feat: Phase 1 - Fragenergänzung + Strukturierter Container
Deploy Development / deploy (push) Successful in 49s
Build Test / lint-backend (push) Successful in 0s
Build Test / build-frontend (push) Successful in 14s

Backend:
- question_augmenter.py (290 Zeilen): Hybrid-Modell für Fragenergänzungen
  * merge_question_augmentations(): Knotengebundene Fragen überschreiben Prompt-Defaults
  * augment_prompt_with_questions(): Markdown-formatierte Fragenergänzung
  * parse_question_augmentations_from_jsonb(): JSONB → QuestionAugmentation[]
- result_container_parser.py (250 Zeilen): Markdown-Sektionen-Parsing
  * parse_result_container(): Extrahiert Analysekern, Entscheidungsanteil, Begründungsanker
  * validate_decision_signal(): Normalisierung gegen answer_spectrum
  * Fallback-Parsing bei unstrukturierten Antworten
- routers/workflow_questions.py (236 Zeilen): CRUD für workflow_question_catalog
  * GET /api/workflow/questions (mit active_only Filter)
  * POST/PUT/DELETE (Admin only, Soft Delete)
- prompt_executor.py: Integration in execute_base_prompt()
  * Fragenergänzung vor LLM-Call (wenn node_questions oder catalog vorhanden)
  * Result-Container-Parsing nach LLM-Response
- main.py: Router-Registrierung (workflow_questions)

Tests:
- test_phase1_question_augmenter.py (8 Tests): Hybrid-Modell, Formatierung, JSONB-Parsing
- test_phase1_result_container_parser.py (17 Tests): Sektion-Extraktion, Decision-Parsing, Validierung

Alle 25 Unit-Tests bestanden.

version: 0.9j (backend)
module:  workflow 0.2.0

Konzept: .claude/task/Workflow_engine_prompting_engine/konzept_workflow_engine_konsolidated.md (Phase 1)
This commit is contained in:
Lars committed 2026-04-03 18:02:25 +02:00
1 parent b5be6e21a5
commit ca562b7130
8 files changed
+1237 -9

No files matched your search

+55 -7
View File
@@ -205,18 +205,54 @@ async def execute_base_prompt(
variables: Dict[str, Any],
openrouter_call_func,
enable_debug: bool = False,
catalog: Optional[Dict] = None
catalog: Optional[Dict] = None,
node_questions: Optional[list] = None # Phase 1: Knotengebundene Fragen
) -> Dict[str, Any]:
"""Execute a base-type prompt (single template)."""
"""
Execute a base-type prompt (single template).
Phase 1: Unterstützt Fragenergänzungen (Hybridmodell)
- node_questions: Knotengebundene Fragen (Priorität 1)
- prompt.question_augmentations: Prompt-Defaults (Priorität 2)
"""
from question_augmenter import (
parse_question_augmentations_from_jsonb,
merge_question_augmentations,
augment_prompt_with_questions
)
from result_container_parser import parse_result_container_robust
template = prompt.get('template')
if not template:
raise HTTPException(400, f"Base prompt missing template: {prompt['slug']}")
debug_info = {} if enable_debug else None
# Phase 1: Load question augmentations (Hybridmodell)
prompt_default_questions = None
if prompt.get('question_augmentations'):
try:
from workflow_models import QuestionAugmentation
prompt_default_questions = parse_question_augmentations_from_jsonb(
prompt['question_augmentations']
)
except Exception as e:
if enable_debug:
debug_info['question_augmentations_error'] = str(e)
# Merge question augmentations (Vorrangregel: Knoten > Prompt)
questions = merge_question_augmentations(node_questions, prompt_default_questions)
# Resolve placeholders (with optional catalog for |d modifier)
prompt_text = resolve_placeholders(template, variables, debug_info, catalog)
# Phase 1: Augment prompt with questions (if any)
if questions:
prompt_text = augment_prompt_with_questions(prompt_text, questions)
if enable_debug:
debug_info['question_augmentations_count'] = len(questions)
debug_info['question_types'] = [q.type for q in questions]
if enable_debug:
debug_info['template'] = template
debug_info['final_prompt'] = prompt_text[:500] + ('...' if len(prompt_text) > 500 else '')
@@ -229,12 +265,24 @@ async def execute_base_prompt(
debug_info['ai_response_length'] = len(response)
debug_info['ai_response_preview'] = response[:200] + ('...' if len(response) > 200 else '')
# Validate JSON if required
output_format = prompt.get('output_format', 'text')
if output_format == 'json':
output = validate_json_output(response, prompt.get('output_schema'), debug_info if enable_debug else None)
# Phase 1: Parse structured result if questions were used
if questions:
expected_question_types = [q.type for q in questions]
container = parse_result_container_robust(response, expected_question_types)
if enable_debug:
debug_info['parsing_status'] = container['parsing_status']
debug_info['parsing_warnings'] = container.get('warnings', [])
output = container
output_format = 'structured_container' # New format type
else:
output = response
# Legacy behavior: Validate JSON if required
output_format = prompt.get('output_format', 'text')
if output_format == 'json':
output = validate_json_output(response, prompt.get('output_schema'), debug_info if enable_debug else None)
else:
output = response
result = {
"type": "base",