Update app.py

app.py

@@ -1,90 +1,59 @@
 import gradio as gr
-from bs4 import BeautifulSoup
-from urllib.parse import urlencode
-import cloudscraper
 from fastmcp import FastMCP
 import logging
+import requests
 
 # Set up logging
 logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")
 logger = logging.getLogger(__name__)
 
 # Initialize FastMCP server
-mcp = FastMCP("Canada Job Bank Scraper Agent")
+mcp = FastMCP("Jobicy Canada Remote Jobs Agent")
 
 @mcp.tool(name="search_jobs")
 def search_jobs_tool(query: str, location: str, limit: int = 10, salary: str = None, job_type: str = None) -> dict:
     """
-    Scrape job listings from the Canada Job Bank website using cloudscraper.
+    Use Jobicy's Remote Jobs API to search remote jobs in Canada.
 
     Args:
-        query (str): Job title or keyword to search for.
-        location (str): Location to filter jobs (e.g., "Toronto, ON").
-        limit (int): Maximum number of job results to return (default: 10).
-        salary (str, optional): Ignored (for API compatibility).
-        job_type (str, optional): Ignored (for API compatibility).
+        query (str): Keyword or job title to search for.
+        location (str): Location filter (ignored since it's remote, but kept for API compatibility).
+        limit (int): Max number of jobs to return (max 50).
+        salary (str): Ignored (for compatibility).
+        job_type (str): Ignored (for compatibility).
 
     Returns:
-        dict: A dictionary with a list of job listings or an error message.
+        dict: Contains 'jobs' list or 'error' message.
     """
-    base_url = "https://www.jobbank.gc.ca/jobsearch/jobsearch"
+    base_url = "https://jobicy.com/api/v2/remote-jobs"
     params = {
-        "searchstring": query.strip(),
-        "locationstring": location.strip(),
-        "sort": "M",  # Sort by most recent
-    }
-    url = base_url + "?" + urlencode(params, safe=",")
-
-    headers = {
-        "User-Agent": (
-            "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
-            "AppleWebKit/537.36 (KHTML, like Gecko) "
-            "Chrome/126.0.0.0 Safari/537.36"
-        ),
-        "Accept-Language": "en-US,en;q=0.9",
-        "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8",
-        "Referer": "https://www.jobbank.gc.ca/",
-        "Connection": "keep-alive",
+        "count": min(max(1, limit), 50),  # ensure between 1 and 50
+        "geo": "canada",
     }
+    if query.strip():
+        params["tag"] = query.strip()
 
     try:
-        logger.info(f"Attempting to scrape: {url}")
-        scraper = cloudscraper.create_scraper()
-        response = scraper.get(url, headers=headers, timeout=10)
+        logger.info(f"Requesting Jobicy API with params: {params}")
+        response = requests.get(base_url, params=params, timeout=10)
         response.raise_for_status()
-        soup = BeautifulSoup(response.text, "html.parser")
-        cards = soup.find_all("article", class_="job-result")  # Verify class name
-
-        if not cards:
-            logger.warning("No job cards found. The website may use JavaScript or the HTML structure may have changed.")
-            logger.debug(f"HTML sample: {soup.prettify()[:2000]}")  # Increased sample size for debugging
-            return {"error": "No job listings found. The website may use JavaScript or the HTML structure may have changed."}
+        data = response.json()
 
+        jobs_raw = data.get("jobs", [])
         jobs = []
-        for card in cards[:limit]:
-            title_elem = card.find("span", class_="job-title")
-            company_elem = card.find("li", class_="employer")
-            location_elem = card.find("li", class_="job-location")
-            link_elem = card.find("a", href=True)
-
-            link = link_elem.get("href") if link_elem else None
-            if link and not link.startswith("http"):
-                link = "https://www.jobbank.gc.ca" + link
-
-            job = {
-                "title": title_elem.get_text(strip=True) if title_elem else "No Title",
-                "company": company_elem.get_text(strip=True).replace("Employer", "").strip() if company_elem else "Unknown Company",
-                "location": location_elem.get_text(strip=True).replace("Location", "").strip() if location_elem else "Unknown Location",
-                "url": link or "#"
-            }
-            jobs.append(job)
-
-        logger.info(f"Found {len(jobs)} job listings.")
+        for job in jobs_raw:
+            jobs.append({
+                "title": job.get("jobTitle", "No Title"),
+                "company": job.get("companyName", "Unknown Company"),
+                "location": job.get("jobGeo", "Remote"),
+                "url": job.get("url", "#"),
+                "pubDate": job.get("pubDate", "Unknown Date"),
+                "salary": f"{job.get('salaryMin', '')} - {job.get('salaryMax', '')} {job.get('salaryCurrency', '')}".strip()
+            })
+
+        logger.info(f"Found {len(jobs)} jobs from Jobicy.")
         return {"jobs": jobs}
 
-    except cloudscraper.exceptions.CloudflareChallengeError as cf_err:
-        logger.error(f"Cloudflare challenge error: {cf_err}")
-        return {"error": f"Cloudflare challenge error: {cf_err}"}
     except requests.exceptions.HTTPError as http_err:
         logger.error(f"HTTP error: {http_err}")
         return {"error": f"HTTP error occurred: {http_err}"}
@@ -100,12 +69,12 @@ def search_jobs_tool(query: str, location: str, limit: int = 10, salary: str = N
 
 def search_jobs_ui(query, location, limit=10, salary=None, job_type=None):
     """
-    Interface function for displaying job search results in markdown format.
+    Gradio UI function for displaying job search results.
     """
-    if not query or not location:
-        return "❌ Please provide both a job title/keyword and location."
+    if not query:
+        return "❌ Please provide a job title or keyword to search for."
 
-    limit = int(max(1, min(limit, 50)))  # Enforce reasonable limit
+    limit = int(max(1, min(limit, 50)))
     result = search_jobs_tool(query, location, limit, salary, job_type)
 
     if "error" in result:
@@ -113,14 +82,16 @@ def search_jobs_ui(query, location, limit=10, salary=None, job_type=None):
 
     jobs = result.get("jobs", [])
     if not jobs:
-        return "No jobs found for your search. Try different keywords or a broader location."
+        return "No jobs found for your search. Try different keywords."
 
-    output = "# Job Search Results\n\n"
+    output = "# Jobicy Canada Remote Job Search Results\n\n"
     for i, job in enumerate(jobs, 1):
         output += (
             f"### {i}. {job['title']}\n"
             f"**Company**: {job['company']}\n"
             f"**Location**: {job['location']}\n"
+            f"**Salary**: {job['salary']}\n"
+            f"**Posted On**: {job['pubDate']}\n"
             f"[Apply Here]({job['url']})\n\n"
         )
 
@@ -131,16 +102,16 @@ app = gr.Interface(
     fn=search_jobs_ui,
     inputs=[
         gr.Textbox(label="Job Title / Keyword", placeholder="e.g., Software Engineer"),
-        gr.Textbox(label="Location", placeholder="e.g., Toronto, ON"),
+        gr.Textbox(label="Location (ignored)", placeholder="Any, e.g., Toronto, ON"),
         gr.Slider(minimum=1, maximum=50, value=10, step=1, label="Number of Results"),
         gr.Textbox(label="Salary (optional, ignored)", placeholder="Not used"),
         gr.Textbox(label="Job Type (optional, ignored)", placeholder="Not used")
     ],
     outputs=gr.Markdown(),
-    title="Canada Job Bank Job Search",
-    description="Search jobs by scraping Canada Job Bank using FastMCP and cloudscraper.",
+    title="Jobicy Canada Remote Job Search",
+    description="Search remote jobs in Canada using Jobicy API with FastMCP and Gradio.",
     theme="huggingface"
 )
 
 if __name__ == "__main__":
-    app.launch(mcp_server=True)
+    app.launch(mcp_server=True)