ai-contextify 0.1.0

package/dist/utils/language.js

@@ -0,0 +1,72 @@
+const LANGUAGE_BY_EXTENSION = {
+    ".ts": "typescript",
+    ".tsx": "tsx",
+    ".js": "javascript",
+    ".jsx": "jsx",
+    ".mjs": "javascript",
+    ".cjs": "javascript",
+    ".py": "python",
+    ".rb": "ruby",
+    ".go": "go",
+    ".rs": "rust",
+    ".java": "java",
+    ".kt": "kotlin",
+    ".swift": "swift",
+    ".c": "c",
+    ".h": "c",
+    ".cpp": "cpp",
+    ".hpp": "cpp",
+    ".cc": "cpp",
+    ".cs": "csharp",
+    ".php": "php",
+    ".sh": "bash",
+    ".bash": "bash",
+    ".zsh": "bash",
+    ".fish": "fish",
+    ".ps1": "powershell",
+    ".sql": "sql",
+    ".yml": "yaml",
+    ".yaml": "yaml",
+    ".toml": "toml",
+    ".ini": "ini",
+    ".env": "dotenv",
+    ".html": "html",
+    ".htm": "html",
+    ".css": "css",
+    ".scss": "scss",
+    ".sass": "sass",
+    ".less": "less",
+    ".vue": "vue",
+    ".svelte": "svelte",
+    ".astro": "astro",
+    ".lua": "lua",
+    ".dart": "dart",
+    ".r": "r",
+    ".scala": "scala",
+    ".clj": "clojure",
+    ".ex": "elixir",
+    ".exs": "elixir",
+    ".erl": "erlang",
+    ".hs": "haskell",
+    ".pl": "perl",
+    ".graphql": "graphql",
+    ".gql": "graphql",
+    ".proto": "protobuf",
+    ".tf": "hcl",
+    ".dockerfile": "dockerfile",
+};
+export function detectLanguage(extension, filename) {
+    const lower = extension.toLowerCase();
+    if (LANGUAGE_BY_EXTENSION[lower])
+        return LANGUAGE_BY_EXTENSION[lower];
+    if (/^dockerfile/i.test(filename))
+        return "dockerfile";
+    if (/^makefile/i.test(filename))
+        return "makefile";
+    return "text";
+}
+const CODE_EXTENSIONS = new Set(Object.keys(LANGUAGE_BY_EXTENSION));
+export function isCodeExtension(extension) {
+    return CODE_EXTENSIONS.has(extension.toLowerCase());
+}
+//# sourceMappingURL=language.js.map

package/dist/utils/language.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"language.js","sourceRoot":"","sources":["../../src/utils/language.ts"],"names":[],"mappings":"AAAA,MAAM,qBAAqB,GAA2B;IACpD,KAAK,EAAE,YAAY;IACnB,MAAM,EAAE,KAAK;IACb,KAAK,EAAE,YAAY;IACnB,MAAM,EAAE,KAAK;IACb,MAAM,EAAE,YAAY;IACpB,MAAM,EAAE,YAAY;IACpB,KAAK,EAAE,QAAQ;IACf,KAAK,EAAE,MAAM;IACb,KAAK,EAAE,IAAI;IACX,KAAK,EAAE,MAAM;IACb,OAAO,EAAE,MAAM;IACf,KAAK,EAAE,QAAQ;IACf,QAAQ,EAAE,OAAO;IACjB,IAAI,EAAE,GAAG;IACT,IAAI,EAAE,GAAG;IACT,MAAM,EAAE,KAAK;IACb,MAAM,EAAE,KAAK;IACb,KAAK,EAAE,KAAK;IACZ,KAAK,EAAE,QAAQ;IACf,MAAM,EAAE,KAAK;IACb,KAAK,EAAE,MAAM;IACb,OAAO,EAAE,MAAM;IACf,MAAM,EAAE,MAAM;IACd,OAAO,EAAE,MAAM;IACf,MAAM,EAAE,YAAY;IACpB,MAAM,EAAE,KAAK;IACb,MAAM,EAAE,MAAM;IACd,OAAO,EAAE,MAAM;IACf,OAAO,EAAE,MAAM;IACf,MAAM,EAAE,KAAK;IACb,MAAM,EAAE,QAAQ;IAChB,OAAO,EAAE,MAAM;IACf,MAAM,EAAE,MAAM;IACd,MAAM,EAAE,KAAK;IACb,OAAO,EAAE,MAAM;IACf,OAAO,EAAE,MAAM;IACf,OAAO,EAAE,MAAM;IACf,MAAM,EAAE,KAAK;IACb,SAAS,EAAE,QAAQ;IACnB,QAAQ,EAAE,OAAO;IACjB,MAAM,EAAE,KAAK;IACb,OAAO,EAAE,MAAM;IACf,IAAI,EAAE,GAAG;IACT,QAAQ,EAAE,OAAO;IACjB,MAAM,EAAE,SAAS;IACjB,KAAK,EAAE,QAAQ;IACf,MAAM,EAAE,QAAQ;IAChB,MAAM,EAAE,QAAQ;IAChB,KAAK,EAAE,SAAS;IAChB,KAAK,EAAE,MAAM;IACb,UAAU,EAAE,SAAS;IACrB,MAAM,EAAE,SAAS;IACjB,QAAQ,EAAE,UAAU;IACpB,KAAK,EAAE,KAAK;IACZ,aAAa,EAAE,YAAY;CAC5B,CAAC;AAEF,MAAM,UAAU,cAAc,CAAC,SAAiB,EAAE,QAAgB;IAChE,MAAM,KAAK,GAAG,SAAS,CAAC,WAAW,EAAE,CAAC;IACtC,IAAI,qBAAqB,CAAC,KAAK,CAAC;QAAE,OAAO,qBAAqB,CAAC,KAAK,CAAC,CAAC;IACtE,IAAI,cAAc,CAAC,IAAI,CAAC,QAAQ,CAAC;QAAE,OAAO,YAAY,CAAC;IACvD,IAAI,YAAY,CAAC,IAAI,CAAC,QAAQ,CAAC;QAAE,OAAO,UAAU,CAAC;IACnD,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,MAAM,eAAe,GAAG,IAAI,GAAG,CAAC,MAAM,CAAC,IAAI,CAAC,qBAAqB,CAAC,CAAC,CAAC;AAEpE,MAAM,UAAU,eAAe,CAAC,SAAiB;IAC/C,OAAO,eAAe,CAAC,GAAG,CAAC,SAAS,CAAC,WAAW,EAAE,CAAC,CAAC;AACtD,CAAC"}

package/dist/utils/logger.d.ts

@@ -0,0 +1,9 @@
+export declare const logger: {
+    info: (msg: string) => void;
+    success: (msg: string) => void;
+    warn: (msg: string) => void;
+    error: (msg: string) => void;
+    dim: (msg: string) => void;
+    step: (msg: string) => void;
+};
+//# sourceMappingURL=logger.d.ts.map

package/dist/utils/logger.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"logger.d.ts","sourceRoot":"","sources":["../../src/utils/logger.ts"],"names":[],"mappings":"AAEA,eAAO,MAAM,MAAM;gBACL,MAAM,KAAG,IAAI;mBACV,MAAM,KAAG,IAAI;gBAChB,MAAM,KAAG,IAAI;iBACZ,MAAM,KAAG,IAAI;eACf,MAAM,KAAG,IAAI;gBACZ,MAAM,KAAG,IAAI;CAC1B,CAAC"}

package/dist/utils/logger.js

@@ -0,0 +1,10 @@
+import chalk from "chalk";
+export const logger = {
+    info: (msg) => console.log(chalk.cyan("ℹ"), msg),
+    success: (msg) => console.log(chalk.green("✔"), msg),
+    warn: (msg) => console.warn(chalk.yellow("⚠"), msg),
+    error: (msg) => console.error(chalk.red("✖"), msg),
+    dim: (msg) => console.log(chalk.dim(msg)),
+    step: (msg) => console.log(chalk.magenta("›"), msg),
+};
+//# sourceMappingURL=logger.js.map

package/dist/utils/logger.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"logger.js","sourceRoot":"","sources":["../../src/utils/logger.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,MAAM,OAAO,CAAC;AAE1B,MAAM,CAAC,MAAM,MAAM,GAAG;IACpB,IAAI,EAAE,CAAC,GAAW,EAAQ,EAAE,CAAC,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,EAAE,GAAG,CAAC;IAC9D,OAAO,EAAE,CAAC,GAAW,EAAQ,EAAE,CAAC,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,KAAK,CAAC,GAAG,CAAC,EAAE,GAAG,CAAC;IAClE,IAAI,EAAE,CAAC,GAAW,EAAQ,EAAE,CAAC,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,GAAG,CAAC,EAAE,GAAG,CAAC;IACjE,KAAK,EAAE,CAAC,GAAW,EAAQ,EAAE,CAAC,OAAO,CAAC,KAAK,CAAC,KAAK,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE,GAAG,CAAC;IAChE,GAAG,EAAE,CAAC,GAAW,EAAQ,EAAE,CAAC,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC;IACvD,IAAI,EAAE,CAAC,GAAW,EAAQ,EAAE,CAAC,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,OAAO,CAAC,GAAG,CAAC,EAAE,GAAG,CAAC;CAClE,CAAC"}

package/dist/utils/tokens.d.ts

@@ -0,0 +1,9 @@
+/**
+ * Cheap, dependency-free token estimator.
+ * Heuristic: ~4 characters per token for English/code,
+ * with a small penalty for whitespace-heavy text.
+ * Good enough for budgeting context windows without bundling a tokenizer.
+ */
+export declare function estimateTokens(text: string): number;
+export declare function formatTokenCount(n: number): string;
+//# sourceMappingURL=tokens.d.ts.map

package/dist/utils/tokens.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"tokens.d.ts","sourceRoot":"","sources":["../../src/utils/tokens.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AACH,wBAAgB,cAAc,CAAC,IAAI,EAAE,MAAM,GAAG,MAAM,CAMnD;AAED,wBAAgB,gBAAgB,CAAC,CAAC,EAAE,MAAM,GAAG,MAAM,CAIlD"}

package/dist/utils/tokens.js

@@ -0,0 +1,22 @@
+/**
+ * Cheap, dependency-free token estimator.
+ * Heuristic: ~4 characters per token for English/code,
+ * with a small penalty for whitespace-heavy text.
+ * Good enough for budgeting context windows without bundling a tokenizer.
+ */
+export function estimateTokens(text) {
+    if (!text)
+        return 0;
+    const charCount = text.length;
+    const whitespace = (text.match(/\s/g) ?? []).length;
+    const effective = charCount - whitespace * 0.3;
+    return Math.max(1, Math.ceil(effective / 4));
+}
+export function formatTokenCount(n) {
+    if (n < 1_000)
+        return `${n}`;
+    if (n < 1_000_000)
+        return `${(n / 1_000).toFixed(1)}k`;
+    return `${(n / 1_000_000).toFixed(2)}M`;
+}
+//# sourceMappingURL=tokens.js.map

package/dist/utils/tokens.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"tokens.js","sourceRoot":"","sources":["../../src/utils/tokens.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AACH,MAAM,UAAU,cAAc,CAAC,IAAY;IACzC,IAAI,CAAC,IAAI;QAAE,OAAO,CAAC,CAAC;IACpB,MAAM,SAAS,GAAG,IAAI,CAAC,MAAM,CAAC;IAC9B,MAAM,UAAU,GAAG,CAAC,IAAI,CAAC,KAAK,CAAC,KAAK,CAAC,IAAI,EAAE,CAAC,CAAC,MAAM,CAAC;IACpD,MAAM,SAAS,GAAG,SAAS,GAAG,UAAU,GAAG,GAAG,CAAC;IAC/C,OAAO,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,IAAI,CAAC,SAAS,GAAG,CAAC,CAAC,CAAC,CAAC;AAC/C,CAAC;AAED,MAAM,UAAU,gBAAgB,CAAC,CAAS;IACxC,IAAI,CAAC,GAAG,KAAK;QAAE,OAAO,GAAG,CAAC,EAAE,CAAC;IAC7B,IAAI,CAAC,GAAG,SAAS;QAAE,OAAO,GAAG,CAAC,CAAC,GAAG,KAAK,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,GAAG,CAAC;IACvD,OAAO,GAAG,CAAC,CAAC,GAAG,SAAS,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,GAAG,CAAC;AAC1C,CAAC"}

package/package.json

@@ -0,0 +1,62 @@
+{
+  "name": "ai-contextify",
+  "version": "0.1.0",
+  "description": "Convert files and entire projects into optimized context for LLMs like Claude, ChatGPT and Gemini.",
+  "keywords": [
+    "ai",
+    "llm",
+    "context",
+    "claude",
+    "chatgpt",
+    "gemini",
+    "rag",
+    "cli",
+    "markdown",
+    "pdf",
+    "docx"
+  ],
+  "license": "MIT",
+  "author": "xean._. <jean24pain@gmail.com>",
+  "type": "module",
+  "main": "dist/index.js",
+  "module": "dist/index.js",
+  "types": "dist/index.d.ts",
+  "bin": {
+    "ai-contextify": "dist/cli.js"
+  },
+  "files": [
+    "dist",
+    "README.md",
+    "LICENSE"
+  ],
+  "engines": {
+    "node": ">=18"
+  },
+  "scripts": {
+    "dev": "tsx src/cli.ts",
+    "build": "tsc -p tsconfig.json && chmod +x dist/cli.js || true",
+    "start": "node dist/cli.js",
+    "clean": "rimraf dist",
+    "prepublishOnly": "npm run clean && npm run build"
+  },
+  "dependencies": {
+    "chalk": "^5.3.0",
+    "commander": "^12.1.0",
+    "fast-glob": "^3.3.2",
+    "fs-extra": "^11.2.0",
+    "mammoth": "^1.8.0",
+    "ora": "^8.1.0",
+    "pdf-parse": "^1.1.1",
+    "remark": "^15.0.1",
+    "remark-gfm": "^4.0.0",
+    "strip-markdown": "^6.0.0"
+  },
+  "devDependencies": {
+    "@types/fs-extra": "^11.0.4",
+    "@types/node": "^20.14.0",
+    "@types/pdf-parse": "^1.1.4",
+    "rimraf": "^5.0.7",
+    "tsx": "^4.16.0",
+    "typescript": "^5.5.0"
+  }
+}