From 2b882870f7073fef070b9c2e7ebe03bd06b9334d Mon Sep 17 00:00:00 2001
From: SangheeSon <sonsanghee3@gmail.com>
Date: Sat, 1 Nov 2025 08:35:54 +0900
Subject: [PATCH] feat(cli): add `--stats` flag to show token savings (#51)

* feat(cli): add --stats flag to show token efficiency

- Add --stats boolean flag to display token count comparison
- Calculate approximate tokens using char length / 4 heuristic
- Show JSON vs TOON token counts with savings percentage
- Opt-in feature, default behavior unchanged

* feat: use tokenx for more accurate estimates

---------

Co-authored-by: Johann Schopplich <mail@johannschopplich.com>
---
 README.md        |  4 ++++
 cli/package.json |  3 +++
 cli/src/index.ts | 19 +++++++++++++++++++
 pnpm-lock.yaml   |  9 +++++++++
 4 files changed, 35 insertions(+)

diff --git a/README.md b/README.md
index 0def22e..a5dfd63 100644
--- a/README.md
+++ b/README.md
@@ -499,11 +499,15 @@ toon input.json
 | `--delimiter <char>` | Array delimiter: `,` (comma), `\t` (tab), `\|` (pipe) |
 | `--indent <number>` | Indentation size (default: `2`) |
 | `--length-marker` | Add `#` prefix to array lengths (e.g., `items[#3]`) |
+| `--stats` | Show token count estimates and savings (encode only) |
 | `--no-strict` | Disable strict validation when decoding |
 
 ### Examples
 
 ```bash
+# Show token savings when encoding
+toon data.json --stats -o output.toon
+
 # Tab-separated output (often more token-efficient)
 toon data.json --delimiter "\t" -o output.toon
 
diff --git a/cli/package.json b/cli/package.json
index 04a9499..89b80b1 100644
--- a/cli/package.json
+++ b/cli/package.json
@@ -8,5 +8,8 @@
   "dependencies": {
     "citty": "^0.1.6",
     "consola": "^3.4.2"
+  },
+  "devDependencies": {
+    "tokenx": "^1.2.0"
   }
 }
diff --git a/cli/src/index.ts b/cli/src/index.ts
index 4c76857..4bbbf88 100644
--- a/cli/src/index.ts
+++ b/cli/src/index.ts
@@ -4,6 +4,7 @@ import * as path from 'node:path'
 import process from 'node:process'
 import { defineCommand, runMain } from 'citty'
 import { consola } from 'consola'
+import { estimateTokenCount } from 'tokenx'
 import { name, version } from '../../package.json' with { type: 'json' }
 import { decode, DEFAULT_DELIMITER, DELIMITERS, encode } from '../../src'
 
@@ -54,6 +55,11 @@ const main = defineCommand({
       description: 'Enable strict mode for decoding',
       default: true,
     },
+    stats: {
+      type: 'boolean',
+      description: 'Show token statistics',
+      default: false,
+    },
   },
   async run({ args }) {
     const input = args.input || args._[0]
@@ -86,6 +92,7 @@ const main = defineCommand({
           delimiter: delimiter as Delimiter,
           indent,
           lengthMarker: args.lengthMarker === true ? '#' : false,
+          printStats: args.stats === true,
         })
       }
       else {
@@ -131,6 +138,7 @@ async function encodeToToon(config: {
   delimiter: Delimiter
   indent: number
   lengthMarker: NonNullable<EncodeOptions['lengthMarker']>
+  printStats: boolean
 }) {
   const jsonContent = await fsp.readFile(config.input, 'utf-8')
 
@@ -159,6 +167,17 @@ async function encodeToToon(config: {
   else {
     console.log(toonOutput)
   }
+
+  if (config.printStats) {
+    const jsonTokens = estimateTokenCount(jsonContent)
+    const toonTokens = estimateTokenCount(toonOutput)
+    const diff = jsonTokens - toonTokens
+    const percent = ((diff / jsonTokens) * 100).toFixed(1)
+
+    console.log()
+    consola.info(`Token estimates: ~${jsonTokens} (JSON) → ~${toonTokens} (TOON)`)
+    consola.success(`Saved ~${diff} tokens (-${percent}%)`)
+  }
 }
 
 async function decodeToJson(config: {
diff --git a/pnpm-lock.yaml b/pnpm-lock.yaml
index 275787a..e600e7e 100644
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
@@ -95,6 +95,10 @@ importers:
       consola:
         specifier: ^3.4.2
         version: 3.4.2
+    devDependencies:
+      tokenx:
+        specifier: ^1.2.0
+        version: 1.2.0
 
 packages:
 
@@ -2200,6 +2204,9 @@ packages:
     resolution: {integrity: sha512-41wJyvKep3yT2tyPqX/4blcfybknGB4D+oETKLs7Q76UiPqRpUJK3hr1nxelyYO0PHKVzJwlu0aCeEAsGI6rpw==}
     engines: {node: '>=20'}
 
+  tokenx@1.2.0:
+    resolution: {integrity: sha512-x4bRrL23b22H+EqW2pbhIkkt3ouj27ZGmAS1QoIqpocEO4m0sAl2H1M4L1UzKqleikY4U9lz/TbEw4jeG8tm2A==}
+
   toml-eslint-parser@0.10.0:
     resolution: {integrity: sha512-khrZo4buq4qVmsGzS5yQjKe/WsFvV8fGfOjDQN0q4iy9FjRfPWRgTFrU8u1R2iu/SfWLhY9WnCi4Jhdrcbtg+g==}
     engines: {node: ^12.22.0 || ^14.17.0 || >=16.0.0}
@@ -4706,6 +4713,8 @@ snapshots:
       '@sindresorhus/base62': 1.0.0
       reserved-identifiers: 1.2.0
 
+  tokenx@1.2.0: {}
+
   toml-eslint-parser@0.10.0:
     dependencies:
       eslint-visitor-keys: 3.4.3