resume data extraction

NorahUniversity/settings.py

@@ -45,7 +45,7 @@ INSTALLED_APPS = [
     'django.contrib.messages',
     'django.contrib.staticfiles',
     'rest_framework',
-    'recruitment',
+    'recruitment.apps.RecruitmentConfig',
     'corsheaders',
     'django.contrib.sites',
     'allauth',

recruitment/admin.py

@@ -1,6 +1,6 @@
 from django.contrib import messages
 from . import models
-from .utils import extract_summary_from_pdf
+# from .utils import extract_summary_from_pdf
 
 from django.contrib import admin
 from django.contrib.auth.admin import UserAdmin as BaseUserAdmin
@@ -27,20 +27,25 @@ class GroupAdmin(BaseGroupAdmin, ModelAdmin):
     pass
 
 
+@admin.register(models.JobPosting)
+class JobPostingAdmin(ModelAdmin):
+    list_display = ('title','description','qualifications')
+
+   
 @admin.register(models.Job)
 class JobAdmin(ModelAdmin):
     list_display = ('title', 'is_published', 'posted_to_linkedin', 'created_at')
     list_filter = ('is_published', 'posted_to_linkedin')
     search_fields = ('title', 'description_en', 'description_ar')
 
-@admin.action(description="Parse selected resumes")
-def parse_resumes(modeladmin, request, queryset):
-    for candidate in queryset:
-        if candidate.resume:
-            summary = extract_summary_from_pdf(candidate.resume.path)
-            candidate.parsed_summary = str(summary)
-            candidate.save()
-    messages.success(request, f"Parsed {queryset.count()} resumes successfully.")
+# @admin.action(description="Parse selected resumes")
+# def parse_resumes(modeladmin, request, queryset):
+#     for candidate in queryset:
+#         if candidate.resume:
+#             summary = extract_summary_from_pdf(candidate.resume.path)
+#             candidate.parsed_summary = str(summary)
+#             candidate.save()
+#     messages.success(request, f"Parsed {queryset.count()} resumes successfully.")
 
 @admin.register(models.Candidate)
 class CandidateAdmin(ModelAdmin):
@@ -48,7 +53,7 @@ class CandidateAdmin(ModelAdmin):
     list_filter = ('applied', 'job')
     search_fields = ('name', 'email')
     # readonly_fields = ('parsed_summary',)
-    actions = [parse_resumes]
+   
 
 @admin.register(models.TrainingMaterial)
 class TrainingMaterialAdmin(ModelAdmin):

recruitment/apps.py

@@ -4,3 +4,5 @@ from django.apps import AppConfig
 class RecruitmentConfig(AppConfig):
     default_auto_field = 'django.db.models.BigAutoField'
     name = 'recruitment'
+    def ready(self):
+         import recruitment.signals 

recruitment/migrations/0013_candidate_criteria_checklist_candidate_match_score_and_more.py

@@ -0,0 +1,33 @@
+# Generated by Django 5.2.7 on 2025-10-05 13:12
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('recruitment', '0012_form_formsubmission_uploadedfile'),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name='candidate',
+            name='criteria_checklist',
+            field=models.JSONField(blank=True, default=dict),
+        ),
+        migrations.AddField(
+            model_name='candidate',
+            name='match_score',
+            field=models.IntegerField(blank=True, null=True),
+        ),
+        migrations.AddField(
+            model_name='candidate',
+            name='strengths',
+            field=models.TextField(blank=True),
+        ),
+        migrations.AddField(
+            model_name='candidate',
+            name='weaknesses',
+            field=models.TextField(blank=True),
+        ),
+    ]

recruitment/migrations/0014_source_jobposting_source.py

@@ -0,0 +1,31 @@
+# Generated by Django 5.2.7 on 2025-10-05 16:11
+
+import django.db.models.deletion
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('recruitment', '0013_candidate_criteria_checklist_candidate_match_score_and_more'),
+    ]
+
+    operations = [
+        migrations.CreateModel(
+            name='Source',
+            fields=[
+                ('id', models.BigAutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')),
+                ('name', models.CharField(choices=[('ATS', 'Applicant Tracking System'), ('ERP', 'ERP system')], max_length=100, verbose_name='Source Type')),
+                ('created_at', models.DateTimeField(auto_now_add=True)),
+            ],
+            options={
+                'verbose_name': 'Source',
+                'verbose_name_plural': 'Sources',
+            },
+        ),
+        migrations.AddField(
+            model_name='jobposting',
+            name='source',
+            field=models.ForeignKey(blank=True, help_text='The system or channel from which this job posting originated or was first published.', null=True, on_delete=django.db.models.deletion.SET_NULL, related_name='job_postings', to='recruitment.source'),
+        ),
+    ]

recruitment/models.py

@@ -103,6 +103,15 @@ class JobPosting(Base):
     start_date = models.DateField(null=True, blank=True, help_text="Desired start date")
     open_positions = models.PositiveIntegerField(default=1, help_text="Number of open positions for this job")
 
+    source = models.ForeignKey(
+        'Source',
+        on_delete=models.SET_NULL, # Recommended: If a source is deleted, job's source is set to NULL
+        related_name='job_postings',
+        null=True, 
+        blank=True, 
+        help_text="The system or channel from which this job posting originated or was first published."
+    )
+
     class Meta:
         ordering = ['-created_at']
         verbose_name = "Job Posting"
@@ -114,7 +123,7 @@ class JobPosting(Base):
     def save(self, *args, **kwargs):
         # Generate unique internal job ID if not exists
         if not self.internal_job_id:
-            prefix = "UNIV"
+            prefix = "KAAUH"
             year = timezone.now().year
             # Get next sequential number
             last_job = JobPosting.objects.filter(
@@ -188,6 +197,12 @@ class Candidate(Base):
     offer_status = models.CharField(choices=Status.choices,max_length=100, null=True, blank=True, verbose_name=_('Offer Status'))
     join_date = models.DateField(null=True, blank=True, verbose_name=_('Join Date'))
 
+    # Scoring fields (populated by signal)
+    match_score = models.IntegerField(null=True, blank=True)
+    strengths = models.TextField(blank=True)
+    weaknesses = models.TextField(blank=True)
+    criteria_checklist = models.JSONField(default=dict, blank=True)
+
     class Meta:
         verbose_name = _('Candidate')
         verbose_name_plural = _('Candidates')
@@ -297,6 +312,8 @@ class Form(models.Model):
     def __str__(self):
         return self.title
 
+
+
 class FormSubmission(models.Model):
     form = models.ForeignKey(Form, on_delete=models.CASCADE, related_name='submissions')
     submission_data = models.JSONField(default=dict)  # Stores form responses
@@ -312,4 +329,32 @@ class UploadedFile(models.Model):
     field_id = models.CharField(max_length=100)
     file = models.FileField(upload_to='form_uploads/%Y/%m/%d/')
     original_filename = models.CharField(max_length=255)
-    uploaded_at = models.DateTimeField(auto_now_add=True)
+    uploaded_at = models.DateTimeField(auto_now_add=True)
+
+
+
+
+class Source(models.Model):
+    class SourceType(models.TextChoices):
+        ATS = 'ATS', _('Applicant Tracking System')
+        CRM = 'ERP', _('ERP system')
+       
+
+    name = models.CharField(
+        max_length=100,
+        choices=SourceType.choices,
+        verbose_name=_('Source Type')
+    )
+    
+    created_at = models.DateTimeField(auto_now_add=True)
+
+    def __str__(self):
+        return f"{self.get_name_display()}"
+
+    class Meta:
+        verbose_name = _('Source')
+        verbose_name_plural = _('Sources')
+
+
+
+     

recruitment/signals.py

@@ -0,0 +1,71 @@
+from django.db.models.signals import post_save
+from django.dispatch import receiver
+from . import models
+
+# @receiver(post_save, sender=models.Candidate)
+# def parse_resume(sender, instance, created, **kwargs):
+#     if instance.resume and not instance.summary:
+#         from .utils import extract_summary_from_pdf,match_resume_with_job_description
+#         summary = extract_summary_from_pdf(instance.resume.path)
+#         if 'error' not in summary:
+#             instance.summary = summary
+#             instance.save()
+
+        # match_resume_with_job_description
+
+import logging
+logger = logging.getLogger(__name__)
+import os
+from .utils import extract_text_from_pdf,score_resume_with_openrouter
+
+
+
+@receiver(post_save, sender=models.Candidate)
+def score_candidate_resume(sender, instance, created, **kwargs):
+    # Skip if no resume or OpenRouter not configured
+    if instance.resume is None:
+        return 
+    if kwargs.get('update_fields') is not None:
+        return
+
+    # Optional: Only re-score if resume changed (advanced: track file hash)
+    # For simplicity, we score on every save with a resume
+
+    try:
+        # Get absolute file path
+        file_path = instance.resume.path
+        if not os.path.exists(file_path):
+            logger.warning(f"Resume file not found: {file_path}")
+            return
+
+        resume_text = extract_text_from_pdf(file_path)
+        # if not resume_text:
+        #     instance.scoring_error = "Could not extract text from resume."
+        #     instance.save(update_fields=['scoring_error'])
+        #     return
+
+        result = score_resume_with_openrouter(resume_text)
+
+        # Update candidate with scoring results
+        instance.match_score = result.get('match_score')
+        instance.strengths = result.get('strengths', '')
+        instance.weaknesses = result.get('weaknesses', '')
+        instance.criteria_checklist = result.get('criteria_checklist', {})
+       
+       
+
+        # Save only scoring-related fields to avoid recursion
+        instance.save(update_fields=[
+            'match_score', 'strengths', 'weaknesses',
+            'criteria_checklist'
+        ])
+
+        logger.info(f"Successfully scored resume for candidate {instance.id}")
+
+    except Exception as e:
+        # error_msg = str(e)[:500]  # Truncate to fit TextField
+        # instance.scoring_error = error_msg
+        # instance.save(update_fields=['scoring_error'])
+        logger.error(f"Failed to score resume for candidate {instance.id}: {e}")
+    
+   

recruitment/utils.py

@@ -1,32 +1,124 @@
-import os
-import fitz  # PyMuPDF
-import spacy
-import requests
+# import os
+# import fitz  # PyMuPDF
+# import spacy
+# import requests
 from recruitment import models
 from django.conf import settings
 
-nlp = spacy.load("en_core_web_sm")
+# nlp = spacy.load("en_core_web_sm")
 
-def extract_text_from_pdf(pdf_path):
+# def extract_text_from_pdf(pdf_path):
+#     text = ""
+#     with fitz.open(pdf_path) as doc:
+#         for page in doc:
+#             text += page.get_text()
+#     return text
+
+# def extract_summary_from_pdf(pdf_path):
+#     if not os.path.exists(pdf_path):
+#         return {'error': 'File not found'}
+
+#     text = extract_text_from_pdf(pdf_path)
+#     doc = nlp(text)
+#     summary = {
+#         'name': doc.ents[0].text if doc.ents else '',
+#         'skills': [chunk.text for chunk in doc.noun_chunks if len(chunk.text.split()) > 1],
+#         'summary': text[:500]
+#     }
+#     return summary
+
+import requests
+from PyPDF2 import PdfReader
+import os
+import json
+import logging
+logger = logging.getLogger(__name__)
+
+OPENROUTER_API_KEY ='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'
+OPENROUTER_MODEL = 'qwen/qwen-2.5-72b-instruct:free'
+
+if not OPENROUTER_API_KEY:
+    logger.warning("OPENROUTER_API_KEY not set. Resume scoring will be skipped.")
+
+def extract_text_from_pdf(file_path):
+    print("text extraction")
     text = ""
-    with fitz.open(pdf_path) as doc:
-        for page in doc:
-            text += page.get_text()
-    return text
+    try:
+        with open(file_path, "rb") as f:
+            reader = PdfReader(f)
+            for page in reader.pages:
+                text += (page.extract_text() or "")
+    except Exception as e:
+        logger.error(f"PDF extraction failed: {e}")
+        raise
+    return text.strip()
 
-def extract_summary_from_pdf(pdf_path):
-    if not os.path.exists(pdf_path):
-        return {'error': 'File not found'}
+def score_resume_with_openrouter(resume_text):    
+    prompt = f"""
+You are an expert technical recruiter. Your task is to score the following candidate for the role of a Senior Data Analyst based on the provided job criteria.
 
-    text = extract_text_from_pdf(pdf_path)
-    doc = nlp(text)
-    summary = {
-        'name': doc.ents[0].text if doc.ents else '',
-        'skills': [chunk.text for chunk in doc.noun_chunks if len(chunk.text.split()) > 1],
-        'summary': text[:500]
-    }
-    return summary
+**Job Criteria:**
+- Must-Have Skills: Python, SQL, 5+ years of experience.
+- Nice-to-Have Skills: Tableau, AWS.
+- Experience: Must have led at least one project.
 
+**Candidate's Extracted Resume Text:**
+\"\"\"
+{resume_text}
+\"\"\"
+
+**Your Task:**
+Provide a response in strict JSON format with the following keys:
+1. 'match_score': A score from 0 to 100 representing how well the candidate fits the role.
+2. 'strengths': A brief summary of why the candidate is a strong fit, referencing specific criteria.
+3. 'weaknesses': A brief summary of where the candidate falls short or what criteria are missing.
+4. 'criteria_checklist': An object where you rate the candidate's match for each specific criterion (e.g., {{'Python': 'Met', 'AWS': 'Not Mentioned'}}).
+
+Only output valid JSON. Do not include any other text.
+"""
+    print("model call")
+    response = requests.post(
+        url="https://openrouter.ai/api/v1/chat/completions",
+        headers={
+            "Authorization": f"Bearer {OPENROUTER_API_KEY}",
+            "Content-Type": "application/json",
+        },
+        data=json.dumps({
+            "model": OPENROUTER_MODEL,
+            "messages": [{"role": "user", "content": prompt}],            
+        },
+        )
+    )
+    # print(response.status_code)
+    # print(response.json())
+    res = {}
+    print(response.status_code)
+    if response.status_code == 200:
+        res = response.json()
+        content = res["choices"][0]['message']['content']
+        try:
+            print(content)
+            content = content.replace("```json","").replace("```","")
+            print(content)
+            res = json.loads(content)
+            print(res)
+        except Exception as e:
+            print(e)
+
+        # res = raw_output["choices"][0]["message"]["content"]        
+    else:
+        print("error response")
+    return res
+    # print(f"rawraw_output)
+    # print(response)
+
+
+
+# def match_resume_with_job_description(resume, job_description,prompt=""):
+#     resume_doc = nlp(resume)
+#     job_doc = nlp(job_description)
+#     similarity = resume_doc.similarity(job_doc)
+#     return similarity
 
 def dashboard_callback(request, context):
     total_jobs = models.Job.objects.count()