vector-inspector 0.2.3__py3-none-any.whl → 0.2.4.3__py3-none-any.whl

vector_inspector/services/import_export_service.py

@@ -5,6 +5,7 @@ import csv
 from typing import Dict, Any, List, Optional
 from pathlib import Path
 import pandas as pd
+import numpy as np
 
 
 class ImportExportService:
@@ -36,9 +37,13 @@ class ImportExportService:
                     "document": documents[i] if i < len(documents) else None,
                     "metadata": metadatas[i] if i < len(metadatas) else {},
                 }
-                # Optionally include embeddings
-                if embeddings and i < len(embeddings):
-                    item["embedding"] = embeddings[i]
+                # Optionally include embeddings (convert numpy arrays to lists)
+                if len(embeddings) > 0 and i < len(embeddings):
+                    embedding = embeddings[i]
+                    # Convert numpy array to list if needed
+                    if isinstance(embedding, np.ndarray):
+                        embedding = embedding.tolist()
+                    item["embedding"] = embedding
                 export_data.append(item)
             
             # Write to file
@@ -82,9 +87,13 @@ class ImportExportService:
                     for key, value in metadatas[i].items():
                         row[f"metadata_{key}"] = value
                 
-                # Optionally add embeddings
-                if include_embeddings and embeddings and i < len(embeddings):
-                    row["embedding"] = json.dumps(embeddings[i])
+                # Optionally add embeddings (convert numpy arrays to lists)
+                if include_embeddings and len(embeddings) > 0 and i < len(embeddings):
+                    embedding = embeddings[i]
+                    # Convert numpy array to list if needed
+                    if isinstance(embedding, np.ndarray):
+                        embedding = embedding.tolist()
+                    row["embedding"] = json.dumps(embedding)
                     
                 rows.append(row)
             
@@ -118,17 +127,24 @@ class ImportExportService:
             # Prepare data for DataFrame
             df_data = {
                 "id": ids,
-                "document": documents if documents else [None] * len(ids),
+                "document": documents if len(documents) > 0 else [None] * len(ids),
             }
             
             # Add metadata fields as separate columns
-            if metadatas and metadatas[0]:
+            if len(metadatas) > 0 and metadatas[0]:
                 for key in metadatas[0].keys():
                     df_data[f"metadata_{key}"] = [m.get(key) if m else None for m in metadatas]
             
-            # Add embeddings as a column
-            if embeddings:
-                df_data["embedding"] = embeddings
+            # Add embeddings as a column (convert numpy arrays to lists for compatibility)
+            if len(embeddings) > 0:
+                # Convert numpy arrays to lists if needed
+                embedding_list = []
+                for emb in embeddings:
+                    if isinstance(emb, np.ndarray):
+                        embedding_list.append(emb.tolist())
+                    else:
+                        embedding_list.append(emb)
+                df_data["embedding"] = embedding_list
             
             # Create DataFrame and save
             df = pd.DataFrame(df_data)
@@ -173,7 +189,7 @@ class ImportExportService:
                 "metadatas": metadatas,
             }
             
-            if embeddings:
+            if len(embeddings) > 0:
                 result["embeddings"] = embeddings
                 
             return result
@@ -227,7 +243,7 @@ class ImportExportService:
                 "metadatas": metadatas,
             }
             
-            if embeddings:
+            if len(embeddings) > 0:
                 result["embeddings"] = embeddings
                 
             return result
@@ -277,7 +293,7 @@ class ImportExportService:
                 "metadatas": metadatas,
             }
             
-            if embeddings:
+            if len(embeddings) > 0:
                 result["embeddings"] = embeddings
                 
             return result

vector_inspector/ui/views/metadata_view.py

@@ -15,6 +15,7 @@ from vector_inspector.ui.components.loading_dialog import LoadingDialog
 from vector_inspector.ui.components.filter_builder import FilterBuilder
 from vector_inspector.services.import_export_service import ImportExportService
 from vector_inspector.services.filter_service import apply_client_side_filters
+from vector_inspector.services.settings_service import SettingsService
 from PySide6.QtWidgets import QApplication
 
 
@@ -29,6 +30,7 @@ class MetadataView(QWidget):
         self.page_size = 50
         self.current_page = 0
         self.loading_dialog = LoadingDialog("Loading data...", self)
+        self.settings_service = SettingsService()
         
         # Debounce timer for filter changes
         self.filter_reload_timer = QTimer()
@@ -426,33 +428,41 @@ class MetadataView(QWidget):
                 QMessageBox.warning(self, "Error", "Failed to update item.")
     
     def _export_data(self, format_type: str):
-        """Export collection data to file."""
+        """Export current table data to file (visible rows or selected rows)."""
         if not self.current_collection:
             QMessageBox.warning(self, "No Collection", "Please select a collection first.")
             return
-            
-        # Get all data (not just current page)
-        self.loading_dialog.show_loading("Exporting data...")
-        QApplication.processEvents()
         
-        try:
-            # Get filter if active
-            where_filter = None
-            if self.filter_group.isChecked() and self.filter_builder.has_filters():
-                where_filter = self.filter_builder.get_filter()
-                
-            # Fetch all data
-            all_data = self.connection.get_all_items(
-                self.current_collection,
-                where=where_filter
-            )
-        finally:
-            self.loading_dialog.hide_loading()
-            
-        if not all_data or not all_data.get("ids"):
+        if not self.current_data or not self.current_data.get("ids"):
             QMessageBox.warning(self, "No Data", "No data to export.")
             return
         
+        # Check if there are selected rows
+        selected_rows = self.table.selectionModel().selectedRows()
+        
+        if selected_rows:
+            # Export only selected rows
+            export_data = {
+                "ids": [],
+                "documents": [],
+                "metadatas": [],
+                "embeddings": []
+            }
+            
+            for index in selected_rows:
+                row = index.row()
+                if row < len(self.current_data["ids"]):
+                    export_data["ids"].append(self.current_data["ids"][row])
+                    if "documents" in self.current_data and row < len(self.current_data["documents"]):
+                        export_data["documents"].append(self.current_data["documents"][row])
+                    if "metadatas" in self.current_data and row < len(self.current_data["metadatas"]):
+                        export_data["metadatas"].append(self.current_data["metadatas"][row])
+                    if "embeddings" in self.current_data and row < len(self.current_data["embeddings"]):
+                        export_data["embeddings"].append(self.current_data["embeddings"][row])
+        else:
+            # Export all visible data from current table
+            export_data = self.current_data
+        
         # Select file path
         file_filters = {
             "json": "JSON Files (*.json)",
@@ -460,10 +470,14 @@ class MetadataView(QWidget):
             "parquet": "Parquet Files (*.parquet)"
         }
         
+        # Get last used directory from settings
+        last_dir = self.settings_service.get("last_import_export_dir", "")
+        default_path = f"{last_dir}/{self.current_collection}.{format_type}" if last_dir else f"{self.current_collection}.{format_type}"
+        
         file_path, _ = QFileDialog.getSaveFileName(
             self,
             f"Export to {format_type.upper()}",
-            f"{self.current_collection}.{format_type}",
+            default_path,
             file_filters[format_type]
         )
         
@@ -475,17 +489,21 @@ class MetadataView(QWidget):
         success = False
         
         if format_type == "json":
-            success = service.export_to_json(all_data, file_path)
+            success = service.export_to_json(export_data, file_path)
         elif format_type == "csv":
-            success = service.export_to_csv(all_data, file_path)
+            success = service.export_to_csv(export_data, file_path)
         elif format_type == "parquet":
-            success = service.export_to_parquet(all_data, file_path)
+            success = service.export_to_parquet(export_data, file_path)
             
         if success:
+            # Save the directory for next time
+            from pathlib import Path
+            self.settings_service.set("last_import_export_dir", str(Path(file_path).parent))
+            
             QMessageBox.information(
                 self,
                 "Export Successful",
-                f"Exported {len(all_data['ids'])} items to {file_path}"
+                f"Exported {len(export_data['ids'])} items to {file_path}"
             )
         else:
             QMessageBox.warning(self, "Export Failed", "Failed to export data.")
@@ -503,10 +521,13 @@ class MetadataView(QWidget):
             "parquet": "Parquet Files (*.parquet)"
         }
         
+        # Get last used directory from settings
+        last_dir = self.settings_service.get("last_import_export_dir", "")
+        
         file_path, _ = QFileDialog.getOpenFileName(
             self,
             f"Import from {format_type.upper()}",
-            "",
+            last_dir,
             file_filters[format_type]
         )
         
@@ -531,6 +552,54 @@ class MetadataView(QWidget):
             if not imported_data:
                 QMessageBox.warning(self, "Import Failed", "Failed to parse import file.")
                 return
+            
+            # Handle Qdrant-specific requirements (similar to backup/restore)
+            from vector_inspector.core.connections.qdrant_connection import QdrantConnection
+            if isinstance(self.connection, QdrantConnection):
+                # Check if embeddings are missing and need to be generated
+                if not imported_data.get("embeddings"):
+                    self.loading_dialog.setLabelText("Generating embeddings for Qdrant...")
+                    QApplication.processEvents()
+                    try:
+                        from sentence_transformers import SentenceTransformer
+                        model = SentenceTransformer("all-MiniLM-L6-v2")
+                        documents = imported_data.get("documents", [])
+                        imported_data["embeddings"] = model.encode(documents, show_progress_bar=False).tolist()
+                    except Exception as e:
+                        QMessageBox.warning(self, "Import Failed", 
+                                          f"Qdrant requires embeddings. Failed to generate: {e}")
+                        return
+                
+                # Convert IDs to Qdrant-compatible format (integers or UUIDs)
+                # Store original IDs in metadata
+                original_ids = imported_data.get("ids", [])
+                qdrant_ids = []
+                metadatas = imported_data.get("metadatas", [])
+                
+                for i, orig_id in enumerate(original_ids):
+                    # Try to convert to integer, otherwise use index
+                    try:
+                        # If it's like "doc_123", extract the number
+                        if isinstance(orig_id, str) and "_" in orig_id:
+                            qdrant_id = int(orig_id.split("_")[-1])
+                        else:
+                            qdrant_id = int(orig_id)
+                    except (ValueError, AttributeError):
+                        # Use index as ID if can't convert
+                        qdrant_id = i
+                    
+                    qdrant_ids.append(qdrant_id)
+                    
+                    # Store original ID in metadata
+                    if i < len(metadatas):
+                        if metadatas[i] is None:
+                            metadatas[i] = {}
+                        metadatas[i]["original_id"] = orig_id
+                    else:
+                        metadatas.append({"original_id": orig_id})
+                
+                imported_data["ids"] = qdrant_ids
+                imported_data["metadatas"] = metadatas
                 
             # Add items to collection
             success = self.connection.add_items(
@@ -544,6 +613,10 @@ class MetadataView(QWidget):
             self.loading_dialog.hide_loading()
             
         if success:
+            # Save the directory for next time
+            from pathlib import Path
+            self.settings_service.set("last_import_export_dir", str(Path(file_path).parent))
+            
             QMessageBox.information(
                 self,
                 "Import Successful",

{vector_inspector-0.2.3.dist-info → vector_inspector-0.2.4.3.dist-info}/METADATA

@@ -1,13 +1,13 @@
 Metadata-Version: 2.1
 Name: vector-inspector
-Version: 0.2.3
+Version: 0.2.4.3
 Summary: A comprehensive desktop application for visualizing, querying, and managing vector database data
 Author-Email: Anthony Dawson <anthonypdawson+github@gmail.com>
 License: MIT
 Project-URL: Homepage, https://vector-inspector.divinedevops.com
-Project-URL: Source, https://github.com/anthony-dawson/vector-inspector
-Project-URL: Issues, https://github.com/anthony-dawson/vector-inspector/issues
-Project-URL: Documentation, https://github.com/anthony-dawson/vector-inspector#readme
+Project-URL: Source, https://github.com/anthonypdawson/vector-inspector
+Project-URL: Issues, https://github.com/anthonypdawson/vector-inspector/issues
+Project-URL: Documentation, https://github.com/anthonypdawson/vector-inspector#readme
 Requires-Python: ==3.12.*
 Requires-Dist: chromadb>=0.4.22
 Requires-Dist: qdrant-client>=1.7.0

{vector_inspector-0.2.3.dist-info → vector_inspector-0.2.4.3.dist-info}/RECORD

@@ -1,6 +1,6 @@
-vector_inspector-0.2.3.dist-info/METADATA,sha256=l_D2Z-QyHHaxIhHUWO-iIocIlkLWbB8IiYytYU86l_k,9384
-vector_inspector-0.2.3.dist-info/WHEEL,sha256=tsUv_t7BDeJeRHaSrczbGeuK-TtDpGsWi_JfpzD255I,90
-vector_inspector-0.2.3.dist-info/entry_points.txt,sha256=u96envMI2NFImZUJDFutiiWl7ZoHrrev9joAgtyvTxo,80
+vector_inspector-0.2.4.3.dist-info/METADATA,sha256=dA5b-TGlPGSnIEqKCXkz2UbmzL1g9U0v-DUEBeycl9I,9386
+vector_inspector-0.2.4.3.dist-info/WHEEL,sha256=tsUv_t7BDeJeRHaSrczbGeuK-TtDpGsWi_JfpzD255I,90
+vector_inspector-0.2.4.3.dist-info/entry_points.txt,sha256=u96envMI2NFImZUJDFutiiWl7ZoHrrev9joAgtyvTxo,80
 vector_inspector/__init__.py,sha256=Q8XbXn98o0eliQWPePhy-aGUz2KNnVg7bQq-sBPl7zQ,119
 vector_inspector/__main__.py,sha256=Vdhw8YA1K3wPMlbJQYL5WqvRzAKVeZ16mZQFO9VRmCo,62
 vector_inspector/core/__init__.py,sha256=hjOqiJwF1P0rXjiOKhK4qDTvBY7G3m4kq8taH-gKrFM,57
@@ -13,7 +13,7 @@ vector_inspector/main.py,sha256=puu1Fur298j6H8fG3_wF85RMhi4tjLZ0Are16kloMqM,479
 vector_inspector/services/__init__.py,sha256=QLgH7oybjHuEYDFNiBgmJxvSpgAzHEuBEPXa3SKJb_I,67
 vector_inspector/services/backup_restore_service.py,sha256=2bbmLamGg7gaYUl3zy_-8wL8xbYvO00kZ2tbImZkzi0,11782
 vector_inspector/services/filter_service.py,sha256=r1y_lPi4Mo4ZqVq_hfbTzRxcCbdH7RlsXg7ytkBTKi8,2334
-vector_inspector/services/import_export_service.py,sha256=76n7Vm02shfa-Fn7AWHWsrmaJVICMMs0JnWI-wXeZ_o,9751
+vector_inspector/services/import_export_service.py,sha256=OPCrBXBewCznu5o8wFGvduU0jGZAcBvp_Fpv15kdoJ4,10712
 vector_inspector/services/settings_service.py,sha256=Y-2eJGyxUPPZid-3S0rCZ13xTRSLWYIOQ3ckSGX7gvE,2078
 vector_inspector/services/visualization_service.py,sha256=Bo9jSihT4axVKOAmefqXHemgm8uw59su2pz-UkXdONc,3901
 vector_inspector/ui/__init__.py,sha256=262ZiXO6Luk8vZnhCIoYxOtGiny0bXK-BTKjxUNBx-w,43
@@ -26,7 +26,7 @@ vector_inspector/ui/main_window.py,sha256=zHL7V_0pxyL_Bw2421GOJ-2X1jNphQdhwTX9s8
 vector_inspector/ui/views/__init__.py,sha256=FeMtVzSbVFBMjdwLQSQqD0FRW4ieJ4ZKXtTBci2e_bw,30
 vector_inspector/ui/views/collection_browser.py,sha256=oG9_YGPoVuMs-f_zSd4EcITmEU9caxvwuubsFUrNf-c,3991
 vector_inspector/ui/views/connection_view.py,sha256=5TL28JMfb0W42eUDDNbj6bIaj6m7WpalpdUEcd37qmM,16903
-vector_inspector/ui/views/metadata_view.py,sha256=1prg0ZFkPbA2PxUzr5Z-WlW8nTXa-_WSNlLIU-RTLME,21286
+vector_inspector/ui/views/metadata_view.py,sha256=aBMHJU61ljttRljOuAkemZkLEowjSaTnMl-lSCZRb88,25453
 vector_inspector/ui/views/search_view.py,sha256=p6vt2heSpEbiRge46VSedTZjg7i3-AWdS0f96JXzlEU,10527
 vector_inspector/ui/views/visualization_view.py,sha256=389lKOOqz4_7I-pUIIu86ZbA2ttq8Tuwflbb5W-W4g0,9038
-vector_inspector-0.2.3.dist-info/RECORD,,
+vector_inspector-0.2.4.3.dist-info/RECORD,,