finetype-cli 0.6.39

CLI for FineType semantic type classification
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
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
# FineType v2 — Technology Domain Definitions
#
# Hierarchy: domain.category.type (locale is a field, not in the key)
# Full label at inference time: domain.category.type.LOCALE
#
# Each definition is a transformation contract:
#   - broad_type:     Target DuckDB type
#   - format_string:  DuckDB strptime format (null if not strptime-based)
#   - transform:      DuckDB SQL expression ({col} = column placeholder)
#   - transform_ext:  Enhanced transform requiring a DuckDB extension
#   - validation:     JSON Schema fragment for data quality checks
#   - tier:           Path from root to parent in the inference graph
#   - decompose:      Optional struct expansion for multi-field output
#
# Technology domain covers:
#   - network formats (IPv4, IPv6, URLs, etc.)
#   - cryptographic formats (UUIDs, hashes, tokens)
#   - code identifiers (ISBN, IMEI, EAN, etc.)
#   - development metadata (versions, licenses, programming languages)
#   - hardware specifications
# ─────────────────────────────────────────────────────────────────────

# ═════════════════════════════════════════════════════════════════════
# CATEGORY: internet
#
# Network addresses, URLs, and web-related formats. These are
# machine-readable identifiers for network resources and protocols.
# ═════════════════════════════════════════════════════════════════════

technology.internet.ip_v4:
  title: "IPv4 Address"
  description: >
    Standard 4-octet IPv4 address format (e.g., 192.168.1.1).
    Resolves to VARCHAR with INET validation. Can optionally
    parse to DuckDB INET type with the inet extension.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: "CAST({col} AS INET)"
  decompose: null
  validation:
    type: string
    pattern: "^(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)$"
  tier: [INET, internet]
  release_priority: 5
  aliases: [ipv4]
  samples:
    - "192.168.1.1"
    - "8.8.8.8"
    - "127.0.0.1"
    - "10.0.0.0"
  references: null
  notes: >
    v1 migration: Was internet.ip_v4. transform_ext uses DuckDB's inet extension
    if available. Without the extension, stored as VARCHAR but validated as IPv4.

technology.internet.ip_v4_with_port:
  title: "IPv4 Address with Port"
  description: >
    IPv4 address with port number separated by colon (e.g., 192.168.1.1:8080).
    Resolves to VARCHAR. Can decompose to separate IP and port fields.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    ip: "REGEXP_EXTRACT({col}, '^([^:]+):')"
    port: "CAST(REGEXP_EXTRACT({col}, ':([0-9]+)$') AS SMALLINT)"
  validation:
    type: string
    pattern: "^(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?):[0-9]{1,5}$"
  tier: [INET, internet]
  release_priority: 4
  aliases: [ipv4_port]
  samples:
    - "192.168.1.1:8080"
    - "10.0.0.1:3306"
    - "127.0.0.1:5432"
  references: null
  notes: >
    v1 migration: Was internet.ip_v4_with_port. Includes decompose to split
    address and port into separate columns during transformation.

technology.internet.ip_v6:
  title: "IPv6 Address"
  description: >
    Standard IPv6 address format with 8 16-bit hex groups separated by colons.
    May use :: notation for zero compression. Resolves to VARCHAR.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: "CAST({col} AS INET)"
  decompose: null
  validation:
    type: string
    pattern: "^([0-9a-fA-F]{1,4}:){7}[0-9a-fA-F]{1,4}$"
  tier: [INET, internet]
  release_priority: 4
  aliases: [ipv6]
  samples:
    - "2001:0db8:85a3:0000:0000:8a2e:0370:7334"
    - "2001:db8:85a3::8a2e:370:7334"
    - "::1"
    - "fe80::1"
  references: null
  notes: >
    v1 migration: Was internet.ip_v6. Complex regex due to :: compression notation.

technology.internet.mac_address:
  title: "MAC Address"
  description: >
    Media Access Control (MAC) address in standard colon-separated hex format
    (e.g., 00:1a:2b:3c:4d:5e). Also accepts hyphen-separated format.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose: null
  validation:
    type: string
    pattern: "^([0-9A-Fa-f]{2}[:-]){5}([0-9A-Fa-f]{2})$"
  tier: [INET, internet]
  release_priority: 4
  aliases: [mac]
  samples:
    - "00:1a:2b:3c:4d:5e"
    - "08-00-27-00-00-00"
    - "a0:b1:c2:d3:e4:f5"
  references: null
  notes: >
    v1 migration: Was internet.mac_address. Pattern accepts both colon and hyphen
    separators. Data typically uppercase, but pattern is case-insensitive.

technology.internet.url:
  title: "URL"
  description: >
    Uniform Resource Locator (complete web address including scheme).
    Includes protocol (http, https, ftp, etc.), hostname, and optional path.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
    format: "uri"
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    scheme: "REGEXP_EXTRACT({col}, '^([^:]+):')"
    hostname: "REGEXP_EXTRACT({col}, '(?:https?://)?([^/?]+)')"
    path: "REGEXP_EXTRACT({col}, '(?:https?://[^/]+)?(/.*)$')"
  validation:
    type: string
    pattern: "^(?:(?:https?|ftp|file):)?//(?:[a-zA-Z0-9](?:[a-zA-Z0-9-]*[a-zA-Z0-9])?\\.)+[a-zA-Z0-9](?:[a-zA-Z0-9-]*[a-zA-Z0-9])?(?::\\d+)?(?:/[^?#]*)?(?:\\?[^#]*)?(?:#.*)?$"
  tier: [VARCHAR, internet]
  release_priority: 5
  aliases: [web_url, uri]
  samples:
    - "https://example.com/path/to/resource"
    - "http://www.google.com"
    - "ftp://ftp.example.org/file.txt"
    - "https://example.com:8080/path?query=value#anchor"
    - "//cdn.example.com/lib.js"
  references: null
  notes: >
    v1 migration: Was internet.url. Includes optional decompose for parsing
    scheme, hostname, and path components.
    Phase 0: Merged technology.internet.uri into this type.
    URI had 37% training data overlap with URL (both use http/https schemes).
    Model could not distinguish them (regression analysis).
    Scheme is optional but `//` is required, so protocol-relative URLs
    (`//cdn.example.com/lib.js`) validate while bare hostnames and bare ids
    (`msg32812262`) do not. Closing the `//` gap makes the validator precise
    enough to gate the url header-hint over-emit (spec
    2026-06-25-sharpen-stage-audit; task t-00007f4d).

technology.internet.hostname:
  title: "Hostname"
  description: >
    Fully qualified domain name or simple hostname. Does not include protocol or path.
    Allows alphanumeric characters, hyphens, and dots. Subdomains supported.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose: null
  validation:
    type: string
    pattern: "^(?:(?:[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?\\.)*[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?)$"
  tier: [VARCHAR, internet]
  release_priority: 4
  aliases: [domain_name]
  samples:
    - "example.com"
    - "www.google.com"
    - "subdomain.example.org"
    - "localhost"
  references: null
  notes: >
    v1 migration: Was internet.hostname. Pattern requires lowercase (use LOWER()
    to normalize). Does not validate DNS resolution.

# technology.internet.port: REMOVED
# False positives on plain integer columns. Range 0-65535 overlaps heavily with integer_number.

technology.internet.top_level_domain:
  title: "Top-Level Domain (TLD)"
  description: >
    Domain extension (e.g., .com, .org, .uk). May include second-level domains
    for country codes (e.g., .co.uk). Does not include the dot.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose: null
  validation:
    type: string
    pattern: "^(?:[a-z]{2,}|xn--[a-z0-9]+)(?:\\.[a-z]{2,})?$"
  tier: [VARCHAR, internet]
  release_priority: 4
  aliases: [tld]
  samples:
    - "com"
    - "org"
    - "uk"
    - "co.uk"
    - "xn--3e0b707e"
  references: null
  notes: >
    v1 migration: Was internet.tld (aliased to top_level_domain for clarity).
    Pattern allows IDN (xn--) format.

# technology.internet.slug: REMOVED in Phase 0
# Collapsed into representation.code.alphanumeric_id.
# Rarely analytically important; confused with hostname in CLDR regression.

technology.internet.user_agent:
  title: "User Agent String"
  description: >
    HTTP User-Agent header string identifying the client application.
    Typically contains browser, OS, and version information.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose: null
  validation:
    type: string
    pattern: "^(Mozilla/|curl/|python-requests/|Wget/|Go-http-client/|axios/|PostmanRuntime/|kube-probe/|Java/|okhttp/|Apache-HttpClient/|libcurl/|node-fetch/|Dalvik/|CFNetwork/|Lynx/|Links |Scrapy/|Googlebot/|Bingbot/|Slackbot|Twitterbot/|facebookexternalhit/|LinkedInBot/|Prometheus/|Datadog/|Ruby/|Dart/|grpc-|HTTPie/|bot|spider|crawl)"
    minLength: 10
    maxLength: 500
  tier: [VARCHAR, internet]
  release_priority: 3
  aliases: null
  samples:
    - "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36"
    - "Mozilla/5.0 (iPhone; CPU iPhone OS 15_0 like Mac OS X)"
    - "curl/7.64.1"
  references: null
  notes: >
    v1 migration: Was internet.user_agent. v13: Added prefix-based pattern
    for validation branch signal. Checks for known UA prefixes (Mozilla/,
    curl/, python-requests/, etc.) and common bot/spider/crawler identifiers.
    Parse with dedicated UA parser for full details.

technology.internet.http_method:
  title: "HTTP Method"
  description: >
    HTTP request method (GET, POST, PUT, DELETE, PATCH, HEAD, OPTIONS, TRACE, CONNECT).
    Broad categorization; lower release priority.
  designation: broad_words
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose: null
  validation:
    type: string
    # 27 case variants = 9 HTTP methods × {UPPER, lower, Title}. Enumerated
    # explicitly in both pattern AND enum because CompiledValidator applies
    # them conjunctively (crates/finetype-core/src/validator.rs delegates to
    # jsonschema::Validator). (?i) on pattern alone does NOT help — enum is
    # exact-string, so "Get" would pass pattern but fail enum.
    # See spec .orbit/specs/2026-04-20-distilled-data-relabel-7-types/spec.yaml (v1.3).
    pattern: "^(GET|Get|get|POST|Post|post|PUT|Put|put|DELETE|Delete|delete|PATCH|Patch|patch|HEAD|Head|head|OPTIONS|Options|options|TRACE|Trace|trace|CONNECT|Connect|connect)$"
    enum: [GET, Get, get, POST, Post, post, PUT, Put, put, DELETE, Delete, delete, PATCH, Patch, patch, HEAD, Head, head, OPTIONS, Options, options, TRACE, Trace, trace, CONNECT, Connect, connect]
  tier: [VARCHAR, internet]
  release_priority: 3
  aliases: null
  samples:
    - "GET"
    - "post"
    - "Delete"
  references: null
  notes: >
    v1 migration: Was internet.http_method marked broad_words. Lower priority
    due to high cardinality but limited semantic information.
    v13: Added explicit pattern for validation branch signal (100% pass rate
    for methods vs 33% for IATA's ^[A-Z]{3}$).
    v17: Expanded enum + pattern to 27 case variants (UPPER/lower/Title per
    method) to fix the v16 training-audit finding where distilled http_method
    rows (SAN JOAQUIN, GOAT, OPERATING, IN PROGRESS, ENROUTE) caused
    misclassification. Cascade: multi-branch validation-branch pass-rate
    feature becomes ~1.0 for real HTTP-method columns regardless of case
    (was ~1/3 when only UPPER variants passed). Realised only after retrain.

# technology.internet.http_status_code: REMOVED
# False positives on plain integer columns. Range 100-599 overlaps heavily with integer_number.

# ═════════════════════════════════════════════════════════════════════
# CATEGORY: cryptographic
#
# Hash functions, tokens, and cryptographic identifiers.
# These are deterministic outputs of cryptographic algorithms or
# random tokens used for security purposes.
# ═════════════════════════════════════════════════════════════════════

# technology.cryptographic.uuid: MOVED to representation.identifier.uuid in v0.5.1
# UUID is a universal identifier, not crypto-specific.

technology.cryptographic.hash:
  title: "Cryptographic Hash"
  description: >
    Output of a cryptographic hash function (MD5, SHA-1, SHA-256, etc.).
    Detected as hex string of fixed length. Broad categorization by length.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    algorithm: "CASE WHEN LENGTH({col}) = 32 THEN 'MD5' WHEN LENGTH({col}) = 40 THEN 'SHA1' WHEN LENGTH({col}) = 64 THEN 'SHA256' WHEN LENGTH({col}) = 128 THEN 'SHA512' ELSE 'UNKNOWN' END"
  validation:
    type: string
    pattern: "^[0-9a-f]{32}$|^[0-9a-f]{40}$|^[0-9a-f]{64}$|^[0-9a-f]{128}$"
  tier: [VARCHAR, cryptographic]
  release_priority: 4
  aliases: null
  samples:
    - "5d41402abc4b2a76b9719d911017c592"
    - "aaf4c61ddcc5e8a2dabede0f3b482cd9aea9434d"
    - "2c26b46911185131006dfb3b57a99d73"
  references: null
  notes: >
    v1 migration: Was cryptographic.hash. Decompose infers algorithm from length:
    32→MD5, 40→SHA1, 64→SHA256, 128→SHA512. No cryptographic validation (can't
    verify true hash without known input).

technology.cryptographic.token_urlsafe:
  title: "URL-Safe Base64 Token"
  description: >
    Random token encoded in URL-safe Base64 (uses - and _ instead of + and /).
    Commonly used for auth tokens, reset links, and secure identifiers.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose: null
  validation:
    type: string
    pattern: "^[A-Za-z0-9_-]{12,128}$"
  tier: [VARCHAR, cryptographic]
  release_priority: 3
  aliases: [urlsafe_token]
  samples:
    - "SFMyNDI5STExOjEwMQ"
    - "dGVzdC10b2tlbg"
    - "aGVsbG8td29ybGQ"
  references: null
  notes: >
    v1 migration: Was cryptographic.token_urlsafe. URL-safe Base64 avoids
    characters that need escaping in URLs. Length typically 12-128 chars.

# ═════════════════════════════════════════════════════════════════════
# CATEGORY: code
#
# Standardized codes and product identifiers (ISBN, IMEI, EAN, etc.).
# These have formal specifications and may include check digits.
# ═════════════════════════════════════════════════════════════════════

# technology.code.isbn: MOVED to identity.commerce.isbn in v0.5.1

technology.code.imei:
  title: "IMEI"
  description: >
    International Mobile Equipment Identity. 15-digit identifier for mobile devices.
    Last digit is a check digit (Luhn algorithm).
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose: null
  validation:
    type: string
    pattern: "^[0-9]{15}$"
  tier: [VARCHAR, code]
  release_priority: 3
  aliases: null
  samples:
    - "490154203237518"
    - "352043068649149"
    - "010000862422488"
  references: null
  notes: >
    v1 migration: Was code.imei. 15 numeric digits. Check digit (last digit)
    uses Luhn algorithm but validation is done via crate, not regex.

# technology.code.ean: MOVED to identity.commerce.ean in v0.5.1

# technology.code.issn: MOVED to identity.commerce.issn in v0.5.1

technology.code.doi:
  title: "DOI"
  description: >
    Digital Object Identifier — a persistent identifier for digital content
    (journal articles, datasets, books). Format: 10.XXXX/suffix where XXXX
    is a registrant code and suffix is assigned by the registrant.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    registrant: "REGEXP_EXTRACT({col}, '^10\\.([0-9]+)/')"
    suffix: "REGEXP_EXTRACT({col}, '^10\\.[0-9]+/(.*)')"
  validation:
    type: string
    pattern: "^10\\.[0-9]{4,9}/[^\\s]+$"
  tier: [VARCHAR, code]
  release_priority: 3
  aliases: null
  samples:
    - "10.1038/nature12373"
    - "10.1000/xyz123"
    - "10.1016/j.cell.2009.01.043"
    - "10.48550/arXiv.2301.07041"
  references: null
  notes: >
    New in v2. DOI prefix always starts with "10." followed by registrant
    code (4-9 digits), then "/" and a suffix. Common registrants include
    10.1038 (Nature), 10.1016 (Elsevier), 10.1126 (Science), 10.1145 (ACM),
    10.48550 (arXiv). Suffix may contain dots, hyphens, slashes.

technology.code.locale_code:
  title: "Locale Code"
  description: >
    Language-region code in IETF BCP 47 format (e.g., en-US, fr-FR, zh-Hans).
    Allows for script subtags (e.g., zh-Hant) and extensions.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    language: "REGEXP_EXTRACT(LOWER({col}), '^([a-z]{2,3})')"
    country: "REGEXP_EXTRACT(LOWER({col}), '-([a-z]{2})(?:$|-)')"
  validation:
    type: string
    pattern: "^[a-zA-Z]{2,3}(?:[-_][a-zA-Z]{2,4})*$"
  tier: [VARCHAR, code]
  release_priority: 4
  aliases: [bcp47, language_tag]
  samples:
    - "en"
    - "en-US"
    - "fr-FR"
    - "zh-Hans-CN"
    - "de-AT"
  references: null
  notes: >
    v1 migration: Was code.locale_code. IETF BCP 47 format. Common codes:
    en, en-US, fr, de, zh-Hans, zh-Hant. Decompose extracts language and country.
    bcp47 alias added — locale_code already covers BCP 47 format.

technology.development.version:
  title: "Version Number"
  description: >
    Software or document version in semantic versioning format (e.g., 1.2.3)
    or variant formats (e.g., v1.2.3, 1.2.3-alpha, 1.2.3+build).
    Resolves to VARCHAR.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    major: "CAST(REGEXP_EXTRACT({col}, '^v?([0-9]+)') AS SMALLINT)"
    minor: "CAST(REGEXP_EXTRACT({col}, '^\\.([0-9]+)') AS SMALLINT)"
    patch: "CAST(REGEXP_EXTRACT({col}, '^\\.([0-9]+)(?:-|\\+|$)') AS SMALLINT)"
    prerelease: "REGEXP_EXTRACT({col}, '-([a-zA-Z0-9.]+)(?:\\+|$)')"
    metadata: "REGEXP_EXTRACT({col}, '\\+([a-zA-Z0-9.]+)$')"
  validation:
    type: string
    pattern: "^v?[0-9]+\\.[0-9]+\\.[0-9]+(?:-[a-zA-Z0-9.]+)?(?:\\+[a-zA-Z0-9.]+)?$"
  tier: [VARCHAR, development]
  release_priority: 4
  aliases: null
  samples:
    - "1.2.3"
    - "v1.2.3"
    - "1.2.3-alpha"
    - "1.2.3-beta.1"
    - "1.2.3+build.123"
  references: null
  notes: >
    v1 migration: Was development.version. Semantic versioning (MAJOR.MINOR.PATCH).
    Decompose extracts major, minor, patch, prerelease, and metadata components.

technology.development.calver:
  title: "Calendar Version"
  description: >
    Calendar versioning (CalVer) following date-based patterns like YYYY.MM.DD
    or YYYY.MM. Used by projects that version based on release calendar.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    year: "CAST(REGEXP_EXTRACT({col}, '^([0-9]{4})') AS SMALLINT)"
    month: "CAST(REGEXP_EXTRACT({col}, '^[0-9]{4}\\.([0-9]{2})') AS SMALLINT)"
    day: "CAST(REGEXP_EXTRACT({col}, '^[0-9]{4}\\.[0-9]{2}\\.([0-9]{2})') AS SMALLINT)"
  validation:
    type: string
    pattern: "^[0-9]{4}\\.[0-9]{2}(?:\\.[0-9]{2})?(?:\\..*)?$"
  tier: [VARCHAR, development]
  release_priority: 3
  aliases: null
  samples:
    - "2024.02"
    - "2024.02.09"
    - "2023.12.31"
    - "2024.01.15.1"
  references: null
  notes: >
    v1 migration: Was development.calver. Decompose extracts year, month, day.
    Format: YYYY.MM or YYYY.MM.DD.

technology.development.docker_ref:
  title: "Docker/OCI Image Reference"
  description: >
    OCI container image reference in the form [registry/]repository[:tag][@digest].
    Used in Dockerfiles, Kubernetes manifests, and CI/CD pipelines.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    registry: "REGEXP_EXTRACT({col}, '^(?:([a-zA-Z0-9.\\-]+(?::\\d+)?)/)')"
    repository: "REGEXP_EXTRACT({col}, '(?:(?:[a-zA-Z0-9.\\-]+(?::\\d+)?)/)?([a-z0-9._\\-/]+?)(?::|@|$)')"
    tag: "REGEXP_EXTRACT({col}, ':([a-zA-Z0-9_.\\-]+)')"
    digest: "REGEXP_EXTRACT({col}, '@(sha256:[a-fA-F0-9]{64})')"
  validation:
    type: string
    pattern: "^(?:(?:[a-zA-Z0-9.\\-]+(?::\\d+)?)/)?(?:[a-z0-9._\\-/]+)(?::[a-zA-Z0-9_.\\-]+)?(?:@sha256:[a-fA-F0-9]{64})?$"
  tier: [VARCHAR, development]
  release_priority: 3
  aliases: [oci_image_ref, container_image]
  samples:
    - "nginx:latest"
    - "ghcr.io/myorg/myapp:v1.2.3"
    - "docker.io/library/postgres:16-alpine"
    - "gcr.io/project/service@sha256:abcdef1234567890abcdef1234567890abcdef1234567890abcdef1234567890"
  references: null
  notes: >
    OCI image reference format. Includes optional registry,
    repository path, tag, and digest components.

# ═════════════════════════════════════════════════════════════════════
# CATEGORY: identifier
#
# Technology-specific identifiers with structured formats that encode
# timestamps or other metadata. Distinguished from representation.identifier
# by having technology-specific semantics.
# ═════════════════════════════════════════════════════════════════════

technology.identifier.ulid:
  title: "ULID"
  description: >
    Universally Unique Lexicographically Sortable Identifier. 26-character
    Crockford Base32 string encoding a 48-bit timestamp and 80 bits of randomness.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: "TRY_CAST({col} AS ULID)"
  decompose:
    timestamp_ms: "EPOCH_MS(({col}::ULID)::BIGINT)"
  validation:
    type: string
    pattern: "^[0123456789ABCDEFGHJKMNPQRSTVWXYZ]{26}$"
  tier: [VARCHAR, identifier]
  release_priority: 4
  aliases: null
  samples:
    - "01ARZ3NDEKTSV4RRFFQ69G5FAV"
    - "01H5J3XZQK0000000000000000"
    - "01FWHE4G8R2SXHM3F07YR3FW0T"
  references: null
  notes: >
    Crockford Base32 excludes I, L, O, U to avoid
    ambiguity. Always uppercase. First 10 chars encode millisecond timestamp.

technology.identifier.tsid:
  title: "TSID"
  description: >
    Time-Sorted ID. 32 lowercase hexadecimal characters with an embedded
    timestamp in the leading digits. Distinguished from UUID by having no
    hyphens, and from cryptographic hashes by the timestamp constraint.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: "tsid_to_timestamp({col})"
  decompose: null
  validation:
    type: string
    pattern: "^[0-9a-f]{32}$"
  tier: [VARCHAR, identifier]
  release_priority: 3
  aliases: null
  samples:
    - "0190a1b2c3d4e5f6a7b8c9d0e1f2a3b4"
    - "018f47d5a6b70000000000000000abcd"
    - "0191c8e4f2a30000fedcba9876543210"
  references: null
  notes: >
    Leading 8-10 hex chars decode to a reasonable
    timestamp (2010-2050). Disambiguation from hash/token_hex requires
    header hints or timestamp validation.

technology.identifier.snowflake_id:
  title: "Snowflake ID"
  description: >
    Twitter/Discord-style snowflake identifier. A 17-20 digit integer
    encoding a millisecond timestamp, worker ID, and sequence number.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: BIGINT
  frictionless:
    type: integer
  format_string: null
  transform: "epoch_ms(({col}::BIGINT >> 22) + 1288834974657)"
  transform_ext: null
  decompose:
    timestamp_ms: "({col}::BIGINT >> 22) + 1288834974657"
    worker_id: "({col}::BIGINT >> 12) & 1023"
    sequence: "{col}::BIGINT & 4095"
  validation:
    type: string
    pattern: "^\\d{17,20}$"
  tier: [BIGINT, identifier]
  release_priority: 3
  aliases: [twitter_id, discord_id]
  samples:
    - "175928847299117063"
    - "1234567890123456789"
    - "80351110224678912"
  references: null
  notes: >
    Twitter epoch: 1288834974657 (2010-11-04).
    Bits: 41 timestamp + 10 worker + 12 sequence. Range ~10^17 to ~10^19.

# ═════════════════════════════════════════════════════════════════════
# CATEGORY: cloud
#
# Cloud provider resource identifiers and URIs.
# ═════════════════════════════════════════════════════════════════════

technology.cloud.aws_arn:
  title: "AWS ARN"
  description: >
    Amazon Resource Name. Globally unique identifier for AWS resources
    in the format arn:partition:service:region:account-id:resource.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    partition: "REGEXP_EXTRACT({col}, '^arn:([^:]+):')"
    service: "REGEXP_EXTRACT({col}, '^arn:[^:]+:([^:]+):')"
    region: "REGEXP_EXTRACT({col}, '^arn:[^:]+:[^:]+:([^:]*):' )"
    account_id: "REGEXP_EXTRACT({col}, '^arn:[^:]+:[^:]+:[^:]*:([^:]*):' )"
    resource: "REGEXP_EXTRACT({col}, '^arn:[^:]+:[^:]+:[^:]*:[^:]*:(.+)$')"
  validation:
    type: string
    pattern: "^arn:(aws|aws-cn|aws-us-gov):[a-zA-Z0-9\\-]+:[a-z0-9\\-]*:\\d{0,12}:.+$"
  tier: [VARCHAR, cloud]
  release_priority: 3
  aliases: [amazon_resource_name]
  samples:
    - "arn:aws:s3:::my-bucket"
    - "arn:aws:iam::123456789012:user/johndoe"
    - "arn:aws:ec2:us-east-1:123456789012:instance/i-1234567890abcdef0"
    - "arn:aws:lambda:eu-west-1:123456789012:function:my-function"
  references: null
  notes: >
    Region and account-id may be empty for global
    resources (e.g., S3 buckets, IAM). Three partitions: aws, aws-cn, aws-us-gov.

technology.cloud.s3_uri:
  title: "S3 URI"
  description: >
    Amazon S3 bucket/key URI in the format s3://bucket-name/key-path.
    Used in data pipelines, ETL tools, and cloud storage references.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    bucket: "REGEXP_EXTRACT({col}, '^s3://([^/]+)')"
    key: "REGEXP_EXTRACT({col}, '^s3://[^/]+/(.*)')"
  validation:
    type: string
    pattern: "^s3://[a-z0-9][a-z0-9.\\-]{1,61}[a-z0-9](/.*)?$"
  tier: [VARCHAR, cloud]
  release_priority: 3
  aliases: null
  samples:
    - "s3://my-bucket/data/file.csv"
    - "s3://production-logs/2024/01/events.parquet"
    - "s3://data-lake-raw/ingestion/batch-001/part-00000.json"
  references: null
  notes: >
    Bucket names: 3-63 chars, lowercase, no underscores.
    Key paths can contain any characters after the bucket name.

technology.cryptographic.jwt:
  title: "JSON Web Token"
  description: >
    JSON Web Token (JWT) consisting of three base64url-encoded segments
    separated by dots: header.payload.signature. Used for authentication
    and information exchange.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    header: "REGEXP_EXTRACT({col}, '^([A-Za-z0-9_-]+)\\.')"
    payload: "REGEXP_EXTRACT({col}, '^\\.([A-Za-z0-9_-]+)\\.')"
    signature: "REGEXP_EXTRACT({col}, '\\.([A-Za-z0-9_-]+)$')"
  validation:
    type: string
    pattern: "^[A-Za-z0-9_-]{10,}\\.[A-Za-z0-9_-]{10,}\\.[A-Za-z0-9_-]{10,}$"
  tier: [VARCHAR, cryptographic]
  release_priority: 3
  aliases: [json_web_token]
  samples:
    - "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiIxMjM0NTY3ODkwIiwibmFtZSI6IkpvaG4gRG9lIiwiaWF0IjoxNTE2MjM5MDIyfQ.SflKxwRJSMeKKF2QT4fwpMeJf36POk6yJV_adQssw5c"
    - "eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJhdXRoLmV4YW1wbGUuY29tIn0.dBjftJeZ4CVP-mB92K27uhbUJU1p1r_wW1gFWFOEjXk"
  references: null
  notes: >
    Each segment is at least 10 base64url chars to
    distinguish from other dot-separated formats. Header typically starts
    with "eyJ" (base64 for '{"').

technology.internet.cidr:
  title: "CIDR Notation"
  description: >
    Classless Inter-Domain Routing notation combining an IPv4 address
    with a prefix length (e.g., 192.168.1.0/24). Used in network
    configuration, firewall rules, and IP access control.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: "CAST({col} AS INET)"
  decompose:
    network: "REGEXP_EXTRACT({col}, '^([^/]+)/')"
    prefix_length: "CAST(REGEXP_EXTRACT({col}, '/(\\d+)$') AS TINYINT)"
  validation:
    type: string
    pattern: "^(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)/([0-9]|[12]\\d|3[0-2])$"
  tier: [INET, internet]
  release_priority: 4
  aliases: [cidr_block, ip_range]
  samples:
    - "192.168.1.0/24"
    - "10.0.0.0/8"
    - "172.16.0.0/12"
    - "0.0.0.0/0"
  references: null
  notes: >
    IPv4 CIDR only. Prefix length 0-32. Distinguished
    from plain IPv4 by the /prefix suffix.

technology.internet.urn:
  title: "URN"
  description: >
    Uniform Resource Name — a persistent, location-independent identifier
    in the format urn:nid:nss. Used for namespaced identifiers like
    ISBN URNs, OIDs, and UUID URNs.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
    format: "uri"
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    nid: "REGEXP_EXTRACT({col}, '^urn:([^:]+):')"
    nss: "REGEXP_EXTRACT({col}, '^urn:[^:]+:(.+)$')"
  validation:
    type: string
    pattern: "^urn:[a-z0-9][a-z0-9\\-]{0,31}:.+$"
  tier: [VARCHAR, internet]
  release_priority: 3
  aliases: [uniform_resource_name]
  samples:
    - "urn:isbn:0451450523"
    - "urn:ietf:rfc:2648"
    - "urn:uuid:6e8bc430-9c3a-11d9-9669-0800200c9a66"
    - "urn:oid:2.16.840"
  references: null
  notes: >
    NID (Namespace ID) is 1-32 lowercase alphanumeric
    or hyphen chars. NSS (Namespace Specific String) is the remainder.

technology.internet.data_uri:
  title: "Data URI"
  description: >
    Inline data URI scheme embedding content directly in a string.
    Format: data:[mediatype][;base64],data. Used for embedding small
    files in HTML, CSS, and API responses.
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
    format: "uri"
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    media_type: "REGEXP_EXTRACT({col}, '^data:([^;,]+)')"
    is_base64: "CASE WHEN {col} LIKE '%base64,%' THEN true ELSE false END"
  validation:
    type: string
    pattern: "^data:(?:[a-zA-Z0-9]+/[a-zA-Z0-9\\-+.]+)?(?:;[a-zA-Z0-9\\-]+=[^;,]*)*(?:;base64)?,.+$"
  tier: [VARCHAR, internet]
  release_priority: 3
  aliases: null
  samples:
    - "data:text/plain;base64,SGVsbG8gV29ybGQ="
    - "data:image/png;base64,iVBORw0KGgoAAAANSUhEUg=="
    - "data:application/json,{\"key\":\"value\"}"
    - "data:text/html,%3Ch1%3EHello%3C%2Fh1%3E"
  references: null
  notes: >
    Media type is optional (defaults to text/plain;charset=US-ASCII).
    Base64 encoding is indicated by ;base64 before the comma.

technology.filesystem.windows_path:
  title: "Windows File Path"
  description: >
    An absolute Microsoft Windows filesystem path: a drive letter and colon
    (C:\, D:\) or a UNC share root (\\server\share) followed by
    backslash-separated segments. Distinct from URLs (no scheme) and POSIX
    paths (forward-slash rooted).
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    drive: "REGEXP_EXTRACT({col}, '^([A-Za-z]:)')"
    filename: "REGEXP_EXTRACT({col}, '([^\\\\]+)$')"
  validation:
    type: string
    pattern: '^([A-Za-z]:\\|\\\\)[^\r\n]*$'
    minLength: 3
  tier: [VARCHAR, filesystem]
  release_priority: 3
  aliases: [path, filepath, directory]
  samples:
    - "C:\\Windows\\System32\\drivers\\WdfLdr.sys"
    - "D:\\research\\repos\\project\\src\\Main.cs"
    - "\\\\server\\share\\reports\\q3.xlsx"
  references: null
  notes: >
    New in v0.6.36 (spec 2026-06-19-plain-text-type-discovery). Highest
    volume x nameability type hiding behind representation.text.plain_text:
    7,651 distinct corpus datasets, blind-panel confidence 0.97 (unanimous).
    Pattern fires on <1.2% of multi-word prose. POSIX-path sibling deferred
    (below volume bar + URL-route ambiguity).

technology.internet.message_id:
  title: "Email Message-ID"
  description: >
    RFC 2822 Message-ID header — a globally-unique identifier for an email
    message: an angle-bracketed left@right token (<unique@domain>).
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    local: "REGEXP_EXTRACT({col}, '^<([^@]+)@')"
    domain: "REGEXP_EXTRACT({col}, '@([^>]+)>$')"
  validation:
    type: string
    pattern: '^<[^<>@\s]+@[^<>@\s]+>$'
  tier: [VARCHAR, internet]
  release_priority: 3
  aliases: [messageid, msgid]
  samples:
    - "<30365805.1075860998985.JavaMail.evans@thyme>"
    - "<CAF2a3b9c@mail.gmail.com>"
    - "<199911abc@server.example.org>"
  references: null
  notes: >
    New in v0.6.36 (spec 2026-06-19-plain-text-type-discovery). 3,032
    distinct corpus datasets, blind-panel confidence 0.95 (unanimous).
    Angle-bracket + @ grammar is unambiguous. Corpus sample is Enron-heavy
    (single provenance) but the RFC 2822 type is universal. Sits beside
    url / urn / mac_address as an internet protocol identifier.

technology.code.qualified_name:
  title: "Qualified Name"
  description: >
    A dotted, reverse-DNS-style fully-qualified identifier: three or more
    dot-separated identifier segments naming a namespaced code symbol
    (Java/.NET package or class, config key, module path).
  designation: universal
  locales: [UNIVERSAL]
  broad_type: VARCHAR
  frictionless:
    type: string
  format_string: null
  transform: "CAST({col} AS VARCHAR)"
  transform_ext: null
  decompose:
    root: "REGEXP_EXTRACT({col}, '^([A-Za-z_][A-Za-z0-9_]*)')"
    leaf: "REGEXP_EXTRACT({col}, '([A-Za-z0-9_]+)$')"
  validation:
    type: string
    pattern: '^[A-Za-z_][A-Za-z0-9_]*(\.[A-Za-z_][A-Za-z0-9_]*){2,}$'
  tier: [VARCHAR, code]
  release_priority: 3
  aliases: [fqn, classname, namespace]
  samples:
    - "com.google.javascript.jscomp.Compiler"
    - "org.apache.commons.math.fraction.Fraction"
    - "Avalonia.UnitTests.Controls.ContentControl"
  references: null
  notes: >
    New in v0.6.36 (spec 2026-06-19-plain-text-type-discovery). 1,677
    distinct corpus datasets, blind-panel confidence 0.79 (unanimous
    "java/dotted FQN"). Pattern fires on ZERO multi-word prose. Boundary
    caveat: structurally overlaps technology.internet.hostname (www.x.com);
    the disambiguating signal is a CamelCase / non-TLD leaf segment — a
    Sharpen-precedence decision for the retrain, not the validator.