如何使用Python PIL库将libvirt流截图转换为其他格式图像？

首先安装依赖库：
pip install pillow libvirt-python

修正后可直接运行的代码 #

import sys
import libvirt
import io
from PIL import Image

domName = 'TestAppliance'
conn = None
try:
    conn = libvirt.open("qemu:///system")
except libvirt.libvirtError as e:
    print(repr(e), file=sys.stderr)
    exit(1)

dom = None
try:
    dom = conn.lookupByName(domName)
except libvirt.libvirtError as e:
    print(repr(e), file=sys.stderr)
    exit(1)

stream = conn.newStream()
# 修正原代码的变量名错误：domain改为dom
imageType = dom.screenshot(stream, 0)

# 所有字节写入内存缓冲区，不生成中间PPM文件
img_buffer = io.BytesIO()
streamBytes = stream.recv(262120)
while streamBytes != b'':
    img_buffer.write(streamBytes)
    streamBytes = stream.recv(262120)

# 重置缓冲区指针到头部，供PIL解析
img_buffer.seek(0)
# 直接读取PPM流转存为PNG，开启优化压缩
with Image.open(img_buffer) as img:
    save_path = f"Screenshot of {dom.name()}.png"
    img.save(save_path, format="PNG", optimize=True)
    print(f'截图已保存至: {save_path}，原始流格式为: {imageType}')

stream.finish()
conn.close()
exit(0)

方案说明 #

• 全程在内存中处理流数据，无本地中间文件生成，减少IO冗余
• 转换后的PNG格式体积仅为原PPM格式的1/10左右，常规虚拟机桌面截图大小可控制在几百KB
• 如需更小体积可改为保存为WebP格式，将save方法参数调整为format="WebP", quality=80即可

内容的提问来源于stack exchange，提问作者PatrickF