Fix Python style violations (E501 line length, E122 indentation)

- Fix E501 line-too-long violations across 8 files by wrapping
  long lines in SQL strings, error messages, docstrings,
  LLM prompt strings, and test code
- Fix E122 continuation line indentation in database.py

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: dpage

web/pgadmin/llm/prompts/explain.py

@@ -9,21 +9,26 @@
 
 """System prompt for EXPLAIN plan analysis."""
 
-EXPLAIN_ANALYSIS_PROMPT = """You are a PostgreSQL performance expert integrated into pgAdmin 4.
-Your task is to analyze EXPLAIN plan output and provide actionable optimization recommendations.
+EXPLAIN_ANALYSIS_PROMPT = """You are a PostgreSQL performance
+expert integrated into pgAdmin 4.
+Your task is to analyze EXPLAIN plan output and provide
+actionable optimization recommendations.
 
 ## Input Format
 
 You will receive:
-1. The EXPLAIN plan output in JSON format (from EXPLAIN (FORMAT JSON, ANALYZE, ...))
+1. The EXPLAIN plan output in JSON format
+(from EXPLAIN (FORMAT JSON, ANALYZE, ...))
 2. The original SQL query that was analyzed
 
 ## Analysis Guidelines
 
 1. **Identify Performance Bottlenecks**:
    - Sequential scans on large tables (consider if an index would help)
-   - Nested loops with high row counts (may indicate missing indexes or poor join order)
-   - Large row estimate variances (actual vs planned) suggesting stale statistics
+   - Nested loops with high row counts
+   (may indicate missing indexes or poor join order)
+   - Large row estimate variances (actual vs planned)
+   suggesting stale statistics
    - Sort operations on large datasets without indexes
    - Hash joins spilling to disk (indicated by batch counts > 1)
    - High startup costs relative to total costs
@@ -49,7 +54,8 @@
 
 ## Response Format
 
-IMPORTANT: Your response MUST be ONLY a valid JSON object with no additional text,
+IMPORTANT: Your response MUST be ONLY a valid JSON object
+with no additional text,
 no markdown formatting, and no code blocks. Return exactly this format:
 
 {
@@ -69,15 +75,17 @@
       "sql": "Exact SQL to execute (if applicable, otherwise null)"
     }
   ],
-  "summary": "One paragraph summary of the overall plan performance and key takeaways"
+  "summary": "One paragraph summary of the overall
+  plan performance and key takeaways"
 }
 
 Rules:
 - Return ONLY the JSON object, nothing before or after it
 - Do NOT wrap the JSON in markdown code blocks (no ```)
 - Order bottlenecks by severity (high first)
 - Order recommendations by priority (1 = highest)
-- If the plan looks optimal, return empty bottlenecks array with a positive summary
+- If the plan looks optimal, return empty bottlenecks
+array with a positive summary
 - Always include at least a summary, even for simple plans
 - The "sql" field should be null if no SQL action is applicable
 """

web/pgadmin/llm/prompts/nlq.py

@@ -9,13 +9,16 @@
 
 """System prompt for Natural Language to SQL translation."""
 
-NLQ_SYSTEM_PROMPT = """You are a PostgreSQL SQL expert integrated into pgAdmin 4.
+NLQ_SYSTEM_PROMPT = """You are a PostgreSQL SQL expert \
+integrated into pgAdmin 4.
 Your task is to generate SQL queries based on natural language requests.
 
 You have access to database inspection tools:
 - get_database_schema: Get list of schemas, tables, and views in the database
-- get_table_info: Get detailed column, constraint, and index information for a table
-- execute_sql_query: Run read-only queries to understand data structure (SELECT only)
+- get_table_info: Get detailed column, constraint, and \
+index information for a table
+- execute_sql_query: Run read-only queries to understand \
+data structure (SELECT only)
 
 Guidelines:
 - Use get_database_schema to discover available tables before writing queries
@@ -31,5 +34,6 @@
 Rules:
 - Return ONLY the JSON object, nothing else
 - No markdown code blocks
-- If you need clarification, set "sql" to null and put your question in "explanation"
+- If you need clarification, set "sql" to null and put \
+your question in "explanation"
 """

web/pgadmin/llm/providers/docker.py

@@ -51,12 +51,15 @@ class DockerClient(LLMClient):
     which provides an OpenAI-compatible API.
     """
 
-    def __init__(self, api_url: Optional[str] = None, model: Optional[str] = None):
+    def __init__(
+        self, api_url: Optional[str] = None, model: Optional[str] = None
+    ):
         """
         Initialize the Docker Model Runner client.
 
         Args:
-            api_url: The Docker Model Runner API URL (default: http://localhost:12434).
+            api_url: The Docker Model Runner API URL
+                (default: http://localhost:12434).
             model: Optional model name. Defaults to ai/qwen3-coder.
         """
         self._api_url = (api_url or DEFAULT_API_URL).rstrip('/')
@@ -321,18 +324,26 @@ def _parse_response(self, data: dict) -> LLMResponse:
             if stop_reason == StopReason.MAX_TOKENS:
                 input_tokens = usage.input_tokens
                 raise LLMClientError(LLMError(
-                    message=f'Response truncated due to token limit '
-                            f'(input: {input_tokens} tokens). '
-                            f'The request is too large for model {self._model}. '
-                            f'Try using a model with a larger context window, '
-                            f'or analyze a smaller scope.',
+                    message=(
+                        f'Response truncated due to token limit '
+                        f'(input: {input_tokens} tokens). '
+                        f'The request is too large for model '
+                        f'{self._model}. '
+                        f'Try using a model with a larger context '
+                        f'window, or analyze a smaller scope.'
+                    ),
                     code='max_tokens',
                     provider=self.provider_name,
                     retryable=False
                 ))
-            elif finish_reason and finish_reason not in ('stop', 'tool_calls'):
+            elif finish_reason and finish_reason not in (
+                'stop', 'tool_calls'
+            ):
                 raise LLMClientError(LLMError(
-                    message=f'Empty response with finish reason: {finish_reason}',
+                    message=(
+                        f'Empty response with finish reason: '
+                        f'{finish_reason}'
+                    ),
                     code=finish_reason,
                     provider=self.provider_name,
                     retryable=False

web/pgadmin/llm/providers/openai.py

@@ -316,17 +316,20 @@ def _parse_response(self, data: dict) -> LLMResponse:
                 raise LLMClientError(LLMError(
                     message=f'Response truncated due to token limit '
                             f'(input: {input_tokens} tokens). '
-                            f'The request is too large for model {self._model}. '
-                            f'Try using a model with a larger context window, '
-                            f'or analyze a smaller scope (e.g., a specific schema '
-                            f'instead of the entire database).',
+                            f'The request is too large for model '
+                            f'{self._model}. '
+                            f'Try using a model with a larger context '
+                            f'window, or analyze a smaller scope (e.g., a '
+                            f'specific schema instead of the entire '
+                            f'database).',
                     code='max_tokens',
                     provider=self.provider_name,
                     retryable=False
                 ))
             elif finish_reason and finish_reason not in ('stop', 'tool_calls'):
                 raise LLMClientError(LLMError(
-                    message=f'Empty response with finish reason: {finish_reason}',
+                    message=(f'Empty response with finish reason: '
+                             f'{finish_reason}'),
                     code=finish_reason,
                     provider=self.provider_name,
                     retryable=False

web/pgadmin/llm/reports/pipeline.py

@@ -58,7 +58,8 @@ def __init__(
             report_type: Type of report ('security', 'performance', 'design').
             sections: List of available Section definitions.
             client: LLM client for making API calls.
-            query_executor: Function to execute queries given query_id and context.
+            query_executor: Function to execute queries given query_id and
+                           context.
             max_retries: Maximum retry attempts for rate-limited calls.
             retry_base_delay: Base delay in seconds for exponential backoff.
         """
@@ -88,7 +89,9 @@ def execute(self, context: dict) -> str:
             if event.get('type') == 'complete':
                 result = event.get('report', '')
             elif event.get('type') == 'error':
-                raise ReportPipelineError(event.get('message', 'Unknown error'))
+                raise ReportPipelineError(
+                    event.get('message', 'Unknown error')
+                )
         return result or ''
 
     def execute_with_progress(
@@ -234,7 +237,8 @@ def _planning_stage(self, context: dict) -> list[str]:
                 if sid in self.sections
             ]
 
-            return valid_ids if valid_ids else [s['id'] for s in available_sections]
+            return (valid_ids if valid_ids else
+                    [s['id'] for s in available_sections])
 
         except (json.JSONDecodeError, LLMClientError):
             # Fallback to all available sections
@@ -312,7 +316,8 @@ def _analyze_section_with_retry(
                     yield {
                         'type': 'retry',
                         'reason': 'rate_limit',
-                        'message': f'Rate limited, retrying in {wait_time}s...',
+                        'message': (f'Rate limited, retrying in '
+                                    f'{wait_time}s...'),
                         'wait_seconds': wait_time
                     }
                     time.sleep(wait_time)
@@ -382,15 +387,20 @@ def _synthesize_with_retry(
                     yield {
                         'type': 'retry',
                         'reason': 'rate_limit',
-                        'message': f'Rate limited, retrying in {wait_time}s...',
+                        'message': (f'Rate limited, retrying in '
+                                    f'{wait_time}s...'),
                         'wait_seconds': wait_time
                     }
                     time.sleep(wait_time)
                 else:
                     # Return partial report with section summaries
-                    partial = "**Note**: Synthesis failed. Section summaries:\n\n"
+                    partial = (
+                        "**Note**: Synthesis failed. Section summaries:\n\n"
+                    )
                     for r in successful_results:
-                        partial += f"## {r['section_name']}\n\n{r['summary']}\n\n"
+                        partial += (
+                            f"## {r['section_name']}\n\n{r['summary']}\n\n"
+                        )
                     yield {'type': 'result', 'result': partial}
                     return
 

web/pgadmin/llm/reports/prompts.py

@@ -14,9 +14,11 @@
 # Planning Stage Prompts
 # =============================================================================
 
-PLANNING_SYSTEM_PROMPT = """You are a PostgreSQL expert helping to plan a database analysis report.
+PLANNING_SYSTEM_PROMPT = """You are a PostgreSQL expert helping to plan a
+database analysis report.
 
-Your task is to select which analysis sections are most relevant for the given report type and database context.
+Your task is to select which analysis sections are most relevant for the
+given report type and database context.
 
 Return ONLY a JSON array of section IDs to analyze, ordered by priority.
 Only include sections that are relevant given the database characteristics.
@@ -55,14 +57,16 @@ def get_planning_user_prompt(
 - Table count: {context.get('table_count', 'Unknown')}
 - Has pg_stat_statements: {context.get('has_stat_statements', False)}
 
-Return a JSON array of section IDs to analyze, e.g.: ["section1", "section2", "section3"]"""
+Return a JSON array of section IDs to analyze, e.g.:
+["section1", "section2", "section3"]"""
 
 
 # =============================================================================
 # Section Analysis Prompts
 # =============================================================================
 
-SECTION_ANALYSIS_SYSTEM_PROMPT = """You are a PostgreSQL expert analyzing database configuration.
+SECTION_ANALYSIS_SYSTEM_PROMPT = """You are a PostgreSQL expert analyzing
+database configuration.
 
 Analyze the provided data and generate a concise summary (max 300 words).
 
@@ -111,7 +115,8 @@ def get_section_analysis_prompt(
 
     data_json = json.dumps(data, indent=2, default=str)
 
-    return f"""Analyze the following {section_name} data for a PostgreSQL {context.get('server_version', '')} server.
+    return f"""Analyze the following {section_name} data for a PostgreSQL
+{context.get('server_version', '')} server.
 
 Section focus: {section_description}
 
@@ -130,19 +135,25 @@ def get_section_analysis_prompt(
 # Synthesis Prompts
 # =============================================================================
 
-SYNTHESIS_SYSTEM_PROMPT = """You are a PostgreSQL expert creating a comprehensive report.
+SYNTHESIS_SYSTEM_PROMPT = """You are a PostgreSQL expert creating a
+comprehensive report.
 
 Combine the section summaries into a cohesive, well-organized report.
 
 Your report MUST:
 1. Start with an **Executive Summary** (3-5 sentences overview)
-2. Include a **Critical Issues** section (aggregate all critical/warning findings)
-3. Include each section's detailed analysis (use the section content as-is, don't add duplicate headers)
-4. End with **Prioritized Recommendations** (numbered list, most important first)
+2. Include a **Critical Issues** section (aggregate all critical/warning
+findings)
+3. Include each section's detailed analysis (use the section content as-is,
+don't add duplicate headers)
+4. End with **Prioritized Recommendations** (numbered list, most important
+first)
 
 IMPORTANT:
-- Do NOT include a report title at the very beginning - start directly with Executive Summary
-- Each section already has its own ### header - do NOT add extra headers around them
+- Do NOT include a report title at the very beginning - start directly with
+Executive Summary
+- Each section already has its own ### header - do NOT add extra headers
+around them
 - Simply organize and flow the sections together naturally
 
 Use severity indicators consistently:
@@ -151,7 +162,8 @@ def get_section_analysis_prompt(
 - 🟡 Advisory - Consider improving
 - 🟢 Good - No issues found
 
-Be professional and actionable. Include SQL commands for recommendations where helpful."""
+Be professional and actionable. Include SQL commands for recommendations
+where helpful."""
 
 
 def get_synthesis_prompt(
@@ -186,7 +198,8 @@ def get_synthesis_prompt(
     if context.get('schema_name'):
         scope_info = f"{context['schema_name']} schema in {scope_info}"
 
-    return f"""Create a comprehensive {report_type_display} Report for {scope_info}.
+    return f"""Create a comprehensive {report_type_display} Report for
+{scope_info}.
 
 Server: PostgreSQL {context.get('server_version', 'Unknown')}