Spaces:
Sleeping
Sleeping
File size: 8,430 Bytes
bda2b5b |
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 |
"""
Chapter-level structural analysis for Tibetan legal manuscripts.
Provides differential highlighting, change detection, and structural alignment.
"""
import difflib
import re
def detect_structural_changes(text1: str, text2: str,
min_change_length: int = 5,
context_window: int = 10) -> dict:
"""
Detect structural changes between two Tibetan text chapters.
Args:
text1: First text chapter
text2: Second text chapter
min_change_length: Minimum length of change to report
context_window: Number of characters to include as context
Returns:
Dictionary with detected changes: insertions, deletions, modifications
"""
# Clean texts for comparison
def clean_text(text):
# Remove extra whitespace and normalize
text = re.sub(r'\s+', ' ', text.strip())
return text
clean1 = clean_text(text1)
clean2 = clean_text(text2)
# Use difflib to detect changes
differ = difflib.Differ()
diff = list(differ.compare(clean1.split(), clean2.split()))
changes = {
'insertions': [],
'deletions': [],
'modifications': [],
'unchanged': []
}
# Track current position in both texts
pos1 = 0
pos2 = 0
for i, line in enumerate(diff):
if line.startswith(' '): # Unchanged
word = line[2:]
changes['unchanged'].append({
'word': word,
'position1': pos1,
'position2': pos2,
'length': len(word)
})
pos1 += len(word) + 1
pos2 += len(word) + 1
elif line.startswith('- '): # Deletion
word = line[2:]
if len(word) >= min_change_length:
changes['deletions'].append({
'word': word,
'position': pos1,
'length': len(word),
'context': get_context(clean1, pos1, context_window)
})
pos1 += len(word) + 1
elif line.startswith('+ '): # Insertion
word = line[2:]
if len(word) >= min_change_length:
changes['insertions'].append({
'word': word,
'position': pos2,
'length': len(word),
'context': get_context(clean2, pos2, context_window)
})
pos2 += len(word) + 1
# Detect modifications (adjacent deletions and insertions)
modifications = detect_modifications(changes['deletions'], changes['insertions'])
changes['modifications'] = modifications
return changes
def get_context(text: str, position: int, window: int) -> str:
"""Get context around a position in text."""
start = max(0, position - window)
end = min(len(text), position + window)
return text[start:end]
def detect_modifications(deletions: list[dict], insertions: list[dict]) -> list[dict]:
"""Detect modifications by pairing nearby deletions and insertions."""
modifications = []
for deletion in deletions[:]: # Copy to avoid modification during iteration
for insertion in insertions[:]:
# If deletion and insertion are close (within 5 positions)
if abs(deletion['position'] - insertion['position']) <= 5:
modifications.append({
'original': deletion['word'],
'replacement': insertion['word'],
'position': deletion['position'],
'deletion_context': deletion['context'],
'insertion_context': insertion['context']
})
# Remove from original lists to avoid duplicates
if deletion in deletions:
deletions.remove(deletion)
if insertion in insertions:
insertions.remove(insertion)
break
return modifications
def generate_structural_alignment(text1: str, text2: str) -> dict[str, list]:
"""
Generate structural alignment between two text chapters.
Returns:
Dictionary with alignment information including gaps and matches
"""
# Split into sentences or clauses for alignment
def split_into_segments(text):
# Split on Tibetan punctuation
segments = re.split(r'[།༎༏༐༑༔]', text)
return [seg.strip() for seg in segments if seg.strip()]
segments1 = split_into_segments(text1)
segments2 = split_into_segments(text2)
# Create alignment using sequence matcher
matcher = difflib.SequenceMatcher(None, segments1, segments2)
alignment = {
'matches': [],
'gaps': [],
'mismatches': [],
'segments1': segments1,
'segments2': segments2
}
for tag, i1, i2, j1, j2 in matcher.get_opcodes():
if tag == 'equal':
alignment['matches'].append({
'segments1': segments1[i1:i2],
'segments2': segments2[j1:j2],
'type': 'match'
})
elif tag == 'delete':
alignment['gaps'].append({
'segments': segments1[i1:i2],
'type': 'deletion',
'position': 'text1'
})
elif tag == 'insert':
alignment['gaps'].append({
'segments': segments2[j1:j2],
'type': 'insertion',
'position': 'text2'
})
elif tag == 'replace':
alignment['mismatches'].append({
'original': segments1[i1:i2],
'replacement': segments2[j1:j2],
'type': 'modification'
})
return alignment
def calculate_structural_similarity_score(text1: str, text2: str) -> dict[str, float]:
"""
Calculate various structural similarity scores between two texts.
Returns:
Dictionary with multiple similarity metrics
"""
changes = detect_structural_changes(text1, text2)
alignment = generate_structural_alignment(text1, text2)
# Calculate scores
total_changes = len(changes['insertions']) + len(changes['deletions']) + len(changes['modifications'])
# Structural similarity score (inverse of changes)
text_length = max(len(text1.split()), len(text2.split()))
structural_score = max(0, 1 - (total_changes / text_length)) if text_length > 0 else 0
# Alignment-based score
total_segments = len(alignment['segments1']) + len(alignment['segments2'])
matches = len(alignment['matches'])
alignment_score = matches / (total_segments / 2) if total_segments > 0 else 0
return {
'structural_similarity': structural_score,
'alignment_score': alignment_score,
'insertions': len(changes['insertions']),
'deletions': len(changes['deletions']),
'modifications': len(changes['modifications']),
'total_changes': total_changes
}
def generate_differential_report(text1: str, text2: str,
file1_name: str = "Text 1",
file2_name: str = "Text 2") -> dict[str, any]:
"""
Generate a comprehensive differential report for two text chapters.
Returns:
Complete report with changes, alignment, and recommendations
"""
changes = detect_structural_changes(text1, text2)
alignment = generate_structural_alignment(text1, text2)
scores = calculate_structural_similarity_score(text1, text2)
report = {
'file1': file1_name,
'file2': file2_name,
'changes': changes,
'alignment': alignment,
'scores': scores,
'summary': {
'significant_differences': len([c for c in changes['modifications'] if len(c['original']) > 10 or len(c['replacement']) > 10]),
'minor_variants': len([c for c in changes['modifications'] if len(c['original']) <= 5 and len(c['replacement']) <= 5]),
'structural_preservation': scores['alignment_score'] > 0.8,
'recommendation': 'Manuscripts are structurally similar' if scores['alignment_score'] > 0.7 else 'Significant structural differences detected'
}
}
return report
|