Merge branch 'main' into release/v4.0.0

.github/workflows/stale.yml

@@ -0,0 +1,42 @@
+name: Mark and Close Stale Issues
+
+on:
+  # Schedule the workflow to run periodically (e.g., daily at 1:30 AM UTC)
+  schedule:
+    - cron: "30 1 * * *"
+  workflow_dispatch:
+
+jobs:
+  stale:
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write
+      pull-requests: write
+
+    steps:
+      - name: Run Stale Action
+        uses: actions/stale@v9
+        with:
+          repo-token: ${{ secrets.GITHUB_TOKEN }}
+          days-before-issue-stale: 10 # Days of inactivity before marking an issue as stale
+          days-before-issue-close: 5 # Days after being marked stale before closing the issue
+          stale-issue-label: "stale" # Label to apply to stale issues
+          exempt-issue-labels: "pinned,important" # Labels to exclude from being marked as stale
+          exempt-issue-assignees: true # Exempt issues with assignees from being marked as stale
+          stale-issue-message: "This issue has been marked as stale due to inactivity. Please comment or update if this is still relevant."
+          close-issue-message: "This issue was closed due to prolonged inactivity."
+          days-before-pr-stale: 10 # Days of inactivity before marking a PR as stale
+          days-before-pr-close: 2 # Days after being marked stale before closing the PR
+          stale-pr-label: "stale" # Label to apply to stale PRs
+          exempt-pr-labels: "pinned,important" # Labels to exclude from being marked as stale
+          stale-pr-message: >
+            "This pull request has been marked as stale due to inactivity. 
+            To keep it open, you can:
+            - Show progress by updating the PR with new commits.
+            - Continue the conversation by adding comments or requesting clarification on any blockers.
+            - Resolve pending feedback by replying to unresolved comments or implementing suggested changes.
+            - Indicate readiness for review by explicitly requesting a review from maintainers or reviewers.
+            If no action is taken within 7 days, this pull request will be closed."
+          close-pr-message: "This PR was closed due to prolonged inactivity."
+          remove-stale-when-updated: true # Remove the stale label if there is new activity
+          operations-per-run: 20 # Number of issues to process per run (default is 30)

README.md

@@ -19,8 +19,8 @@ Join our community: [Telegram](https://t.me/AIhawkCommunity) (for Normal user) |
 
 </div>
 
-**Creator** [feder-cr](https://github.com/feder-cr), Co-Founder of Ai Hawk </br>
-As AI Hawk is focusing on their proprietary product - solving problems in hiring for companies, currently this project is led, managed, and maintained by a group of open-source contributors, with a focus on building tools to help job seekers land the jobs they deserve.
+**Creator** [feder-cr](https://github.com/feder-cr), Co-Founder of AIHawk </br>
+As AIHawk is focusing on their proprietary product - solving problems in hiring for companies, currently this project is led, managed, and maintained by a group of open-source contributors, with a focus on building tools to help job seekers land the jobs they deserve.
 
 **Project Maintainers / Leads**: [surapuramakhil](https://github.com/surapuramakhil), [sarob](https://github.com/sarob), [cjbbb](https://github.com/cjbbb)
 
@@ -755,7 +755,12 @@ Made with [contrib.rocks](https://contrib.rocks).
 
 ## License
 
-This project is licensed under the  MIT + Commons Clause License - see the [LICENSE](LICENSE) file for details.
+This project is licensed under the AGPL License. Documentation is licensed under CC BY - see the [AGPL LICENSE](LICENSE) and [CC BY LICENSE](docs/LICENSE) files for details.
+
+The AGPL License requires that any derivative work must also be open source and distributed under the same license.
+
+The CC BY License permits others to distribute, remix, adapt, and build upon your work, even for commercial purposes, as long as they credit you for the original creation. 
+
 
 ## Disclaimer
 

requirements.txt

@@ -8,7 +8,7 @@ langchain==0.2.11
 langchain-anthropic
 langchain-huggingface
 langchain-community==0.2.10
-langchain-core===0.2.36
+langchain-core==0.2.36
 langchain-google-genai==1.0.10
 langchain-ollama==0.1.3
 langchain-openai==0.1.17

src/ai_hawk/job_manager.py

@@ -4,6 +4,7 @@
 import time
 from itertools import product
 from pathlib import Path
+import traceback
 from turtle import color
 
 from inputimeout import inputimeout, TimeoutOccurred
@@ -13,6 +14,7 @@
 
 from ai_hawk.linkedIn_easy_applier import AIHawkEasyApplier
 from config import JOB_MAX_APPLICATIONS, JOB_MIN_APPLICATIONS, MINIMUM_WAIT_TIME_IN_SECONDS
+
 from src.job import Job
 from src.logging import logger
 
@@ -155,7 +157,7 @@ def start_applying(self):
                     logger.debug("Starting the application process for this page...")
 
                     try:
-                        jobs = self.get_jobs_from_page()
+                        jobs = self.get_jobs_from_page(scroll=True)
                         if not jobs:
                             logger.debug("No more jobs found on this page. Exiting loop.")
                             break
@@ -166,7 +168,7 @@ def start_applying(self):
                     try:
                         self.apply_jobs()
                     except Exception as e:
-                        logger.error(f"Error during job application: {e}")
+                        logger.error(f"Error during job application: {e} {traceback.format_exc()}")
                         continue
 
                     logger.debug("Applying to jobs on this page has been completed!")
@@ -239,10 +241,9 @@ def start_applying(self):
                     time.sleep(sleep_time)
                 page_sleep += 1
 
-    def get_jobs_from_page(self):
+    def get_jobs_from_page(self, scroll=False):
 
         try:
-
             no_jobs_element = self.driver.find_element(By.CLASS_NAME, 'jobs-search-two-pane__no-results-banner--expand')
             if 'No matching jobs found' in no_jobs_element.text or 'unfortunately, things aren' in self.driver.page_source.lower():
                 logger.debug("No matching jobs found on this page, skipping.")
@@ -252,41 +253,37 @@ def get_jobs_from_page(self):
             pass
 
         try:
-            job_results = self.driver.find_element(By.CLASS_NAME, "jobs-search-results-list")
-            browser_utils.scroll_slow(self.driver, job_results)
-            browser_utils.scroll_slow(self.driver, job_results, step=300, reverse=True)
+            # XPath query to find the ul tag with class scaffold-layout__list-container
+            jobs_xpath_query = "//ul[contains(@class, 'scaffold-layout__list-container')]"
+            jobs_container = self.driver.find_element(By.XPATH, jobs_xpath_query)
+
+            if scroll:
+                jobs_container_scrolableElement = jobs_container.find_element(By.XPATH,"..")
+                logger.warning(f'is scrollable: {browser_utils.is_scrollable(jobs_container_scrolableElement)}')
+
+                browser_utils.scroll_slow(self.driver, jobs_container_scrolableElement)
+                browser_utils.scroll_slow(self.driver, jobs_container_scrolableElement, step=300, reverse=True)
 
-            job_list_elements = self.driver.find_elements(By.CLASS_NAME, 'scaffold-layout__list-container')[
-                0].find_elements(By.CLASS_NAME, 'jobs-search-results__list-item')
-            if not job_list_elements:
+            job_element_list = jobs_container.find_elements(By.XPATH, ".//li[contains(@class, 'jobs-search-results__list-item') and contains(@class, 'ember-view')]")
+
+            if not job_element_list:
                 logger.debug("No job class elements found on page, skipping.")
                 return []
 
-            return job_list_elements
+            return job_element_list
 
-        except NoSuchElementException:
-            logger.debug("No job results found on the page.")
+        except NoSuchElementException as e:
+            logger.warning(f'No job results found on the page. \n expection: {traceback.format_exc()}')
             return []
 
         except Exception as e:
-            logger.error(f"Error while fetching job elements: {e}")
+            logger.error(f"Error while fetching job elements: {e} {traceback.format_exc()}")
             return []
 
     def read_jobs(self):
-        try:
-            no_jobs_element = self.driver.find_element(By.CLASS_NAME, 'jobs-search-two-pane__no-results-banner--expand')
-            if 'No matching jobs found' in no_jobs_element.text or 'unfortunately, things aren' in self.driver.page_source.lower():
-                raise Exception("No more jobs on this page")
-        except NoSuchElementException:
-            pass
-
-        job_results = self.driver.find_element(By.CLASS_NAME, "jobs-search-results-list")
-        browser_utils.scroll_slow(self.driver, job_results)
-        browser_utils.scroll_slow(self.driver, job_results, step=300, reverse=True)
-        job_list_elements = self.driver.find_elements(By.CLASS_NAME, 'scaffold-layout__list-container')[0].find_elements(By.CLASS_NAME, 'jobs-search-results__list-item')
-        if not job_list_elements:
-            raise Exception("No job class elements found on page")
-        job_list = [self.job_tile_to_job(job_element) for job_element in job_list_elements] 
+
+        job_element_list = self.get_jobs_from_page()
+        job_list = [self.job_tile_to_job(job_element) for job_element in job_element_list] 
         for job in job_list:            
             if self.is_blacklisted(job.title, job.company, job.link, job.location):
                 logger.info(f"Blacklisted {job.title} at {job.company} in {job.location}, skipping...")
@@ -299,22 +296,9 @@ def read_jobs(self):
                 continue
 
     def apply_jobs(self):
-        try:
-            no_jobs_element = self.driver.find_element(By.CLASS_NAME, 'jobs-search-two-pane__no-results-banner--expand')
-            if 'No matching jobs found' in no_jobs_element.text or 'unfortunately, things aren' in self.driver.page_source.lower():
-                logger.debug("No matching jobs found on this page, skipping")
-                return
-        except NoSuchElementException:
-            pass
-
-        job_list_elements = self.driver.find_elements(By.CLASS_NAME, 'scaffold-layout__list-container')[
-            0].find_elements(By.CLASS_NAME, 'jobs-search-results__list-item')
-
-        if not job_list_elements:
-            logger.debug("No job class elements found on page, skipping")
-            return
+        job_element_list = self.get_jobs_from_page()
 
-        job_list = [self.job_tile_to_job(job_element) for job_element in job_list_elements]
+        job_list = [self.job_tile_to_job(job_element) for job_element in job_element_list]
 
         for job in job_list:
 
@@ -487,12 +471,12 @@ def job_tile_to_job(self, job_tile) -> Job:
             logger.debug(f"Job link extracted: {job.link}")
         except NoSuchElementException:
             logger.warning("Job link is missing.")
-        
+
         try:
-            job.company = job_tile.find_element(By.CLASS_NAME, 'job-card-container__primary-description').text
+            job.company = job_tile.find_element(By.XPATH, ".//div[contains(@class, 'artdeco-entity-lockup__subtitle')]//span").text
             logger.debug(f"Job company extracted: {job.company}")
-        except NoSuchElementException:
-            logger.warning("Job company is missing.")
+        except NoSuchElementException as e:
+            logger.warning(f'Job company is missing. {e} {traceback.format_exc()}')
 
         # Extract job ID from job url
         try:
@@ -510,11 +494,17 @@ def job_tile_to_job(self, job_tile) -> Job:
         except NoSuchElementException:
             logger.warning("Job location is missing.")
 
+
         try:
-            job.apply_method = job_tile.find_element(By.CLASS_NAME, 'job-card-container__apply-method').text
-        except NoSuchElementException:
-            job.apply_method = "Applied"
-            logger.warning("Apply method not found, assuming 'Applied'.")
+            job_state = job_tile.find_element(By.XPATH, ".//ul[contains(@class, 'job-card-list__footer-wrapper')]//li[contains(@class, 'job-card-container__apply-method')]").text
+        except NoSuchElementException as e:
+            try:
+                # Fetching state when apply method is not found
+                job_state = job_tile.find_element(By.XPATH, ".//ul[contains(@class, 'job-card-list__footer-wrapper')]//li[contains(@class, 'job-card-container__footer-job-state')]").text
+                job.apply_method = "Applied"
+                logger.warning(f'Apply method not found, state {job_state}. {e} {traceback.format_exc()}')
+            except NoSuchElementException as e:
+                logger.warning(f'Apply method and state not found. {e} {traceback.format_exc()}')
 
         return job
 

src/ai_hawk/linkedIn_easy_applier.py

@@ -376,8 +376,8 @@ def fill_up(self, job_context : JobContext) -> None:
                 EC.presence_of_element_located((By.CLASS_NAME, 'jobs-easy-apply-content'))
             )
 
-            pb4_elements = easy_apply_content.find_elements(By.CLASS_NAME, 'pb4')
-            for element in pb4_elements:
+            input_elements = easy_apply_content.find_elements(By.CLASS_NAME, 'jobs-easy-apply-form-section__grouping')
+            for element in input_elements:
                 self._process_form_element(element, job_context)
         except Exception as e:
             logger.error(f"Failed to find form elements: {e}")

src/ai_hawk/llm/llm_manager.py

@@ -695,8 +695,14 @@ def is_job_suitable(self):
         )
         output = self._clean_llm_output(raw_output)
         logger.debug(f"Job suitability output: {output}")
-        score = re.search(r"Score: (\d+)", output).group(1)
-        reasoning = re.search(r"Reasoning: (.+)", output, re.DOTALL).group(1)
+
+        try:
+            score = re.search(r"Score:\s*(\d+)", output, re.IGNORECASE).group(1)
+            reasoning = re.search(r"Reasoning:\s*(.+)", output, re.IGNORECASE | re.DOTALL).group(1)
+        except AttributeError:
+            logger.warning("Failed to extract score or reasoning from LLM. Proceeding with application, but job may or may not be suitable.")
+            return True
+
         logger.info(f"Job suitability score: {score}")
         if int(score) < JOB_SUITABILITY_SCORE:
             logger.debug(f"Job is not suitable: {reasoning}")

tests/test_aihawk_job_manager.py

@@ -71,21 +71,33 @@ def test_get_jobs_from_page_no_jobs(mocker, job_manager):
 
 def test_get_jobs_from_page_with_jobs(mocker, job_manager):
     """Test get_jobs_from_page when job elements are found."""
-    # Mock the no_jobs_element to behave correctly
-    mock_no_jobs_element = mocker.Mock()
-    mock_no_jobs_element.text = "No matching jobs found"
+    # Mock no_jobs_element to simulate the absence of "No matching jobs found" banner
+    no_jobs_element_mock = mocker.Mock()
+    no_jobs_element_mock.text = ""  # Empty text means "No matching jobs found" is not present
 
-    # Mocking the find_element to return the mock no_jobs_element
-    mocker.patch.object(job_manager.driver, 'find_element',
-                        return_value=mock_no_jobs_element)
+    # Mock the driver to simulate the page source
+    mocker.patch.object(job_manager.driver, 'page_source', return_value="")
 
-    # Mock the page_source
-    mocker.patch.object(job_manager.driver, 'page_source',
-                        return_value="some page content")
+    # Mock the outer find_element
+    container_mock = mocker.Mock()
 
-    # Ensure jobs are returned as empty list due to "No matching jobs found"
-    jobs = job_manager.get_jobs_from_page()
-    assert jobs == []  # No jobs expected due to "No matching jobs found"
+    # Mock the inner find_elements to return job list items
+    job_element_mock = mocker.Mock()
+    # Simulating two job items
+    job_elements_list = [job_element_mock, job_element_mock]
+
+    # Return the container mock, which itself returns the job elements list
+    container_mock.find_elements.return_value = job_elements_list
+    mocker.patch.object(job_manager.driver, 'find_element', side_effect=[
+        no_jobs_element_mock,
+        container_mock
+    ])
+
+    job_manager.get_jobs_from_page()
+
+    assert job_manager.driver.find_element.call_count == 2
+    assert container_mock.find_elements.call_count == 1
+
 
 
 def test_apply_jobs_with_no_jobs(mocker, job_manager):
@@ -94,9 +106,6 @@ def test_apply_jobs_with_no_jobs(mocker, job_manager):
     mock_element = mocker.Mock()
     mock_element.text = "No matching jobs found"
 
-    # Mock the driver to simulate the page source
-    mocker.patch.object(job_manager.driver, 'page_source', return_value="")
-
     # Mock the driver to return the mock element when find_element is called
     mocker.patch.object(job_manager.driver, 'find_element',
                         return_value=mock_element)
@@ -111,28 +120,15 @@ def test_apply_jobs_with_no_jobs(mocker, job_manager):
 def test_apply_jobs_with_jobs(mocker, job_manager):
     """Test apply_jobs when jobs are present."""
 
-    # Mock no_jobs_element to simulate the absence of "No matching jobs found" banner
-    no_jobs_element = mocker.Mock()
-    no_jobs_element.text = ""  # Empty text means "No matching jobs found" is not present
-    mocker.patch.object(job_manager.driver, 'find_element',
-                        return_value=no_jobs_element)
-
     # Mock the page_source to simulate what the page looks like when jobs are present
     mocker.patch.object(job_manager.driver, 'page_source',
                         return_value="some job content")
 
-    # Mock the outer find_elements (scaffold-layout__list-container)
-    container_mock = mocker.Mock()
-
-    # Mock the inner find_elements to return job list items
+    # Simulating two job elements
     job_element_mock = mocker.Mock()
-    # Simulating two job items
     job_elements_list = [job_element_mock, job_element_mock]
-
-    # Return the container mock, which itself returns the job elements list
-    container_mock.find_elements.return_value = job_elements_list
-    mocker.patch.object(job_manager.driver, 'find_elements',
-                        return_value=[container_mock])
+
+    mocker.patch.object(job_manager, 'get_jobs_from_page', return_value=job_elements_list)
 
     job = Job(
         title="Title",
@@ -181,7 +177,7 @@ def test_apply_jobs_with_jobs(mocker, job_manager):
     job_manager.apply_jobs()
 
     # Assertions
-    assert job_manager.driver.find_elements.call_count == 1
+    assert job_manager.get_jobs_from_page.call_count == 1
     # Called for each job element
     assert job_manager.job_tile_to_job.call_count == 2
     # Called for each job element