Spaces:
Running
Running
File size: 7,439 Bytes
97a0620 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 |
import gradio as gr
import requests
from PIL import Image
import io
import os
BASE_URL = "https://api.jigsawstack.com/v1"
headers = {"x-api-key": os.getenv("")}
# ----------------- JigsawStack API Wrappers ------------------
def enhanced_ai_scrape(input_method, url, html, prompts_str, selector, page_pos):
def error_response(message):
return (
message,
gr.update(visible=False),
gr.update(visible=False),
gr.update(visible=False),
gr.update(visible=False),
gr.update(visible=False),
)
try:
# Validate element prompts
prompts = [p.strip() for p in prompts_str.split(",") if p.strip()]
if not prompts:
return error_response("Error: No element prompts provided.")
if len(prompts) > 5:
return error_response("Error: Maximum 5 element prompts allowed.")
payload = {
"element_prompts": prompts,
"root_element_selector": selector or "main",
"page_position": int(page_pos) if str(page_pos).strip().isdigit() else 1
}
# Add URL or HTML based on input method
if input_method == "URL":
if not url or not url.strip():
return error_response("Error: URL is required when using URL input method.")
payload["url"] = url.strip()
elif input_method == "HTML Content":
if not html or not html.strip():
return error_response("Error: HTML content is required when using HTML input method.")
payload["html"] = html.strip()
response = requests.post(f"{BASE_URL}/ai/scrape", headers=headers, json=payload)
response.raise_for_status()
result = response.json()
if not result.get("success"):
return error_response(f"Error: Scraping failed - {result.get('message', 'Unknown error')}")
# Extract all the data
context = result.get("context", {})
selectors = result.get("selectors", {})
data = result.get("data", [])
links = result.get("link", [])
current_page = result.get("page_position", 1)
total_pages = result.get("page_position_length", 1)
# Format pagination info
pagination_text = f"Page {current_page} of {total_pages}"
if total_pages > 1:
pagination_text += f" (Total pages available: {total_pages})"
status_text = f"✅ Successfully scraped {len(data)} data items"
if context:
status_text += f" with {len(context)} context elements"
return (
status_text,
gr.update(value=context, visible=True if context else False),
gr.update(value=selectors, visible=True if selectors else False),
gr.update(value=data, visible=True if data else False),
gr.update(value=links, visible=True if links else False),
gr.update(value=pagination_text, visible=True),
)
except requests.exceptions.RequestException as req_err:
return error_response(f"Request failed: {str(req_err)}")
except Exception as e:
return error_response(f"Unexpected error: {str(e)}")
# ----------------- Gradio UI ------------------
with gr.Blocks() as demo:
gr.Markdown("""
<div style='text-align: center; margin-bottom: 24px;'>
<h1 style='font-size:2.2em; margin-bottom: 0.2em;'>🧩 JigsawStack AI Scraper</h1>
<p style='font-size:1.2em; margin-top: 0;'>Extract structured data from web pages with advanced AI models.</p>
<p style='font-size:1em; margin-top: 0.5em;'>For more details and API usage, see the <a href='https://jigsawstack.com/docs/api-reference/ai/scrape' target='_blank'>documentation</a>.</p>
</div>
""")
with gr.Row():
with gr.Column():
gr.Markdown("#### Input Method")
input_method_scraper = gr.Radio(
choices=["URL", "HTML Content"],
label="Choose Input Method",
value="URL"
)
# Conditional inputs based on selection
url_scraper = gr.Textbox(
label="Page URL",
placeholder="https://example.com/pricing",
info="URL of the page to scrape"
)
html_content = gr.Textbox(
label="HTML Content",
lines=8,
placeholder="<html>...</html>",
visible=False,
info="Raw HTML content to scrape"
)
gr.Markdown("#### Scraping Configuration")
element_prompts = gr.Textbox(
label="Element Prompts (comma-separated)",
lines=3,
placeholder="Plan title, Plan price, Features, Button text",
info="Items to scrape (max 5). E.g., 'Plan price', 'Plan title'"
)
root_selector = gr.Textbox(
label="Root Element Selector",
value="main",
placeholder="main, .container, #content",
info="CSS selector to limit scraping scope (default: main)"
)
page_position = gr.Number(
label="Page Position",
value=1,
minimum=1,
info="For pagination, current page number (min: 1)"
)
with gr.Column():
gr.Markdown("#### Results")
scrape_status = gr.Textbox(
label="Status",
interactive=False,
placeholder="Ready to scrape..."
)
gr.Markdown("#### Extracted Data")
context_output = gr.JSON(
label="Context Data",
visible=False
)
selectors_output = gr.JSON(
label="CSS Selectors Used",
visible=False
)
detailed_data = gr.JSON(
label="Detailed Scrape Data",
visible=False
)
links_data = gr.JSON(
label="Detected Links",
visible=False
)
gr.Markdown("#### Pagination Info")
pagination_info = gr.Textbox(
label="Page Information",
interactive=False,
visible=False
)
scrape_btn = gr.Button("Scrape with AI", variant="primary")
# Function to show/hide input groups based on selection
def update_scraper_input_visibility(method):
if method == "URL":
return gr.Textbox(visible=True), gr.Textbox(visible=False)
elif method == "HTML Content":
return gr.Textbox(visible=False), gr.Textbox(visible=True)
else:
return gr.Textbox(visible=True), gr.Textbox(visible=False)
input_method_scraper.change(
update_scraper_input_visibility,
inputs=input_method_scraper,
outputs=[url_scraper, html_content]
)
scrape_btn.click(
enhanced_ai_scrape,
inputs=[input_method_scraper, url_scraper, html_content, element_prompts, root_selector, page_position],
outputs=[scrape_status, context_output, selectors_output, detailed_data, links_data, pagination_info]
)
demo.launch()
|