Update gdpr_filter.py
Browse files- gdpr_filter.py +13 -7
gdpr_filter.py
CHANGED
|
@@ -1,7 +1,5 @@
|
|
| 1 |
-
|
| 2 |
import re
|
| 3 |
import logging
|
| 4 |
-
from config import Config
|
| 5 |
|
| 6 |
logger = logging.getLogger(__name__)
|
| 7 |
|
|
@@ -31,6 +29,13 @@ class GDPRFilter:
|
|
| 31 |
r'(ol|in|stat|pril|zol|am|um|ase|mycin|axin|cillin)$',
|
| 32 |
]
|
| 33 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
@staticmethod
|
| 35 |
def has_medication_context(text: str) -> bool:
|
| 36 |
for pattern in GDPRFilter.MEDICATION_CONTEXT_KEYWORDS:
|
|
@@ -70,13 +75,14 @@ class GDPRFilter:
|
|
| 70 |
text = re.sub(r'(\[PERSONNR\])\s+\d{1,4}\b', r'\1', text)
|
| 71 |
|
| 72 |
has_med_context = GDPRFilter.has_medication_context(text)
|
| 73 |
-
exclusions = Config.GDPR_EXCLUSION_LIST
|
| 74 |
|
| 75 |
def replace_name(match):
|
| 76 |
word = match.group(0)
|
| 77 |
-
if
|
| 78 |
-
|
| 79 |
-
|
|
|
|
|
|
|
| 80 |
return word
|
| 81 |
return '[NAMN]'
|
| 82 |
|
|
@@ -113,4 +119,4 @@ if __name__ == "__main__":
|
|
| 113 |
print(f"\nINPUT: {text}")
|
| 114 |
print(f"OUTPUT: {result}")
|
| 115 |
|
| 116 |
-
print("\n" + "="*80)
|
|
|
|
|
|
|
| 1 |
import re
|
| 2 |
import logging
|
|
|
|
| 3 |
|
| 4 |
logger = logging.getLogger(__name__)
|
| 5 |
|
|
|
|
| 29 |
r'(ol|in|stat|pril|zol|am|um|ase|mycin|axin|cillin)$',
|
| 30 |
]
|
| 31 |
|
| 32 |
+
MEDICATION_EXCLUSIONS = [
|
| 33 |
+
'Metoprolol', 'Atorvastatin', 'Losartan', 'Omeprazol',
|
| 34 |
+
'Lisinopril', 'Simvastatin', 'Ramipril', 'Bisoprolol',
|
| 35 |
+
'Warfarin', 'Insulin', 'Aspirin', 'Paracetamol', 'Ibuprofen',
|
| 36 |
+
'Amoxicillin', 'Penicillin', 'Tetracyclin', 'Doxycyclin',
|
| 37 |
+
]
|
| 38 |
+
|
| 39 |
@staticmethod
|
| 40 |
def has_medication_context(text: str) -> bool:
|
| 41 |
for pattern in GDPRFilter.MEDICATION_CONTEXT_KEYWORDS:
|
|
|
|
| 75 |
text = re.sub(r'(\[PERSONNR\])\s+\d{1,4}\b', r'\1', text)
|
| 76 |
|
| 77 |
has_med_context = GDPRFilter.has_medication_context(text)
|
|
|
|
| 78 |
|
| 79 |
def replace_name(match):
|
| 80 |
word = match.group(0)
|
| 81 |
+
if GDPRFilter.is_likely_medication(word):
|
| 82 |
+
if has_med_context:
|
| 83 |
+
return word
|
| 84 |
+
return '[LÄKEMEDEL]'
|
| 85 |
+
if any(word.startswith(ex) for ex in GDPRFilter.MEDICATION_EXCLUSIONS):
|
| 86 |
return word
|
| 87 |
return '[NAMN]'
|
| 88 |
|
|
|
|
| 119 |
print(f"\nINPUT: {text}")
|
| 120 |
print(f"OUTPUT: {result}")
|
| 121 |
|
| 122 |
+
print("\n" + "="*80)
|