Spaces:

ResumeSelectors
/

resume-analyzer

Sleeping

mohit kumar

Initial Resume Analyzer deploy

a4db4ad 5 months ago

8.62 kB

# =============================================================================
# engine/__init__.py — Resume Engine Package Initialiser
# =============================================================================
# PURPOSE:
#   Makes engine/ a proper Python package.
#   Exposes all public functions from every module in one clean import.
#
# BENEFIT — instead of writing this in main.py:
#   from engine.parser    import load_resumes_from_folder
#   from engine.database  import init_db, insert_resume
#   from engine.extractor import build_resume_record
#   ... (7 separate import lines)
#
# You can write this:
#   from engine import load_resumes_from_folder, init_db, build_resume_record
#
# STATUS LEGEND (updated as each step is completed):
#   ✅ Built and tested
#   🔲 Not built yet — imported conditionally so missing modules don't crash
#
# MODULES:
#   ✅ parser.py      Step 3  — PDF/DOCX text extraction
#   ✅ database.py    Step 2  — SQLite read/write operations
#   🔲 extractor.py   Step 4  — Structured field extraction (skills, exp, etc)
#   🔲 jd_parser.py   Step 5  — Job description parsing
#   🔲 matcher.py     Step 6  — Scoring engine
#   🔲 explainer.py   Step 7  — Summary + interview question generation
# =============================================================================

# ── Standard library ──────────────────────────────────────────────────────────
import os
import sys

# Ensure project root is always on the path so config.py is importable
# regardless of where you run Python from
_ENGINE_DIR  = os.path.dirname(os.path.abspath(__file__))
_PROJECT_ROOT = os.path.dirname(_ENGINE_DIR)
if _PROJECT_ROOT not in sys.path:
    sys.path.insert(0, _PROJECT_ROOT)


# =============================================================================
# ✅ STEP 3 — parser.py
# Text extraction from PDF and DOCX resume files
# =============================================================================
try:
    from engine.parser import (
        extract_text,               # Route any file to the correct extractor
        extract_text_from_pdf,      # PDF-specific extractor (pdfplumber)
        extract_text_from_docx,     # DOCX-specific extractor (python-docx)
        load_resumes_from_folder,   # Batch-load all resumes from a folder
        clean_text,                 # Internal text cleaner (useful for JD text too)
    )
    _PARSER_OK = True
except (ImportError, SyntaxError, Exception) as e:
    _PARSER_OK = False
    import warnings
    warnings.warn(f"[engine] parser.py not available: {e}", ImportWarning)


# =============================================================================
# ✅ STEP 2 — database.py
# All SQLite read/write operations
# =============================================================================
try:
    from engine.database import (
        init_db,                # Create all tables (run once on startup)
        insert_resume,          # Insert one parsed resume → returns resume_id
        insert_jd,              # Insert one job description → returns jd_id
        insert_match_result,    # Insert one scored match → returns match_id
        get_all_resumes,        # Fetch all resumes as list of dicts
        get_resume_by_id,       # Fetch one resume by resume_id
        get_all_jds,            # Fetch all job descriptions
        get_matches_by_jd,      # Fetch all matches for a JD, sorted by score
        export_matches_to_csv,  # Save match results to a CSV file
    )
    _DATABASE_OK = True
except (ImportError, SyntaxError, Exception) as e:
    _DATABASE_OK = False
    import warnings
    warnings.warn(f"[engine] database.py not available: {e}", ImportWarning)


# =============================================================================
# 🔲 STEP 4 — extractor.py
# Structured field extraction from raw resume text
# (Skills, experience, education, stability, recency etc.)
# =============================================================================
try:
    from engine.extractor import (
        build_resume_record,    # Master function — returns complete DB-ready dict
        extract_candidate_name,
        extract_contact,
        extract_skills,
        extract_soft_skills,
        extract_experience,
        extract_education,
        extract_projects,
        extract_certifications,
        extract_recency,
    )
    _EXTRACTOR_OK = True
except (ImportError, SyntaxError, Exception):
    _EXTRACTOR_OK = False   # Silently skip — not built yet


# =============================================================================
# 🔲 STEP 5 — jd_parser.py
# Job description parsing
# =============================================================================
try:
    from engine.jd_parser import (
        parse_jd_text,          # Parse raw JD text → structured dict
        load_jd_from_file,      # Load JD from .txt or .pdf file
    )
    _JD_PARSER_OK = True
except (ImportError, SyntaxError, Exception):
    _JD_PARSER_OK = False


# =============================================================================
# 🔲 STEP 6 — matcher.py
# Scoring engine — computes all 5 component scores + final weighted score
# =============================================================================
try:
    from engine.matcher import (
        match_resume_to_jd,     # Master scorer — returns complete match dict
        score_skill_match,
        score_experience,
        score_stability,
        score_recency,
        score_education,
        compute_final_score,
    )
    _MATCHER_OK = True
except (ImportError, SyntaxError, Exception):
    _MATCHER_OK = False


# =============================================================================
# 🔲 STEP 7 — explainer.py
# Plain-language summaries + interview question generation
# =============================================================================
try:
    from engine.explainer import (
        generate_summary,           # Build structured text explanation
        generate_interview_questions, # Call OpenAI to generate questions
        build_full_output,          # Combine match result + summary + questions
    )
    _EXPLAINER_OK = True
except (ImportError, SyntaxError, Exception):
    _EXPLAINER_OK = False


# =============================================================================
# PACKAGE STATUS — what is ready vs pending
# =============================================================================

# Dictionary you can query programmatically: engine.STATUS
STATUS = {
    "parser"    : _PARSER_OK,
    "database"  : _DATABASE_OK,
    "extractor" : _EXTRACTOR_OK,
    "jd_parser" : _JD_PARSER_OK,
    "matcher"   : _MATCHER_OK,
    "explainer" : _EXPLAINER_OK,
}

# Convenience: list of modules that are fully ready
READY_MODULES = [name for name, ok in STATUS.items() if ok]


# =============================================================================
# SELF-TEST
# Run this file directly to see which modules are loaded:
#   python engine/__init__.py
# =============================================================================

if __name__ == "__main__":

    GREEN  = "\033[92m"
    RED    = "\033[91m"
    YELLOW = "\033[93m"
    BOLD   = "\033[1m"
    RESET  = "\033[0m"

    step_map = {
        "parser"    : "Step 3",
        "database"  : "Step 2",
        "extractor" : "Step 4",
        "jd_parser" : "Step 5",
        "matcher"   : "Step 6",
        "explainer" : "Step 7",
    }

    print(f"\n{BOLD}{'='*55}{RESET}")
    print(f"{BOLD}  RESUME ENGINE — MODULE STATUS{RESET}")
    print(f"{BOLD}{'='*55}{RESET}\n")

    all_built = ["parser", "database"]   # steps confirmed built
    for module, ok in STATUS.items():
        step  = step_map[module]
        if ok:
            tag = f"{GREEN}✅ LOADED{RESET}"
        elif module in all_built:
            tag = f"{RED}✗  FAILED TO IMPORT — check for errors in {module}.py{RESET}"
        else:
            tag = f"{YELLOW}🔲 NOT BUILT YET{RESET}"

        print(f"  {step:<8}  {module:<12}  {tag}")

    built_count = sum(STATUS.values())
    total       = len(STATUS)

    print(f"\n{BOLD}  Progress: {built_count}/{total} modules loaded{RESET}")

    if STATUS["parser"] and STATUS["database"]:
        print(f"\n  {GREEN}{BOLD}✓ Core modules ready — proceed to Step 4 (extractor.py){RESET}")
    else:
        print(f"\n  {RED}Fix import errors above before proceeding.{RESET}")

    print(f"\n{'='*55}\n")

	# =============================================================================
	# engine/__init__.py — Resume Engine Package Initialiser
	# =============================================================================
	# PURPOSE:
	# Makes engine/ a proper Python package.
	# Exposes all public functions from every module in one clean import.
	#
	# BENEFIT — instead of writing this in main.py:
	# from engine.parser import load_resumes_from_folder
	# from engine.database import init_db, insert_resume
	# from engine.extractor import build_resume_record
	# ... (7 separate import lines)
	#
	# You can write this:
	# from engine import load_resumes_from_folder, init_db, build_resume_record
	#
	# STATUS LEGEND (updated as each step is completed):
	# ✅ Built and tested
	# 🔲 Not built yet — imported conditionally so missing modules don't crash
	#
	# MODULES:
	# ✅ parser.py Step 3 — PDF/DOCX text extraction
	# ✅ database.py Step 2 — SQLite read/write operations
	# 🔲 extractor.py Step 4 — Structured field extraction (skills, exp, etc)
	# 🔲 jd_parser.py Step 5 — Job description parsing
	# 🔲 matcher.py Step 6 — Scoring engine
	# 🔲 explainer.py Step 7 — Summary + interview question generation
	# =============================================================================

	# ── Standard library ──────────────────────────────────────────────────────────
	import os
	import sys

	# Ensure project root is always on the path so config.py is importable
	# regardless of where you run Python from
	_ENGINE_DIR = os.path.dirname(os.path.abspath(__file__))
	_PROJECT_ROOT = os.path.dirname(_ENGINE_DIR)
	if _PROJECT_ROOT not in sys.path:
	sys.path.insert(0, _PROJECT_ROOT)


	# =============================================================================
	# ✅ STEP 3 — parser.py
	# Text extraction from PDF and DOCX resume files
	# =============================================================================
	try:
	from engine.parser import (
	extract_text, # Route any file to the correct extractor
	extract_text_from_pdf, # PDF-specific extractor (pdfplumber)
	extract_text_from_docx, # DOCX-specific extractor (python-docx)
	load_resumes_from_folder, # Batch-load all resumes from a folder
	clean_text, # Internal text cleaner (useful for JD text too)
	)
	_PARSER_OK = True
	except (ImportError, SyntaxError, Exception) as e:
	_PARSER_OK = False
	import warnings
	warnings.warn(f"[engine] parser.py not available: {e}", ImportWarning)


	# =============================================================================
	# ✅ STEP 2 — database.py
	# All SQLite read/write operations
	# =============================================================================
	try:
	from engine.database import (
	init_db, # Create all tables (run once on startup)
	insert_resume, # Insert one parsed resume → returns resume_id
	insert_jd, # Insert one job description → returns jd_id
	insert_match_result, # Insert one scored match → returns match_id
	get_all_resumes, # Fetch all resumes as list of dicts
	get_resume_by_id, # Fetch one resume by resume_id
	get_all_jds, # Fetch all job descriptions
	get_matches_by_jd, # Fetch all matches for a JD, sorted by score
	export_matches_to_csv, # Save match results to a CSV file
	)
	_DATABASE_OK = True
	except (ImportError, SyntaxError, Exception) as e:
	_DATABASE_OK = False
	import warnings
	warnings.warn(f"[engine] database.py not available: {e}", ImportWarning)


	# =============================================================================
	# 🔲 STEP 4 — extractor.py
	# Structured field extraction from raw resume text
	# (Skills, experience, education, stability, recency etc.)
	# =============================================================================
	try:
	from engine.extractor import (
	build_resume_record, # Master function — returns complete DB-ready dict
	extract_candidate_name,
	extract_contact,
	extract_skills,
	extract_soft_skills,
	extract_experience,
	extract_education,
	extract_projects,
	extract_certifications,
	extract_recency,
	)
	_EXTRACTOR_OK = True
	except (ImportError, SyntaxError, Exception):
	_EXTRACTOR_OK = False # Silently skip — not built yet


	# =============================================================================
	# 🔲 STEP 5 — jd_parser.py
	# Job description parsing
	# =============================================================================
	try:
	from engine.jd_parser import (
	parse_jd_text, # Parse raw JD text → structured dict
	load_jd_from_file, # Load JD from .txt or .pdf file
	)
	_JD_PARSER_OK = True
	except (ImportError, SyntaxError, Exception):
	_JD_PARSER_OK = False


	# =============================================================================
	# 🔲 STEP 6 — matcher.py
	# Scoring engine — computes all 5 component scores + final weighted score
	# =============================================================================
	try:
	from engine.matcher import (
	match_resume_to_jd, # Master scorer — returns complete match dict
	score_skill_match,
	score_experience,
	score_stability,
	score_recency,
	score_education,
	compute_final_score,
	)
	_MATCHER_OK = True
	except (ImportError, SyntaxError, Exception):
	_MATCHER_OK = False


	# =============================================================================
	# 🔲 STEP 7 — explainer.py
	# Plain-language summaries + interview question generation
	# =============================================================================
	try:
	from engine.explainer import (
	generate_summary, # Build structured text explanation
	generate_interview_questions, # Call OpenAI to generate questions
	build_full_output, # Combine match result + summary + questions
	)
	_EXPLAINER_OK = True
	except (ImportError, SyntaxError, Exception):
	_EXPLAINER_OK = False


	# =============================================================================
	# PACKAGE STATUS — what is ready vs pending
	# =============================================================================

	# Dictionary you can query programmatically: engine.STATUS
	STATUS = {
	"parser" : _PARSER_OK,
	"database" : _DATABASE_OK,
	"extractor" : _EXTRACTOR_OK,
	"jd_parser" : _JD_PARSER_OK,
	"matcher" : _MATCHER_OK,
	"explainer" : _EXPLAINER_OK,
	}

	# Convenience: list of modules that are fully ready
	READY_MODULES = [name for name, ok in STATUS.items() if ok]


	# =============================================================================
	# SELF-TEST
	# Run this file directly to see which modules are loaded:
	# python engine/__init__.py
	# =============================================================================

	if __name__ == "__main__":

	GREEN = "\033[92m"
	RED = "\033[91m"
	YELLOW = "\033[93m"
	BOLD = "\033[1m"
	RESET = "\033[0m"

	step_map = {
	"parser" : "Step 3",
	"database" : "Step 2",
	"extractor" : "Step 4",
	"jd_parser" : "Step 5",
	"matcher" : "Step 6",
	"explainer" : "Step 7",
	}

	print(f"\n{BOLD}{'='*55}{RESET}")
	print(f"{BOLD} RESUME ENGINE — MODULE STATUS{RESET}")
	print(f"{BOLD}{'='*55}{RESET}\n")

	all_built = ["parser", "database"] # steps confirmed built
	for module, ok in STATUS.items():
	step = step_map[module]
	if ok:
	tag = f"{GREEN}✅ LOADED{RESET}"
	elif module in all_built:
	tag = f"{RED}✗ FAILED TO IMPORT — check for errors in {module}.py{RESET}"
	else:
	tag = f"{YELLOW}🔲 NOT BUILT YET{RESET}"

	print(f" {step:<8} {module:<12} {tag}")

	built_count = sum(STATUS.values())
	total = len(STATUS)

	print(f"\n{BOLD} Progress: {built_count}/{total} modules loaded{RESET}")

	if STATUS["parser"] and STATUS["database"]:
	print(f"\n {GREEN}{BOLD}✓ Core modules ready — proceed to Step 4 (extractor.py){RESET}")
	else:
	print(f"\n {RED}Fix import errors above before proceeding.{RESET}")

	print(f"\n{'='*55}\n")