifier(Modifier::DIM)),
+                        Span::from(format!(
+                            "{} tokens used",
+                            format_si_suffix(token_usage.blended_total())
+                        ))
+                        .style(Style::default().add_modifier(Modifier::DIM)),
                    );
                    let last_token_usage = &token_usage_info.last_token_usage;
                    if let Some(context_window) = token_usage_info.model_context_window {
--- a/codex-rs/tui/src/history_cell.rs
+++ b/codex-rs/tui/src/history_cell.rs
@@ -28,6 +28,7 @@ use codex_core::protocol::SandboxPolicy;
 use codex_core::protocol::SessionConfiguredEvent;
 use codex_core::protocol::TokenUsage;
 use codex_protocol::mcp_protocol::ConversationId;
+use codex_protocol::num_format::format_with_separators;
 use codex_protocol::parse_command::ParsedCommand;
 use image::DynamicImage;
 use image::ImageReader;
@@ -964,7 +965,7 @@ pub(crate) fn new_status_output(
    // Input: <input> [+ <cached> cached]
    let mut input_line_spans: Vec<Span<'static>> = vec![
        "  • Input: ".into(),
-        usage.non_cached_input().to_string().into(),
+        format_with_separators(usage.non_cached_input()).into(),
    ];
    if usage.cached_input_tokens > 0 {
        let cached = usage.cached_input_tokens;
@@ -974,12 +975,12 @@ pub(crate) fn new_status_output(
    // Output: <output>
    lines.push(Line::from(vec![
        "  • Output: ".into(),
-        usage.output_tokens.to_string().into(),
+        format_with_separators(usage.output_tokens).into(),
    ]));
    // Total: <total>
    lines.push(Line::from(vec![
        "  • Total: ".into(),
-        usage.blended_total().to_string().into(),
+        format_with_separators(usage.blended_total()).into(),
    ]));

    PlainHistoryCell { lines }