geutebruck/parse_all_actions.py

"""
Parse all action documentation HTML files to extract complete action catalog
"""
import re
import json
from pathlib import Path
from bs4 import BeautifulSoup

# Base path for action documentation
ACTION_DOCS_PATH = Path(r"C:\DEV\COPILOT\SOURCES\CODEX\GeViScope\GeViScopeSDK_HTML5\Content\ActionDoku")

# Category mapping from filename to display name
CATEGORY_MAPPING = {
    "ATMActions.htm": "ATM",
    "AudioActions.htm": "Audio",
    "BackupActions.htm": "Backup",
    "CameraControlActions.htm": "Camera Control",
    "CashManagementActions.htm": "Cash Management",
    "DeviceActions.htm": "Device",
    "DigitalContactsActions.htm": "Digital Contacts",
    "LogisticActions.htm": "Logistics",
    "LPSActions.htm": "License Plate System",
    "POSActions.htm": "Point of Sale",
    "RemoteExportActions.htm": "Remote Export",
    "SkiDataActions.htm": "Ski Data",
    "SystemActions.htm": "System",
    "VideoActions.htm": "Video",
    "ViewerActions.htm": "Viewer",
    "ViewerNotificationActions.htm": "Viewer Notification",
    "Lenelactions.htm": "Lenel Access Control"
}

def extract_action_name(action_name_str):
    """
    Extract action name and parameters from string like:
    'CustomAction(Int, String)' -> ('CustomAction', ['Int', 'String'])
    """
    match = re.match(r'([A-Za-z0-9_]+)\((.*?)\)', action_name_str)
    if match:
        name = match.group(1)
        params_str = match.group(2)
        if params_str.strip():
            params = [p.strip() for p in params_str.split(',')]
        else:
            params = []
        return name, params
    return action_name_str, []

def parse_action_html(html_file, category):
    """Parse a single HTML file and extract all actions"""
    actions = {}

    try:
        with open(html_file, 'r', encoding='utf-8') as f:
            soup = BeautifulSoup(f.read(), 'html.parser')

        # Find all h2 headers (each action starts with h2)
        for h2 in soup.find_all('h2'):
            # Get the next p tag which should contain "Action name:"
            action_name_p = h2.find_next('p')
            if not action_name_p:
                continue

            action_name_text = action_name_p.get_text()
            if not action_name_text.startswith('Action name:'):
                continue

            # Extract action name
            action_full_name = action_name_text.replace('Action name:', '').strip()
            action_name, param_types = extract_action_name(action_full_name)

            # Get action category (logical, notification, command)
            category_p = action_name_p.find_next('p')
            action_category = "unknown"
            if category_p and 'Action category:' in category_p.get_text():
                action_category = category_p.get_text().replace('Action category:', '').strip()

            # Get description
            desc_p = category_p.find_next('p') if category_p else h2.find_next('p', class_=None)
            description = desc_p.get_text().strip() if desc_p else ""

            # Find parameter table
            param_table = None
            current = h2.find_next_sibling()
            while current and current.name != 'h2':
                if current.name == 'table':
                    param_table = current
                    break
                current = current.find_next_sibling()

            # Extract parameters from table
            parameters = []
            if param_table:
                rows = param_table.find_all('tr')
                for row in rows[1:]:  # Skip header row
                    cols = row.find_all('td')
                    if len(cols) >= 2:
                        param_name = cols[0].get_text().strip()
                        param_type = cols[1].get_text().strip()
                        if param_name and param_type:
                            parameters.append(param_name)

            # Store action info
            actions[action_name] = {
                "parameters": parameters,
                "description": description,
                "category": category,
                "action_category": action_category,
                "required_caption": True,  # Default, can be refined
                "supports_delay": True     # Default, can be refined
            }

            print(f"  Found: {action_name} ({len(parameters)} params)")

    except Exception as e:
        print(f"Error parsing {html_file}: {e}")

    return actions

def main():
    """Parse all action HTML files and create comprehensive action catalog"""
    all_actions = {}

    print("Parsing action documentation files...")
    print("=" * 60)

    for filename, category_name in CATEGORY_MAPPING.items():
        html_file = ACTION_DOCS_PATH / filename

        if not html_file.exists():
            print(f"⚠ Skipping {filename} (not found)")
            continue

        print(f"\n{category_name} ({filename})")
        print("-" * 60)

        actions = parse_action_html(html_file, category_name)
        all_actions.update(actions)

        print(f"  Total: {len(actions)} actions")

    print("\n" + "=" * 60)
    print(f"TOTAL ACTIONS FOUND: {len(all_actions)}")
    print("=" * 60)

    # Save to JSON
    output_file = Path(r"C:\DEV\COPILOT\all_actions_catalog.json")
    with open(output_file, 'w', encoding='utf-8') as f:
        json.dump(all_actions, f, indent=2, ensure_ascii=False)

    print(f"\nSaved to: {output_file}")

    # Print category breakdown
    print("\nActions by category:")
    category_counts = {}
    for action_name, action_info in all_actions.items():
        cat = action_info['category']
        category_counts[cat] = category_counts.get(cat, 0) + 1

    for cat, count in sorted(category_counts.items(), key=lambda x: x[1], reverse=True):
        print(f"  {cat}: {count} actions")

if __name__ == "__main__":
    main()