lsj_dict.py

import xml.etree.ElementTree as ET
from collections import defaultdict
from autocomplete import load_compressed_word_list
import json
import streamlit as st
import re

def read_xml(file):
    """
    Read an XML file of the Greek LSJ dictionary 
    and return a dictionary with the words and their definitions.
    """
    tree = ET.parse(file)
    root = tree.getroot()
    
    xml_info = defaultdict(dict)
    
    for entry in root.findall('.//entryFree'):
        entry_info = extract_entry_info(entry)
        
        xml_info[entry_info['lemma']] = entry_info
        
    return xml_info


def extract_entry_info(entry):
    """
    Extract information from an entry in the LSJ dictionary.
    """
    definitions = defaultdict(dict)
    
    # Save the lemma in the dictionary and remove digits
    lemma = ''.join([i for i in entry.get('key') if not i.isdigit()])
    
    # Save the orthographies in the dictionary
    orthographies = [orth.text for orth in entry.findall('orth')]
    definitions[lemma]['orthographies'] = orthographies
    
    # Check if there is a tr element with a definition
    definition = ' '.join(entry.itertext()).strip()
    definitions[lemma]['definitions'] = {'tr': definition}


    # text = get_descendants_text(entry)
    
    text = get_all_text(entry)
    
    cleaned_text = prettify_text(text)
    
    definitions[lemma]['definitions']['text'] = cleaned_text
        
    
    return {'lemma': lemma, 'orthographies': orthographies, 'definitions': definitions[lemma]['definitions']}   
            

def get_all_text(element):
    """Recursively collect text from an element and all its descendants."""
    text = (element.text or "")
    for child in element:
        if child.tag == 'sense':
            level = child.get('n')
            text += f"[SENSE_SEPARATOR][level={level}]\n\n"
        elif child.tag == 'tr' and element.tag == 'sense':
            if child.text is not None:
                text += f"<tr>{child.text.strip()}</tr>\n"
            # Skip further recursion for this child since we are already handling its text
            text += (child.tail or "") + " "
            continue
        text += get_all_text(child) + " "
        text += (child.tail or "")  + " "
    return text
   

def get_descendants_text(element):
    """
    Get all the text of the descendants of a given element, separating every 'sense' element.
    """
    text = ""

    for child in element:
        if child.tag == 'sense':
            # Add a separator before each 'sense' element
            text += f"[SENSE_SEPARATOR]\n\n"
        if child.tag == 'tr' and element.tag == 'sense':
            # Add [tr] tags around text inside 'tr' tags within 'sense' tags
            if child.text is not None:
                text += f"<tr>{child.text.strip()}</tr>\n"
                text += child.tail
        else:
            if child.get('n') and len(child.get('n')) <= 2:
                text += f"{child.get('n')}. "
            text += child.text or ""
            text += get_descendants_text(child)
    return text
    



def prettify_text(text):
    """
    Prettify the text of the definitions into a readable format, 
    adding [tr] tags to text inside 'tr' tags within 'sense' tags.
    """
    # Split text using the sense separator
    parts = text.split("[SENSE_SEPARATOR]")
    
    # Prettify each part separately
    prettified_parts = []
    
    for part in parts:       
        # Remove leading and trailing whitespace and join lines with a space
        cleaned_part = ' '.join(line.strip() for line in part.split('\n') if line.strip())
        prettified_parts.append(cleaned_part)
    
    # Join prettified parts using sense separator and newline
    prettified_text = "\n\n[SENSE_SEPARATOR] ".join(prettified_parts)
    
    return prettified_text


def full_dictionary():
    """
    Return the full dictionary of the LSJ dictionary.
    """
    merged_info = {}
    for i in range(1, 28):  # eng1 to eng27
        file = f"LSJ_GreekUnicode/grc.lsj.perseus-eng{i}.xml"
        xml_info = read_xml(file)
        for lemma, info in xml_info.items():
            # Merge dictionaries, assuming word is unique across all files
            merged_info.setdefault(lemma, {}).update(info)
    
    return merged_info
    

def format_text(data):
    """
        Modify text to desired template
    """
    text = data['definitions']['text']
    
    # Change <tr> tags to bold
    text = text.replace("<tr>", "<span class='tr'> ").replace("</tr>", "</span>")
    
    pattern = r"\s+([,;:.()\"\'\[\]])\s+"
    text = re.sub(pattern, r"\1 ", text)
    
    pattern_2 = r"(\S)([,;:.()\"\'\[\]])(\S)"
    text = re.sub(pattern_2, r"\1\2 \3", text)
    # .replace(",", ", ").replace(";", "; ").replace(":", ": ").replace("(", " (").replace(")", ") ").replace("[", " [").replace("]", "] ").replace(" ,", ", ").replace(" ; ", "; ").replace(" : ", ": ").replace(" ." , ". ")
    
    formatted_text = []
    
    primary_indicators = [
        "A", "B", "C", "D", "E", "F", "G", "H", "I", "J"
    ]
    
    secondary_indicators = [
        "I", "II", "III", "IV", "V", "VI", "VII", "VIII", "IX", "X"
    ]
    
    tertiary_indicators = [
        "2", "3", "4", "5", "6", "7", "8", "9", "10"
    ]
    
    quaternary_indicators = [
        "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", 
        "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", 
        "u", "v", "w", "x", "y", "z"
    ]
    
    header = text.split("\n\n[SENSE_SEPARATOR]")[0]
    formatted_text.append(header)
    
    for text_part in text.split("[SENSE_SEPARATOR]")[1:]:
        level = text_part.split("level=")[1].split("]")[0]
        text_part = text_part.replace(f"[level={level}]", "")
        if level:
            if level == "A":
                formatted_text.append(f"<div class='list-class primary-class'> {text_part.replace('[SENSE_SEPARATOR]', '')} </div>")
            elif level in secondary_indicators:
                formatted_text.append(f"<div class='list-class secondary-class'><span class='secondary-indicator'>{level}. </span> {text_part.replace('[SENSE_SEPARATOR]', '')} </div>")
            elif level in tertiary_indicators:
                formatted_text.append(f"<div class='list-class tertiary-class'> <span class='tertiary-indicator'>{level}. </span> {text_part.replace('[SENSE_SEPARATOR]', '')} </div>")
            elif level in quaternary_indicators:
                formatted_text.append(f"<div class='list-class quaternary-class'> <span class='quaternary-indicator'>{level}. </span> {text_part.replace('[SENSE_SEPARATOR]', '')} </div> ")
            elif level in primary_indicators:
                formatted_text.append(f"<div class='list-class primary-class'> <span class='primary-indicator'>{level}. </span> {text_part.replace('[SENSE_SEPARATOR]', '')} </div>")
    
    return '\n'.join(formatted_text)






def main(): 
    
    # This code is used to convert the .xml files into a .json file
    # TO DO: Make seperate function
    download = True
    
    if download is True:
        merged_info = {}
        for i in range(1, 28):  # eng1 to eng27
            file = f"LSJ_GreekUnicode/grc.lsj.perseus-eng{i}.xml"
            xml_info = read_xml(file)
            for word, info in xml_info.items():
                # Merge dictionaries, assuming word is unique across all files
                merged_info.setdefault(word, {}).update(info)

        
        # Store merged dictionaries as .json file with pretty print
        with open("lsj_dict.json", "w", encoding="utf-8") as file:
            json.dump(merged_info, file, ensure_ascii=False, indent=4)
        
    

    lemma_dict = json.load(open('lsj_dict.json', 'r'))



if __name__ == "__main__":
    main()