File size: 19,373 Bytes
46df5f0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
39b06be
46df5f0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
"""
Metadata comparison between bib entries and fetched metadata.
"""
from dataclasses import dataclass
from typing import Optional

from ..parsers.bib_parser import BibEntry
from ..fetchers.arxiv_fetcher import ArxivMetadata
from ..fetchers.scholar_fetcher import ScholarResult
from ..fetchers.crossref_fetcher import CrossRefResult
from ..fetchers.semantic_scholar_fetcher import SemanticScholarResult
from ..fetchers.openalex_fetcher import OpenAlexResult
from ..fetchers.dblp_fetcher import DBLPResult
from ..utils.normalizer import TextNormalizer


@dataclass
class ComparisonResult:
    """Result of comparing bib entry with fetched metadata."""
    entry_key: str
    
    # Title comparison
    title_match: bool
    title_similarity: float
    bib_title: str
    fetched_title: str
    
    # Author comparison
    author_match: bool
    author_similarity: float
    bib_authors: list[str]
    fetched_authors: list[str]
    
    # Year comparison
    year_match: bool
    bib_year: str
    fetched_year: str
    
    # Overall assessment
    is_match: bool
    confidence: float
    issues: list[str]
    source: str  # 'arxiv', 'crossref', 'scholar', 'semantic_scholar', 'openalex', 'dblp', or 'unable'
    
    @property
    def has_issues(self) -> bool:
        return len(self.issues) > 0


class MetadataComparator:
    """Compares bibliography entries with fetched metadata."""
    
    # Thresholds for matching
    TITLE_THRESHOLD = 0.8
    AUTHOR_THRESHOLD = 0.6
    
    def __init__(self):
        self.normalizer = TextNormalizer
    
    def compare_with_arxiv(self, bib_entry: BibEntry, arxiv_meta: ArxivMetadata) -> ComparisonResult:
        """Compare bib entry with arXiv metadata."""
        issues = []
        
        # Compare titles
        bib_title_norm = self.normalizer.normalize_for_comparison(bib_entry.title)
        arxiv_title_norm = self.normalizer.normalize_for_comparison(arxiv_meta.title)
        
        title_similarity = self.normalizer.similarity_ratio(bib_title_norm, arxiv_title_norm)
        # Also try Levenshtein for short titles
        if len(bib_title_norm) < 100:
            lev_sim = self.normalizer.levenshtein_similarity(bib_title_norm, arxiv_title_norm)
            title_similarity = max(title_similarity, lev_sim)
        
        title_match = title_similarity >= self.TITLE_THRESHOLD
        
        if not title_match:
            issues.append(f"Title mismatch. Bib: '{bib_entry.title}', Retrieved: '{arxiv_meta.title}'")
        
        # Compare authors
        bib_authors = self.normalizer.normalize_author_list(bib_entry.author)
        arxiv_authors = [self.normalizer.normalize_author_name(a) for a in arxiv_meta.authors]
        
        author_similarity = self._compare_author_lists(bib_authors, arxiv_authors)
        author_match = author_similarity >= self.AUTHOR_THRESHOLD
        
        if not author_match:
            issues.append(f"Author mismatch. Bib: {', '.join(bib_authors)}, Retrieved: {', '.join(arxiv_authors)}")
        
        # Compare years
        bib_year = bib_entry.year.strip()
        arxiv_year = arxiv_meta.year
        year_match = bib_year == arxiv_year
        
        if not year_match and bib_year and arxiv_year:
            issues.append(f"Year mismatch. Bib: {bib_year}, Retrieved: {arxiv_year}")
        
        # Overall assessment
        is_match = title_match and author_match
        confidence = (title_similarity * 0.5 + author_similarity * 0.3 + (1.0 if year_match else 0.5) * 0.2)
        
        return ComparisonResult(
            entry_key=bib_entry.key,
            title_match=title_match,
            title_similarity=title_similarity,
            bib_title=bib_entry.title,
            fetched_title=arxiv_meta.title,
            author_match=author_match,
            author_similarity=author_similarity,
            bib_authors=bib_authors,
            fetched_authors=arxiv_authors,
            year_match=year_match,
            bib_year=bib_year,
            fetched_year=arxiv_year,
            is_match=is_match,
            confidence=confidence,
            issues=issues,
            source="arxiv"
        )
    
    def compare_with_scholar(self, bib_entry: BibEntry, scholar_result: ScholarResult) -> ComparisonResult:
        """Compare bib entry with Scholar search result."""
        issues = []
        
        # Compare titles
        bib_title_norm = self.normalizer.normalize_for_comparison(bib_entry.title)
        scholar_title_norm = self.normalizer.normalize_for_comparison(scholar_result.title)
        
        title_similarity = self.normalizer.similarity_ratio(bib_title_norm, scholar_title_norm)
        if len(bib_title_norm) < 100:
            lev_sim = self.normalizer.levenshtein_similarity(bib_title_norm, scholar_title_norm)
            title_similarity = max(title_similarity, lev_sim)
        
        title_match = title_similarity >= self.TITLE_THRESHOLD
        
        if not title_match:
            issues.append(f"Title mismatch. Bib: '{bib_entry.title}', Retrieved: '{scholar_result.title}'")
        
        # Compare authors (Scholar format is less structured)
        bib_authors = self.normalizer.normalize_author_list(bib_entry.author)
        # Scholar authors are comma-separated
        scholar_authors_raw = scholar_result.authors.split(',')
        scholar_authors = [self.normalizer.normalize_author_name(a.strip()) for a in scholar_authors_raw]
        
        author_similarity = self._compare_author_lists(bib_authors, scholar_authors)
        author_match = author_similarity >= self.AUTHOR_THRESHOLD
        
        if not author_match:
            issues.append(f"Author mismatch. Bib: {', '.join(bib_authors)}, Retrieved: {', '.join(scholar_authors)}")
        
        # Compare years
        bib_year = bib_entry.year.strip()
        scholar_year = scholar_result.year
        year_match = bib_year == scholar_year
        
        if not year_match and bib_year and scholar_year:
            issues.append(f"Year mismatch. Bib: {bib_year}, Retrieved: {scholar_year}")
        
        # Overall assessment
        is_match = title_match and author_match
        confidence = (title_similarity * 0.5 + author_similarity * 0.3 + (1.0 if year_match else 0.5) * 0.2)
        
        return ComparisonResult(
            entry_key=bib_entry.key,
            title_match=title_match,
            title_similarity=title_similarity,
            bib_title=bib_entry.title,
            fetched_title=scholar_result.title,
            author_match=author_match,
            author_similarity=author_similarity,
            bib_authors=bib_authors,
            fetched_authors=scholar_authors,
            year_match=year_match,
            bib_year=bib_year,
            fetched_year=scholar_year,
            is_match=is_match,
            confidence=confidence,
            issues=issues,
            source="scholar"
        )
    
    def compare_with_crossref(self, bib_entry: BibEntry, crossref_result: CrossRefResult) -> ComparisonResult:
        """Compare bib entry with CrossRef search result."""
        issues = []
        
        # Compare titles
        bib_title_norm = self.normalizer.normalize_for_comparison(bib_entry.title)
        crossref_title_norm = self.normalizer.normalize_for_comparison(crossref_result.title)
        
        title_similarity = self.normalizer.similarity_ratio(bib_title_norm, crossref_title_norm)
        if len(bib_title_norm) < 100:
            lev_sim = self.normalizer.levenshtein_similarity(bib_title_norm, crossref_title_norm)
            title_similarity = max(title_similarity, lev_sim)
        
        title_match = title_similarity >= self.TITLE_THRESHOLD
        
        if not title_match:
            issues.append(f"Title mismatch. Bib: '{bib_entry.title}', Retrieved: '{crossref_result.title}'")
        
        # Compare authors
        bib_authors = self.normalizer.normalize_author_list(bib_entry.author)
        crossref_authors = [self.normalizer.normalize_author_name(a) for a in crossref_result.authors]
        
        author_similarity = self._compare_author_lists(bib_authors, crossref_authors)
        author_match = author_similarity >= self.AUTHOR_THRESHOLD
        
        if not author_match:
            issues.append(f"Author mismatch. Bib: {', '.join(bib_authors)}, Retrieved: {', '.join(crossref_authors)}")
        
        # Compare years
        bib_year = bib_entry.year.strip()
        crossref_year = crossref_result.year
        year_match = bib_year == crossref_year
        
        if not year_match and bib_year and crossref_year:
            issues.append(f"Year mismatch. Bib: {bib_year}, Retrieved: {crossref_year}")
        
        # Overall assessment
        is_match = title_match and author_match
        confidence = (title_similarity * 0.5 + author_similarity * 0.3 + (1.0 if year_match else 0.5) * 0.2)
        
        return ComparisonResult(
            entry_key=bib_entry.key,
            title_match=title_match,
            title_similarity=title_similarity,
            bib_title=bib_entry.title,
            fetched_title=crossref_result.title,
            author_match=author_match,
            author_similarity=author_similarity,
            bib_authors=bib_authors,
            fetched_authors=crossref_authors,
            year_match=year_match,
            bib_year=bib_year,
            fetched_year=crossref_year,
            is_match=is_match,
            confidence=confidence,
            issues=issues,
            source="crossref"
        )
    
    def create_unable_result(self, bib_entry: BibEntry, reason: str = "Unable to fetch metadata") -> ComparisonResult:
        """Create result when metadata couldn't be fetched."""
        return ComparisonResult(
            entry_key=bib_entry.key,
            title_match=False,
            title_similarity=0.0,
            bib_title=bib_entry.title,
            fetched_title="",
            author_match=False,
            author_similarity=0.0,
            bib_authors=self.normalizer.normalize_author_list(bib_entry.author),
            fetched_authors=[],
            year_match=False,
            bib_year=bib_entry.year,
            fetched_year="",
            is_match=False,
            confidence=0.0,
            issues=[reason],
            source="unable"
        )
    
    def _compare_author_lists(self, list1: list[str], list2: list[str]) -> float:
        """Compare two author lists."""
        if not list1 and not list2:
            return 1.0
        if not list1 or not list2:
            return 0.0
        
        # Find best matches for each author in list1
        total_similarity = 0.0
        for author1 in list1:
            best_match = 0.0
            for author2 in list2:
                # Check if one name contains the other (handle abbreviated names)
                if self._names_match(author1, author2):
                    best_match = 1.0
                    break
                sim = self.normalizer.similarity_ratio(author1, author2)
                best_match = max(best_match, sim)
            total_similarity += best_match
        
        return total_similarity / len(list1)
    
    def _names_match(self, name1: str, name2: str) -> bool:
        """Check if two names match (handles abbreviated names)."""
        words1 = name1.split()
        words2 = name2.split()
        
        if not words1 or not words2:
            return False
        
        # Check if last names match
        if words1[-1] != words2[-1]:
            # Try first word as last name too
            if words1[0] != words2[-1] and words1[-1] != words2[0]:
                return False
        
        return True
    
    def compare_with_semantic_scholar(self, bib_entry: BibEntry, ss_result: SemanticScholarResult) -> ComparisonResult:
        """Compare bib entry with Semantic Scholar result."""
        issues = []
        
        # Compare titles
        bib_title_norm = self.normalizer.normalize_for_comparison(bib_entry.title)
        ss_title_norm = self.normalizer.normalize_for_comparison(ss_result.title)
        
        title_similarity = self.normalizer.similarity_ratio(bib_title_norm, ss_title_norm)
        if len(bib_title_norm) < 100:
            lev_sim = self.normalizer.levenshtein_similarity(bib_title_norm, ss_title_norm)
            title_similarity = max(title_similarity, lev_sim)
        
        title_match = title_similarity >= self.TITLE_THRESHOLD
        
        if not title_match:
            issues.append(f"Title mismatch. Bib: '{bib_entry.title}', Retrieved: '{ss_result.title}'")
        
        # Compare authors
        bib_authors = self.normalizer.normalize_author_list(bib_entry.author)
        ss_authors = [self.normalizer.normalize_author_name(a) for a in ss_result.authors]
        
        author_similarity = self._compare_author_lists(bib_authors, ss_authors)
        author_match = author_similarity >= self.AUTHOR_THRESHOLD
        
        if not author_match:
            issues.append(f"Author mismatch. Bib: {', '.join(bib_authors)}, Retrieved: {', '.join(ss_authors)}")
        
        # Compare years
        bib_year = bib_entry.year.strip()
        ss_year = ss_result.year
        year_match = bib_year == ss_year
        
        if not year_match and bib_year and ss_year:
            issues.append(f"Year mismatch. Bib: {bib_year}, Retrieved: {ss_year}")
        
        # Overall assessment
        is_match = title_match and author_match
        confidence = (title_similarity * 0.5 + author_similarity * 0.3 + (1.0 if year_match else 0.5) * 0.2)
        
        return ComparisonResult(
            entry_key=bib_entry.key,
            title_match=title_match,
            title_similarity=title_similarity,
            bib_title=bib_entry.title,
            fetched_title=ss_result.title,
            author_match=author_match,
            author_similarity=author_similarity,
            bib_authors=bib_authors,
            fetched_authors=ss_authors,
            year_match=year_match,
            bib_year=bib_year,
            fetched_year=ss_year,
            is_match=is_match,
            confidence=confidence,
            issues=issues,
            source="semantic_scholar"
        )
    
    def compare_with_openalex(self, bib_entry: BibEntry, oa_result: OpenAlexResult) -> ComparisonResult:
        """Compare bib entry with OpenAlex result."""
        issues = []
        
        # Compare titles
        bib_title_norm = self.normalizer.normalize_for_comparison(bib_entry.title)
        oa_title_norm = self.normalizer.normalize_for_comparison(oa_result.title)
        
        title_similarity = self.normalizer.similarity_ratio(bib_title_norm, oa_title_norm)
        if len(bib_title_norm) < 100:
            lev_sim = self.normalizer.levenshtein_similarity(bib_title_norm, oa_title_norm)
            title_similarity = max(title_similarity, lev_sim)
        
        title_match = title_similarity >= self.TITLE_THRESHOLD
        
        if not title_match:
            issues.append(f"Title mismatch. Bib: '{bib_entry.title}', Retrieved: '{oa_result.title}'")
        
        # Compare authors
        bib_authors = self.normalizer.normalize_author_list(bib_entry.author)
        oa_authors = [self.normalizer.normalize_author_name(a) for a in oa_result.authors]
        
        author_similarity = self._compare_author_lists(bib_authors, oa_authors)
        author_match = author_similarity >= self.AUTHOR_THRESHOLD
        
        if not author_match:
            issues.append(f"Author mismatch. Bib: {', '.join(bib_authors)}, Retrieved: {', '.join(oa_authors)}")
        
        # Compare years
        bib_year = bib_entry.year.strip()
        oa_year = oa_result.year
        year_match = bib_year == oa_year
        
        if not year_match and bib_year and oa_year:
            issues.append(f"Year mismatch. Bib: {bib_year}, Retrieved: {oa_year}")
        
        # Overall assessment
        is_match = title_match and author_match
        confidence = (title_similarity * 0.5 + author_similarity * 0.3 + (1.0 if year_match else 0.5) * 0.2)
        
        return ComparisonResult(
            entry_key=bib_entry.key,
            title_match=title_match,
            title_similarity=title_similarity,
            bib_title=bib_entry.title,
            fetched_title=oa_result.title,
            author_match=author_match,
            author_similarity=author_similarity,
            bib_authors=bib_authors,
            fetched_authors=oa_authors,
            year_match=year_match,
            bib_year=bib_year,
            fetched_year=oa_year,
            is_match=is_match,
            confidence=confidence,
            issues=issues,
            source="openalex"
        )
    
    def compare_with_dblp(self, bib_entry: BibEntry, dblp_result: DBLPResult) -> ComparisonResult:
        """Compare bib entry with DBLP result."""
        issues = []
        
        # Compare titles
        bib_title_norm = self.normalizer.normalize_for_comparison(bib_entry.title)
        dblp_title_norm = self.normalizer.normalize_for_comparison(dblp_result.title)
        
        title_similarity = self.normalizer.similarity_ratio(bib_title_norm, dblp_title_norm)
        if len(bib_title_norm) < 100:
            lev_sim = self.normalizer.levenshtein_similarity(bib_title_norm, dblp_title_norm)
            title_similarity = max(title_similarity, lev_sim)
        
        title_match = title_similarity >= self.TITLE_THRESHOLD
        
        if not title_match:
            issues.append(f"Title mismatch. Bib: '{bib_entry.title}', Retrieved: '{dblp_result.title}'")
        
        # Compare authors
        bib_authors = self.normalizer.normalize_author_list(bib_entry.author)
        dblp_authors = [self.normalizer.normalize_author_name(a) for a in dblp_result.authors]
        
        author_similarity = self._compare_author_lists(bib_authors, dblp_authors)
        author_match = author_similarity >= self.AUTHOR_THRESHOLD
        
        if not author_match:
            issues.append(f"Author mismatch. Bib: {', '.join(bib_authors)}, Retrieved: {', '.join(dblp_authors)}")
        
        # Compare years
        bib_year = bib_entry.year.strip()
        dblp_year = dblp_result.year
        year_match = bib_year == dblp_year
        
        if not year_match and bib_year and dblp_year:
            issues.append(f"Year mismatch. Bib: {bib_year}, Retrieved: {dblp_year}")
        
        # Overall assessment
        is_match = title_match and author_match
        confidence = (title_similarity * 0.5 + author_similarity * 0.3 + (1.0 if year_match else 0.5) * 0.2)
        
        return ComparisonResult(
            entry_key=bib_entry.key,
            title_match=title_match,
            title_similarity=title_similarity,
            bib_title=bib_entry.title,
            fetched_title=dblp_result.title,
            author_match=author_match,
            author_similarity=author_similarity,
            bib_authors=bib_authors,
            fetched_authors=dblp_authors,
            year_match=year_match,
            bib_year=bib_year,
            fetched_year=dblp_year,
            is_match=is_match,
            confidence=confidence,
            issues=issues,
            source="dblp"
        )