-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathsl5_overlay_extractor.py
More file actions
executable file
·429 lines (349 loc) · 19.6 KB
/
Copy pathsl5_overlay_extractor.py
File metadata and controls
executable file
·429 lines (349 loc) · 19.6 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
#!/usr/bin/env python3
"""
SL5 Overlay Control Extractor
This script extracts control information from the sl5_overlay PDF.
Key differences from insider threat overlay:
- Attributes are plain text (not underlined)
- Controls are "selected" if they have "Justification to Select" rather than just "Open Questions"
- Different attribute patterns
- Enhanced to properly extract Open Questions for control enhancements
- Updated to exclude controls with empty attributes from final JSON output
"""
import fitz # PyMuPDF
import json
import re
import sys
from typing import Dict, List
class SL5ControlExtractor:
def __init__(self, pdf_path: str):
self.pdf_path = pdf_path
self.controls = {}
self.current_control = None
self.current_attribute = None
self.debug_mode = False
# Known attributes for SL5 overlay (case-sensitive)
self.known_attributes = [
"Justification to Select",
"Open Questions",
"Inherit From",
"Parameter Value(s)",
"Test Method",
"Control Enhancements",
"Supplemental Guidance",
"Regulatory/Statutory Reference(s)"
]
def extract_controls(self) -> Dict:
"""Extract all controls from the PDF document."""
try:
doc = fitz.open(self.pdf_path)
print(f"Processing {len(doc)} pages...")
for page_num in range(len(doc)):
page = doc[page_num]
self._process_page(page, page_num + 1)
doc.close()
# Post-process to determine selected status
self._determine_selected_status()
return self.controls
except Exception as e:
print(f"Error processing PDF: {e}")
return {}
def _process_page(self, page, page_num: int):
"""Process a single page."""
try:
text_dict = page.get_text("dict")
formatted_text = self._extract_formatted_text(text_dict)
self._find_controls_and_attributes(formatted_text, page_num)
except Exception as e:
if self.debug_mode:
print(f"Error processing page {page_num}: {e}")
def _extract_formatted_text(self, text_dict: dict) -> List[dict]:
"""Extract text with formatting information."""
formatted_text = []
for block in text_dict.get("blocks", []):
if block.get("type") == 0: # Text block
for line in block.get("lines", []):
line_text = ""
line_formats = []
for span in line.get("spans", []):
text = span.get("text", "")
flags = span.get("flags", 0)
font = span.get("font", "")
# Extract formatting information
bold = bool(flags & 16)
italic = bool(flags & 2)
underline = bool(flags & 4)
line_text += text
if text.strip(): # Don't add format info for whitespace
line_formats.append({
"text": text,
"bold": bold,
"italic": italic,
"underline": underline,
"font": font
})
if line_text.strip():
formatted_text.append({
"text": line_text,
"formats": line_formats
})
return formatted_text
def _find_controls_and_attributes(self, formatted_text: List[dict], page_num: int):
"""Find controls and their attributes in the formatted text."""
processing_enhancements = False
parent_control = None
for text_info in formatted_text:
line_text = text_info["text"]
formats = text_info["formats"]
# Look for control IDs (typically bold)
is_bold = any(fmt.get("bold", False) for fmt in formats)
if is_bold:
control_match = re.match(r'^([A-Z]{2}-\d{1,2}(?:\(\d+\))?)', line_text)
if control_match:
control_id = control_match.group(1)
# Extract control name (everything after the ID and comma/space)
remaining_text = line_text[len(control_id):].strip()
if remaining_text.startswith(','):
remaining_text = remaining_text[1:].strip()
control_name = remaining_text
# Initialize new control
self.controls[control_id] = {
"name": control_name,
"attributes": {},
"selected": False # Will be determined later
}
self.current_control = control_id
self.current_attribute = None
processing_enhancements = False
parent_control = None
print(f"Found control: {control_id} - {control_name[:50]}...")
continue
# Check if we're entering a Control Enhancements section
if self.current_control and line_text.strip() == "Control Enhancements:":
# Don't create an empty attribute, just set the processing flag
processing_enhancements = True
parent_control = self.current_control # Remember the parent control
if self.debug_mode:
print(f" Entering Control Enhancements section for {self.current_control}")
continue
# Look for control enhancements (indicated by bullet points and enhancement format)
if processing_enhancements and (line_text.strip().startswith('●') or line_text.strip().startswith('•')):
# Extract enhancement pattern
enhancement_match = re.search(r'([A-Z]{2}-\d{1,2}\(\d+\))\s+(.+?)\s*\|\s*(.+)', line_text)
if enhancement_match:
enhancement_id = enhancement_match.group(1)
base_control_name = enhancement_match.group(2).strip()
enhancement_name = enhancement_match.group(3).strip()
# Create full enhancement name combining base control and enhancement
full_enhancement_name = f"{base_control_name} | {enhancement_name}"
# Initialize new enhancement as a separate control
self.controls[enhancement_id] = {
"name": full_enhancement_name,
"attributes": {},
"selected": False # Will be determined later
}
self.current_control = enhancement_id
self.current_attribute = None
print(f"Found enhancement: {enhancement_id} - {full_enhancement_name[:50]}...")
continue
# Look for attributes (not necessarily bold in SL5 overlay)
if self.current_control and ':' in line_text:
attribute_result = self._extract_attribute_from_line(line_text)
if attribute_result:
attribute_name, attribute_content = attribute_result
# Store the attribute content - NO LONGER redirect Open Questions from enhancements
self.controls[self.current_control]["attributes"][attribute_name] = attribute_content.strip()
self.current_attribute = attribute_name
if self.debug_mode:
print(f" Added attribute to {self.current_control}: {attribute_name}")
# Special handling: if we encounter Open Questions that are not part of an enhancement
# and we were processing enhancements, we may need to stop processing enhancements
# This handles cases where Open Questions appear after all enhancements
if (attribute_name == "Open Questions" and
processing_enhancements and
parent_control and
not '(' in self.current_control):
# This is an Open Questions for the parent control, not an enhancement
processing_enhancements = False
self.current_control = parent_control
parent_control = None
continue
# Continue building current attribute content if we have one
if (self.current_control and
self.current_attribute and
line_text.strip() and
not self._is_control_or_enhancement(line_text) and
not self._extract_attribute_from_line(line_text)):
# Append to current attribute
current_content = self.controls[self.current_control]["attributes"][self.current_attribute]
self.controls[self.current_control]["attributes"][self.current_attribute] = current_content + " " + line_text.strip()
def _extract_attribute_from_line(self, line_text: str):
"""Extract attribute name and content from a line if it matches known attributes."""
for attr in self.known_attributes:
if line_text.startswith(attr + ":"):
content = line_text[len(attr) + 1:].strip()
return attr, content
return None
def _is_control_or_enhancement(self, line_text: str) -> bool:
"""Check if line contains a control ID or enhancement pattern."""
# Check for control ID pattern
if re.match(r'^[A-Z]{2}-\d{1,2}(?:\(\d+\))?', line_text.strip()):
return True
# Check for enhancement pattern with bullet points
if (line_text.strip().startswith('●') or line_text.strip().startswith('•')):
enhancement_match = re.search(r'([A-Z]{2}-\d{1,2}\(\d+\))', line_text)
return enhancement_match is not None
return False
def _determine_selected_status(self):
"""Determine if each control is selected based on presence of 'Justification to Select'."""
for control_id, control_info in self.controls.items():
# A control is selected if it has "Justification to Select"
control_info["selected"] = "Justification to Select" in control_info["attributes"]
def save_to_json(self, filename: str):
"""Save extracted controls to JSON file, excluding controls with empty attributes."""
try:
# Filter out controls with empty attributes
filtered_controls = {
control_id: control_info
for control_id, control_info in self.controls.items()
if control_info["attributes"] # Only include if attributes dict is not empty
}
# Count how many controls were filtered out
excluded_count = len(self.controls) - len(filtered_controls)
with open(filename, 'w', encoding='utf-8') as f:
json.dump(filtered_controls, f, indent=2, ensure_ascii=False)
print(f"Controls saved to {filename}")
if excluded_count > 0:
print(f"Excluded {excluded_count} controls with empty attributes from JSON output")
except Exception as e:
print(f"Error saving to JSON: {e}")
def print_summary(self):
"""Print a summary of extracted controls."""
print(f"\n=== SL5 OVERLAY EXTRACTION SUMMARY ===")
print(f"Total controls extracted: {len(self.controls)}")
# Count controls with and without attributes
controls_with_attributes = sum(1 for control in self.controls.values() if control["attributes"])
controls_without_attributes = len(self.controls) - controls_with_attributes
print(f"Controls with attributes: {controls_with_attributes}")
print(f"Controls without attributes: {controls_without_attributes}")
if not self.controls:
print("No controls found!")
return
# Count selected vs not selected (only for controls with attributes)
controls_with_attrs = {k: v for k, v in self.controls.items() if v["attributes"]}
selected_count = sum(1 for control in controls_with_attrs.values() if control["selected"])
not_selected_count = len(controls_with_attrs) - selected_count
print(f"Selected controls (with Justification): {selected_count}")
print(f"Not selected controls (Open Questions only): {not_selected_count}")
# Count controls by family
families = {}
enhancements = 0
base_controls = 0
enhancements_with_open_questions = 0
for control_id in self.controls.keys():
family = control_id.split('-')[0]
families[family] = families.get(family, 0) + 1
if '(' in control_id:
enhancements += 1
# Check if this enhancement has Open Questions
if "Open Questions" in self.controls[control_id]["attributes"]:
enhancements_with_open_questions += 1
else:
base_controls += 1
print(f"Base controls: {base_controls}")
print(f"Control enhancements: {enhancements}")
print(f"Control enhancements with Open Questions: {enhancements_with_open_questions}")
print("\nControls by family:")
for family, count in sorted(families.items()):
print(f" {family}: {count} controls")
# Show sample of selected and not selected controls (only those with attributes)
print("\nSample selected controls:")
selected_controls = [(cid, cdata) for cid, cdata in self.controls.items()
if cdata["selected"] and cdata["attributes"]]
for i, (control_id, control_info) in enumerate(selected_controls[:5]):
control_type = "Enhancement" if '(' in control_id else "Base Control"
print(f" {control_id} ({control_type}): {control_info['name'][:50]}...")
print("\nSample not selected controls (Open Questions):")
not_selected_controls = [(cid, cdata) for cid, cdata in self.controls.items()
if not cdata["selected"] and cdata["attributes"]]
for i, (control_id, control_info) in enumerate(not_selected_controls[:5]):
control_type = "Enhancement" if '(' in control_id else "Base Control"
print(f" {control_id} ({control_type}): {control_info['name'][:50]}...")
# Show sample of enhancements with Open Questions
print("\nSample control enhancements with Open Questions:")
enhancement_open_questions = [(cid, cdata) for cid, cdata in self.controls.items()
if '(' in cid and "Open Questions" in cdata["attributes"]]
for i, (control_id, control_info) in enumerate(enhancement_open_questions[:5]):
open_questions = control_info["attributes"]["Open Questions"]
print(f" {control_id}: {open_questions[:60]}...")
if controls_without_attributes > 0:
print(f"\nNote: {controls_without_attributes} controls with empty attributes will be excluded from JSON output")
def debug_page(self, page_num: int):
"""Debug a specific page."""
try:
doc = fitz.open(self.pdf_path)
if page_num > len(doc):
print(f"Page {page_num} does not exist. Document has {len(doc)} pages.")
doc.close()
return
page = doc[page_num - 1] # Convert to 0-based index
text_dict = page.get_text("dict")
print(f"=== DEBUG PAGE {page_num} ===")
for block_idx, block in enumerate(text_dict.get("blocks", [])):
if block.get("type") == 0: # Text block
print(f"\nBlock {block_idx}:")
for line_idx, line in enumerate(block.get("lines", [])):
line_text = ""
for span in line.get("spans", []):
line_text += span.get("text", "")
if line_text.strip():
print(f" Line {line_idx}: '{line_text.strip()}'")
# Check for potential control patterns
if re.match(r'^[A-Z]{2}-\d{1,2}(?:\(\d+\))?', line_text.strip()):
print(f" *** POTENTIAL CONTROL ID ***")
# Check if potential enhancement
if (line_text.strip().startswith('●') or line_text.strip().startswith('•')):
enhancement_match = re.search(r'([A-Z]{2}-\d{1,2}\(\d+\))', line_text)
if enhancement_match:
print(f" *** POTENTIAL ENHANCEMENT: {enhancement_match.group(1)} ***")
for span in line.get("spans", []):
text = span.get("text", "")
flags = span.get("flags", 0)
font = span.get("font", "")
if text.strip():
bold = bool(flags & 16)
underline = bool(flags & 2)
print(f" Span: '{text}' | Font: {font} | Bold: {bold} | Underline: {underline}")
doc.close()
except Exception as e:
print(f"Error debugging page: {e}")
def main():
if len(sys.argv) < 2:
print("Usage: python sl5_overlay_extractor.py <pdf_file> [--debug-page N]")
print("Example: python sl5_overlay_extractor.py sl5_overlay.pdf")
print("Example: python sl5_overlay_extractor.py sl5_overlay.pdf --debug-page 10")
sys.exit(1)
pdf_file = sys.argv[1]
# Check for debug mode
if len(sys.argv) == 4 and sys.argv[2] == "--debug-page":
try:
debug_page_num = int(sys.argv[3])
extractor = SL5ControlExtractor(pdf_file)
extractor.debug_mode = True
extractor.debug_page(debug_page_num)
return
except ValueError:
print("Invalid page number for debug mode")
sys.exit(1)
output_file = "extracted_sl5_overlay.json"
print("SL5 Overlay Control Extractor")
print("=" * 40)
extractor = SL5ControlExtractor(pdf_file)
controls = extractor.extract_controls()
if controls:
extractor.save_to_json(output_file)
extractor.print_summary()
else:
print("No controls were extracted.")
print("Try using --debug-page N to see the formatting details.")
if __name__ == "__main__":
main()