Unicode_description_generator.py

"""
=========================================================
Smart Unicode Description Generator for Wikimedia Commons
=========================================================
[ SETUP & USAGE ]
1. Install reqs: pip install pywikibot
2. Login config: Ensure 'user-config.py' is set up.
3. Purpose     : Scans filenames for Unicode codepoints (U+XXXX).
                 Automatically detects the character and its official 
                 Unicode name. Generates a smart description with options
                 for custom Font Names and Glyph Styles.
4. Run         : Set CONFIGURATION below and execute.
=========================================================
"""

import pywikibot
from pywikibot import pagegenerators
import re
import time
import unicodedata

# ================= CONFIGURATION =================

# 1. Target Categories
TARGET_CATEGORIES = [
    'Category:YOUR_CATEGORY_NAME_HERE'
]

# 2. Font & Style Settings
# If FONT_NAME is provided, it will be used directly. 
# If left empty (""), it will check EXTRACT_FONT_FROM_FILENAME.
FONT_NAME = ""                                 # e.g., "GNU Unifont" or "Noto Sans Bengali"

# Set to True to extract font name from the filename (e.g., "... in Noto Sans - U+...").
# Only works if FONT_NAME above is left empty.
EXTRACT_FONT_FROM_FILENAME = True              

# Optional: Prepends a style to the description (e.g., "Dotted style Bengali letter...").
GLYPH_TYPE = ""                                # e.g., "Dotted", "Pixelated", "Bold". Leave empty for default.

# 3. Advanced Controls
PROCESS_SUBCATEGORIES = False  # Set True to check inside subcategories
SLEEP_TIME = 5                 # Seconds to pause after saving
DRY_RUN = False                # Set True to test without making real edits

# =================================================

def generate_smart_description(filename):
    """
    Extracts Unicode data and builds a smart description based on user configuration.
    """
    # 1. Extract the Unicode Hex (Handles U+XXXX, U-XXXX, U+XXXXXX)
    code_match = re.search(r'U[-+]([0-9A-Fa-f]{4,6})', filename, re.IGNORECASE)
    
    if not code_match:
        return None # Skip if no Unicode pattern is found
    
    hex_str = code_match.group(1).upper()
    
    # 2. Get Character and Official Unicode Name
    try:
        char = chr(int(hex_str, 16))
        char_name = unicodedata.name(char).capitalize()
    except ValueError:
        char = "?"
        char_name = "Unknown Unicode character"

    # Avoid printing invisible control characters
    char_display = f" ({char})" if not unicodedata.category(char).startswith('C') else ""

    # 3. Determine Font Name
    final_font_name = ""
    
    if FONT_NAME.strip():
        # Priority 1: Use hardcoded font name from config
        final_font_name = FONT_NAME.strip()
    elif EXTRACT_FONT_FROM_FILENAME:
        # Priority 2: Try to extract from filename
        font_match = re.search(r'\bin\s+(.*?)\s*(?:-|–|—)?\s*U[-+]', filename, re.IGNORECASE)
        if font_match:
            extracted_name = font_match.group(1).strip()
            final_font_name = re.sub(r'[-–—]\s*$', '', extracted_name).strip()

    # 4. Determine Glyph Style Prefix
    style_prefix = f"{GLYPH_TYPE.strip()} style " if GLYPH_TYPE.strip() else ""

    # 5. Build Final Description String with Compart Link
    compart_link = f"[https://www.compart.com/en/unicode/U+{hex_str} U+{hex_str}]"
    
    if final_font_name:
        # Format: [Style] Bengali letter ka (ক) in Kalpurush font - [Link]
        return f"{style_prefix}{char_name}{char_display} in {final_font_name} font - {compart_link}"
    else:
        # Format: [Style] Bengali letter ka (ক) - [Link]
        return f"{style_prefix}{char_name}{char_display} - {compart_link}"

def main():
    try:
        site = pywikibot.Site('commons', 'commons')
        site.login()
        print(f"Logged in as: {site.user()}")
    except Exception as e:
        print(f"Login Failed: {e}")
        return

    for cat_name in TARGET_CATEGORIES:
        print(f"\n[{'DRY RUN' if DRY_RUN else 'LIVE'}] Processing Category: {cat_name}...")
        
        try:
            cat = pywikibot.Category(site, cat_name)
            gen = pagegenerators.CategorizedPageGenerator(cat, recurse=PROCESS_SUBCATEGORIES)
        except pywikibot.exceptions.InvalidTitleError:
            print(f"[ERROR] Invalid category name: {cat_name}")
            continue

        for page in gen:
            if not page.is_filepage():
                continue

            file_title = page.title(with_ns=False)

            # Generate the dynamic description string
            smart_desc = generate_smart_description(file_title)

            if not smart_desc:
                continue

            current_text = page.text

            # Regex pattern to find |description={{en|...}}
            desc_pattern = r'(\|description\s*=\s*\{\{en\|)(.*?)(\}\})'

            if re.search(desc_pattern, current_text):
                
                # Replace existing description with the new smart description
                new_text = re.sub(desc_pattern, r'\1' + smart_desc + r'\3', current_text)

                if new_text != current_text:
                    if DRY_RUN:
                        print(f"[-] DRY RUN: {file_title}")
                        print(f"    -> Would update to: '{smart_desc}'")
                        continue

                    print(f"[\u2713] Updating: {file_title}")
                    print(f"    -> New Desc: {smart_desc}")
                    
                    page.text = new_text
                    
                    try:
                        page.save(summary=f"Smart description update: {smart_desc}")
                        time.sleep(SLEEP_TIME)
                        
                    except pywikibot.exceptions.LockedPageError:
                        print(f"    [SKIP] Page is locked.")
                    except pywikibot.exceptions.OtherPageSaveError as e:
                        print(f"    [ERROR] Save failed: {e}")
                    except Exception as e:
                        print(f"    [ERROR] Generic error: {e}")
                else:
                    pass
            else:
                print(f"[SKIP] No English description template found: {file_title}")

    if DRY_RUN:
        print("\n\u2705 DRY RUN complete. No files were actually modified.")
    else:
        print("\n\u2705 Batch processing completed.")

if __name__ == '__main__':
    main()