idk

diff.ts

@@ -6,12 +6,29 @@ function estimateTokens(text: string): number {
 }
 
 const SKIP_EXTENSIONS = new Set([
-  "lock", "sum",
+  "lock",
+  "sum",
   "map",
-  "png", "jpg", "jpeg", "gif", "svg", "ico", "webp",
-  "woff", "woff2", "ttf", "eot",
-  "pdf", "zip", "tar", "gz",
-  "bin", "exe", "dll", "so", "dylib",
+  "png",
+  "jpg",
+  "jpeg",
+  "gif",
+  "svg",
+  "ico",
+  "webp",
+  "woff",
+  "woff2",
+  "ttf",
+  "eot",
+  "pdf",
+  "zip",
+  "tar",
+  "gz",
+  "bin",
+  "exe",
+  "dll",
+  "so",
+  "dylib",
 ]);
 
 const SKIP_PATTERNS = [
@@ -25,35 +42,53 @@ const SKIP_PATTERNS = [
 function shouldSkip(filename: string): boolean {
   if (!filename.includes(".")) return true; // no extension = likely binary
   const ext = filename.split(".").pop()!.toLowerCase();
-  return SKIP_EXTENSIONS.has(ext) || SKIP_PATTERNS.some((p) => p.test(filename));
+  return (
+    SKIP_EXTENSIONS.has(ext) || SKIP_PATTERNS.some((p) => p.test(filename))
+  );
 }
 
 const EXT_TO_LANG: Record<string, string> = {
-  ts: "TypeScript", tsx: "TypeScript",
-  js: "JavaScript", jsx: "JavaScript", mjs: "JavaScript", cjs: "JavaScript",
+  ts: "TypeScript",
+  tsx: "TypeScript",
+  js: "JavaScript",
+  jsx: "JavaScript",
+  mjs: "JavaScript",
+  cjs: "JavaScript",
   py: "Python",
   go: "Go",
   rs: "Rust",
   rb: "Ruby",
   java: "Java",
-  kt: "Kotlin", kts: "Kotlin",
+  kt: "Kotlin",
+  kts: "Kotlin",
   swift: "Swift",
   cs: "C#",
-  cpp: "C++", cc: "C++", cxx: "C++", hpp: "C++",
+  cpp: "C++",
+  cc: "C++",
+  cxx: "C++",
+  hpp: "C++",
   c: "C",
   php: "PHP",
-  sh: "Shell", bash: "Shell", zsh: "Shell",
-  yaml: "YAML", yml: "YAML",
+  sh: "Shell",
+  bash: "Shell",
+  zsh: "Shell",
+  yaml: "YAML",
+  yml: "YAML",
   json: "JSON",
   md: "Markdown",
   sql: "SQL",
-  html: "HTML", htm: "HTML",
-  css: "CSS", scss: "CSS", sass: "CSS", less: "CSS",
+  html: "HTML",
+  htm: "HTML",
+  css: "CSS",
+  scss: "CSS",
+  sass: "CSS",
+  less: "CSS",
   vue: "Vue",
   svelte: "Svelte",
   toml: "TOML",
   xml: "XML",
-  tf: "Terraform", hcl: "HCL",
+  tf: "Terraform",
+  hcl: "HCL",
 };
 
 function detectLanguage(filename: string): string {
@@ -68,6 +103,7 @@ export interface DiffFile {
 }
 
 export function parseDiff(rawDiff: string): DiffFile[] {
+  console.log("Parsing diff...");
   const files: DiffFile[] = [];
 
   const sections = rawDiff.split(/^diff --git /m).filter((s) => s.trim());
@@ -82,7 +118,10 @@ export function parseDiff(rawDiff: string): DiffFile[] {
     let filename: string;
     if (plusLine?.startsWith("+++ b/")) {
       filename = plusLine.slice(6);
-    } else if (plusLine === "+++ /dev/null" && minusLine?.startsWith("--- a/")) {
+    } else if (
+      plusLine === "+++ /dev/null" &&
+      minusLine?.startsWith("--- a/")
+    ) {
       filename = minusLine.slice(6); // deleted file — use old path
     } else {
       continue;
@@ -113,6 +152,9 @@ export function parseDiff(rawDiff: string): DiffFile[] {
 }
 
 export function chunkFile(file: DiffFile, maxTokens: number): DiffChunk[] {
+  console.log(
+    `Chunking file ${file.filename} with ${file.hunks.length} hunks...`,
+  );
   return file.hunks.map(({ header, body }) => {
     let hunk = body;
 
@@ -122,13 +164,21 @@ export function chunkFile(file: DiffFile, maxTokens: number): DiffChunk[] {
       const charBudget = maxTokens * 4;
       let chars = 0;
       let i = 0;
-      while (i < lines.length && chars + (lines[i]?.length ?? 0) + 1 <= charBudget) {
+      while (
+        i < lines.length &&
+        chars + (lines[i]?.length ?? 0) + 1 <= charBudget
+      ) {
         chars += (lines[i]?.length ?? 0) + 1;
         i++;
       }
       hunk = lines.slice(0, i).join("\n");
     }
 
-    return { filename: file.filename, language: file.language, hunk, context: header };
+    return {
+      filename: file.filename,
+      language: file.language,
+      hunk,
+      context: header,
+    };
   });
 }