Spaces:
Running
Running
Daniel Amendoeira
commited on
Update tools.py
Browse files
tools.py
CHANGED
@@ -4,6 +4,7 @@ import requests
|
|
4 |
import openai
|
5 |
import os
|
6 |
import tempfile
|
|
|
7 |
from urllib.parse import urlparse, parse_qs
|
8 |
from openai import OpenAI
|
9 |
from youtube_transcript_api import YouTubeTranscriptApi
|
@@ -95,7 +96,6 @@ def transcribe_audio(audio_file: str, file_extension: str) -> str:
|
|
95 |
Args:
|
96 |
audio_file (str): local file path to the audio file (.mp3, .m4a, etc.)
|
97 |
file_extension (str): file extension of the audio, e.g. mp3
|
98 |
-
|
99 |
Returns:
|
100 |
str: The transcribed text from the audio.
|
101 |
"""
|
@@ -181,7 +181,6 @@ def webpage_content(url: str) -> str:
|
|
181 |
""" Fetch text from a webpage or PDF file.
|
182 |
Args:
|
183 |
url (str): The URL of the webpage to fetch.
|
184 |
-
|
185 |
Returns:
|
186 |
str: Extracted text.
|
187 |
"""
|
@@ -204,3 +203,23 @@ def webpage_content(url: str) -> str:
|
|
204 |
|
205 |
except Exception as e:
|
206 |
return f"webpage_content failed: {e}"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
import openai
|
5 |
import os
|
6 |
import tempfile
|
7 |
+
import pandas as pd
|
8 |
from urllib.parse import urlparse, parse_qs
|
9 |
from openai import OpenAI
|
10 |
from youtube_transcript_api import YouTubeTranscriptApi
|
|
|
96 |
Args:
|
97 |
audio_file (str): local file path to the audio file (.mp3, .m4a, etc.)
|
98 |
file_extension (str): file extension of the audio, e.g. mp3
|
|
|
99 |
Returns:
|
100 |
str: The transcribed text from the audio.
|
101 |
"""
|
|
|
181 |
""" Fetch text from a webpage or PDF file.
|
182 |
Args:
|
183 |
url (str): The URL of the webpage to fetch.
|
|
|
184 |
Returns:
|
185 |
str: Extracted text.
|
186 |
"""
|
|
|
203 |
|
204 |
except Exception as e:
|
205 |
return f"webpage_content failed: {e}"
|
206 |
+
|
207 |
+
@tool
|
208 |
+
def read_excel(file_url: str) -> str:
|
209 |
+
""" Reads an Excel file from a URL and returns the content as CSV text.
|
210 |
+
Args:
|
211 |
+
file_url (str): URL to the Excel file (.xlsx, .xls)
|
212 |
+
Returns:
|
213 |
+
str: Content of the Excel file as CSV text.
|
214 |
+
"""
|
215 |
+
try:
|
216 |
+
response = requests.get(file_url)
|
217 |
+
response.raise_for_status()
|
218 |
+
|
219 |
+
excel_content = BytesIO(response.content)
|
220 |
+
df = pd.read_excel(excel_content)
|
221 |
+
|
222 |
+
return df.to_csv(index=False) # convert dataframe to CSV string for easy processing
|
223 |
+
|
224 |
+
except Exception as e:
|
225 |
+
return f"read_excel failed: {str(e)}"
|