agent/pi
2
0
Fork 0
mirror of https://github.com/earendil-works/pi.git synced 2026-06-18 15:54:04 +08:00
Code Issues Packages Projects Releases Wiki Activity

refactor(coding-agent): resize images from bytes

Browse Source
This commit is contained in:
Armin Ronacher
2026-05-19 23:11:51 +02:00
parent 956f20a572
commit 2d40f5aa71
4 changed files with 42 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files
packages/coding-agent/src/cli/file-processor.ts
+2 -3
View File
@@ -50,13 +50,12 @@ export async function processFileArguments(fileArgs: string[], options?: Process
if (mimeType) {
// Handle image file
const content = await readFile(absolutePath);
const base64Content = content.toString("base64");
let attachment: ImageContent;
let dimensionNote: string | undefined;
if (autoResizeImages) {
const resized = await resizeImage({ type: "image", data: base64Content, mimeType });
const resized = await resizeImage(content, mimeType);
if (!resized) {
text += `<file name="${absolutePath}">[Image omitted: could not be resized below the inline image size limit.]</file>\n`;
continue;
@@ -71,7 +70,7 @@ export async function processFileArguments(fileArgs: string[], options?: Process
attachment = {
type: "image",
mimeType,
data: base64Content,
data: content.toString("base64"),
};
}
packages/coding-agent/src/core/tools/read.ts
+2 -3
View File
@@ -250,10 +250,9 @@ export function createReadToolDefinition(
if (mimeType) {
// Read image as binary.
const buffer = await ops.readFile(absolutePath);
const base64 = buffer.toString("base64");
if (autoResizeImages) {
// Resize image if needed before sending it back to the model.
const resized = await resizeImage({ type: "image", data: base64, mimeType });
const resized = await resizeImage(buffer, mimeType);
if (!resized) {
let textNote = `Read image file [${mimeType}]\n[Image omitted: could not be resized below the inline image size limit.]`;
if (nonVisionImageNote) textNote += `\n${nonVisionImageNote}`;
@@ -273,7 +272,7 @@ export function createReadToolDefinition(
if (nonVisionImageNote) textNote += `\n${nonVisionImageNote}`;
content = [
{ type: "text", text: textNote },
{ type: "image", data: base64, mimeType },
{ type: "image", data: buffer.toString("base64"), mimeType },
];
}
} else {
packages/coding-agent/src/utils/image-resize.ts
+9 -8
View File
@@ -1,4 +1,3 @@
import type { ImageContent } from "@earendil-works/pi-ai";
import { applyExifOrientation } from "./exif-orientation.js";
import { loadPhoton } from "./photon.js";
@@ -57,10 +56,13 @@ function encodeCandidate(buffer: Uint8Array, mimeType: string): EncodedCandidate
* 3. If still too large, try JPEG with decreasing quality
* 4. If still too large, progressively reduce dimensions until 1x1
*/
export async function resizeImage(img: ImageContent, options?: ImageResizeOptions): Promise<ResizedImage | null> {
export async function resizeImage(
inputBytes: Uint8Array,
mimeType: string,
options?: ImageResizeOptions,
): Promise<ResizedImage | null> {
const opts = { ...DEFAULT_OPTIONS, ...options };
const inputBuffer = Buffer.from(img.data, "base64");
const inputBase64Size = Buffer.byteLength(img.data, "utf-8");
const inputBase64Size = Math.ceil(inputBytes.byteLength / 3) * 4;
const photon = await loadPhoton();
if (!photon) {
@@ -69,20 +71,19 @@ export async function resizeImage(img: ImageContent, options?: ImageResizeOption
let image: ReturnType<typeof photon.PhotonImage.new_from_byteslice> | undefined;
try {
const inputBytes = new Uint8Array(inputBuffer);
const rawImage = photon.PhotonImage.new_from_byteslice(inputBytes);
image = applyExifOrientation(photon, rawImage, inputBytes);
if (image !== rawImage) rawImage.free();
const originalWidth = image.get_width();
const originalHeight = image.get_height();
const format = img.mimeType?.split("/")[1] ?? "png";
const format = mimeType.split("/")[1] ?? "png";
// Check if already within all limits (dimensions AND encoded size)
if (originalWidth <= opts.maxWidth && originalHeight <= opts.maxHeight && inputBase64Size < opts.maxBytes) {
return {
data: img.data,
mimeType: img.mimeType ?? `image/${format}`,
data: Buffer.from(inputBytes).toString("base64"),
mimeType: mimeType || `image/${format}`,
originalWidth,
originalHeight,
width: originalWidth,