dbagent-cli 0.8.5__tar.gz → 0.8.7__tar.gz

{dbagent_cli-0.8.5 → dbagent_cli-0.8.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dbagent-cli
-Version: 0.8.5
+Version: 0.8.7
 Summary: Enterprise AI Database Copilot, Schema Graph Introspector & Autonomous Recovery Agent (CLI)
 Author: Santhosh Gupta
 Requires-Python: >=3.9

{dbagent_cli-0.8.5 → dbagent_cli-0.8.7}/dbagent/__init__.py

@@ -2,5 +2,5 @@
 DB-Agent: Universal Database Introspection and Script Generation AI Agent (CLI).
 """
 
-__version__ = "0.8.5"
+__version__ = "0.8.7"
 __author__ = "Santhosh Gupta"

{dbagent_cli-0.8.5 → dbagent_cli-0.8.7}/dbagent/agent/generator.py

@@ -84,6 +84,43 @@ You are chatting with the user in the interactive database shell.
 """
 
 
+# --- Architectural / Linkage analysis prompt ---
+
+ARCHITECTURE_SYSTEM_PROMPT = """You are DB-Agent, a Principal Database Architect, Data Modeler, and Systems Engineer.
+Target Database Context:
+- **Dialect**: {dialect_name}
+- **Database Name**: {database_name}
+
+### Detailed Schema Sub-Graph Context (Tables, Columns, Primary Keys, Foreign Keys, Sample Data):
+{schema_context}
+
+The user is asking for architectural analysis, entity relationships, linkages, mappings, or how tables relate to each other in this database.
+
+Provide a comprehensive, high-quality, professional Architectural Report structured as follows:
+
+### 1. 🗺️ Entity-Relationship & Linkage Architecture Diagram
+Provide a clean, readable ASCII / Unicode box diagram showing the entities, relationship paths, cardinality (1:1, 1:N, N:M junction tables), and the exact connecting foreign keys / join keys.
+Example structure:
+```text
+┌──────────────┐          ┌──────────────────────┐          ┌───────────────┐
+│  ivis_user   │ 1 ──── * │ ivis_tenant_user_map │ * ──── 1 │  ivis_tenant  │
+└──────────────┘          └──────────────────────┘          └───────────────┘
+  (pk_user_id)                  (fk_user_id)                  (pk_tenant_id)
+```
+
+### 2. 📐 Architectural & Theoretical Breakdown
+- Explain in clear theoretical and domain terms how the entities map to each other.
+- Highlight the mapping mechanism (direct foreign key vs junction/mapping table vs hierarchical tree).
+- Detail the cardinality, tenant isolation, and user access models represented in the schema.
+
+### 3. 🔍 How to Query & Verify Mappings (SQL Examples)
+Provide ready-to-run, dialect-accurate SQL JOIN queries answering the user's specific questions (e.g. how to find all sites mapped to a user, how to get tenant membership for a user). Format all SQL in ```sql blocks.
+
+### 4. 💡 Architectural Insights & Best Practices
+Mention indexing recommendations, cascade rules, or performance considerations for these join paths.
+"""
+
+
 # --- Retry prompt when SQL fails ---
 
 RETRY_PROMPT_TEMPLATE = """The previous SQL query failed with this error:
@@ -111,7 +148,7 @@ class ScriptGenerator:
     @staticmethod
     def classify_intent(user_prompt: str) -> str:
         """
-        Classify user's natural language intent as 'chat', 'read', 'write', or 'ddl'.
+        Classify user's natural language intent as 'chat', 'analysis', 'read', 'write', or 'ddl'.
         Used before SQL generation to predict the query type.
         """
         prompt_lower = user_prompt.lower().strip()
@@ -132,6 +169,21 @@ class ScriptGenerator:
             if clean_prompt == trig or prompt_lower.startswith(f"{trig} ") or prompt_lower.startswith(f"{trig},") or prompt_lower.startswith(f"{trig}!"):
                 return "chat"
 
+        # Architecture / Relationship / Linkage / Analysis indicators
+        analysis_keywords = [
+            "linkage architecture", "linkage", "architecture", "relationship between",
+            "how are", "how is", "how were", "how they were related", "how were they related",
+            "how they were mapped", "how to map", "mapped to", "er diagram", "diagram of",
+            "data model", "explain relationship", "explain schema", "schema analysis",
+            "analyze the database", "analyze database", "table relationship",
+            "foreign key relationship", "how are tables related", "which sites in",
+            "how to know that which", "how to know which", "mapping architecture",
+            "entity relationship", "theoritical information", "theoretical information",
+        ]
+        for kw in analysis_keywords:
+            if kw in prompt_lower:
+                return "analysis"
+
         # DDL indicators
         ddl_keywords = [
             "create table", "alter table", "drop table", "add column",
@@ -194,6 +246,39 @@ class ScriptGenerator:
             model=model,
         )
 
+    def generate_analysis_response(
+        self,
+        schema: DatabaseSchema,
+        user_prompt: str,
+        conversation_history: Optional[str] = None,
+        model: Optional[str] = None,
+    ) -> str:
+        """Generate a comprehensive architectural analysis report with ASCII diagrams, theory, and SQL queries."""
+        # 1. Filter relevant tables including foreign keys and junction tables
+        relevant_tables = SchemaSelector.select_relevant_tables(schema, user_prompt, max_tables=20)
+        schema_markdown = SchemaFormatter.to_markdown(
+            schema,
+            selected_tables=relevant_tables,
+            include_samples=True,
+            max_sample_rows=2,
+        )
+
+        system_prompt = ARCHITECTURE_SYSTEM_PROMPT.format(
+            dialect_name=schema.dialect_name,
+            database_name=schema.database_name,
+            schema_context=schema_markdown,
+        )
+
+        full_prompt = user_prompt
+        if conversation_history:
+            full_prompt = f"{conversation_history}\n\nUser: {user_prompt}"
+
+        return self.llm.generate(
+            prompt=full_prompt,
+            system_prompt=system_prompt,
+            model=model,
+        )
+
     # --- Full generation (for `generate` command) ---
 
     def generate_script(

{dbagent_cli-0.8.5 → dbagent_cli-0.8.7}/dbagent/agent/pipeline.py

@@ -202,6 +202,34 @@ class QueryPipeline:
                 result.execution_time_ms = (time.time() - t_start) * 1000
                 return result
 
+        # --- 1.2. Architectural / Linkage Analysis Path ---
+        if result.intent == "analysis":
+            try:
+                cached = self.cache_mgr.load_cached_schema(self.db_id)
+                if cached:
+                    schema = cached
+                else:
+                    schema = self.connector.inspect_schema(include_samples=True, max_samples=2)
+
+                result.schema = schema
+                conv_history = self.conversation.format_context_for_prompt()
+                result.message = self.generator.generate_analysis_response(
+                    schema=schema,
+                    user_prompt=user_prompt,
+                    conversation_history=conv_history,
+                    model=self.model,
+                )
+                result.is_chat = True
+                result.was_executed = True
+                result.query_type = "analysis"
+                result.execution_time_ms = (time.time() - t_start) * 1000
+                self.conversation.add_turn(user_prompt=user_prompt, sql="", assistant_response=result.message, is_chat=True)
+                return result
+            except Exception as e:
+                result.error = f"Architecture analysis failed: {str(e)}"
+                result.execution_time_ms = (time.time() - t_start) * 1000
+                return result
+
         # --- 1.5. Direct Raw SQL Fast-Path (0ms AI latency, 0 tokens) ---
         if SQLValidator.is_direct_sql(user_prompt):
             raw_sql = user_prompt.strip().rstrip(";") + ";"

{dbagent_cli-0.8.5 → dbagent_cli-0.8.7}/dbagent/ui/chat_screen.py

@@ -91,13 +91,12 @@ def render_chat_header(
 
 
 def render_user_turn(user_prompt: str) -> None:
-    """Render a user turn message bubble/card."""
-    user_text = Text()
-    user_text.append("💬 You: ", style="bold blue")
-    user_text.append(user_prompt, style="bold white")
+    """Render a clean highlighted user prompt bar like Claude Code."""
     console.print("")
-    console.print(Rule(style="dim blue"))
-    console.print(user_text)
+    prompt_text = Text()
+    prompt_text.append("❯ ", style="bold cyan")
+    prompt_text.append(f"{user_prompt}", style="bold white")
+    console.print(Panel(prompt_text, box=None, style="on grey15", padding=(0, 1)))
 
 
 def render_assistant_turn(
@@ -107,38 +106,38 @@ def render_assistant_turn(
     ai_model: Optional[str] = None,
 ) -> None:
     """Render the assistant response card with SQL, data grid, and execution timing."""
+    timing_str = f"⚡ {result.execution_time_ms / 1000:.2f}s" if result.execution_time_ms > 0 else ""
+    
     if result.is_direct_sql and result.retries == 0:
-        agent_title = "[bold cyan]🤖 DB-Agent[/bold cyan] [dim](Direct SQL • 0 tokens)[/dim]"
+        agent_title = f"[bold cyan]● DB-Agent[/bold cyan] [dim](Direct SQL • 0 tokens · {timing_str})[/dim]"
     else:
         model_display = f" ({ai_model})" if ai_model else ""
-        agent_title = f"[bold cyan]🤖 DB-Agent{model_display}[/bold cyan]"
-        
-    timing_str = f"⚡ {result.execution_time_ms / 1000:.2f}s" if result.execution_time_ms > 0 else ""
+        agent_title = f"[bold cyan]● DB-Agent{model_display}[/bold cyan] [dim]{timing_str}[/dim]"
     
     # Conversational Chat Response
     if result.is_chat and result.message:
-        console.print(f"\n{agent_title} [dim]{timing_str}[/dim]")
+        console.print(f"\n{agent_title}")
         console.print(Markdown(result.message))
         console.print("")
         return
 
     # Error Response
     if result.error:
-        console.print(f"\n[bold red]🤖 DB-Agent (Error)[/bold red] [dim]{timing_str}[/dim]")
+        console.print(f"\n[bold red]● DB-Agent (Error)[/bold red] [dim]{timing_str}[/dim]")
         if result.friendly_error:
-            print_error(result.friendly_error)
+            console.print(f"  [dim]└[/dim] [bold yellow]{result.friendly_error}[/bold yellow]")
         else:
-            print_error(f"Error: {result.error}")
+            console.print(f"  [dim]└[/dim] [bold red]{result.error}[/bold red]")
 
         if result.error_suggestions:
-            console.print(f"[dim]Did you mean: {', '.join(result.error_suggestions)}[/dim]")
+            console.print(f"  [dim]└ Did you mean: {', '.join(result.error_suggestions)}[/dim]")
 
         if result.sql:
-            console.print("[dim]Generated SQL:[/dim]")
+            console.print("  [dim]Generated SQL:[/dim]")
             syntax = Syntax(result.sql.strip(), "sql", theme="monokai", line_numbers=False)
             console.print(Panel(syntax, border_style="red", padding=(0, 1)))
         if result.retries > 0:
-            console.print(f"[dim](Retried {result.retries} time(s))[/dim]")
+            console.print(f"  [dim]└ (Retried {result.retries} time(s))[/dim]")
         console.print("")
         return
 

{dbagent_cli-0.8.5 → dbagent_cli-0.8.7}/dbagent_cli.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dbagent-cli
-Version: 0.8.5
+Version: 0.8.7
 Summary: Enterprise AI Database Copilot, Schema Graph Introspector & Autonomous Recovery Agent (CLI)
 Author: Santhosh Gupta
 Requires-Python: >=3.9

{dbagent_cli-0.8.5 → dbagent_cli-0.8.7}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "dbagent-cli"
-version = "0.8.5"
+version = "0.8.7"
 description = "Enterprise AI Database Copilot, Schema Graph Introspector & Autonomous Recovery Agent (CLI)"
 readme = "README.md"
 requires-python = ">=3.9"

{dbagent_cli-0.8.5 → dbagent_cli-0.8.7}/tests/test_generator.py

@@ -71,3 +71,23 @@ def test_validator_destructive_delete_without_where():
     is_safe, warnings = ScriptValidator.analyze_safety("DELETE FROM orders")
     assert is_safe is False
     assert any("without WHERE" in w for w in warnings)
+
+
+def test_classify_intent_analysis():
+    prompt = "analyze the database and give me the linkage architecture that how ivis_user and ivis_tenant was related and how they were mapped and how to know that which sites in ivis_site were mapped to which user"
+    assert ScriptGenerator.classify_intent(prompt) == "analysis"
+
+    assert ScriptGenerator.classify_intent("how are users and orders related?") == "analysis"
+    assert ScriptGenerator.classify_intent("explain the relationship between customers and payments") == "analysis"
+    assert ScriptGenerator.classify_intent("give me the ER diagram and linkage architecture") == "analysis"
+
+
+def test_generate_analysis_response(mock_schema):
+    provider = MockProvider()
+    generator = ScriptGenerator(provider)
+    result = generator.generate_analysis_response(
+        mock_schema,
+        "analyze the linkage architecture between events and users",
+    )
+    assert result is not None
+    assert len(result) > 0