intern/unsorted-folder-full-text-search
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Refactor logging and error handling; implement model loading in a separate thread for improved UI responsiveness

Browse Source
This commit is contained in:
Konstantin
2026-01-09 18:36:41 +01:00
parent de3eedc5c0
commit d24f6591bd
1 changed files with 161 additions and 119 deletions
Download Patch File Download Diff File Expand all files Collapse all files

280
uff_app.py
View File

@@ -5,16 +5,20 @@ import pdfplumber
import numpy as np import numpy as np
import zipfile import zipfile
import io import io
from sentence_transformers import SentenceTransformer, util import traceback
from sentence_transformers import SentenceTransformer, util
from rapidfuzz import process, fuzz from rapidfuzz import process, fuzz
# Wichtige Importe für UI und Signale
from PyQt6.QtCore import qInstallMessageHandler, QtMsgType, Qt, QThread, pyqtSignal, QUrl
from PyQt6.QtWidgets import (QApplication, QMainWindow, QWidget, QVBoxLayout, from PyQt6.QtWidgets import (QApplication, QMainWindow, QWidget, QVBoxLayout,
QHBoxLayout, QLineEdit, QPushButton, QLabel, QHBoxLayout, QLineEdit, QPushButton, QLabel,
QFileDialog, QTextBrowser, QProgressBar, QMessageBox, QFileDialog, QTextBrowser, QProgressBar, QMessageBox,
QListWidget, QListWidgetItem, QSplitter, QFrame) QListWidget, QListWidgetItem, QSplitter, QFrame, QSplashScreen)
from PyQt6.QtCore import Qt, QThread, pyqtSignal, QUrl from PyQt6.QtGui import QDesktopServices, QPixmap
from PyQt6.QtGui import QDesktopServices
# --- 0. LOGGING & SYSTEM-SETUP ---
if os.name == 'nt': if os.name == 'nt':
base_dir = os.getenv('LOCALAPPDATA') base_dir = os.getenv('LOCALAPPDATA')
@@ -27,14 +31,14 @@ if not os.path.exists(log_dir):
log_file_path = os.path.join(log_dir, "uff.log") log_file_path = os.path.join(log_dir, "uff.log")
# Logger-Klasse, die alles in die Datei schreibt # Logger-Klasse
class Logger(object): class Logger(object):
def __init__(self): def __init__(self):
self.log = open(log_file_path, "w", encoding="utf-8") # "w" überschreibt bei jedem Neustart self.log = open(log_file_path, "w", encoding="utf-8")
def write(self, message): def write(self, message):
self.log.write(message) self.log.write(message)
self.log.flush() # Sofort schreiben, damit nichts verloren geht self.log.flush()
def flush(self): def flush(self):
self.log.flush() self.log.flush()
@@ -46,11 +50,52 @@ sys.stderr = sys.stdout
print(f"--- START LOGGING ---") print(f"--- START LOGGING ---")
print(f"Logfile liegt hier: {log_file_path}") print(f"Logfile liegt hier: {log_file_path}")
# Font-Warnungen unterdrücken # --- QT MESSAGE HANDLER (Der Filter für C++ Errors) ---
os.environ["QT_LOGGING_RULES"] = "qt.qpa.fonts.warning=false;qt.text.fonts.db.warning=false" def qt_message_handler(mode, context, message):
"""
Fängt interne Qt-Nachrichten ab und filtert Font-Fehler heraus.
"""
msg_lower = message.lower()
# FILTER-LISTE: Erweitert basierend auf deinen Logs
ignore_keywords = [
"qt.text.font",
"qt.qpa.fonts",
"opentype support missing",
"directwrite",
"unable to create font",
"fontbbox",
"script 66",
"script 9",
"script 10",
"script 20",
"script 32"
]
# Wenn eines der Keywords vorkommt -> Nachricht ignorieren (return)
if any(keyword in msg_lower for keyword in ignore_keywords):
return
# Formatierung für das Logfile
mode_str = "INFO"
if mode == QtMsgType.QtWarningMsg: mode_str = "WARNING"
elif mode == QtMsgType.QtCriticalMsg: mode_str = "CRITICAL"
elif mode == QtMsgType.QtFatalMsg: mode_str = "FATAL"
# Nur relevante Nachrichten ins Log schreiben
try:
sys.stdout.write(f"[Qt {mode_str}] {message}\n")
except:
pass
# Handler installieren (Muss VOR der App-Erstellung passieren)
qInstallMessageHandler(qt_message_handler)
# Zusätzlich Environment Variable setzen
os.environ["QT_LOGGING_RULES"] = "qt.text.font.db=false;qt.qpa.fonts=false"
# --- 1. DATENBANK MANAGER (Mit Hybrid Search Scoring) --- # --- 1. DATENBANK MANAGER ---
class DatabaseHandler: class DatabaseHandler:
def __init__(self): def __init__(self):
@@ -65,31 +110,24 @@ class DatabaseHandler:
os.makedirs(self.app_data_dir) os.makedirs(self.app_data_dir)
self.db_name = os.path.join(self.app_data_dir, "uff_index.db") self.db_name = os.path.join(self.app_data_dir, "uff_index.db")
print(f"Datenbank Pfad: {self.db_name}") print(f"Datenbank Pfad: {self.db_name}")
self.model = None
print("Lade das semantische Modell (all-MiniLM-L6-v2)...")
self.model = SentenceTransformer('all-MiniLM-L6-v2')
print("Modell geladen.")
self.init_db() self.init_db()
def init_db(self): def init_db(self):
conn = sqlite3.connect(self.db_name) conn = sqlite3.connect(self.db_name)
cursor = conn.cursor() cursor = conn.cursor()
# FTS-Tabelle für die Stichwortsuche
cursor.execute(""" cursor.execute("""
CREATE VIRTUAL TABLE IF NOT EXISTS documents CREATE VIRTUAL TABLE IF NOT EXISTS documents
USING fts5(filename, path, content); USING fts5(filename, path, content);
""") """)
# Tabelle für die Ordner
cursor.execute(""" cursor.execute("""
CREATE TABLE IF NOT EXISTS folders ( CREATE TABLE IF NOT EXISTS folders (
path TEXT PRIMARY KEY, path TEXT PRIMARY KEY,
alias TEXT alias TEXT
); );
""") """)
# Tabelle für die Vektor-Embeddings
cursor.execute(""" cursor.execute("""
CREATE TABLE IF NOT EXISTS embeddings ( CREATE TABLE IF NOT EXISTS embeddings (
doc_id INTEGER PRIMARY KEY, doc_id INTEGER PRIMARY KEY,
@@ -115,11 +153,9 @@ class DatabaseHandler:
cursor = conn.cursor() cursor = conn.cursor()
cursor.execute("SELECT rowid FROM documents WHERE path LIKE ?", (f"{path}%",)) cursor.execute("SELECT rowid FROM documents WHERE path LIKE ?", (f"{path}%",))
ids_to_delete = [row[0] for row in cursor.fetchall()] ids_to_delete = [row[0] for row in cursor.fetchall()]
if ids_to_delete: if ids_to_delete:
cursor.execute("DELETE FROM documents WHERE path LIKE ?", (f"{path}%",)) cursor.execute("DELETE FROM documents WHERE path LIKE ?", (f"{path}%",))
cursor.execute(f"DELETE FROM embeddings WHERE doc_id IN ({','.join('?'*len(ids_to_delete))})", ids_to_delete) cursor.execute(f"DELETE FROM embeddings WHERE doc_id IN ({','.join('?'*len(ids_to_delete))})", ids_to_delete)
cursor.execute("DELETE FROM folders WHERE path = ?", (path,)) cursor.execute("DELETE FROM folders WHERE path = ?", (path,))
conn.commit() conn.commit()
conn.close() conn.close()
@@ -131,43 +167,32 @@ class DatabaseHandler:
return [r[0] for r in rows] return [r[0] for r in rows]
def search(self, query): def search(self, query):
if not query.strip(): return [] if not query.strip() or not self.model: return []
# --- PHASE 1: SEMANTISCHE SUCHE (Vektor) --- # PHASE 1: SEMANTIK
query_embedding = self.model.encode(query, convert_to_tensor=False) query_embedding = self.model.encode(query, convert_to_tensor=False)
conn = sqlite3.connect(self.db_name) conn = sqlite3.connect(self.db_name)
cursor = conn.cursor() cursor = conn.cursor()
cursor.execute("SELECT doc_id, vec FROM embeddings") cursor.execute("SELECT doc_id, vec FROM embeddings")
all_embeddings_data = cursor.fetchall() all_embeddings_data = cursor.fetchall()
doc_ids = [item[0] for item in all_embeddings_data] doc_ids = [item[0] for item in all_embeddings_data]
if not doc_ids: if not doc_ids:
conn.close() conn.close()
return [] return []
# BLOBs zurück zu Vektoren
all_embeddings = np.array([np.frombuffer(item[1], dtype=np.float32) for item in all_embeddings_data]) all_embeddings = np.array([np.frombuffer(item[1], dtype=np.float32) for item in all_embeddings_data])
# Cosine Similarity (Werte zwischen -1 und 1)
# clip auf 0, da negative Werte hier irrelevant sind
cos_scores = util.cos_sim(query_embedding, all_embeddings)[0].numpy() cos_scores = util.cos_sim(query_embedding, all_embeddings)[0].numpy()
cos_scores = np.clip(cos_scores, 0, 1) cos_scores = np.clip(cos_scores, 0, 1)
# Map: doc_id -> Semantic Score (0.0 - 1.0)
semantic_map = {doc_id: float(score) for doc_id, score in zip(doc_ids, cos_scores)} semantic_map = {doc_id: float(score) for doc_id, score in zip(doc_ids, cos_scores)}
# --- PHASE 2: STICHWORTSUCHE (FTS & Fuzzy) --- # PHASE 2: LEXIKALISCH
words = query.replace('"', '').split() words = query.replace('"', '').split()
if not words: words = [query] if not words: words = [query]
sql_query_parts = [f'"{w}"*' for w in words] sql_query_parts = [f'"{w}"*' for w in words]
sql_query_string = " OR ".join(sql_query_parts) sql_query_string = " OR ".join(sql_query_parts)
try: try:
# Wir holen Kandidaten, die die Wörter enthalten
fts_rows = cursor.execute(""" fts_rows = cursor.execute("""
SELECT rowid, filename, content SELECT rowid, filename, content
FROM documents FROM documents
@@ -178,55 +203,53 @@ class DatabaseHandler:
fts_rows = [] fts_rows = []
lexical_map = {} lexical_map = {}
for doc_id, filename, content in fts_rows: for doc_id, filename, content in fts_rows:
# Fuzzy-Score berechnen (0 bis 100) -> normalisieren auf 0.0 - 1.0
ratio_name = fuzz.partial_ratio(query.lower(), filename.lower()) ratio_name = fuzz.partial_ratio(query.lower(), filename.lower())
ratio_content = fuzz.partial_token_set_ratio(query.lower(), content[:5000].lower()) ratio_content = fuzz.partial_token_set_ratio(query.lower(), content[:5000].lower())
best_ratio = max(ratio_name, ratio_content) best_ratio = max(ratio_name, ratio_content)
lexical_map[doc_id] = best_ratio / 100.0 lexical_map[doc_id] = best_ratio / 100.0
# --- PHASE 3: HYBRID FUSION (Kombination) --- # PHASE 3: HYBRID
final_scores = {} final_scores = {}
ALPHA = 0.65
# Gewichtung anpassen BETA = 0.35
ALPHA = 0.65 # 65% Semantik
BETA = 0.35 # 35% Stichwort
for doc_id, sem_score in semantic_map.items(): for doc_id, sem_score in semantic_map.items():
# Filter: Nur Ergebnisse mit minimaler Relevanz betrachten
if sem_score < 0.15 and doc_id not in lexical_map: if sem_score < 0.15 and doc_id not in lexical_map:
continue continue
lex_score = lexical_map.get(doc_id, 0.0) lex_score = lexical_map.get(doc_id, 0.0)
# Hybrid Score
hybrid_score = (sem_score * ALPHA) + (lex_score * BETA) hybrid_score = (sem_score * ALPHA) + (lex_score * BETA)
# Bonus: Wenn beides hoch ist (Semantik UND Keyword)
if sem_score > 0.4 and lex_score > 0.6: if sem_score > 0.4 and lex_score > 0.6:
hybrid_score += 0.1 hybrid_score += 0.1
final_scores[doc_id] = hybrid_score final_scores[doc_id] = hybrid_score
# --- PHASE 4: SORTIEREN & AUSGEBEN --- # PHASE 4: SORT
sorted_ids = sorted(final_scores.keys(), key=lambda x: final_scores[x], reverse=True) sorted_ids = sorted(final_scores.keys(), key=lambda x: final_scores[x], reverse=True)
results = [] results = []
for doc_id in sorted_ids[:50]: # Top 50 Ergebnisse for doc_id in sorted_ids[:50]:
row = cursor.execute( row = cursor.execute(
"SELECT filename, path, snippet(documents, 2, '<b>', '</b>', '...', 15) FROM documents WHERE rowid = ?", "SELECT filename, path, snippet(documents, 2, '<b>', '</b>', '...', 15) FROM documents WHERE rowid = ?",
(doc_id,) (doc_id,)
).fetchone() ).fetchone()
if row: if row:
results.append(row) results.append(row)
conn.close() conn.close()
return results return results
# --- 2. INDEXER (Mit ZIP Support & Recursion) --- # --- 2. MODEL LOADER ---
class ModelLoaderThread(QThread):
model_loaded = pyqtSignal(object)
def run(self):
print("Lade das semantische Modell (all-MiniLM-L6-v2)...")
try:
model = SentenceTransformer('all-MiniLM-L6-v2')
print("Modell geladen.")
self.model_loaded.emit(model)
except Exception as e:
print(f"Fehler beim Laden des Modells: {e}")
self.model_loaded.emit(None)
# --- 3. INDEXER ---
class IndexerThread(QThread): class IndexerThread(QThread):
progress_signal = pyqtSignal(str) progress_signal = pyqtSignal(str)
finished_signal = pyqtSignal(int, int, bool) finished_signal = pyqtSignal(int, int, bool)
@@ -242,58 +265,41 @@ class IndexerThread(QThread):
self.is_running = False self.is_running = False
def _extract_text_from_stream(self, file_stream, filename): def _extract_text_from_stream(self, file_stream, filename):
"""
Liest Text aus einem Dateiobjekt (Stream) oder Pfad, basierend auf der Endung.
Robuster gegen defekte PDF-Seiten.
"""
ext = os.path.splitext(filename)[1].lower() ext = os.path.splitext(filename)[1].lower()
text = "" text = ""
try: try:
if ext == ".pdf": if ext == ".pdf":
# pdfplumber kann direkt Dateiobjekte (BytesIO) lesen
try: try:
with pdfplumber.open(file_stream) as pdf: with pdfplumber.open(file_stream) as pdf:
for page in pdf.pages: for page in pdf.pages:
try: try:
# Versuch, Text von der einzelnen Seite zu holen
if page_text := page.extract_text(): if page_text := page.extract_text():
text += page_text + "\n" text += page_text + "\n"
except Exception as e: except Exception as e:
# Wenn eine Seite defekt ist (z.B. FontBBox Fehler), überspringen wir nur diese Seite print(f"Warnung: Konnte eine Seite in '{filename}' nicht lesen. Fehler: {e}")
print(f"Warnung: Konnte eine Seite in '{filename}' nicht lesen (übersprungen). Fehler: {e}")
continue continue
except Exception as e: except Exception as e:
# Wenn die ganze PDF nicht geöffnet werden kann print(f"Warnung: PDF '{filename}' defekt. Fehler: {e}")
print(f"Warnung: PDF '{filename}' konnte nicht geöffnet werden. Fehler: {e}")
return None return None
elif ext in [".txt", ".md", ".py", ".json", ".csv", ".html", ".log", ".ini", ".xml"]: elif ext in [".txt", ".md", ".py", ".json", ".csv", ".html", ".log", ".ini", ".xml"]:
# Wir lesen die Bytes und decodieren sie
if hasattr(file_stream, 'read'): if hasattr(file_stream, 'read'):
content_bytes = file_stream.read() content_bytes = file_stream.read()
if isinstance(content_bytes, str): if isinstance(content_bytes, str):
# Fallback
with open(file_stream, 'r', encoding='utf-8', errors='ignore') as f: with open(file_stream, 'r', encoding='utf-8', errors='ignore') as f:
text = f.read() text = f.read()
else: else:
text = content_bytes.decode('utf-8', errors='ignore') text = content_bytes.decode('utf-8', errors='ignore')
else: else:
# Echter Dateipfad
with open(file_stream, "r", encoding="utf-8", errors="ignore") as f: with open(file_stream, "r", encoding="utf-8", errors="ignore") as f:
text = f.read() text = f.read()
except Exception as e: except Exception as e:
# Allgemeiner Fehler beim Lesen
# print(f"Lese-Fehler bei {filename}: {e}")
return None return None
return text return text
def run(self): def run(self):
conn = sqlite3.connect(self.db_name) conn = sqlite3.connect(self.db_name)
cursor = conn.cursor() cursor = conn.cursor()
# Bereinigen alter Einträge
cursor.execute("SELECT rowid FROM documents WHERE path LIKE ?", (f"{self.folder_path}%",)) cursor.execute("SELECT rowid FROM documents WHERE path LIKE ?", (f"{self.folder_path}%",))
ids_to_delete = [row[0] for row in cursor.fetchall()] ids_to_delete = [row[0] for row in cursor.fetchall()]
if ids_to_delete: if ids_to_delete:
@@ -305,7 +311,6 @@ class IndexerThread(QThread):
skipped = 0 skipped = 0
was_cancelled = False was_cancelled = False
# --- REKURSIVES DURCHSUCHEN ---
for root, dirs, files in os.walk(self.folder_path): for root, dirs, files in os.walk(self.folder_path):
if not self.is_running: if not self.is_running:
was_cancelled = True was_cancelled = True
@@ -319,30 +324,21 @@ class IndexerThread(QThread):
file_path = os.path.join(root, file) file_path = os.path.join(root, file)
self.progress_signal.emit(f"Prüfe: {file}...") self.progress_signal.emit(f"Prüfe: {file}...")
# A. ZIP-DATEIEN BEHANDELN
if file.lower().endswith('.zip'): if file.lower().endswith('.zip'):
try: try:
with zipfile.ZipFile(file_path, 'r') as z: with zipfile.ZipFile(file_path, 'r') as z:
for z_info in z.infolist(): for z_info in z.infolist():
if z_info.is_dir(): continue if z_info.is_dir(): continue
# Virtueller Pfad: C:\Ordner\Archiv.zip :: innen/datei.txt
virtual_path = f"{file_path} :: {z_info.filename}" virtual_path = f"{file_path} :: {z_info.filename}"
with z.open(z_info) as z_file: with z.open(z_info) as z_file:
# Inhalt in RAM laden (BytesIO)
file_in_memory = io.BytesIO(z_file.read()) file_in_memory = io.BytesIO(z_file.read())
content = self._extract_text_from_stream(file_in_memory, z_info.filename) content = self._extract_text_from_stream(file_in_memory, z_info.filename)
if content and len(content.strip()) > 20: if content and len(content.strip()) > 20:
self._save_to_db(cursor, z_info.filename, virtual_path, content) self._save_to_db(cursor, z_info.filename, virtual_path, content)
indexed += 1 indexed += 1
except Exception as e: except Exception as e:
print(f"Zip Error {file}: {e}") print(f"Zip Error {file}: {e}")
skipped += 1 skipped += 1
# B. NORMALE DATEIEN
else: else:
content = self._extract_text_from_stream(file_path, file) content = self._extract_text_from_stream(file_path, file)
if content and len(content.strip()) > 20: if content and len(content.strip()) > 20:
@@ -358,32 +354,26 @@ class IndexerThread(QThread):
self.finished_signal.emit(indexed, skipped, was_cancelled) self.finished_signal.emit(indexed, skipped, was_cancelled)
def _save_to_db(self, cursor, filename, path, content): def _save_to_db(self, cursor, filename, path, content):
# 1. Text speichern cursor.execute("INSERT INTO documents (filename, path, content) VALUES (?, ?, ?)", (filename, path, content))
cursor.execute(
"INSERT INTO documents (filename, path, content) VALUES (?, ?, ?)",
(filename, path, content)
)
doc_id = cursor.lastrowid doc_id = cursor.lastrowid
# 2. Embedding erstellen (Max 8000 chars)
embedding = self.model.encode(content[:8000], convert_to_tensor=False) embedding = self.model.encode(content[:8000], convert_to_tensor=False)
embedding_blob = embedding.tobytes() embedding_blob = embedding.tobytes()
# 3. Vektor speichern
cursor.execute("INSERT INTO embeddings (doc_id, vec) VALUES (?, ?)", (doc_id, embedding_blob)) cursor.execute("INSERT INTO embeddings (doc_id, vec) VALUES (?, ?)", (doc_id, embedding_blob))
# --- 3. UI (Unverändert) ---
# --- 4. UI ---
class UffWindow(QMainWindow): class UffWindow(QMainWindow):
def __init__(self): def __init__(self, splash=None):
super().__init__() super().__init__()
self.splash = splash
self.db = DatabaseHandler() self.db = DatabaseHandler()
self.indexer_thread = None self.indexer_thread = None
self.initUI() self.initUI()
self.load_saved_folders() self.load_saved_folders()
def initUI(self): def initUI(self):
self.setWindowTitle("UFF Text Search v5.0 (Hybrid Zip)") self.setWindowTitle("UFF Text Search")
self.resize(1000, 700) self.resize(1000, 700)
central = QWidget() central = QWidget()
@@ -395,22 +385,18 @@ class UffWindow(QMainWindow):
left_panel.setFixedWidth(250) left_panel.setFixedWidth(250)
left_layout = QVBoxLayout(left_panel) left_layout = QVBoxLayout(left_panel)
left_layout.setContentsMargins(0, 0, 0, 0) left_layout.setContentsMargins(0, 0, 0, 0)
lbl_folders = QLabel("📂 Meine Ordner") lbl_folders = QLabel("📂 Meine Ordner")
lbl_folders.setStyleSheet("font-weight: bold; font-size: 14px;") lbl_folders.setStyleSheet("font-weight: bold; font-size: 14px;")
self.folder_list = QListWidget() self.folder_list = QListWidget()
self.folder_list.setSelectionMode(QListWidget.SelectionMode.SingleSelection) self.folder_list.setSelectionMode(QListWidget.SelectionMode.SingleSelection)
btn_add = QPushButton(" + Hinzufügen") self.btn_add = QPushButton(" + Hinzufügen")
btn_add.clicked.connect(self.add_new_folder) self.btn_add.clicked.connect(self.add_new_folder)
self.btn_remove = QPushButton(" - Entfernen")
btn_remove = QPushButton(" - Entfernen") self.btn_remove.clicked.connect(self.delete_selected_folder)
btn_remove.clicked.connect(self.delete_selected_folder)
self.btn_rescan = QPushButton(" ↻ Neu scannen") self.btn_rescan = QPushButton(" ↻ Neu scannen")
self.btn_rescan.clicked.connect(self.rescan_selected_folder) self.btn_rescan.clicked.connect(self.rescan_selected_folder)
self.btn_cancel = QPushButton("🛑 Abbrechen") self.btn_cancel = QPushButton("🛑 Abbrechen")
self.btn_cancel.setStyleSheet("background-color: #ffcccc; color: #cc0000; font-weight: bold;") self.btn_cancel.setStyleSheet("background-color: #ffcccc; color: #cc0000; font-weight: bold;")
self.btn_cancel.clicked.connect(self.cancel_indexing) self.btn_cancel.clicked.connect(self.cancel_indexing)
@@ -418,8 +404,8 @@ class UffWindow(QMainWindow):
left_layout.addWidget(lbl_folders) left_layout.addWidget(lbl_folders)
left_layout.addWidget(self.folder_list) left_layout.addWidget(self.folder_list)
left_layout.addWidget(btn_add) left_layout.addWidget(self.btn_add)
left_layout.addWidget(btn_remove) left_layout.addWidget(self.btn_remove)
left_layout.addStretch() left_layout.addStretch()
left_layout.addWidget(self.btn_rescan) left_layout.addWidget(self.btn_rescan)
left_layout.addWidget(self.btn_cancel) left_layout.addWidget(self.btn_cancel)
@@ -434,21 +420,25 @@ class UffWindow(QMainWindow):
self.input_search.returnPressed.connect(self.perform_search) self.input_search.returnPressed.connect(self.perform_search)
self.input_search.setStyleSheet("padding: 8px; font-size: 14px;") self.input_search.setStyleSheet("padding: 8px; font-size: 14px;")
btn_go = QPushButton("Suchen") self.btn_go = QPushButton("Suchen")
btn_go.setFixedWidth(100) self.btn_go.setFixedWidth(100)
btn_go.clicked.connect(self.perform_search) self.btn_go.clicked.connect(self.perform_search)
search_container.addWidget(self.input_search) search_container.addWidget(self.input_search)
search_container.addWidget(btn_go) search_container.addWidget(self.btn_go)
self.lbl_status = QLabel("Bereit. Hybrid-Modell geladen.") self.lbl_status = QLabel("Initialisiere...")
self.lbl_status.setStyleSheet("color: #666;") self.lbl_status.setStyleSheet("color: #666;")
self.progress_bar = QProgressBar() self.progress_bar = QProgressBar()
self.progress_bar.hide() self.progress_bar.hide()
# STANDARD BROWSER MIT RICHTIGEN EINSTELLUNGEN
self.result_browser = QTextBrowser() self.result_browser = QTextBrowser()
self.result_browser.setOpenExternalLinks(False) # WICHTIG: Interne Links deaktivieren, damit wir sie abfangen können
self.result_browser.anchorClicked.connect(self.link_clicked) self.result_browser.setOpenExternalLinks(False)
# Wenn wir darauf klicken, wird unser Slot aufgerufen
self.result_browser.anchorClicked.connect(self.link_clicked)
self.result_browser.setStyleSheet("background-color: white; border: 1px solid #ccc;") self.result_browser.setStyleSheet("background-color: white; border: 1px solid #ccc;")
right_layout.addLayout(search_container) right_layout.addLayout(search_container)
@@ -462,8 +452,39 @@ class UffWindow(QMainWindow):
splitter.setSizes([250, 750]) splitter.setSizes([250, 750])
main_layout.addWidget(splitter) main_layout.addWidget(splitter)
self.set_main_ui_enabled(False)
def set_main_ui_enabled(self, enabled):
self.input_search.setEnabled(enabled)
self.btn_go.setEnabled(enabled)
self.folder_list.setEnabled(enabled)
self.btn_add.setEnabled(enabled)
self.btn_remove.setEnabled(enabled)
self.btn_rescan.setEnabled(enabled)
def start_model_loading(self):
if self.splash:
self.splash.showMessage("Lade semantisches Modell...", Qt.AlignmentFlag.AlignBottom | Qt.AlignmentFlag.AlignHCenter, Qt.GlobalColor.white)
self.model_loader = ModelLoaderThread()
self.model_loader.model_loaded.connect(self.on_model_loaded)
self.model_loader.start()
def on_model_loaded(self, model):
if self.splash:
self.splash.showMessage("Modell geladen. Starte Benutzeroberfläche...", Qt.AlignmentFlag.AlignBottom | Qt.AlignmentFlag.AlignHCenter, Qt.GlobalColor.white)
if model is None:
self.lbl_status.setText("Fehler: Modell konnte nicht geladen werden.")
QMessageBox.critical(self, "Kritischer Fehler", "Das semantische Modell konnte nicht geladen werden.")
self.close()
else:
self.db.model = model
self.lbl_status.setText("Bereit. Hybrid-Modell geladen.")
self.set_main_ui_enabled(True)
if self.splash:
self.splash.finish(self)
# LOGIK
def load_saved_folders(self): def load_saved_folders(self):
self.folder_list.clear() self.folder_list.clear()
folders = self.db.get_folders() folders = self.db.get_folders()
@@ -500,11 +521,14 @@ class UffWindow(QMainWindow):
self.start_indexing(item.text()) self.start_indexing(item.text())
def start_indexing(self, folder): def start_indexing(self, folder):
if not self.db.model:
QMessageBox.warning(self, "Bitte warten", "Das Suchmodell wird noch geladen.")
return
self.set_ui_busy(True) self.set_ui_busy(True)
self.lbl_status.setText(f"Starte... {os.path.basename(folder)}") self.lbl_status.setText(f"Starte... {os.path.basename(folder)}")
self.indexer_thread = IndexerThread(folder, db_name=self.db.db_name, model=self.db.model) self.indexer_thread = IndexerThread(folder, db_name=self.db.db_name, model=self.db.model)
self.indexer_thread.progress_signal.connect(lambda msg: self.lbl_status.setText(msg)) self.indexer_thread.progress_signal.connect(lambda msg: self.lbl_status.setText(msg))
self.indexer_thread.finished_signal.connect(self.indexing_finished) self.indexer_thread.finished_signal.connect(self.indexing_finished)
self.indexer_thread.start() self.indexer_thread.start()
@@ -526,6 +550,9 @@ class UffWindow(QMainWindow):
def set_ui_busy(self, busy): def set_ui_busy(self, busy):
self.input_search.setEnabled(not busy) self.input_search.setEnabled(not busy)
self.folder_list.setEnabled(not busy) self.folder_list.setEnabled(not busy)
self.btn_add.setEnabled(not busy)
self.btn_remove.setEnabled(not busy)
self.btn_go.setEnabled(not busy)
self.btn_rescan.setVisible(not busy) self.btn_rescan.setVisible(not busy)
self.btn_cancel.setVisible(busy) self.btn_cancel.setVisible(busy)
if busy: if busy:
@@ -549,15 +576,14 @@ class UffWindow(QMainWindow):
html = "<h3 style='color: gray; text-align: center; margin-top: 20px;'>Nichts gefunden.</h3>" html = "<h3 style='color: gray; text-align: center; margin-top: 20px;'>Nichts gefunden.</h3>"
for filename, filepath, snippet in results: for filename, filepath, snippet in results:
# Falls es eine Datei im Zip ist, müssen wir den Link anpassen,
# damit er zumindest das Zip öffnet.
if " :: " in filepath: if " :: " in filepath:
real_path = filepath.split(" :: ")[0] real_path = filepath.split(" :: ")[0]
display_path = filepath # Zeige den virtuellen Pfad display_path = filepath
else: else:
real_path = filepath real_path = filepath
display_path = filepath display_path = filepath
# Link für QTextBrowser
file_url = QUrl.fromLocalFile(real_path).toString() file_url = QUrl.fromLocalFile(real_path).toString()
html += f""" html += f"""
@@ -571,11 +597,27 @@ class UffWindow(QMainWindow):
""" """
self.result_browser.setHtml(html) self.result_browser.setHtml(html)
# --- DIE FUNKTION ZUM ÖFFNEN DER LINKS ---
def link_clicked(self, url): def link_clicked(self, url):
print(f"Versuche zu öffnen: {url.toString()}")
QDesktopServices.openUrl(url) QDesktopServices.openUrl(url)
if __name__ == "__main__": if __name__ == "__main__":
app = QApplication(sys.argv) app = QApplication(sys.argv)
window = UffWindow()
splash = None
try:
pixmap = QPixmap("assets/uff_banner.jpeg")
splash = QSplashScreen(pixmap)
splash.show()
splash.showMessage("Initialisiere Anwendung...", Qt.AlignmentFlag.AlignBottom | Qt.AlignmentFlag.AlignHCenter, Qt.GlobalColor.white)
except:
pass
app.processEvents()
window = UffWindow(splash)
window.show() window.show()
window.start_model_loading()
sys.exit(app.exec()) sys.exit(app.exec())