vicary 0.2.18 → 0.2.19

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 96894f65cf1e88e86b89131843ae808215e94c6df4ce8e62b47286fd8ba60431
-  data.tar.gz: 3054de303e6ba58b370a27d34d96b046ba27878f2bcbe8e40efa0c421c09f1b3
+  metadata.gz: a5bf0874dcc6fb688fb61bc76441886d82f709b45017a97e41a71a5de67a8be3
+  data.tar.gz: 21b6dcb645bd2a449ba89b8971d25e3c0cf650a6cc6e2515e0daff3c5e22e8a7
 SHA512:
-  metadata.gz: 80246adb714cb8a2d32794187fafdd80cc58b573bf633cf5f6c89528f89cc71624e1bba013eea26f6b0c7fdf238f8d7e033b25391b60fd7338c7439ad8e823fd
-  data.tar.gz: 9c1b2a572734f144b49cc770be0d731f4b3559d88d020b9a6be77d78c9dd24faa1a25219a5a1274e46c1e3697c87c569e36d7b9228acd84769e24fcb1faed1ce
+  metadata.gz: a3f1e0e9ed8bafb67ceb4812dce7c56cac60e3f58fb2fbc2ad0ecd5fce15741ebd9f9404c3ccf5a23912f27f849642bf867d805ad627981fcfcff56ce03162d9
+  data.tar.gz: f86e63597d16e1aa83ba33fa41bfe9a8fdff3ef59ecaa12e2d7fcaec2d38e00b837a0578fbafe25a53b90acb5017cc128f3929d02cfc6ade4cd25f3ea0e05d88

data/lib/vicary/candidates.rb

@@ -1993,6 +1993,30 @@ module Vicary
         out
       end
 
+      # A title abbreviation and the word it introduces, the title in any case.
+      # With a period every title in the set counts; without one only `mr` and
+      # `mrs`, because a bare `ms` or `dr` is as often "ms paint" or "dr" for drive.
+      TITLED_WORD_RE = /\b(?:(?:mrs|mr|ms|dr|prof)\.[ \t]+|(?:mrs|mr)[ \t]+)([A-Za-z][A-Za-z'’]*(?:-[A-Za-z][A-Za-z'’]*)*)/i
+
+      # A surname written in lowercase behind a title: "mr. kowalski". The
+      # lowercase route seeds on the given-name tier, so a surname alone never
+      # reaches it, and the capitalised route needs the capital the writer left
+      # off. A title is the one place the word after it is a name whatever its
+      # case, so the span is the title and that one word. A stop word or a single
+      # letter after the title is skipped.
+      def find_titled_lowercase_candidates(text, protected_span, settlement = nil)
+        out = []
+        each_match(text, TITLED_WORD_RE) do |m|
+          word = m[1]
+          start = m.begin(0)
+          stop = start + m[0].length
+          next if word.length < 2 || stop?(word.downcase) || protected_span.call(start, stop)
+
+          out << Candidate.new(m[0], start, stop, classify(words(m[0]), settlement))
+        end
+        out
+      end
+
       # Every name-shaped span, before any notability decision.
       #
       # High recall and deliberately poor precision — precision is what the
@@ -2166,6 +2190,12 @@ module Vicary
           ).each do |candidate|
             next if claimed.any? { |s, e| candidate.start < e && candidate.end > s }
 
+            out << candidate
+          end
+          claimed = out.map { |candidate| [candidate.start, candidate.end] }
+          find_titled_lowercase_candidates(text, is_protected, settlement).each do |candidate|
+            next if claimed.any? { |s, e| candidate.start < e && candidate.end > s }
+
             out << candidate
           end
         end

data/lib/vicary/version.rb

@@ -6,5 +6,5 @@ module Vicary
   # Shared across all three front doors on purpose: one detector, one number. A
   # gem 0.3.0 that corresponds to nothing on PyPI cannot be reasoned about, and
   # the parity claim is between *versions*, not between package names.
-  VERSION = "0.2.18"
+  VERSION = "0.2.19"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: vicary
 version: !ruby/object:Gem::Version
-  version: 0.2.18
+  version: 0.2.19
 platform: ruby
 authors:
 - Blake Thomas
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-10-22 00:00:00.000000000 Z
+date: 2026-10-28 00:00:00.000000000 Z
 dependencies: []
 description: |
   Finds the names a student writes about — classmates, teachers, relatives — and