pipeline.py

from tabulate import tabulate
from modules import visual_checks, text_checks, content_checks

def classify(image_path):
    """Perform complete classification with detailed results."""
    # Components to check
    components = [
        visual_checks.image_quality,
        visual_checks.ribbon,
        text_checks.tagline,
        text_checks.tooMuchText,
        content_checks.theme,
        content_checks.body,
        text_checks.cta,
        text_checks.tnc,
        visual_checks.gnc
    ]

    # Collect all results
    all_results = {}
    for component in components:
        try:
            results = component(image_path)
            all_results.update(results)
        except Exception as e:
            print(f"Error in component {component.__name__}: {e}")
            # Optionally set default values or log error
            pass

    # Calculate final classification
    # Check if any result value is 1 (or starts with '1' for string results like "1 [Religious]")
    final_classification = 0
    for result in all_results.values():
        if isinstance(result, int):
            if result == 1:
                final_classification = 1
                break
        elif isinstance(result, str):
            if result.startswith('1'):
                final_classification = 1
                break

    # Determine Pass or Fail
    classification_result = "Fail" if final_classification == 1 else "Pass"

    # Prepare the table data
    table_data = []
    labels = [
        "Bad Image Quality", "No Ribbon", "Empty/Illegible/Black Tagline", "Multiple Taglines",
        "Incomplete Tagline", "Hyperlink", "Price Tag", "Excessive Emojis", "Too Much Text",
        "Inappropriate Content", "Religious Content", "High Risk Content",
        "Illegal Content", "Competitor References", "Bad CTA", "Terms & Conditions",
        "Visual Gesture or Icon"
    ]

    # Collect labels responsible for failure
    failure_labels = []
    for label in labels:
        result = all_results.get(label, 0)
        
        is_fail = False
        if isinstance(result, int) and result == 1:
            is_fail = True
        elif isinstance(result, str) and result.startswith('1'):
            is_fail = True
            
        if is_fail:
            failure_labels.append(label)

        table_data.append([label, result])

    # Format the results as a table
    result_table = tabulate(table_data, headers=["LABEL", "RESULT"], tablefmt="fancy_grid")

    # Return the final classification, result table, and failure labels (if any)
    return classification_result, result_table, failure_labels

# Dummy interface for testing (can be enabled if needed)
def classify_dummy(image_path):
    import random
    all_results = {
        "Bad Image Quality": 0,
        "No Ribbon": random.choice([0, 1]),
        "Empty/Illegible/Black Tagline": 0,
        "Multiple Taglines": 0,
        "Incomplete Tagline": 0,
        "Hyperlink": 0,
        "Price Tag": 0,
        "Excessive Emojis": 0,
        "Too Much Text": 0,
        "Inappropriate Content": 0,
        "Religious Content": 0,
        "High Risk Content": 0,
        "Illegal Content": 0,
        "Competitor References": 0,
        "Bad CTA": 0,
        "Terms & Conditions": 0,
        "Visual Gesture or Icon": 0
    }
    
    final_classification = 1 if any(result == 1 for result in all_results.values()) else 0
    classification_result = "Fail" if final_classification == 1 else "Pass"
    
    table_data = []
    labels = list(all_results.keys())
    failure_labels = [label for label in labels if all_results[label] == 1]
    
    for label in labels:
        table_data.append([label, all_results[label]])
        
    result_table = tabulate(table_data, headers=["LABEL", "RESULT"], tablefmt="fancy_grid")
    return classification_result, result_table, failure_labels