Files

535 lines
19 KiB
Python
Raw Permalink Normal View History

"""
NX Parameter Updater
Updates design variable values in NX .prt files.
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
This module can read expressions in two ways:
1. Parse .exp files (NX native export format) - RECOMMENDED, captures ALL expressions
2. Parse binary .prt files directly - fallback method, may miss some expressions
For updating values:
1. Binary .prt file modification (current implementation)
2. Future: Use NXOpen API if NX is running
The .exp format is preferred for reading because it captures:
- All expression types (formulas, references, constants)
- Unitless expressions
- Complete accuracy
"""
from pathlib import Path
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
from typing import Dict, List, Optional
import re
import shutil
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
import subprocess
from datetime import datetime
refactor: Centralize NX and environment configuration in config.py MAJOR IMPROVEMENT: Single source of truth for all system paths Now to change NX version or Python environment, edit ONE file (config.py): NX_VERSION = "2412" # Change this for NX updates PYTHON_ENV_NAME = "atomizer" # Change this for env updates All code automatically uses new paths - no manual file hunting! New Central Configuration (config.py): - NX_VERSION: Automatically updates all NX paths - NX_INSTALLATION_DIR: Derived from version - NX_RUN_JOURNAL: Path to run_journal.exe - NX_MATERIAL_LIBRARY: Path to physicalmateriallibrary.xml - NX_PYTHON_STUBS: Path to Python stubs for intellisense - PYTHON_ENV_NAME: Python environment name - PROJECT_ROOT: Auto-detected project root - Helper functions: get_nx_journal_command(), validate_config(), print_config() Updated Files to Use Config: - optimization_engine/nx_updater.py: Uses NX_RUN_JOURNAL from config - dashboard/api/app.py: Uses NX_RUN_JOURNAL from config - Both have fallbacks if config unavailable Benefits: 1. Change NX version in 1 place, not 10+ files 2. Automatic validation of paths on import 3. Helper functions for common operations 4. Clear error messages if paths missing 5. Easy to add new Simcenter versions Future NX Update Process: 1. Edit config.py: NX_VERSION = "2506" 2. Run: python config.py (verify paths) 3. Done! All code uses NX 2506 Migration Scripts Included: - migrate_to_config.py: Full migration with documentation - apply_config_migration.py: Applied to update dashboard 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 14:31:33 -05:00
import sys
# Import centralized configuration
sys.path.insert(0, str(Path(__file__).parent.parent))
from config import NX_RUN_JOURNAL
class NXParameterUpdater:
"""
Updates parametric expression values in NX .prt files.
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
NX Expression Formats:
Binary .prt format:
#(Number [mm]) tip_thickness: 20.0;
*(Number [degrees]) support_angle: 30.0;
.exp export format (RECOMMENDED for reading):
[MilliMeter]beam_length=5000
[Kilogram]mass=973.968443678471
hole_count=10
Pattern_p7=hole_count
"""
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
def __init__(self, prt_file_path: Path, backup: bool = True, nx_run_journal_path: Optional[Path] = None):
"""
Initialize updater for a specific .prt file.
Args:
prt_file_path: Path to NX .prt file
backup: If True, create backup before modifying
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
nx_run_journal_path: Path to NX run_journal.exe (for .exp export)
If None, uses default NX 2412 path
"""
self.prt_path = Path(prt_file_path)
if not self.prt_path.exists():
raise FileNotFoundError(f".prt file not found: {prt_file_path}")
self.backup_enabled = backup
self.content = None
self.text_content = None
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
# Default NX run_journal.exe path
if nx_run_journal_path is None:
refactor: Centralize NX and environment configuration in config.py MAJOR IMPROVEMENT: Single source of truth for all system paths Now to change NX version or Python environment, edit ONE file (config.py): NX_VERSION = "2412" # Change this for NX updates PYTHON_ENV_NAME = "atomizer" # Change this for env updates All code automatically uses new paths - no manual file hunting! New Central Configuration (config.py): - NX_VERSION: Automatically updates all NX paths - NX_INSTALLATION_DIR: Derived from version - NX_RUN_JOURNAL: Path to run_journal.exe - NX_MATERIAL_LIBRARY: Path to physicalmateriallibrary.xml - NX_PYTHON_STUBS: Path to Python stubs for intellisense - PYTHON_ENV_NAME: Python environment name - PROJECT_ROOT: Auto-detected project root - Helper functions: get_nx_journal_command(), validate_config(), print_config() Updated Files to Use Config: - optimization_engine/nx_updater.py: Uses NX_RUN_JOURNAL from config - dashboard/api/app.py: Uses NX_RUN_JOURNAL from config - Both have fallbacks if config unavailable Benefits: 1. Change NX version in 1 place, not 10+ files 2. Automatic validation of paths on import 3. Helper functions for common operations 4. Clear error messages if paths missing 5. Easy to add new Simcenter versions Future NX Update Process: 1. Edit config.py: NX_VERSION = "2506" 2. Run: python config.py (verify paths) 3. Done! All code uses NX 2506 Migration Scripts Included: - migrate_to_config.py: Full migration with documentation - apply_config_migration.py: Applied to update dashboard 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 14:31:33 -05:00
self.nx_run_journal_path = NX_RUN_JOURNAL
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
else:
self.nx_run_journal_path = Path(nx_run_journal_path)
self._load_file()
def _load_file(self):
"""Load .prt file as binary."""
with open(self.prt_path, 'rb') as f:
self.content = bytearray(f.read())
# Decode as latin-1 for text operations (preserves all bytes)
self.text_content = self.content.decode('latin-1', errors='ignore')
def _create_backup(self):
"""Create timestamped backup of original file."""
timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
backup_path = self.prt_path.with_suffix(f'.prt.bak_{timestamp}')
shutil.copy2(self.prt_path, backup_path)
print(f"Backup created: {backup_path}")
return backup_path
def find_expressions(self) -> List[Dict[str, any]]:
"""
Find all expressions in the .prt file.
Returns:
List of dicts with name, value, units
"""
expressions = []
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
# Pattern for NX expressions (with optional units):
# #(Number [mm]) tip_thickness: 20.0; - with units
# *(Number [mm]) p3: 10.0; - with units
# ((Number [degrees]) support_angle: 30.0; - with units
# (Number) hole_count: 5.0; - without units (unitless)
pattern = r'[#*\(]*\((\w+)(?:\s*\[([^\]]*)\])?\)\s*([a-zA-Z_][a-zA-Z0-9_]*)\s*:\s*([-+]?\d*\.?\d+(?:[eE][-+]?\d+)?)'
for match in re.finditer(pattern, self.text_content):
expr_type, units, name, value = match.groups()
expressions.append({
'name': name,
'value': float(value),
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
'units': units if units else '', # Empty string if no units
'type': expr_type
})
return expressions
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
def export_expressions_to_exp(self, output_path: Optional[Path] = None) -> Path:
"""
Export expressions to .exp file using NX journal.
This is the RECOMMENDED method for reading expressions because it:
- Captures ALL expressions (formulas, references, constants)
- Includes unitless expressions
- Uses NX's native export, ensuring 100% accuracy
Args:
output_path: Path for .exp file (without .exp extension)
If None, uses temp file in same directory as .prt
Returns:
Path to the .exp file created
"""
if output_path is None:
# Create temp file in same directory
output_path = self.prt_path.with_suffix('') # Remove .prt
output_path = Path(str(output_path) + "_expressions")
# Get paths
journal_script = Path(__file__).parent / "export_expressions.py"
if not journal_script.exists():
raise FileNotFoundError(f"Export journal script not found: {journal_script}")
if not self.nx_run_journal_path.exists():
raise FileNotFoundError(f"NX run_journal.exe not found: {self.nx_run_journal_path}")
# Run NX journal to export expressions
print(f"[NX] Exporting expressions from {self.prt_path.name} to .exp format...")
# NX run_journal.exe syntax: run_journal.exe <journal-file> -args <arg1> <arg2> ...
# Build command string with proper quoting
cmd_str = f'"{self.nx_run_journal_path}" "{journal_script}" -args "{self.prt_path}" "{output_path}"'
result = subprocess.run(cmd_str, capture_output=True, text=True, shell=True)
exp_file = Path(str(output_path) + ".exp")
# NOTE: NX run_journal.exe treats sys.exit(0) as a "syntax error" even though
# it's a successful exit. We check if the file was created instead of return code.
if not exp_file.exists():
print(f"[ERROR] NX journal failed to create .exp file:")
print(result.stdout)
print(result.stderr)
raise FileNotFoundError(f"Expected .exp file not created: {exp_file}")
print(f"[OK] Expressions exported to: {exp_file}")
return exp_file
def parse_exp_file(self, exp_file_path: Path) -> Dict[str, Dict[str, any]]:
"""
Parse a .exp file and return all expressions.
.exp format examples:
[MilliMeter]beam_length=5000
[Kilogram]p173=973.968443678471
hole_count=10
Pattern_p7=hole_count
Args:
exp_file_path: Path to .exp file
Returns:
Dict mapping expression name to info dict with 'value', 'units', 'formula'
"""
expressions = {}
with open(exp_file_path, 'r', encoding='utf-8') as f:
for line in f:
line = line.strip()
# Skip empty lines and comments
if not line or line.startswith('//'):
continue
# Pattern: [Unit]name=value or name=value
# [MilliMeter]beam_length=5000
# hole_count=10
# Pattern_p7=hole_count (formula reference)
match = re.match(r'(?:\[([^\]]+)\])?([a-zA-Z_][a-zA-Z0-9_]*)=(.*)', line)
if match:
units, name, value_str = match.groups()
# Try to parse as number
try:
value = float(value_str)
formula = None
except ValueError:
# It's a formula/reference (e.g., "hole_count")
value = None
formula = value_str
expressions[name] = {
'value': value,
'units': units if units else '',
'formula': formula,
'type': 'Number' # All .exp expressions are Number type
}
return expressions
def get_all_expressions(self, use_exp_export: bool = True) -> Dict[str, Dict[str, any]]:
feat: Add substudy system with live history tracking and workflow fixes Major Features: - Hierarchical substudy system (like NX Solutions/Subcases) * Shared model files across all substudies * Independent configuration per substudy * Continuation support from previous substudies * Real-time incremental history updates - Live history tracking with optimization_history_incremental.json - Complete bracket_displacement_maximizing study with substudy examples Core Fixes: - Fixed expression update workflow to pass design_vars through simulation_runner * Restored working NX journal expression update mechanism * OP2 timestamp verification instead of file deletion * Resolved issue where all trials returned identical objective values - Fixed LLMOptimizationRunner to pass design variables to simulation runner - Enhanced NXSolver with timestamp-based file regeneration verification New Components: - optimization_engine/llm_optimization_runner.py - LLM-driven optimization runner - optimization_engine/optimization_setup_wizard.py - Phase 3.3 setup wizard - studies/bracket_displacement_maximizing/ - Complete substudy example * run_substudy.py - Substudy runner with continuation * run_optimization.py - Standalone optimization runner * config/substudy_template.json - Template for new substudies * substudies/coarse_exploration/ - 20-trial coarse search * substudies/fine_tuning/ - 50-trial refinement (continuation example) * SUBSTUDIES_README.md - Complete substudy documentation Technical Improvements: - Incremental history saving after each trial (optimization_history_incremental.json) - Expression update workflow: .prt update → NX journal receives values → geometry update → FEM update → solve - Trial indexing fix in substudy result saving - Updated README with substudy system documentation Testing: - Successfully ran 20-trial coarse_exploration substudy - Verified different objective values across trials (workflow fix validated) - Confirmed live history updates in real-time - Tested shared model file usage across substudies 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-16 21:29:54 -05:00
"""
Get all expressions as a dictionary.
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
Args:
use_exp_export: If True, uses NX .exp export (RECOMMENDED)
If False, uses binary .prt parsing (may miss expressions)
feat: Add substudy system with live history tracking and workflow fixes Major Features: - Hierarchical substudy system (like NX Solutions/Subcases) * Shared model files across all substudies * Independent configuration per substudy * Continuation support from previous substudies * Real-time incremental history updates - Live history tracking with optimization_history_incremental.json - Complete bracket_displacement_maximizing study with substudy examples Core Fixes: - Fixed expression update workflow to pass design_vars through simulation_runner * Restored working NX journal expression update mechanism * OP2 timestamp verification instead of file deletion * Resolved issue where all trials returned identical objective values - Fixed LLMOptimizationRunner to pass design variables to simulation runner - Enhanced NXSolver with timestamp-based file regeneration verification New Components: - optimization_engine/llm_optimization_runner.py - LLM-driven optimization runner - optimization_engine/optimization_setup_wizard.py - Phase 3.3 setup wizard - studies/bracket_displacement_maximizing/ - Complete substudy example * run_substudy.py - Substudy runner with continuation * run_optimization.py - Standalone optimization runner * config/substudy_template.json - Template for new substudies * substudies/coarse_exploration/ - 20-trial coarse search * substudies/fine_tuning/ - 50-trial refinement (continuation example) * SUBSTUDIES_README.md - Complete substudy documentation Technical Improvements: - Incremental history saving after each trial (optimization_history_incremental.json) - Expression update workflow: .prt update → NX journal receives values → geometry update → FEM update → solve - Trial indexing fix in substudy result saving - Updated README with substudy system documentation Testing: - Successfully ran 20-trial coarse_exploration substudy - Verified different objective values across trials (workflow fix validated) - Confirmed live history updates in real-time - Tested shared model file usage across substudies 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-16 21:29:54 -05:00
Returns:
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
Dict mapping expression name to info dict with 'value', 'units', 'type', 'formula'
feat: Add substudy system with live history tracking and workflow fixes Major Features: - Hierarchical substudy system (like NX Solutions/Subcases) * Shared model files across all substudies * Independent configuration per substudy * Continuation support from previous substudies * Real-time incremental history updates - Live history tracking with optimization_history_incremental.json - Complete bracket_displacement_maximizing study with substudy examples Core Fixes: - Fixed expression update workflow to pass design_vars through simulation_runner * Restored working NX journal expression update mechanism * OP2 timestamp verification instead of file deletion * Resolved issue where all trials returned identical objective values - Fixed LLMOptimizationRunner to pass design variables to simulation runner - Enhanced NXSolver with timestamp-based file regeneration verification New Components: - optimization_engine/llm_optimization_runner.py - LLM-driven optimization runner - optimization_engine/optimization_setup_wizard.py - Phase 3.3 setup wizard - studies/bracket_displacement_maximizing/ - Complete substudy example * run_substudy.py - Substudy runner with continuation * run_optimization.py - Standalone optimization runner * config/substudy_template.json - Template for new substudies * substudies/coarse_exploration/ - 20-trial coarse search * substudies/fine_tuning/ - 50-trial refinement (continuation example) * SUBSTUDIES_README.md - Complete substudy documentation Technical Improvements: - Incremental history saving after each trial (optimization_history_incremental.json) - Expression update workflow: .prt update → NX journal receives values → geometry update → FEM update → solve - Trial indexing fix in substudy result saving - Updated README with substudy system documentation Testing: - Successfully ran 20-trial coarse_exploration substudy - Verified different objective values across trials (workflow fix validated) - Confirmed live history updates in real-time - Tested shared model file usage across substudies 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-16 21:29:54 -05:00
"""
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
if use_exp_export:
# Use NX native .exp export (captures ALL expressions)
try:
exp_file = self.export_expressions_to_exp()
expressions = self.parse_exp_file(exp_file)
# Clean up temp file
exp_file.unlink()
return expressions
except Exception as e:
print(f"[WARNING] .exp export failed: {e}")
print("[WARNING] Falling back to binary .prt parsing...")
# Fall through to binary parsing
# Fallback: Binary .prt parsing
feat: Add substudy system with live history tracking and workflow fixes Major Features: - Hierarchical substudy system (like NX Solutions/Subcases) * Shared model files across all substudies * Independent configuration per substudy * Continuation support from previous substudies * Real-time incremental history updates - Live history tracking with optimization_history_incremental.json - Complete bracket_displacement_maximizing study with substudy examples Core Fixes: - Fixed expression update workflow to pass design_vars through simulation_runner * Restored working NX journal expression update mechanism * OP2 timestamp verification instead of file deletion * Resolved issue where all trials returned identical objective values - Fixed LLMOptimizationRunner to pass design variables to simulation runner - Enhanced NXSolver with timestamp-based file regeneration verification New Components: - optimization_engine/llm_optimization_runner.py - LLM-driven optimization runner - optimization_engine/optimization_setup_wizard.py - Phase 3.3 setup wizard - studies/bracket_displacement_maximizing/ - Complete substudy example * run_substudy.py - Substudy runner with continuation * run_optimization.py - Standalone optimization runner * config/substudy_template.json - Template for new substudies * substudies/coarse_exploration/ - 20-trial coarse search * substudies/fine_tuning/ - 50-trial refinement (continuation example) * SUBSTUDIES_README.md - Complete substudy documentation Technical Improvements: - Incremental history saving after each trial (optimization_history_incremental.json) - Expression update workflow: .prt update → NX journal receives values → geometry update → FEM update → solve - Trial indexing fix in substudy result saving - Updated README with substudy system documentation Testing: - Successfully ran 20-trial coarse_exploration substudy - Verified different objective values across trials (workflow fix validated) - Confirmed live history updates in real-time - Tested shared model file usage across substudies 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-16 21:29:54 -05:00
expressions_list = self.find_expressions()
return {
expr['name']: {
'value': expr['value'],
'units': expr['units'],
'type': expr['type'],
'formula': None # Binary .prt files don't have formulas accessible
}
for expr in expressions_list
}
def update_expression(self, name: str, new_value: float) -> bool:
"""
Update a single expression value.
Args:
name: Expression name
new_value: New value
Returns:
True if updated, False if not found
"""
# Find the expression pattern
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
# Match: (Type [units]) name: old_value; OR (Type) name: old_value; (unitless)
# We need to be careful to match the exact name and preserve formatting
# Pattern that captures the full expression line
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
# Units are optional (unitless expressions like hole_count don't have [units])
pattern = rf'([#*\(]*\(\w+(?:\s*\[[^\]]*\])?\)\s*)({re.escape(name)})\s*:\s*([-+]?\d*\.?\d+(?:[eE][-+]?\d+)?)'
matches = list(re.finditer(pattern, self.text_content))
if not matches:
print(f"Warning: Expression '{name}' not found in .prt file")
return False
if len(matches) > 1:
print(f"Warning: Multiple matches for '{name}', updating first occurrence")
# Get the first match
match = matches[0]
prefix, expr_name, old_value = match.groups()
# Format new value (preserve decimal places if possible)
# Check if old value had decimal point
if '.' in old_value or 'e' in old_value.lower():
# Use same precision as old value
decimal_places = len(old_value.split('.')[-1]) if '.' in old_value else 2
new_value_str = f"{new_value:.{decimal_places}f}"
else:
# Integer format
new_value_str = f"{int(new_value)}"
# Build replacement string
replacement = f"{prefix}{expr_name}: {new_value_str}"
# Replace in text content
old_match = match.group(0)
self.text_content = self.text_content.replace(old_match, replacement, 1)
# Also update in binary content
old_bytes = old_match.encode('latin-1')
new_bytes = replacement.encode('latin-1')
# Find and replace in binary content
start_pos = self.content.find(old_bytes)
if start_pos != -1:
# Replace bytes
self.content[start_pos:start_pos+len(old_bytes)] = new_bytes
print(f"Updated: {name} = {old_value} -> {new_value_str}")
return True
else:
print(f"Warning: Could not update binary content for '{name}'")
return False
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
def update_expressions(self, updates: Dict[str, float], use_nx_import: bool = True):
"""
Update multiple expressions at once.
Args:
updates: Dict mapping expression name to new value
{'tip_thickness': 22.5, 'support_angle': 35.0}
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
use_nx_import: If True, uses NX journal to import .exp file (RECOMMENDED for all expressions)
If False, uses binary .prt editing (may miss some expressions)
"""
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
if use_nx_import:
# Use NX journal to import expressions
return self.update_expressions_via_import(updates)
# Fallback: Binary .prt editing
print(f"\nUpdating {len(updates)} expressions in {self.prt_path.name}:")
updated_count = 0
for name, value in updates.items():
if self.update_expression(name, value):
updated_count += 1
print(f"Successfully updated {updated_count}/{len(updates)} expressions")
feat: Add robust NX expression import system for all expression types Major Enhancement: - Implemented .exp file-based expression updates via NX journal scripts - Fixes critical issue with feature-linked expressions (e.g., hole_count) - Supports ALL NX expression types including binary-stored ones - Full 4D design space validation completed successfully New Components: 1. import_expressions.py - NX journal for .exp file import - Uses NXOpen.ExpressionCollection.ImportFromFile() - Replace mode overwrites existing values - Automatic model update and save - Comprehensive error handling 2. export_expressions.py - NX journal for .exp file export - Exports all expressions to text format - Used for unit detection and verification 3. Enhanced nx_updater.py - New update_expressions_via_import() method - Automatic unit detection from .exp export - Creates study-variable-only .exp files - Replaces fragile binary .prt editing Technical Details: - .exp Format: [Units]name=value (e.g., [MilliMeter]beam_length=5000) - Unitless expressions: name=value (e.g., hole_count=10) - Robustness: Native NX functionality, no regex failures - Performance: < 1 second per update operation Validation: - Simple Beam Optimization study (4D design space) * beam_half_core_thickness: 10-40 mm * beam_face_thickness: 10-40 mm * holes_diameter: 150-450 mm * hole_count: 5-15 (integer) Results: ✅ 3-trial validation completed successfully ✅ All 4 variables update correctly in all trials ✅ Mesh adaptation verified (hole_count: 6, 15, 11 → different mesh sizes) ✅ Trial 0: 5373 CQUAD4 elements (6 holes) ✅ Trial 1: 5158 CQUAD4 + 1 CTRIA3 (15 holes) ✅ Trial 2: 5318 CQUAD4 (11 holes) Problem Solved: - hole_count expression was not updating with binary .prt editing - Expression stored in feature parameter, not accessible via text regex - Binary format prevented reliable text-based updates Solution: - Use NX native expression import/export - Works for ALL expressions (text and binary-stored) - Automatic unit handling - Model update integrated in journal Documentation: - New: docs/NX_EXPRESSION_IMPORT_SYSTEM.md (comprehensive guide) - Updated: CHANGELOG.md with Phase 3.2 progress - Study: studies/simple_beam_optimization/ (complete example) Files Added: - optimization_engine/import_expressions.py - optimization_engine/export_expressions.py - docs/NX_EXPRESSION_IMPORT_SYSTEM.md - studies/simple_beam_optimization/ (full study) Files Modified: - optimization_engine/nx_updater.py - CHANGELOG.md Compatibility: - NX 2412 tested and verified - Python 3.10+ - Works with all NX expression types 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
2025-11-17 12:34:06 -05:00
def update_expressions_via_import(self, updates: Dict[str, float]):
"""
Update expressions by creating a .exp file and importing it via NX journal.
This method works for ALL expressions including those not stored in text format
in the binary .prt file (like hole_count).
Args:
updates: Dict mapping expression name to new value
"""
print(f"\nUpdating {len(updates)} expressions via NX .exp import:")
# Get all expressions to determine units
all_expressions = self.get_all_expressions(use_exp_export=True)
# Create .exp file with ONLY the study variables
exp_file = self.prt_path.parent / f"{self.prt_path.stem}_study_variables.exp"
with open(exp_file, 'w', encoding='utf-8') as f:
for name, value in updates.items():
if name in all_expressions:
units = all_expressions[name].get('units', '')
if units:
# Expression with units: [MilliMeter]beam_length=5000
f.write(f"[{units}]{name}={value}\n")
else:
# Unitless expression: hole_count=10
f.write(f"{name}={value}\n")
print(f" {name}: {value} {units if units else ''}")
else:
print(f" Warning: {name} not found in part expressions, skipping")
print(f"\n[EXP] Created: {exp_file}")
# Run NX journal to import expressions
journal_script = Path(__file__).parent / "import_expressions.py"
if not journal_script.exists():
raise FileNotFoundError(f"Import journal script not found: {journal_script}")
if not self.nx_run_journal_path.exists():
raise FileNotFoundError(f"NX run_journal.exe not found: {self.nx_run_journal_path}")
print(f"[NX] Importing expressions into {self.prt_path.name}...")
# Build command
cmd_str = f'"{self.nx_run_journal_path}" "{journal_script}" -args "{self.prt_path}" "{exp_file}"'
result = subprocess.run(cmd_str, capture_output=True, text=True, shell=True)
# Clean up .exp file
exp_file.unlink()
# Check if import succeeded
if result.returncode != 0 and "successfully" not in result.stdout.lower():
print(f"[ERROR] NX journal failed:")
print(result.stdout)
print(result.stderr)
raise RuntimeError(f"Expression import failed")
print(f"[OK] All {len(updates)} expressions updated successfully!")
def save(self, output_path: Path = None):
"""
Save modified .prt file.
Args:
output_path: Optional different path to save to.
If None, overwrites original (with backup if enabled)
"""
if output_path is None:
output_path = self.prt_path
if self.backup_enabled:
self._create_backup()
# Write updated binary content
with open(output_path, 'wb') as f:
f.write(self.content)
print(f"Saved to: {output_path}")
def verify_update(self, name: str, expected_value: float, tolerance: float = 1e-6) -> bool:
"""
Verify that an expression was updated correctly.
Args:
name: Expression name
expected_value: Expected value
tolerance: Acceptable difference
Returns:
True if value matches (within tolerance)
"""
expressions = self.find_expressions()
expr = next((e for e in expressions if e['name'] == name), None)
if expr is None:
print(f"Expression '{name}' not found")
return False
actual_value = expr['value']
difference = abs(actual_value - expected_value)
if difference <= tolerance:
print(f"OK Verified: {name} = {actual_value} (expected {expected_value})")
return True
else:
print(f"FAIL Verification failed: {name} = {actual_value}, expected {expected_value} (diff: {difference})")
return False
# Convenience function for optimization loop
def update_nx_model(prt_file_path: Path, design_variables: Dict[str, float], backup: bool = False):
"""
Convenience function to update NX model parameters.
Args:
prt_file_path: Path to .prt file
design_variables: Dict of parameter name -> value
backup: Whether to create backup
Example:
>>> update_nx_model(
... Path("Bracket.prt"),
... {'tip_thickness': 22.5, 'support_angle': 35.0}
... )
"""
updater = NXParameterUpdater(prt_file_path, backup=backup)
updater.update_expressions(design_variables)
updater.save()
# Example usage
if __name__ == "__main__":
import sys
if len(sys.argv) < 2:
print("Usage: python nx_updater.py <path_to_prt_file>")
sys.exit(1)
prt_path = Path(sys.argv[1])
# Test: Find and print all expressions
print("="*60)
print("NX PARAMETER UPDATER TEST")
print("="*60)
updater = NXParameterUpdater(prt_path, backup=True)
print("\nCurrent expressions in file:")
expressions = updater.find_expressions()
for expr in expressions:
print(f" {expr['name']}: {expr['value']} {expr['units']}")
# Test update (if expressions found)
if expressions:
print("\n" + "="*60)
print("TEST UPDATE")
print("="*60)
# Update first expression
first_expr = expressions[0]
test_name = first_expr['name']
test_new_value = first_expr['value'] * 1.1 # Increase by 10%
print(f"\nUpdating {test_name} from {first_expr['value']} to {test_new_value}")
updater.update_expression(test_name, test_new_value)
# Save to test file
test_output = prt_path.with_suffix('.prt.test')
updater.save(test_output)
# Verify by re-reading
print("\n" + "="*60)
print("VERIFICATION")
print("="*60)
verifier = NXParameterUpdater(test_output, backup=False)
verifier.verify_update(test_name, test_new_value)
print(f"\nTest complete. Modified file: {test_output}")
else:
print("\nNo expressions found in file. Nothing to test.")