vicary 0.2.17 → 0.2.19

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 381ed16c17b2aa427a72b173af92825ad08db621b13e562fdc7977d41d760a26
-  data.tar.gz: 9e7a68508af7f405888c9a0e5ebc9b7993b3f0e7de6149c91156cf7c23b9fe29
+  metadata.gz: a5bf0874dcc6fb688fb61bc76441886d82f709b45017a97e41a71a5de67a8be3
+  data.tar.gz: 21b6dcb645bd2a449ba89b8971d25e3c0cf650a6cc6e2515e0daff3c5e22e8a7
 SHA512:
-  metadata.gz: 87ffdbf4edb505a3c1d5faca88c817a5a1ad44b093fbfc329f99958c05a528b078c3fdcc1d97c9e6ffca2d255be617cb3912d6d6e2a9735f02861600aa0f0ce6
-  data.tar.gz: ec871699f441d07106e5f32981eb9a34f35ce2935f75a7f78163fc0ae277dc1afbe8fdd042e3dbf9676b8a4493dc9fb8078c36a0e3ba9cb65804cad79b17c93b
+  metadata.gz: a3f1e0e9ed8bafb67ceb4812dce7c56cac60e3f58fb2fbc2ad0ecd5fce15741ebd9f9404c3ccf5a23912f27f849642bf867d805ad627981fcfcff56ce03162d9
+  data.tar.gz: f86e63597d16e1aa83ba33fa41bfe9a8fdff3ef59ecaa12e2d7fcaec2d38e00b837a0578fbafe25a53b90acb5017cc128f3929d02cfc6ade4cd25f3ea0e05d88

data/assets/MANIFEST.json

@@ -22,11 +22,11 @@
       }
     },
     "stop_words_never_capitalised.txt": {
-      "bytes": 8017,
-      "entries": 499,
+      "bytes": 8108,
+      "entries": 511,
       "format": 1,
       "min_package_version": "0.2.10",
-      "sha256": "23798cb88672c126d71a9d0988b9adf0aa044bdc291847012d76deae71fe1d33"
+      "sha256": "809e92aa74f001533be8bd33bfe119f5edab4bef2fbde687140f16c998c1039f"
     },
     "stop_words_sometimes_capitalised.txt": {
       "bytes": 6351,
@@ -37,5 +37,5 @@
     }
   },
   "manifest_version": 1,
-  "written_by": "vicary 0.2.13"
+  "written_by": "vicary 0.2.17"
 }

data/assets/stop_words_never_capitalised.txt

@@ -1,5 +1,5 @@
 #!lexicon 1
-#!list stop_words_never_capitalised 499
+#!list stop_words_never_capitalised 511
 # Ordinary words that must never become name candidates, and that English never
 # capitalises in the middle of a sentence.
 #
@@ -58,6 +58,7 @@
   a an the this that these those there here it its it's
   i me my mine myself we us our ours ourselves you your yours
   he him his she her hers they them their theirs who whom whose which what
+  how why where wherever whatever whoever
   and or but so because although though however therefore thus hence yet
   if then else when while until since before after during once whenever
   for from to into onto out off over under above below between among across
@@ -109,18 +110,19 @@
   cants coulds didnts dids doeses doings donts durings eithers elses etcs
   everies everybodies everyones everythings everywheres feels fews finallies
   finds froms gaves gets gives gos gots growings hads hases havings helpeds
-  hences hims hises howevers ids ifs ills ins insteads intos ises isnts justs
-  keeps kepts knews knows learneds learnings learns leaves lefts lets lookeds
-  lookings looks makes maybes mights mosts movings muches musts myselfs
-  neededs neithers nobodies nos nothings nots nowheres nows offs oftens okays
-  onces onlies ons ontos ourselveses outs perhapses playings pluses puts
-  quites rarelies rathers reads reallies remembereds remembers runnings saids
-  saws says sets severals shalls shes shoulds sinces sittings somebodies
-  someones somethings sometimeses somewheres standings starteds stoppeds
-  suches talkings taughts teaches tells thans thems thens therefores thes
-  theses thinkings thinks thises thoses thoughs thoughts throughs thuses
+  hences hims hises howevers hows ids ifs ills ins insteads intos ises isnts
+  justs keeps kepts knews knows learneds learnings learns leaves lefts lets
+  lookeds lookings looks makes maybes mights mosts movings muches musts
+  myselfs neededs neithers nobodies nos nothings nots nowheres nows offs
+  oftens okays onces onlies ons ontos ourselveses outs perhapses playings
+  pluses puts quites rarelies rathers reads reallies remembereds remembers
+  runnings saids saws says sets severals shalls shes shoulds sinces sittings
+  somebodies someones somethings sometimeses somewheres standings starteds
+  stoppeds suches talkings taughts teaches tells thans thems thens therefores
+  thes theses thinkings thinks thises thoses thoughs thoughts throughs thuses
   todays togethers tolds tomorrows toos trieds tries tryings unders unlesses
   untils ups uses usings usuallies veries versuses walkings wanteds wants
-  wases wents weres whenevers whens whethers whiches whoms whos whoses withins
-  withouts wonts workeds woulds writes wrotes yeahs yeses yesterdays yets yous
+  wases wents weres whatevers whenevers whens wheres wherevers whethers
+  whiches whies whoevers whoms whos whoses withins withouts wonts workeds
+  woulds writes wrotes yeahs yeses yesterdays yets yous
 # <<< end generated inflections

data/lib/vicary/candidates.rb

@@ -257,6 +257,16 @@ module Vicary
     # Matches the abbreviation a break candidate sits directly behind.
     TRAILING_WORD = /([A-Za-z]+)\.\s*\z/
 
+    # Words that name the letter after them as a letter: "a capital C", "a big M".
+    # Feedback on spelling and capitals says this all the time, and the letter
+    # then stood alone mid-sentence with nothing to tell it from an initial.
+    # Measured on 696 generated replies: 4 of the 39 placeholders a host's guard
+    # had to remove.
+    LETTER_NAMING_WORDS = Set.new(%w[capital big small lowercase uppercase letter]).freeze
+
+    # The word directly before a span, read back over whitespace and an opening quote.
+    PRECEDING_WORD = /([A-Za-z]+)[\s'"‘“]*\z/
+
     # {SENTENCE_BREAK} with the `\A` arm removed, so it cannot match empty.
     #
     # A dialect difference, and it moved a number. `SENTENCE_BREAK` matches the
@@ -925,6 +935,18 @@ module Vicary
       # Reading the document's own orthography
       # ---------------------------------------------------------------------
 
+      # A lone capital that is a label or a letter, never an initial: glued to a
+      # digit (`E9`, a quote label the word pattern used to cut at the digit) or
+      # named as a letter by the word before it ({LETTER_NAMING_WORDS}). Only a
+      # single character is ever released; `Jake2009` stays a candidate.
+      def letter_not_a_name?(text, start, finish)
+        return false unless finish - start == 1
+        return true if finish < text.length && text[finish].match?(/[0-9]/)
+
+        preceding = PRECEDING_WORD.match(text[0...start])
+        !preceding.nil? && LETTER_NAMING_WORDS.include?(preceding[1].downcase)
+      end
+
       # Offsets at which a sentence begins.
       #
       # A break directly behind a title abbreviation is not one — see
@@ -1971,6 +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
@@ -2078,6 +2124,7 @@ module Vicary
 
             start = m.begin(0) + offset
             next if is_protected.call(start, start + joined.length)
+            next if letter_not_a_name?(text, start, start + joined.length)
 
             # Cheapest of the three capital-discounting rules and the only one
             # that reads neither position nor list, so it goes first.
@@ -2143,6 +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 = "0.2.17"
+  VERSION = "0.2.19"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: vicary
 version: !ruby/object:Gem::Version
-  version: 0.2.17
+  version: 0.2.19
 platform: ruby
 authors:
 - Blake Thomas
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-10-21 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