Build and run agents you can see, understand and trust.
Python agentchatbotlarge-language-modelsllmllm-agentmcpmulti-agentmulti-modalreact-agentrealtime-agentvoice-agentvoice-assistant
feat(formatter): supprot local audio path start with file:// (#1349)
qbcfa441b5a0710
diff --git a/src/agentscope/formatter/_openai_formatter.py b/src/agentscope/formatter/_openai_formatter.pyindex 57aa01a42..fef61e60a 100644--- a/src/agentscope/formatter/_openai_formatter.py+++ b/src/agentscope/formatter/_openai_formatter.py@@ -116,21 +116,22 @@ def _to_openai_image_url(url: str) -> str:def _to_openai_audio_data(source: URLSource | Base64Source) -> dict:"""Covert an audio source to OpenAI format."""if source["type"] == "url":- extension = source["url"].split(".")[-1].lower()+ raw_url = source["url"].removeprefix("file://")+ extension = raw_url.split(".")[-1].lower()if extension not in ["wav", "mp3"]:raise TypeError(f"Unsupported audio file extension: {extension}, ""wav and mp3 are supported.",)- parsed_url = urlparse(source["url"])+ parsed_url = urlparse(raw_url)- if os.path.exists(source["url"]):- with open(source["url"], "rb") as audio_file:+ if os.path.exists(raw_url):+ with open(raw_url, "rb") as audio_file:data = base64.b64encode(audio_file.read()).decode("utf-8")# web url- elif parsed_url.scheme != "":+ elif parsed_url.scheme not in ["", "file"]:response = requests.get(source["url"])response.raise_for_status()data = base64.b64encode(response.content).decode("utf-8")