Merge pull request #5 from JobSpy-ai/features/scraper-refactor

Features/scraper refactor

api/core/jobs/__init__.py

@@ -58,8 +58,12 @@ class JobPost(BaseModel):
 
 
 class JobResponse(BaseModel):
-    job_count: int
-    page: int = 1
-    total_pages: int
+    success: bool
+    error: str = None
+
+    total_pages: int = None
+    job_count: int = None
+
+    page: int = None
+    jobs: list[JobPost] = []
 
-    jobs: list[JobPost]

api/core/scrapers/__init__.py

@@ -1,6 +1,6 @@
 from pydantic import BaseModel
 from enum import Enum
-from ..jobs import JobResponse
+from ..jobs import JobResponse, JobPost
 
 
 class Site(Enum):
@@ -13,13 +13,11 @@ class ScraperInput(BaseModel):
     location: str
     search_term: str
     distance: int = 25
-
-    page: int = 1
+    results_wanted: int = 15  #: TODO: implement
 
 
 class Scraper:  #: to be used as a child class
     def __init__(self, site: Site):
         self.site = site
 
-    def scrape(self, scraper_input: ScraperInput) -> JobResponse:
-        ...
+    def scrape(self, scraper_input: ScraperInput) -> JobResponse: ...

api/core/scrapers/indeed/__init__.py

@@ -10,6 +10,10 @@ from api.core.jobs import *
 from api.core.scrapers import Scraper, ScraperInput, Site
 
 
+class ParsingException(Exception):
+    pass
+
+
 class IndeedScraper(Scraper):
     def __init__(self):
         site = Site(Site.INDEED)
@@ -25,7 +29,7 @@ class IndeedScraper(Scraper):
             "q": scraper_input.search_term,
             "l": scraper_input.location,
             "filter": 0,
-            "start": 0 if scraper_input.page is None else (scraper_input.page - 1) * 10,
+            "start": 0,
             "radius": scraper_input.distance,
         }
 
@@ -38,12 +42,25 @@ class IndeedScraper(Scraper):
 
         soup = BeautifulSoup(response.content, "html.parser")
 
-        jobs = IndeedScraper.parse_jobs(soup)
+        try:
+            jobs = IndeedScraper.parse_jobs(soup)
+        except ParsingException:
+            return JobResponse(
+                success=False,
+                error="Failed to parse jobs.",
+            )
+
         total_num_jobs = IndeedScraper.total_jobs(soup)
         total_pages = ceil(total_num_jobs / 15)
 
         job_list: list[JobPost] = []
-        # page_number = jobs["metaData"]["mosaicProviderJobCardsModel"]["pageNumber"]
+        if not jobs.get('metaData', {}).get("mosaicProviderJobCardsModel", {}).get("results"):
+            return JobResponse(
+                success=False,
+                error="No jobs found",
+            )
+
+        page_number = jobs["metaData"]["mosaicProviderJobCardsModel"]["pageNumber"]
         for job in jobs["metaData"]["mosaicProviderJobCardsModel"]["results"]:
             snippet_html = BeautifulSoup(job["snippet"], "html.parser")
 
@@ -94,9 +111,10 @@ class IndeedScraper(Scraper):
             job_list.append(job_post)
 
         job_response = JobResponse(
+            success=True,
             jobs=job_list,
             job_count=total_num_jobs,
-            page=scraper_input.page,
+            page=page_number,
             total_pages=total_pages,
         )
         return job_response
@@ -116,7 +134,14 @@ class IndeedScraper(Scraper):
         return None
 
     @staticmethod
-    def parse_jobs(soup):
+    def parse_jobs(soup: BeautifulSoup) -> dict:
+        """
+        Parses the jobs from the soup object
+
+        :param soup:
+        :return: jobs
+        """
+
         script_tag = IndeedScraper.find_mosaic_script(soup)
 
         if script_tag:
@@ -130,11 +155,9 @@ class IndeedScraper(Scraper):
                 jobs = json.loads(m.group(1).strip())
                 return jobs
             else:
-                return {"message": f"Could not find mosaic provider job cards data"}
+                raise ParsingException("Could not find mosaic provider job cards data")
         else:
-            return {
-                "message": f"Could not find a script tag containing mosaic provider data"
-            }
+            raise ParsingException("Could not find a script tag containing mosaic provider data")
 
     @staticmethod
     def total_jobs(soup):

api/core/scrapers/linkedin/__init__.py

@@ -16,8 +16,10 @@ class LinkedInScraper(Scraper):
         self.url = "https://www.linkedin.com/jobs"
 
     def scrape(self, scraper_input: ScraperInput) -> JobResponse:
+        current_page = 0
+
         params = {
-            "pageNum": scraper_input.page - 1,
+            "pageNum": current_page,
             "location": scraper_input.location,
             "distance": scraper_input.distance,
         }
@@ -58,6 +60,8 @@ class LinkedInScraper(Scraper):
             if datetime_tag:
                 datetime_str = datetime_tag["datetime"]
                 date_posted = datetime.strptime(datetime_str, "%Y-%m-%d")
+            else:
+                date_posted = None
 
             job_post = JobPost(
                 title=title,
@@ -74,9 +78,11 @@ class LinkedInScraper(Scraper):
         job_count = int("".join(filter(str.isdigit, job_count_text)))
         total_pages = ceil(job_count / 25)
         job_response = JobResponse(
+            success=True,
+
             jobs=job_list,
             job_count=job_count,
-            page=scraper_input.page,
+            page=current_page + 1,
             total_pages=total_pages,
         )
         return job_response

api/core/scrapers/ziprecruiter/__init__.py

@@ -21,10 +21,12 @@ class ZipRecruiterScraper(Scraper):
             client_identifier="chrome112", random_tls_extension_order=True
         )
 
+        current_page = 1
+
         params = {
             "search": scraper_input.search_term,
             "location": scraper_input.location,
-            "page": min(scraper_input.page, 10),
+            "page": min(current_page, 10),
             "radius": scraper_input.distance,
         }
 
@@ -80,6 +82,7 @@ class ZipRecruiterScraper(Scraper):
         job_count = job_count.replace(",", "")
         total_pages = data["maxPages"]
         job_response = JobResponse(
+            success=True,
             jobs=job_list,
             job_count=job_count,
             page=params["page"],
@@ -87,6 +90,7 @@ class ZipRecruiterScraper(Scraper):
         )
         return job_response
 
+    @staticmethod
     def get_interval(interval_str):
         interval_alias = {"annually": CompensationInterval.YEARLY}
         interval_str = interval_str.lower()
@@ -97,7 +101,7 @@ class ZipRecruiterScraper(Scraper):
         return CompensationInterval(interval_str)
 
     @staticmethod
-    def get_date_posted(job: str):
+    def get_date_posted(job: BeautifulSoup):
         button = job.find(
             "button", {"class": "action_input save_job zrs_btn_secondary_200"}
         )
@@ -107,7 +111,7 @@ class ZipRecruiterScraper(Scraper):
         return params.get("posted_time", [None])[0]
 
     @staticmethod
-    def get_compensation(job):
+    def get_compensation(job: BeautifulSoup):
         pay_element = job.find("li", {"class": "perk_item perk_pay"})
         if pay_element is None:
             return None
@@ -116,7 +120,7 @@ class ZipRecruiterScraper(Scraper):
         return ZipRecruiterScraper.create_compensation_object(pay)
 
     @staticmethod
-    def get_location(job):
+    def get_location(job: BeautifulSoup):
         location_string = job.find("a", {"class": "company_location"}).text.strip()
         parts = location_string.split(", ")
         city, state = parts