this will load data from Opensearch.

it doesn't have prompts as well
This commit is contained in:
2025-08-14 13:49:46 -07:00
parent a2adc8b958
commit 9cf2f0e6fa
16 changed files with 1694 additions and 7 deletions

View File

@@ -1,10 +1,14 @@
from dash import callback, Input, Output, State
from dash import callback, Input, Output, State, no_update
from ...data.processor import DataProcessor
from ...data.sources.opensearch import OpenSearchClient
from ...models.field_mapper import FieldMapper
from ...config.settings import AppSettings
class DataProcessingCallbacks:
def __init__(self):
self.processor = DataProcessor()
self.opensearch_client = OpenSearchClient()
self._register_callbacks()
def _register_callbacks(self):
@@ -67,6 +71,283 @@ class DataProcessingCallbacks:
"embeddings": processed_data.embeddings.tolist(),
}
# OpenSearch callbacks
@callback(
[
Output("tab-content", "children"),
],
[Input("data-source-tabs", "active_tab")],
prevent_initial_call=False,
)
def render_tab_content(active_tab):
from ...ui.components.datasource import DataSourceComponent
datasource = DataSourceComponent()
if active_tab == "opensearch-tab":
return [datasource.create_opensearch_tab()]
else:
return [datasource.create_file_upload_tab()]
@callback(
Output("auth-collapse", "is_open"),
[Input("auth-toggle", "n_clicks")],
[State("auth-collapse", "is_open")],
prevent_initial_call=True,
)
def toggle_auth(n_clicks, is_open):
if n_clicks:
return not is_open
return is_open
@callback(
Output("auth-toggle", "children"),
[Input("auth-collapse", "is_open")],
prevent_initial_call=False,
)
def update_auth_button_text(is_open):
return "Hide Authentication" if is_open else "Show Authentication"
@callback(
[
Output("connection-status", "children"),
Output("field-mapping-section", "children"),
Output("field-mapping-section", "style"),
Output("load-data-section", "style"),
Output("load-opensearch-data-btn", "disabled"),
Output("embedding-field-dropdown", "options"),
Output("text-field-dropdown", "options"),
Output("id-field-dropdown", "options"),
Output("category-field-dropdown", "options"),
Output("subcategory-field-dropdown", "options"),
Output("tags-field-dropdown", "options"),
],
[Input("test-connection-btn", "n_clicks")],
[
State("opensearch-url", "value"),
State("opensearch-index", "value"),
State("opensearch-username", "value"),
State("opensearch-password", "value"),
State("opensearch-api-key", "value"),
],
prevent_initial_call=True,
)
def test_opensearch_connection(
n_clicks, url, index_name, username, password, api_key
):
if not n_clicks or not url or not index_name:
return no_update, no_update, no_update, no_update, no_update, no_update, no_update, no_update, no_update, no_update, no_update
# Test connection
success, message = self.opensearch_client.connect(
url=url,
username=username,
password=password,
api_key=api_key,
verify_certs=AppSettings.OPENSEARCH_VERIFY_CERTS,
)
if not success:
return (
self._create_status_alert(f"{message}", "danger"),
[],
{"display": "none"},
{"display": "none"},
True,
[], # empty options for hidden dropdowns
[],
[],
[],
[],
[],
)
# Analyze fields
success, field_analysis, analysis_message = (
self.opensearch_client.analyze_fields(index_name)
)
if not success:
return (
self._create_status_alert(f"{analysis_message}", "danger"),
[],
{"display": "none"},
{"display": "none"},
True,
[], # empty options for hidden dropdowns
[],
[],
[],
[],
[],
)
# Generate field suggestions
field_suggestions = FieldMapper.suggest_mappings(field_analysis)
from ...ui.components.datasource import DataSourceComponent
datasource = DataSourceComponent()
field_mapping_ui = datasource.create_field_mapping_interface(
field_suggestions
)
return (
self._create_status_alert(f"{message}", "success"),
field_mapping_ui,
{"display": "block"},
{"display": "block"},
False,
[{"label": field, "value": field} for field in field_suggestions.get("embedding", [])],
[{"label": field, "value": field} for field in field_suggestions.get("text", [])],
[{"label": field, "value": field} for field in field_suggestions.get("id", [])],
[{"label": field, "value": field} for field in field_suggestions.get("category", [])],
[{"label": field, "value": field} for field in field_suggestions.get("subcategory", [])],
[{"label": field, "value": field} for field in field_suggestions.get("tags", [])],
)
@callback(
[
Output("processed-data", "data", allow_duplicate=True),
Output("opensearch-success-alert", "children", allow_duplicate=True),
Output("opensearch-success-alert", "is_open", allow_duplicate=True),
Output("opensearch-error-alert", "children", allow_duplicate=True),
Output("opensearch-error-alert", "is_open", allow_duplicate=True),
],
[Input("load-opensearch-data-btn", "n_clicks")],
[
State("opensearch-index", "value"),
State("embedding-field-dropdown", "value"),
State("text-field-dropdown", "value"),
State("id-field-dropdown", "value"),
State("category-field-dropdown", "value"),
State("subcategory-field-dropdown", "value"),
State("tags-field-dropdown", "value"),
],
prevent_initial_call=True,
)
def load_opensearch_data(
n_clicks,
index_name,
embedding_field,
text_field,
id_field,
category_field,
subcategory_field,
tags_field,
):
if not n_clicks or not index_name or not embedding_field or not text_field:
return no_update, no_update, no_update, no_update, no_update
try:
# Create field mapping
field_mapping = FieldMapper.create_mapping_from_dict(
{
"embedding": embedding_field,
"text": text_field,
"id": id_field,
"category": category_field,
"subcategory": subcategory_field,
"tags": tags_field,
}
)
# Fetch data from OpenSearch
success, raw_documents, message = self.opensearch_client.fetch_data(
index_name, size=AppSettings.OPENSEARCH_DEFAULT_SIZE
)
if not success:
return (
no_update,
"",
False,
f"❌ Failed to fetch data: {message}",
True,
)
# Process the data
processed_data = self.processor.process_opensearch_data(
raw_documents, field_mapping
)
if processed_data.error:
return (
{"error": processed_data.error},
"",
False,
f"❌ Data processing error: {processed_data.error}",
True,
)
success_message = f"✅ Successfully loaded {len(processed_data.documents)} documents from OpenSearch"
return (
{
"documents": [
self._document_to_dict(doc)
for doc in processed_data.documents
],
"embeddings": processed_data.embeddings.tolist(),
},
success_message,
True,
"",
False,
)
except Exception as e:
return (no_update, "", False, f"❌ Unexpected error: {str(e)}", True)
# Sync callbacks to update hidden dropdowns from UI dropdowns
@callback(
Output("embedding-field-dropdown", "value"),
Input("embedding-field-dropdown-ui", "value"),
prevent_initial_call=True,
)
def sync_embedding_dropdown(value):
return value
@callback(
Output("text-field-dropdown", "value"),
Input("text-field-dropdown-ui", "value"),
prevent_initial_call=True,
)
def sync_text_dropdown(value):
return value
@callback(
Output("id-field-dropdown", "value"),
Input("id-field-dropdown-ui", "value"),
prevent_initial_call=True,
)
def sync_id_dropdown(value):
return value
@callback(
Output("category-field-dropdown", "value"),
Input("category-field-dropdown-ui", "value"),
prevent_initial_call=True,
)
def sync_category_dropdown(value):
return value
@callback(
Output("subcategory-field-dropdown", "value"),
Input("subcategory-field-dropdown-ui", "value"),
prevent_initial_call=True,
)
def sync_subcategory_dropdown(value):
return value
@callback(
Output("tags-field-dropdown", "value"),
Input("tags-field-dropdown-ui", "value"),
prevent_initial_call=True,
)
def sync_tags_dropdown(value):
return value
@staticmethod
def _document_to_dict(doc):
return {
@@ -118,3 +399,10 @@ class DataProcessingCallbacks:
f"❌ Error processing file{file_part}: {error}. "
"Please check that your file is valid NDJSON with required 'text' and 'embedding' fields."
)
@staticmethod
def _create_status_alert(message: str, color: str):
"""Create a status alert component."""
import dash_bootstrap_components as dbc
return dbc.Alert(message, color=color, className="mb-2")

View File

@@ -0,0 +1,320 @@
from dash import dcc, html
import dash_bootstrap_components as dbc
from .upload import UploadComponent
class DataSourceComponent:
def __init__(self):
self.upload_component = UploadComponent()
def create_tabbed_interface(self):
"""Create tabbed interface for different data sources."""
return dbc.Card(
[
dbc.CardHeader(
[
dbc.Tabs(
[
dbc.Tab(label="File Upload", tab_id="file-tab"),
dbc.Tab(label="OpenSearch", tab_id="opensearch-tab"),
],
id="data-source-tabs",
active_tab="file-tab",
)
]
),
dbc.CardBody([html.Div(id="tab-content")]),
]
)
def create_file_upload_tab(self):
"""Create file upload tab content."""
return html.Div(
[
self.upload_component.create_error_alert(),
self.upload_component.create_data_upload(),
self.upload_component.create_prompts_upload(),
self.upload_component.create_reset_button(),
]
)
def create_opensearch_tab(self):
"""Create OpenSearch tab content."""
return html.Div(
[
# Connection section
html.H6("Connection", className="mb-2"),
dbc.Row(
[
dbc.Col(
[
dbc.Label("OpenSearch URL:"),
dbc.Input(
id="opensearch-url",
type="text",
placeholder="https://opensearch.example.com:9200",
className="mb-2",
),
],
width=12,
),
]
),
dbc.Row(
[
dbc.Col(
[
dbc.Label("Index Name:"),
dbc.Input(
id="opensearch-index",
type="text",
placeholder="my-embeddings-index",
className="mb-2",
),
],
width=6,
),
dbc.Col(
[
dbc.Button(
"Test Connection",
id="test-connection-btn",
color="primary",
size="sm",
className="mt-4",
),
],
width=6,
className="d-flex align-items-end",
),
]
),
# Authentication section (collapsible)
dbc.Collapse(
[
html.Hr(),
html.H6("Authentication (Optional)", className="mb-2"),
dbc.Row(
[
dbc.Col(
[
dbc.Label("Username:"),
dbc.Input(
id="opensearch-username",
type="text",
className="mb-2",
),
],
width=6,
),
dbc.Col(
[
dbc.Label("Password:"),
dbc.Input(
id="opensearch-password",
type="password",
className="mb-2",
),
],
width=6,
),
]
),
dbc.Label("OR"),
dbc.Input(
id="opensearch-api-key",
type="text",
placeholder="API Key",
className="mb-2",
),
],
id="auth-collapse",
is_open=False,
),
dbc.Button(
"Show Authentication",
id="auth-toggle",
color="link",
size="sm",
className="p-0 mb-3",
),
# Connection status
html.Div(id="connection-status", className="mb-3"),
# Field mapping section (hidden initially)
html.Div(id="field-mapping-section", style={"display": "none"}),
# Hidden dropdowns to prevent callback errors
html.Div([
dcc.Dropdown(id="embedding-field-dropdown", style={"display": "none"}),
dcc.Dropdown(id="text-field-dropdown", style={"display": "none"}),
dcc.Dropdown(id="id-field-dropdown", style={"display": "none"}),
dcc.Dropdown(id="category-field-dropdown", style={"display": "none"}),
dcc.Dropdown(id="subcategory-field-dropdown", style={"display": "none"}),
dcc.Dropdown(id="tags-field-dropdown", style={"display": "none"}),
], style={"display": "none"}),
# Load data button (hidden initially)
html.Div(
[
dbc.Button(
"Load Data",
id="load-opensearch-data-btn",
color="success",
className="mb-2",
disabled=True,
),
],
id="load-data-section",
style={"display": "none"},
),
# OpenSearch status/results
html.Div(id="opensearch-status", className="mb-3"),
]
)
def create_field_mapping_interface(self, field_suggestions):
"""Create field mapping interface based on detected fields."""
return html.Div(
[
html.Hr(),
html.H6("Field Mapping", className="mb-2"),
html.P(
"Map your OpenSearch fields to the required format:",
className="text-muted small",
),
# Required fields
dbc.Row(
[
dbc.Col(
[
dbc.Label(
"Embedding Field (required):", className="fw-bold"
),
dcc.Dropdown(
id="embedding-field-dropdown-ui",
options=[
{"label": field, "value": field}
for field in field_suggestions.get("embedding", [])
],
value=field_suggestions.get("embedding", [None])[0], # Default to first suggestion
placeholder="Select embedding field...",
className="mb-2",
),
],
width=6,
),
dbc.Col(
[
dbc.Label(
"Text Field (required):", className="fw-bold"
),
dcc.Dropdown(
id="text-field-dropdown-ui",
options=[
{"label": field, "value": field}
for field in field_suggestions.get("text", [])
],
value=field_suggestions.get("text", [None])[0], # Default to first suggestion
placeholder="Select text field...",
className="mb-2",
),
],
width=6,
),
]
),
# Optional fields
html.H6("Optional Fields", className="mb-2 mt-3"),
dbc.Row(
[
dbc.Col(
[
dbc.Label("ID Field:"),
dcc.Dropdown(
id="id-field-dropdown-ui",
options=[
{"label": field, "value": field}
for field in field_suggestions.get("id", [])
],
value=field_suggestions.get("id", [None])[0], # Default to first suggestion
placeholder="Select ID field...",
className="mb-2",
),
],
width=6,
),
dbc.Col(
[
dbc.Label("Category Field:"),
dcc.Dropdown(
id="category-field-dropdown-ui",
options=[
{"label": field, "value": field}
for field in field_suggestions.get("category", [])
],
value=field_suggestions.get("category", [None])[0], # Default to first suggestion
placeholder="Select category field...",
className="mb-2",
),
],
width=6,
),
]
),
dbc.Row(
[
dbc.Col(
[
dbc.Label("Subcategory Field:"),
dcc.Dropdown(
id="subcategory-field-dropdown-ui",
options=[
{"label": field, "value": field}
for field in field_suggestions.get("subcategory", [])
],
value=field_suggestions.get("subcategory", [None])[0], # Default to first suggestion
placeholder="Select subcategory field...",
className="mb-2",
),
],
width=6,
),
dbc.Col(
[
dbc.Label("Tags Field:"),
dcc.Dropdown(
id="tags-field-dropdown-ui",
options=[
{"label": field, "value": field}
for field in field_suggestions.get("tags", [])
],
value=field_suggestions.get("tags", [None])[0], # Default to first suggestion
placeholder="Select tags field...",
className="mb-2",
),
],
width=6,
),
]
),
]
)
def create_error_alert(self):
"""Create error alert component for OpenSearch issues."""
return dbc.Alert(
id="opensearch-error-alert",
dismissable=True,
is_open=False,
color="danger",
className="mb-3",
)
def create_success_alert(self):
"""Create success alert component for OpenSearch operations."""
return dbc.Alert(
id="opensearch-success-alert",
dismissable=True,
is_open=False,
color="success",
className="mb-3",
)

View File

@@ -1,21 +1,22 @@
from dash import dcc, html
import dash_bootstrap_components as dbc
from .upload import UploadComponent
from .datasource import DataSourceComponent
class SidebarComponent:
def __init__(self):
self.upload_component = UploadComponent()
self.datasource_component = DataSourceComponent()
def create_layout(self):
return dbc.Col(
[
html.H5("Upload Data", className="mb-3"),
self.upload_component.create_error_alert(),
self.upload_component.create_data_upload(),
self.upload_component.create_prompts_upload(),
self.upload_component.create_reset_button(),
html.H5("Visualization Controls", className="mb-3"),
html.H5("Data Sources", className="mb-3"),
self.datasource_component.create_error_alert(),
self.datasource_component.create_success_alert(),
self.datasource_component.create_tabbed_interface(),
html.H5("Visualization Controls", className="mb-3 mt-4"),
]
+ self._create_method_dropdown()
+ self._create_color_dropdown()