db implementation, phrase amount limit

app.py

@@ -0,0 +1,143 @@
+import gradio as gr
+import json
+import re
+import sqlite3
+import logging
+from util import process_json_files
+from gematria import calculate_gematria
+
+logging.basicConfig(level=logging.INFO, format='%(message)s')
+
+def flatten_text(text):
+    """Hilfsfunktion, um verschachtelte Listen zu flachen Listen zu machen."""
+    if isinstance(text, list):
+        return " ".join(flatten_text(item) if isinstance(item, list) else item for item in text)
+    return text
+
+def initialize_database():
+    conn = sqlite3.connect('gematria.db')
+    c = conn.cursor()
+    c.execute('''
+        CREATE TABLE IF NOT EXISTS results (
+            gematria_sum INTEGER,
+            words TEXT,
+            book INTEGER,
+            title TEXT,
+            chapter INTEGER,
+            verse INTEGER
+        )
+    ''')
+    conn.commit()
+    conn.close()
+
+def insert_phrase_to_db(c, gematria_sum, phrase_candidate, book_id, title, chapter_id, verse_id):
+    logging.info(f"Inserting: {gematria_sum}, {phrase_candidate}, {book_id}, {title}, {chapter_id + 1}, {verse_id + 1}")
+    c.execute('''
+        INSERT INTO results (gematria_sum, words, book, title, chapter, verse)
+        VALUES (?, ?, ?, ?, ?, ?)
+    ''', (gematria_sum, phrase_candidate, book_id, title, chapter_id + 1, verse_id + 1))
+
+def populate_database(tanach_texts, max_phrase_length=1):
+    conn = sqlite3.connect('gematria.db')
+    c = conn.cursor()
+    for book_id, text in enumerate(tanach_texts):
+        if not isinstance(text.get('text'), list):
+            continue
+        title = text.get('title', 'Unknown')
+        chapters = text['text']
+        for chapter_id, chapter in enumerate(chapters):
+            if not isinstance(chapter, list):
+                continue
+            chapter_text = ' '.join([flatten_text(verse) for verse in chapter])
+            chapter_text = re.sub(r"[^\u05D0-\u05EA ]+", "", chapter_text)
+            chapter_text = re.sub(r" +", " ", chapter_text)
+            words = chapter_text.split()
+            max_length = min(max_phrase_length, len(words))
+            for length in range(1, max_length + 1):
+                for start in range(len(words) - length + 1):
+                    phrase_candidate = " ".join(words[start:start + length])
+                    gematria_sum = calculate_gematria(phrase_candidate.replace(" ", ""))
+                    insert_phrase_to_db(c, gematria_sum, phrase_candidate, book_id, title, chapter_id, start + 1)
+    conn.commit()
+    conn.close()
+
+def search_gematria_in_db(gematria_sum):
+    conn = sqlite3.connect('gematria.db')
+    c = conn.cursor()
+    c.execute('''
+        SELECT words, book, title, chapter, verse FROM results WHERE gematria_sum = ?
+    ''', (gematria_sum,))
+    results = c.fetchall()
+    conn.close()
+    return results
+
+def db(tanach_texts, max_phrase_length=1):
+    initialize_database()
+    populate_database(tanach_texts, max_phrase_length)
+    logging.info("Datenbank erfolgreich erstellt und gefüllt.")
+
+def gematria_search_interface(phrase):
+    debug_output = []
+
+    def debug_callback(message):
+        debug_output.append(message)
+        logging.info(message)
+
+    if not phrase.strip():
+        return "Bitte geben Sie eine Phrase ein.", "\n".join(debug_output)
+
+    phrase_gematria = calculate_gematria(phrase.replace(" ", ""))
+    debug_callback(f"Debug: Gematria der Suchphrase '{phrase}' ist {phrase_gematria}")
+
+    matching_phrases = search_gematria_in_db(phrase_gematria)
+
+    if not matching_phrases:
+        return "Keine passenden Phrasen gefunden.", "\n".join(debug_output)
+
+    result = "Passende Phrasen:\n"
+    for match in matching_phrases:
+        result += f"Buch: {match[1]} ({match[2]})\nKapitel: {match[3]}, Vers: {match[4]}\nPhrase: {match[0]}\n\n"
+    
+    return result, "\n".join(debug_output)
+
+def run_test():
+    debug_output = []
+    test_phrase = "אבגדהוזחטיכלמנסעפצקרשת"
+    expected_gematria = 1495
+
+    def debug_callback(message):
+        debug_output.append(message)
+        logging.info(message)
+
+    # Lade die Test-JSON-Inhalte für 00.json
+    test_texts_00 = process_json_files(0, 0)
+    db(test_texts_00, max_phrase_length=1)  # Populate the database with 1-word phrases
+    matching_phrases_00 = search_gematria_in_db(expected_gematria)
+    assert matching_phrases_00, "Keine passenden Phrasen gefunden in 00.json."
+    assert matching_phrases_00[0][0].replace(" ", "") == test_phrase, f"Gefundene Phrase stimmt nicht überein: {matching_phrases_00[0][0]}"
+    print("Test erfolgreich: Die Phrase wurde korrekt gefunden und die Gematria stimmt überein in 00.json.")
+
+    # Lade die Test-JSON-Inhalte für 01.json
+    test_texts_01 = process_json_files(1, 1)
+    db(test_texts_01, max_phrase_length=2)  # Populate the database with 1-word phrases
+    search_phrase_01 = "בארון במצרים"
+    expected_gematria_01 = calculate_gematria(search_phrase_01.replace(" ", ""))
+
+    matching_phrases_01 = search_gematria_in_db(expected_gematria_01)
+    assert matching_phrases_01, "Keine passenden Phrasen gefunden in 01.json."
+    assert matching_phrases_01[0][0].replace(" ", "") == search_phrase_01.replace(" ", ""), f"Gefundene Phrase stimmt nicht überein: {matching_phrases_01[0][0]}"
+    print("Test erfolgreich: Die Phrase wurde korrekt gefunden und die Gematria stimmt überein in 01.json.")
+    print("\n".join(debug_output))
+
+iface = gr.Interface(
+    fn=gematria_search_interface,
+    inputs="text",
+    outputs=["text", "text"],
+    title="Gematria-Suche im Tanach",
+    description="Suche nach Phrasen im Tanach, die denselben Gematria-Wert haben wie die eingegebene Phrase.",
+    live=True  # Aktiviert das Live-Update
+)
+
+if __name__ == "__main__":
+    run_test()  # Test ausführen
+    iface.launch()