pubid 2.0.10 → 2.0.12

data/lib/pubid/idf/pubid_urn_schema.rb

@@ -0,0 +1,44 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Idf
+    # The pubid-scheme URN of IDF (issue #670; docs/URN-SCHEMES.adoc).
+    #
+    #   IDF 146:2003/AMD 1:2023  ->
+    #   urn:pubid:idf:is:146:pubbody-IDF:year-2003:amd:1:year-2023
+    #
+    # The current urn:idf:… form of #to_urn is pubid-invented and stays
+    # there. `type`, `stage` and `typed_stage` serialize every registry
+    # field of their components, so each field is its own atom (the NIST
+    # ruling): reducing them to one identifying value would change to_hash
+    # and need a relaton-data re-crawl — a separate decision.
+    PubidUrnSchema = Pubid::PubidUrn.define("idf") do
+      atom "publisher.body", "pubbody"
+      atom "type.abbr", "typeabbr"
+      atom "type.name", "typename"
+      atom "type.type_code", "typetypecode"
+      atom "part", "part"
+      atom "year", "year"
+      atom "stage.abbr", "stageabbr"
+      atom "stage.name", "stagename"
+      atom "stage.stage_code", "stagestagecode"
+      atom "stage.harmonized_stages", "stageharmonizedstages", kind: :list
+      atom "typed_stage.abbr", "typedstageabbr", kind: :list
+      atom "typed_stage.name", "typedstagename"
+      atom "typed_stage.code", "typedstage"
+      atom "typed_stage.long_abbr", "typedstagelongabbr"
+      atom "typed_stage.short_abbr", "typedstageshortabbr"
+      atom "typed_stage.stage_code", "typedstagestagecode"
+      atom "typed_stage.type_code", "typedstagetypecode"
+      atom "typed_stage.harmonized_stages", "typedstageharmonizedstages",
+           kind: :list
+
+      type "pubid:idf:all-parts", "all"
+      type "pubid:idf:amendment", "amd"
+      type "pubid:idf:corrigendum", "cor"
+      type "pubid:idf:identifier", "id"
+      type "pubid:idf:international-standard", "is"
+      type "pubid:idf:reviewed-method", "rm"
+    end
+  end
+end

data/lib/pubid/idf.rb

@@ -17,6 +17,7 @@ module Pubid
     autoload :SupplementIdentifier, "#{__dir__}/idf/supplement_identifier"
     autoload :UrnGenerator, "#{__dir__}/idf/urn_generator"
     autoload :UrnParser, "#{__dir__}/idf/urn_parser"
+    autoload :PubidUrnSchema, "#{__dir__}/idf/pubid_urn_schema"
 
     def self.parse(identifier)
       unless identifier.is_a?(String)

data/lib/pubid/iec/CLAUDE.md

@@ -31,3 +31,16 @@ These notes were part of the root `CLAUDE.md`. Read them before you change `lib/
   **Two parent-slot gaps recorded by the cross-flavor `base` survey** (`lib/pubid/bsi/CLAUDE.md` has the measured version). `Identifiers::TestReportForm#cispr_identifier` is a polymorphic `Identifier` parent slot with no `#root` override, so `root` returns the form itself. And `Identifiers::ConsolidatedIdentifier#base_document` returns `identifiers&.first` **without** the recursive `&.base_document` that its BSI, CSA and CEN counterparts use — a one-level peel where the others peel fully, so a consolidated edition of a supplement does not reduce to the bare standard. Neither is chased here; both are pinned by nothing, which is why they are written down.
 
   **Stage codes — the SSOT and what is settled.** The authoritative source for IEC stage abbreviations is committed in this repo: **`data/iec/stage-codes.md`** — the 52-row IEC/ISO stage-code correspondence table transcribed from IEC's official *IEC/ISO Harmonized APIs* documents (v0.0.8 2023-06, api-portal.iec.ch; v0.0.7 2021-05, iec-apiportal.apigee.io — the successors that replace the IEC Data Feeds), the legacy Data-Feeds field examples, and the IS-lifecycle ladder from `StageCodes_IS_v2.pdf` (2023-06). `www.iec.ch` itself is fully WAF-gated to non-browsers and the Wayback Machine holds no snapshots of its PDFs, so the extract — not the unreachable original — is the reference. **`Pubid::Iec.locate_stage` reads `TYPED_STAGES` only** (`lib/pubid/iec.rb`); `PROJECT_STAGES` (AFDIS, TCDV, PRVC…) are not queryable through it. Every one of metanorma-iec's 55 `STAGE_CODES` — the list quoted in pubid#237 — resolves, locked by `spec/pubid/iec/typed_stages_spec.rb` in the issue's own order; aliases are always **appended, never prepended**, so `abbr.first` (what renders) is unchanged. Notable rulings from that pass: **ADIS** is the legacy Data-Feeds spelling of AFDIS ("Approved for FDIS circulation", ISO_LEVEL 40.99) and is an alias on `cdv` (the typed stage claiming 40.99); **CDISH/CDPAS** are the 50.20 rows of the DISH/DPAS deliverables, not "committee draft ISH/PAS" (30.xx); **Reconciliation items recorded in the extract, deliberately not applied:** DELPUB sits at 99.60 there while the `is` ladder does not claim 99.60; CAN carries two codes (30.98 IS/TR/TS, 50.98 PAS/ISH/COR); the rows CDISH-table-siblings DECDISH/MERGED/SPLIT/PCC/PRVN/DEL/RPUB/TPUB and the per-deliverable translation/preparation families are not registered as typed-stage abbreviations.
+
+## pubid-scheme URN (issue #670)
+
+`lib/pubid/iec/pubid_urn_schema.rb` completes the partial schema the IEEE
+and GOST waves landed; rulings in `docs/URN-SCHEMES.adoc` (`[[urn-iec]]`).
+Invariant: every atom added after those waves sits after `database-` and
+before `lang-`, so a landed URN encodes identically (`database` became the
+bool the corpus serializes before any URN carried it). The URN inherits
+the unserialized-identity gaps pinned above: a WorkingDocument's URN is
+the type-only `urn:pubid:iec:wd` (all WD rows share one hash — a shared
+hash, not a collision), and a TestReportForm's CISPR document rides
+`cispr_identifier`. Mapping them stays the wire-format decision it
+already was.

data/lib/pubid/iec/pubid_urn_schema.rb

@@ -2,15 +2,28 @@
 
 module Pubid
   module Iec
-    # The pubid-scheme URN of IEC — PARTIAL (issue #656;
+    # The pubid-scheme URN of IEC (issue #670 completes the flavor;
     # docs/URN-SCHEMES.adoc).
     #
-    # Only the types other flavors nest are declared: the International
-    # Standard (IEC/IEEE co-publications, GOST adoptions), and the TS, TR,
-    # amendment and consolidated identifier a GOST adoption can name
-    # ("IEC 60079-0:2017+AMD1:2020"). IEC's own wave declares the rest of the
-    # flavor; until then IEC stays pending in the contract spec. The official
-    # RFC 5141-style urn:iec:std: form stays on #to_urn.
+    #   CISPR 12:2007/AMD1:2009  ->
+    #   urn:pubid:iec:is:12:pub-CISPR:year-2007:amd:1:pub-CISPR:year-2009
+    #
+    # The partial schema landed with IEEE named the one type IEEE nests; the
+    # GOST wave added the TS, TR, amendment and consolidated identifier a
+    # GOST adoption can name. This wave declares the rest of the flavor and
+    # the atoms its types serialize (`vap`, the sheet/fragment numbers);
+    # every new atom sits after the ones earlier waves placed, so the
+    # published URNs are unchanged. `database` was a string atom before any
+    # URN carried it — the corpus holds only `true`, so it is a bool.
+    #
+    # Known identity gaps the URN inherits, because it carries exactly what
+    # to_hash carries (pinned in lib/pubid/iec/CLAUDE.md): a WorkingDocument
+    # serializes only `_type`, and a TestReportForm's CISPR document sits in
+    # the unserialized `cispr_identifier` — such rows emit type-only URNs
+    # (`urn:pubid:iec:wd`) and are one hash, so they are not a collision.
+    # Mapping them is a wire-format change and stays the separate decision
+    # it already was. The official RFC 5141-style urn:iec:std: form stays
+    # on #to_urn.
     PubidUrnSchema = Pubid::PubidUrn.define("iec") do
       atom "publisher", "pub"
       atom "copublishers", "co", kind: :list
@@ -23,17 +36,42 @@ module Pubid
       atom "edition", "ed"
       atom "stage", "stage"
       atom "stage_iteration", "iter"
-      atom "database", "database"
+      atom "database", "database", kind: :bool
+      atom "fragment_number", "frag"
+      atom "sheet_number", "sheet"
+      atom "sheet_year", "sheetyear"
+      atom "vap", "vap", kind: :list
       atom "languages.*.code", "lang"
       atom "identifiers", "ids", kind: :ids
 
       facet "languages.*.original_code", "langsrc"
 
+      type "pubid:iec:all-parts", "all"
       type "pubid:iec:amendment", "amd"
+      type "pubid:iec:base", "base"
+      type "pubid:iec:component-specification", "cs"
+      type "pubid:iec:conformity-assessment", "ca"
       type "pubid:iec:consolidated-identifier", "consol", core: []
+      type "pubid:iec:corrigendum", "cor"
+      type "pubid:iec:fragment-identifier", "fi", core: []
+      type "pubid:iec:guide", "guide"
+      type "pubid:iec:identifier", "id"
       type "pubid:iec:international-standard", "is"
+      type "pubid:iec:interpretation-sheet", "is2"
+      type "pubid:iec:operational-document", "od"
+      type "pubid:iec:publicly-available-specification", "pas"
+      type "pubid:iec:sheet-identifier", "si", core: []
+      type "pubid:iec:single-identifier", "si2"
+      type "pubid:iec:societal-technology-trend-report", "sttr"
+      type "pubid:iec:systems-reference-document", "srd"
+      type "pubid:iec:technical-group", "tg"
       type "pubid:iec:technical-report", "tr"
       type "pubid:iec:technical-specification", "ts"
+      type "pubid:iec:technology-report", "tr2"
+      type "pubid:iec:test-report-form", "trf"
+      type "pubid:iec:vap-identifier", "vi"
+      type "pubid:iec:white-paper", "wp"
+      type "pubid:iec:working-document", "wd"
     end
   end
 end

data/lib/pubid/ietf/pubid_urn_schema.rb

@@ -0,0 +1,24 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Ietf
+    # The pubid-scheme URN of IETF (issue #670; docs/URN-SCHEMES.adoc).
+    #
+    #   draft-3gpp-collaboration-01  ->
+    #   urn:pubid:ietf:id2:3gpp-collaboration:v-01
+    #
+    # The official urn:ietf:rfc:… namespace (RFC 2648) stays on #to_urn.
+    # The internet-draft token is `id2`, because `id` names the flavor's
+    # bare identifier.
+    PubidUrnSchema = Pubid::PubidUrn.define("ietf") do
+      atom "version", "v"
+
+      type "pubid:ietf:bcp", "bcp"
+      type "pubid:ietf:fyi", "fyi"
+      type "pubid:ietf:identifier", "id"
+      type "pubid:ietf:internet-draft", "id2"
+      type "pubid:ietf:rfc", "rfc"
+      type "pubid:ietf:std", "std"
+    end
+  end
+end

data/lib/pubid/ietf.rb

@@ -24,6 +24,7 @@ module Pubid
     autoload :Renderer, "#{__dir__}/ietf/renderer"
     autoload :UrnGenerator, "#{__dir__}/ietf/urn_generator"
     autoload :UrnParser, "#{__dir__}/ietf/urn_parser"
+    autoload :PubidUrnSchema, "#{__dir__}/ietf/pubid_urn_schema"
 
     # Parse an IETF identifier string
     def self.parse(identifier)

data/lib/pubid/iho/CLAUDE.md

@@ -172,3 +172,14 @@ edition rename (the serialized key moved `version` → `edition`) and
 the volume unification; relaton-cli's `MachineIndex::REFINABLE_KEYS`
 needs `edition`/`volume` beside the retired `version`/`year` keys; the
 pubid pin moves in step with the relaton-data-iho re-key.
+
+## pubid-scheme URN (issue #670)
+
+`lib/pubid/iho/pubid_urn_schema.rb`; rulings in `docs/URN-SCHEMES.adoc`
+(`[[urn-iho]]`). The atoms follow the containment order of
+`UrnGenerator::SEGMENTS` — the document's `ed-`/`year-` first, then each
+component with its own edition/year right after it (`sup-`, `part-`,
+`parted-`, `appendix-`, `appsub-`, `vol-`, `issue-` …). Case is preserved
+(`part-A`); the yearbook year is a volume (`vol-2024`, no `year-` value —
+the legacy `y.` marker stays decodable in the official URN parser, never
+emitted). The official `urn:mrn:iho:…` form stays on `#to_urn`.

data/lib/pubid/iho/pubid_urn_schema.rb

@@ -0,0 +1,49 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Iho
+    # The pubid-scheme URN of IHO (issue #670; docs/URN-SCHEMES.adoc).
+    #
+    #   IHO S-100 5.0.0 Part 2 (2024)  ->
+    #   urn:pubid:iho:std:100:ed-5.0.0:year-2024:part-2
+    #
+    # The atoms follow the containment order of the official URN
+    # (urn:iho:s:100:ed-5.2.0:part-2, Pubid::Iho::UrnGenerator): the
+    # document's edition and year first, then each component with its own
+    # edition/year right after it — supplement, part, annex, appendix (the
+    # two-level Appendix D-2 keeps the subnumber as its own atom), volume
+    # (the unified locator, yearbook volumes included: `vol-2024`), issue.
+    # Letters stay case-preserved (`part-A`). The legacy `y.` yearbook
+    # marker is decodable in the official URN parser, never emitted here:
+    # a periodical has no `year-` atom value, its volume carries the year.
+    # The official urn:mrn:iho:… form stays on #to_urn.
+    PubidUrnSchema = Pubid::PubidUrn.define("iho") do
+      atom "edition", "ed"
+      atom "year", "year"
+      atom "supplement", "sup"
+      atom "supplement_edition", "suped"
+      atom "supplement_year", "supyear"
+      atom "part", "part"
+      atom "part_edition", "parted"
+      atom "part_year", "party"
+      atom "annex", "annex"
+      atom "annex_edition", "anned"
+      atom "annex_year", "annyear"
+      atom "appendix", "appendix"
+      atom "appendix_subnumber", "appsub"
+      atom "appendix_edition", "apped"
+      atom "appendix_year", "appyear"
+      atom "volume", "vol"
+      atom "issue", "issue"
+      atom "volume_edition", "voled"
+      atom "volume_year", "voly"
+
+      type "pubid:iho:bibliographic", "b"
+      type "pubid:iho:circular-letter", "cl"
+      type "pubid:iho:identifier", "id"
+      type "pubid:iho:miscellaneous", "m"
+      type "pubid:iho:publication", "pub"
+      type "pubid:iho:standard", "std"
+    end
+  end
+end

data/lib/pubid/iho.rb

@@ -14,6 +14,7 @@ module Pubid
     autoload :Renderer,     "#{__dir__}/iho/renderer"
     autoload :UrnGenerator, "#{__dir__}/iho/urn_generator"
     autoload :UrnParser, "#{__dir__}/iho/urn_parser"
+    autoload :PubidUrnSchema, "#{__dir__}/iho/pubid_urn_schema"
 
     # Parse an IHO identifier string into an identifier object
     # @param identifier [String] The IHO identifier string to parse

data/lib/pubid/isbn/pubid_urn_schema.rb

@@ -0,0 +1,20 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Isbn
+    # The pubid-scheme URN of ISBN (issue #670; docs/URN-SCHEMES.adoc).
+    #
+    #   ISBN 0-306-40615-2  ->  urn:pubid:isbn:book:hyphenated-0-306-40615-2:raw-0306406152
+    #
+    # There is no positional core: the hyphenated and raw spellings are both
+    # optional atoms (an ISBN given bare serializes only `raw`). The official
+    # urn:isbn: namespace (RFC 3187) stays on #to_urn.
+    PubidUrnSchema = Pubid::PubidUrn.define("isbn") do
+      atom "hyphenated", "hyphenated"
+      atom "raw", "raw"
+
+      type "pubid:isbn:book", "book", core: []
+      type "pubid:isbn:identifier", "id", core: []
+    end
+  end
+end

data/lib/pubid/isbn.rb

@@ -18,6 +18,7 @@ module Pubid
     autoload :Identifiers, "#{__dir__}/isbn/identifiers"
     autoload :Parser, "#{__dir__}/isbn/parser"
     autoload :Renderer, "#{__dir__}/isbn/renderer"
+    autoload :PubidUrnSchema, "#{__dir__}/isbn/pubid_urn_schema"
 
     def self.parse(identifier)
       Identifier.parse(identifier)

data/lib/pubid/jis/pubid_urn_schema.rb

@@ -0,0 +1,29 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Jis
+    # The pubid-scheme URN of JIS (issue #670; docs/URN-SCHEMES.adoc).
+    #
+    #   JIS A 0001:1999  ->  urn:pubid:jis:jis:0001:series-A:year-1999
+    #
+    # The current urn:jis:… form of #to_urn is pubid-invented and stays
+    # there.
+    PubidUrnSchema = Pubid::PubidUrn.define("jis") do
+      atom "series", "series"
+      atom "parts", "part", kind: :list
+      atom "year", "year", kind: :int
+      atom "reaffirmed", "reaffirmed", kind: :bool
+
+      type "pubid:jis:all-parts", "all"
+      type "pubid:jis:amendment", "amd"
+      type "pubid:jis:corrigendum", "cor"
+      type "pubid:jis:explanation", "e"
+      type "pubid:jis:identifier", "id"
+      type "pubid:jis:japanese-industrial-standard", "jis"
+      type "pubid:jis:standard", "std"
+      type "pubid:jis:technical-report", "tr"
+      type "pubid:jis:technical-specification", "ts"
+      type "pubid:jis:welding-engineering-standard", "wes"
+    end
+  end
+end

data/lib/pubid/jis.rb

@@ -19,6 +19,7 @@ module Pubid
     autoload :SupplementIdentifier, "#{__dir__}/jis/supplement_identifier"
     autoload :UrnGenerator, "#{__dir__}/jis/urn_generator"
     autoload :UrnParser, "#{__dir__}/jis/urn_parser"
+    autoload :PubidUrnSchema, "#{__dir__}/jis/pubid_urn_schema"
 
     # Parse a JIS identifier string
     def self.parse(identifier)

data/lib/pubid/omg/pubid_urn_schema.rb

@@ -0,0 +1,21 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Omg
+    # The pubid-scheme URN of OMG (issue #670; docs/URN-SCHEMES.adoc).
+    #
+    #   OMG AMI4CCM 1.0  ->  urn:pubid:omg:spec:acronym-AMI4CCM:v-1.0
+    #
+    # No positional core: an OMG specification names itself by its acronym.
+    # The current urn:omg:… form of #to_urn is pubid-invented and stays
+    # there.
+    PubidUrnSchema = Pubid::PubidUrn.define("omg") do
+      atom "part", "part"
+      atom "version", "v"
+      atom "acronym", "acronym"
+
+      type "pubid:omg:identifier", "id", core: []
+      type "pubid:omg:specification", "spec", core: []
+    end
+  end
+end

data/lib/pubid/omg.rb

@@ -17,6 +17,7 @@ module Pubid
     autoload :Identifiers, "#{__dir__}/omg/identifiers"
     autoload :Parser, "#{__dir__}/omg/parser"
     autoload :Renderer, "#{__dir__}/omg/renderer"
+    autoload :PubidUrnSchema, "#{__dir__}/omg/pubid_urn_schema"
 
     def self.parse(identifier)
       Identifier.parse(identifier)

data/lib/pubid/sae/pubid_urn_schema.rb

@@ -0,0 +1,33 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Sae
+    # The pubid-scheme URN of SAE (issue #670; docs/URN-SCHEMES.adoc).
+    #
+    #   SAE J3016_202104   ->  urn:pubid:sae:id:3016:typeabbr-J:year-2021:month-4
+    #   SAE ARP4754A:2010  ->  urn:pubid:sae:id:4754:typeabbr-ARP:rev-A:yearflat-2010
+    #
+    # The current urn:sae:… form of #to_urn is pubid-invented and stays
+    # there. The type (`J`, `AS`, `ARP`, `AMS`) is an atom, because the
+    # identifier type is one class carrying every series code; the date
+    # fields are Integers.
+    #
+    # SAE's hash carries two date shapes (the canonical flattening keeps a
+    # year-alone date flat: pubid#360), so a year gets one prefix per shape
+    # — the ashrae ruling. `year-` stays the nested year+month form the
+    # first release emitted; `yearflat-` is the flat year-alone form the
+    # corpus round (pubid#669) added. A part is top-level in the hash
+    # (AMS-QQ-A-250/4B -> part-4).
+    PubidUrnSchema = Pubid::PubidUrn.define("sae") do
+      atom "type.abbr", "typeabbr"
+      atom "revision", "rev"
+      atom "amendment", "amd"
+      atom "part", "part"
+      atom "date.year", "year", kind: :int
+      atom "date.month", "month", kind: :int
+      atom "year", "yearflat"
+
+      type "pubid:sae:identifier", "id"
+    end
+  end
+end

data/lib/pubid/sae.rb

@@ -17,6 +17,7 @@ module Pubid
     autoload :Renderer, "#{__dir__}/sae/renderer"
     autoload :UrnGenerator, "#{__dir__}/sae/urn_generator"
     autoload :UrnParser, "#{__dir__}/sae/urn_parser"
+    autoload :PubidUrnSchema, "#{__dir__}/sae/pubid_urn_schema"
 
     def self.parse(input)
       Identifier.parse(input)

data/lib/pubid/tgpp/pubid_urn_schema.rb

@@ -0,0 +1,23 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Tgpp
+    # The pubid-scheme URN of 3GPP (issue #670; docs/URN-SCHEMES.adoc).
+    #
+    #   TS 01.01:REL-96/5.0.0  ->  urn:pubid:3gpp:ts:01.01:part-01:release-96:v-5.0.0
+    #
+    # The official urn:3gpp: namespace (RFC 5279) stays on #to_urn. The
+    # parts ("01.01" splits to part "01" and part "01"…) are the list the
+    # flavor models; the release and the version close the URN.
+    PubidUrnSchema = Pubid::PubidUrn.define("3gpp") do
+      atom "parts", "part", kind: :list
+      atom "version", "v"
+      atom "release", "release"
+      atom "suffix", "suffix"
+
+      type "pubid:3gpp:identifier", "id"
+      type "pubid:3gpp:technical-report", "tr"
+      type "pubid:3gpp:technical-specification", "ts"
+    end
+  end
+end

data/lib/pubid/tgpp.rb

@@ -34,6 +34,7 @@ module Pubid
     autoload :Renderer, "#{__dir__}/tgpp/renderer"
     autoload :UrnGenerator, "#{__dir__}/tgpp/urn_generator"
     autoload :UrnParser, "#{__dir__}/tgpp/urn_parser"
+    autoload :PubidUrnSchema, "#{__dir__}/tgpp/pubid_urn_schema"
 
     # Parse a 3GPP identifier string
     def self.parse(identifier)

data/lib/pubid/un/pubid_urn_schema.rb

@@ -0,0 +1,19 @@
+# frozen_string_literal: true
+
+module Pubid
+  module Un
+    # The pubid-scheme URN of UN (issue #670; docs/URN-SCHEMES.adoc).
+    #
+    #   A/RES/78/1  ->  urn:pubid:un:document:A,RES,78:1
+    #
+    # The path segments are the list before the final number; the number is
+    # the positional core. The current urn:un:… form of #to_urn is
+    # pubid-invented and stays there.
+    PubidUrnSchema = Pubid::PubidUrn.define("un") do
+      atom "path", "path", kind: :list
+
+      type "pubid:un:document", "document"
+      type "pubid:un:identifier", "id"
+    end
+  end
+end

data/lib/pubid/un.rb

@@ -12,6 +12,7 @@ module Pubid
     autoload :Identifiers, "#{__dir__}/un/identifiers"
     autoload :Parser, "#{__dir__}/un/parser"
     autoload :Renderer, "#{__dir__}/un/renderer"
+    autoload :PubidUrnSchema, "#{__dir__}/un/pubid_urn_schema"
 
     def self.parse(identifier)
       Identifier.parse(identifier)

data/lib/pubid/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Pubid
-  VERSION = "2.0.10"
+  VERSION = "2.0.12"
 end

data/lib/pubid.rb

@@ -238,6 +238,7 @@ module Pubid
   # @param format [Symbol] :auto, :human, :mr_string, or :urn
   # @return [Identifier] The parsed identifier
   # @raise [ArgumentError] for a non-String, or one over {MAX_INPUT_LENGTH}
+  #   (as {Errors::InvalidInputError}, which is one)
   # @raise [Pubid::Errors::ParseError] when no flavor can parse the string — the
   #   class every flavor's own +parse+ raises (see the parse-failure contract)
   def self.parse(string, format: :auto)
@@ -249,6 +250,15 @@ module Pubid
       raise Pubid::Errors::InvalidInputError,
             Pubid::INPUT_TOO_LONG_MESSAGE
     end
+    # Invalid encoding is bad input, not a grammar failure: the detectors
+    # and normalization regexes raise raw `ArgumentError` (invalid byte
+    # sequence) or `Encoding::CompatibilityError` on it, which is the leak
+    # the contract closes (issue #670). The bytes are read as UTF-8 — a
+    # binary string of ASCII bytes (".b") is fine, high bytes are not.
+    unless string.dup.force_encoding(Encoding::UTF_8).valid_encoding?
+      raise Pubid::Errors::InvalidInputError,
+            "input string has invalid encoding"
+    end
 
     format = FormatDetector.detect(string) if format == :auto
 
@@ -282,22 +292,38 @@ module Pubid
   #
   # @param string [String] the URN
   # @return [Identifier] the parsed identifier
+  # @raise [Pubid::Errors::InvalidInputError] for a non-String
   #
   # A pubid-scheme URN ("urn:pubid:…", see {PubidUrn}) is decoded by the
   # shared codec; every other namespace goes to its flavor's UrnParser.
+  # Every decode failure on this surface is a pubid error (issue #670): an
+  # empty string, a missing or unknown namespace, invalid encoding, and a
+  # URN a flavor's parser rejects all raise {Errors::UrnParseError} — the
+  # raw `ArgumentError` the detection and the flavor parsers raise is the
+  # wrapped failure's cause, never the raised class.
   def self.parse_urn(string)
+    unless string.is_a?(String)
+      raise Pubid::Errors::InvalidInputError,
+            Pubid::INPUT_NOT_A_STRING_MESSAGE
+    end
     eager_load_flavors!
     return PubidUrn.parse(string) if PubidUrn.pubid_urn?(string)
 
     flavor = detect_flavor_from_urn(string)
     flavor_module = Registry.get(flavor)
     unless flavor_module
-      raise ArgumentError,
-            "Unknown flavor in URN: #{flavor}"
+      raise Errors::UrnParseError,
+            "unknown URN namespace: #{flavor.inspect} (in #{string.inspect})"
     end
 
     urn_parser = flavor_module.const_get(:UrnParser)
     urn_parser.parse(string)
+  rescue Errors::Error
+    raise
+  rescue StandardError => e
+    raise Errors::UrnParseError,
+          "URN #{string.inspect} does not parse as #{flavor.inspect}: #{e.message}",
+          cause: e
   end
 
   # The slug form (relaton#441): the pubid-scheme URN of +source+ in