Spaces:
Sleeping
Sleeping
Initial commit
Browse files- Dockerfile +49 -0
- README.md +6 -7
- app.py +156 -0
- requirements.txt +4 -0
Dockerfile
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FROM python:3.11-slim
|
| 2 |
+
|
| 3 |
+
WORKDIR /app
|
| 4 |
+
|
| 5 |
+
RUN apt-get update && apt-get install -y \
|
| 6 |
+
wget \
|
| 7 |
+
gnupg \
|
| 8 |
+
ca-certificates \
|
| 9 |
+
fonts-liberation \
|
| 10 |
+
libasound2 \
|
| 11 |
+
libatk-bridge2.0-0 \
|
| 12 |
+
libatk1.0-0 \
|
| 13 |
+
libc6 \
|
| 14 |
+
libcairo2 \
|
| 15 |
+
libcups2 \
|
| 16 |
+
libdbus-1-3 \
|
| 17 |
+
libexpat1 \
|
| 18 |
+
libfontconfig1 \
|
| 19 |
+
libgcc1 \
|
| 20 |
+
libglib2.0-0 \
|
| 21 |
+
libgtk-3-0 \
|
| 22 |
+
libnspr4 \
|
| 23 |
+
libnss3 \
|
| 24 |
+
libx11-6 \
|
| 25 |
+
libx11-xcb1 \
|
| 26 |
+
libxcb1 \
|
| 27 |
+
libxcomposite1 \
|
| 28 |
+
libxcursor1 \
|
| 29 |
+
libxdamage1 \
|
| 30 |
+
libxext6 \
|
| 31 |
+
libxfixes3 \
|
| 32 |
+
libxi6 \
|
| 33 |
+
libxrandr2 \
|
| 34 |
+
libxrender1 \
|
| 35 |
+
libxss1 \
|
| 36 |
+
libxtst6 \
|
| 37 |
+
xdg-utils \
|
| 38 |
+
&& apt-get clean && rm -rf /var/lib/apt/lists/*
|
| 39 |
+
|
| 40 |
+
COPY requirements.txt .
|
| 41 |
+
RUN pip install --no-cache-dir -r requirements.txt
|
| 42 |
+
|
| 43 |
+
RUN playwright install chromium
|
| 44 |
+
|
| 45 |
+
COPY . .
|
| 46 |
+
|
| 47 |
+
EXPOSE 7860
|
| 48 |
+
|
| 49 |
+
CMD ["python3" "./app.py"]
|
README.md
CHANGED
|
@@ -1,12 +1,11 @@
|
|
| 1 |
---
|
| 2 |
title: SERPent
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: docker
|
| 7 |
-
|
| 8 |
-
license: mit
|
| 9 |
-
short_description: Reusable SERP scrapping API for AI projects
|
| 10 |
---
|
| 11 |
|
| 12 |
-
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: SERPent
|
| 3 |
+
emoji: 🐍
|
| 4 |
+
colorFrom: green
|
| 5 |
+
colorTo: yellow
|
| 6 |
sdk: docker
|
| 7 |
+
app_port: 7860
|
|
|
|
|
|
|
| 8 |
---
|
| 9 |
|
| 10 |
+
|
| 11 |
+
# `SERPent`
|
app.py
ADDED
|
@@ -0,0 +1,156 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from contextlib import asynccontextmanager
|
| 2 |
+
from typing import Optional
|
| 3 |
+
from fastapi import FastAPI
|
| 4 |
+
from pydantic import BaseModel, Field
|
| 5 |
+
from playwright.async_api import async_playwright, Browser, BrowserContext, Page
|
| 6 |
+
from urllib.parse import quote_plus
|
| 7 |
+
import logging
|
| 8 |
+
import re
|
| 9 |
+
import uvicorn
|
| 10 |
+
|
| 11 |
+
logging.basicConfig(level=logging.INFO)
|
| 12 |
+
|
| 13 |
+
# playwright global context
|
| 14 |
+
playwright = None
|
| 15 |
+
pw_browser: Browser = None
|
| 16 |
+
|
| 17 |
+
|
| 18 |
+
@asynccontextmanager
|
| 19 |
+
async def api_lifespan(app: FastAPI):
|
| 20 |
+
global playwright, pw_browser
|
| 21 |
+
playwright = await async_playwright().start()
|
| 22 |
+
pw_browser = await playwright.chromium.launch(headless=True)
|
| 23 |
+
|
| 24 |
+
yield
|
| 25 |
+
|
| 26 |
+
await pw_browser.close()
|
| 27 |
+
await playwright.stop()
|
| 28 |
+
|
| 29 |
+
app = FastAPI(lifespan=api_lifespan)
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
class APISearchParams(BaseModel):
|
| 33 |
+
queries: list[str] = Field(...,
|
| 34 |
+
description="The list of queries to search for")
|
| 35 |
+
n_results: int = Field(
|
| 36 |
+
10, description="Number of results to return for each query. Valid values are 10, 25, 50 and 100")
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
class APIPatentResults(BaseModel):
|
| 40 |
+
"""Response of /search_patents endpoint"""
|
| 41 |
+
error: Optional[str]
|
| 42 |
+
results: Optional[list[dict]]
|
| 43 |
+
|
| 44 |
+
|
| 45 |
+
class APIBraveResults(BaseModel):
|
| 46 |
+
"""Response of /search_brave endpoint"""
|
| 47 |
+
error: Optional[str]
|
| 48 |
+
results: Optional[list[dict]]
|
| 49 |
+
|
| 50 |
+
|
| 51 |
+
async def query_google_patents(browser: Browser, q: str, n_results: int = 10):
|
| 52 |
+
"""Queries google patents for the specified query and number of results. Returns relevant patents"""
|
| 53 |
+
context: BrowserContext = await browser.new_context()
|
| 54 |
+
page: Page = await context.new_page()
|
| 55 |
+
|
| 56 |
+
async def _block_resources(route, request):
|
| 57 |
+
if request.resource_type in ["stylesheet", "image"]:
|
| 58 |
+
await route.abort()
|
| 59 |
+
else:
|
| 60 |
+
await route.continue_()
|
| 61 |
+
|
| 62 |
+
await page.route("**/*", _block_resources)
|
| 63 |
+
|
| 64 |
+
url = f"https://patents.google.com/?q=({quote_plus(q)})&oq={quote_plus(q)}&num={n_results}"
|
| 65 |
+
await page.goto(url)
|
| 66 |
+
|
| 67 |
+
await page.wait_for_function(
|
| 68 |
+
f"""() => document.querySelectorAll('search-result-item').length >= {n_results}""",
|
| 69 |
+
timeout=30_000
|
| 70 |
+
)
|
| 71 |
+
|
| 72 |
+
# regex to locate a patent id
|
| 73 |
+
PATENT_ID_REGEX = r"\b[A-Z]{2}\d{6,}(?:[A-Z]\d?)?\b"
|
| 74 |
+
|
| 75 |
+
items = await page.locator("search-result-item").all()
|
| 76 |
+
matches = []
|
| 77 |
+
for item in items:
|
| 78 |
+
all_text = " ".join(await item.locator("span").all_inner_texts())
|
| 79 |
+
found = re.findall(PATENT_ID_REGEX, all_text)
|
| 80 |
+
if found:
|
| 81 |
+
matches.append(found[0])
|
| 82 |
+
|
| 83 |
+
await context.close()
|
| 84 |
+
return matches
|
| 85 |
+
|
| 86 |
+
|
| 87 |
+
async def query_brave_search(browser: Browser, q: str, n_results: int = 10):
|
| 88 |
+
"""Queries brave search for the specified query"""
|
| 89 |
+
context: BrowserContext = await browser.new_context()
|
| 90 |
+
page: Page = await context.new_page()
|
| 91 |
+
|
| 92 |
+
async def _block_resources(route, request):
|
| 93 |
+
if request.resource_type in ["stylesheet", "image"]:
|
| 94 |
+
await route.abort()
|
| 95 |
+
else:
|
| 96 |
+
await route.continue_()
|
| 97 |
+
|
| 98 |
+
await page.route("**/*", _block_resources)
|
| 99 |
+
|
| 100 |
+
url = f"https://search.brave.com/search?q={quote_plus(q)}"
|
| 101 |
+
await page.goto(url)
|
| 102 |
+
|
| 103 |
+
results_cards = await page.locator('.snippet').all()
|
| 104 |
+
|
| 105 |
+
results = []
|
| 106 |
+
|
| 107 |
+
for result in results_cards:
|
| 108 |
+
title = await result.locator('.title').all_inner_texts()
|
| 109 |
+
description = await result.locator('.snippet-description').all_inner_texts()
|
| 110 |
+
url = await result.locator('a').nth(0).get_attribute('href')
|
| 111 |
+
|
| 112 |
+
if url.startswith('/'):
|
| 113 |
+
continue
|
| 114 |
+
|
| 115 |
+
results.append({"title": title[0] if len(title) > 0 else "", "body": description[0] if len(
|
| 116 |
+
description) > 0 else "", "href": url})
|
| 117 |
+
|
| 118 |
+
return results[:n_results]
|
| 119 |
+
|
| 120 |
+
|
| 121 |
+
@app.post("/search_scholar")
|
| 122 |
+
async def query_google_scholar(params: APISearchParams):
|
| 123 |
+
"""Queries google scholar for the specified query"""
|
| 124 |
+
return {"error": "Unimplemented"}
|
| 125 |
+
|
| 126 |
+
|
| 127 |
+
@app.post("/search_patents")
|
| 128 |
+
async def search_patents(params: APISearchParams) -> APIPatentResults:
|
| 129 |
+
"""Searches google patents for the specified queries and returns the found documents."""
|
| 130 |
+
results = []
|
| 131 |
+
for q in params.queries:
|
| 132 |
+
logging.info(f"Searching Google Patents with query `{q}`")
|
| 133 |
+
try:
|
| 134 |
+
res = await query_google_patents(pw_browser, q, params.n_results)
|
| 135 |
+
results.extend(res)
|
| 136 |
+
except Exception as e:
|
| 137 |
+
logging.error(
|
| 138 |
+
f"Failed to query Google Patents with query `{q}`: {e}")
|
| 139 |
+
return APIPatentResults(results=[{"href": f"https://patents.google.com/patent/{id}/en", "id": id} for id in results], error=None)
|
| 140 |
+
|
| 141 |
+
|
| 142 |
+
@app.post("/search_brave")
|
| 143 |
+
async def search_brave(params: APISearchParams) -> APIBraveResults:
|
| 144 |
+
results = []
|
| 145 |
+
for q in params.queries:
|
| 146 |
+
logging.info(f"Searching Brave search with query `{q}`")
|
| 147 |
+
try:
|
| 148 |
+
res = await query_brave_search(pw_browser, q, params.n_results)
|
| 149 |
+
results.extend(res)
|
| 150 |
+
except Exception as e:
|
| 151 |
+
logging.error(
|
| 152 |
+
f"Failed to query Brave search with query `{q}`: {e}")
|
| 153 |
+
|
| 154 |
+
return APIBraveResults(results=results, error=None)
|
| 155 |
+
|
| 156 |
+
uvicorn.run(app, port=7860)
|
requirements.txt
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
fastapi
|
| 2 |
+
uvicorn
|
| 3 |
+
pydantic
|
| 4 |
+
playwright
|