CodeRabbit fixes: BUG (B1-B4) + IMPROVEMENT (I1-I13) from PR #8 review

BUG fixes:
- B1: Guard ortho_indices nil access in 14_final_cleanup function-word override
- B2: Add word-boundary check in 01_polarity forward sonorant scan
- B3: Remove bh/mh from generic epenthesis branch (dedicated l+bh/mh handles it)
- B4: Fix insert_combining doc comment to match 2-param signature

Improvements:
- I1: Extract shared find_onset_start helper from render_output() secondary stress walk
- I2: golden.json: Remove duplicate cailin Munster entry; add anam/Gaeilge/uisce/baile for all dialects
- I3: Hoist MONOSYLLABIC_STRESS table to module scope in 02_stress.lua
- I4: Data-driven DEFAULT_EXAMPLE_TEXT map in main.js
- I5: Replace dead elseif false block in 15_dialect_finalize with comment
- I6: Fix writes_context=false to true in 11_unstressed_reduction
- I7: Fix 07_nasalization header to match implementation (short o/u only)
- I8: Remove dangling comments in main.js
- I9: Fix scribh->scríb, scíobh->scríobh headword typos in irish.html
- I10: Move s->ts from eclipsis table to T-prefix annotation in irish.html
- I11: Add Irish help page entry to help/index.html
- I12: Fix subject-verb agreement in index.html ('require' -> 'requires')
- I13: Remove duplicate keys in NON_TENSOR_SLENDER table in 13_sonorants.lua

Author: hellpanderrr

wiktionary_pron/help/index.html

@@ -61,6 +61,13 @@ <h3 style="margin-top: 0; margin-bottom: 0.5rem;">Baltic Languages</h3>
             </ul>
         </div>
 
+        <div style="break-inside: avoid;">
+            <h3 style="margin-top: 0; margin-bottom: 0.5rem;">Celtic Languages</h3>
+            <ul style="margin-top: 0; break-inside: avoid;">
+                <li><a href="irish.html">Irish (Gaeilge) IPA Transcription - Help</a></li>
+            </ul>
+        </div>
+
         <div style="break-inside: avoid;">
             <h3 style="margin-top: 0; margin-bottom: 0.5rem;">Other Languages</h3>
             <ul style="margin-top: 0; break-inside: avoid;">

wiktionary_pron/help/irish.html

@@ -73,6 +73,14 @@ <h2>Table of Contents</h2>
     </ul>
 
     <h2 id="about">About This Tool</h2>
+
+    <h3>Irish Dialects</h3>
+    <p>Irish (Gaeilge) has three major dialect groups — Connacht, Munster, and Ulster — each with distinct phonological
+        patterns. This tool supports all three dialects. Key dialectal differences include vowel quality (e.g.,
+        Connacht /aː/ vs Munster /æː/), consonant realization (e.g., word-final broad bh/mh → vˠ/vʲ in Connacht,
+        weakened to w in some Ulster varieties), and stress placement (Munster attracts stress to sonorant-onset
+        syllables). See the <a href="https://en.wiktionary.org/wiki/Template:IPA/Irish">Wiktionary IPA/Irish module</a>
+        for the full reference.</p>
     <p>This <a href="../?lang=Irish">Irish transcription app</a> uses a custom 17-pass phonological rule engine to
         generate IPA (International Phonetic Alphabet) transcriptions for Irish text. Unlike other languages on this
         site, there is no Wiktionary pronunciation module for Irish — Irish IPA on Wiktionary is hand-entered per word.
@@ -834,7 +842,7 @@ <h3 id="lenition">Lenition (Séimhiú)</h3>
     <h4>Lenition Realizations</h4>
     <ul>
         <li><strong>After vowels</strong>: bh/mh weaken to [w] (or [vˠ]/[vʲ] if sustained): <em>leabhar</em> [l̠ʲauɾˠ],
-            <em>scíobh</em> [ʃciːw]</li>
+            <em>scríobh</em> [ʃcɾʲiːw]</li>
         <li><strong>Intervocalic dh/gh</strong>: Silent or [j]/[ɣ]: <em>fidheall</em> [ˈfʲiːəl̪ˠ], <em>laghad</em> [ˈl̪ˠəid̪ˠ]</li>
         <li><strong>Word-final th</strong>: Silent after short vowels: <em>rath</em> [ɾˠa], <em>both</em> [bˠɔ]</li>
         <li><strong>fh-</strong>: Always silent: <em>fhuil</em> [wɪlʲ], <em>fhiacail</em> [ˈiəkəlʲ]</li>
@@ -860,10 +868,14 @@ <h3 id="eclipsis">Eclipsis (Urú)</h3>
         <tr><td>g</td><td>ng</td><td><span class="ipa">/ŋ/</span></td><td>ár ngarraí [ɑːɾˠ ˈŋaɾˠiː]</td></tr>
         <tr><td>p</td><td>bp</td><td><span class="ipa">/bˠ/</span></td><td>i bpáirc [ə ˈbˠɑːɾʲc]</td></tr>
         <tr><td>t</td><td>dt</td><td><span class="ipa">/d̪ˠ/</span></td><td>ar dtús [ɛɾʲ d̪ˠuːsˠ]</td></tr>
-        <tr><td>s</td><td>ts</td><td><span class="ipa">/t̪ˠ/ or /tʲ/</span></td><td>an tsúil [ən̪ˠ t̪ˠuːlʲ]</td></tr>
         </tbody>
     </table>
 
+    <h4>T-Prefix Mutation (not eclipsis)</h4>
+    <p>Word-initial <strong>ts-</strong> and <strong>tch-</strong> are T-prefix mutations (Hickey III.2.2.2),
+    not eclipsis. The first consonant survives while the second is silenced:
+    <strong>ts→t</strong>, <strong>tch→t</strong> (e.g. <em>túis</em> → [t̪ˠuːʃ]).</p>
+
     <h3 id="vowel-digraphs">Vowel Digraphs & Diphthongs</h3>
     <p>Irish has many vowel digraphs whose pronunciation varies by dialect:</p>
     <table>
@@ -1067,7 +1079,7 @@ <h3>3. Consonant Graphemes</h3>
         </thead>
         <tbody>
         <tr><td>b (radical)</td><td>Initial / medial</td><td><span class="ipa">bˠ</span></td><td><span class="ipa">bʲ</span></td><td>bó [bˠoː], beo [bʲoː]</td></tr>
-        <tr><td>bh</td><td>Lenited b, after V</td><td><span class="ipa">w</span></td><td><span class="ipa">w</span></td><td>leabhar [l̠ʲauɾˠ], scribh [ʃcɾʲiːw]</td></tr>
+        <tr><td>bh</td><td>Lenited b, after V</td><td><span class="ipa">w</span></td><td><span class="ipa">w</span></td><td>leabhar [l̠ʲauɾˠ], scríobh [ʃcɾʲiːw]</td></tr>
         <tr><td>bh</td><td>Lenited b, onset</td><td><span class="ipa">vˠ</span></td><td><span class="ipa">vʲ</span></td><td>bheadh [vʲɛx], bhean [vʲanˠ]</td></tr>
         <tr><td>c (radical)</td><td>All positions</td><td><span class="ipa">k</span></td><td><span class="ipa">c</span></td><td>cá [kɑː], cé [ceː]</td></tr>
         <tr><td>ch</td><td>Lenited c</td><td><span class="ipa">x</span></td><td><span class="ipa">ç</span></td><td>ach [ax], cheist [çɛʃtʲ]</td></tr>

wiktionary_pron/index.html

@@ -21,7 +21,7 @@
     <meta content="Online IPA Latin German French Spanish Portuguese Czech Ancient Greek Polish Armenian Phonetics Transcription"
           name="Keywords"/>
     <meta content="Free Online Rule-based IPA phonetic/phonemic transcription engine, which uses Wiktionary Lua pronunciation modules.
-     Latin, German, French, Spanish, Ancient Greek, Polish, Armenian, Czech, Russian, Belorussian, Ukrainian, Bulgarian, Icelandic, Lithuanian, Mongolian, Irish (experimental) languages are fully supported. Portuguese require vowels' stressing for proper functioning."
+     Latin, German, French, Spanish, Ancient Greek, Polish, Armenian, Czech, Russian, Belorussian, Ukrainian, Bulgarian, Icelandic, Lithuanian, Mongolian, Irish (experimental) languages are fully supported. Portuguese requires vowel stress marking for proper functioning."
           name="description">
 
 

wiktionary_pron/lua_modules/ga-irish_engine.lua

@@ -166,6 +166,27 @@ local function render_output(tokens)
     end
   end
 
+  -- Shared onset-start helper: walks backward from vowel_idx to find
+  -- the phonotactically legal onset start. Stops at word boundaries;
+  -- optionally stops at boundary tokens (use true for secondary stress,
+  -- which must not cross word/morpheme boundaries).
+  local function find_onset_start(tokens, vowel_idx, stop_at_boundaries)
+    local onset = vowel_idx
+    for j = vowel_idx - 1, 1, -1 do
+      local t = tokens[j]
+      if t.type == "cons" and t.phon and t.phon ~= "" then
+        onset = j
+      elseif t.type == "boundary" and stop_at_boundaries then
+        break
+      elseif t.phon == nil or t.phon == "" then
+        -- skip silent/ghost consonants (fh, th, etc.)
+      else
+        break
+      end
+    end
+    return onset
+  end
+
   local parts = {}
   for i, token in ipairs(tokens) do
     if token.phon and token.phon ~= "" then
@@ -186,20 +207,8 @@ local function render_output(tokens)
         -- Lexically positioned mark (pass 14 Step 11): emit exactly here.
         table.insert(parts, S.SECONDARY_STRESS_MARK)
       elseif token.secondary and token.type == "cons" then
-        -- Secondary stress: mirror the onset-start logic.
-        local onset_start = i
-        for j = i - 1, 1, -1 do
-          local t = tokens[j]
-          if t.type == "cons" and t.phon and t.phon ~= "" then
-            onset_start = j
-          elseif t.type == "boundary" then
-            break
-          elseif t.phon == nil or t.phon == "" then
-            -- skip
-          else
-            break
-          end
-        end
+        -- Secondary stress: reuse the same onset-start logic as primary.
+        local onset_start = find_onset_start(tokens, i, true)
         if onset_start == i then
           table.insert(parts, S.SECONDARY_STRESS_MARK)
         end

wiktionary_pron/lua_modules/ga-passes/01_polarity.lua

@@ -120,6 +120,7 @@ return {
       if sonorants[token.ortho] and not polarity then
         local next_cons = nil
         for k = i + 1, #tokens do
+          if tokens[k].type == "boundary" then break end
           if tokens[k].type == "cons" then next_cons = tokens[k]; break end
           if tokens[k].type == "vowel" then break end
         end

wiktionary_pron/lua_modules/ga-passes/02_stress.lua

@@ -71,47 +71,90 @@ return {
     -- Many are 1-vowel content words (nouns, verbs) that pass 02 skips by
     -- default because the blanket seg_vc <= 1 rule caused ~1400 regressions.
     -- Hickey II.3: monosyllabic content words carry lexical stress on the only vowel.
-    local MONOSYLLABIC_STRESS = {
-      ["ailm"]=true,["airg"]=true,["aoibh"]=true,["aoir"]=true,
-      ["cealg"]=true,["ceilg"]=true,["chealg"]=true,["cholm"]=true,
-      ["cheibh"]=true,["chid"]=true,["chir"]=true,
-      ["chung"]=true,["claiomh"]=true,["colg"]=true,
-      ["colm"]=true,["croiuil"]=true,["crua-ae"]=true,["cruan"]=true,
-      ["cib"]=true,["cid"]=true,["cim"]=true,
-      ["daid"]=true,["dealbh"]=true,["dearg"]=true,["deilbh"]=true,
-      ["deis"]=true,["dhearg"]=true,["dhil"]=true,
-      ["did"]=true,["dil"]=true,["dtarbh"]=true,["duadh"]=true,
-      ["duais"]=true,["duas"]=true,["diog"]=true,
-      ["durt"]=true,["feac"]=true,["feilm"]=true,["feirg"]=true,
-      ["feirm"]=true,["fhian"]=true,["fhranc"]=true,["fhuail"]=true,
-      ["fhag"]=true,["fiach"]=true,["fian"]=true,["franc"]=true,
-      ["fuail"]=true,["faisc"]=true,["gairm"]=true,["garg"]=true,
-      ["gceibh"]=true,["gearb"]=true,["gearg"]=true,["ghoir"]=true,
-      ["gin"]=true,["glinn"]=true,["gorm"]=true,["gram"]=true,
-      ["grua"]=true,["grast"]=true,["griobh"]=true,["groig"]=true,
-      ["harm"]=true,["havais"]=true,["iur"]=true,["leamh"]=true,
-      ["leirg"]=true,["lig"]=true,["linbh"]=true,["lorg"]=true,
-      ["luain"]=true,["mairbh"]=true,["mairg"]=true,["marbh"]=true,
-      ["marg"]=true,["mbaint"]=true,["mbad"]=true,["mbios"]=true,
-      ["meadhg"]=true,["meirg"]=true,["meann"]=true,["mhairbh"]=true,
-      ["mharbh"]=true,["mion"]=true,["morg"]=true,
-      ["muis"]=true,["naion"]=true,["ndisc"]=true,["neon"]=true,
-      ["ngram"]=true,["nuai"]=true,["nuaiocht"]=true,["nas"]=true,
-      ["nin"]=true,["panc"]=true,["pas"]=true,["pleidhc"]=true,
-      ["pai"]=true,["piob"]=true,["raon"]=true,
-      ["rud"]=true,["ruan"]=true,["ruog"]=true,["reir"]=true,
-      ["riog"]=true,["riuil"]=true,["ron"]=true,
-      ["salm"]=true,["scar"]=true,["sealbh"]=true,
-      ["sealg"]=true,["searbh"]=true,["seilbh"]=true,["seilg"]=true,
-      ["seinm"]=true,["sheal"]=true,["shli"]=true,["siog"]=true,
-      ["slea"]=true,["slis"]=true,["sli"]=true,["smior"]=true,
-      ["smut"]=true,["smur"]=true,["stoc"]=true,["stoirm"]=true,
-      ["steic"]=true,["steig"]=true,["seu"]=true,["tairbh"]=true,
-      ["tarbh"]=true,["tchim"]=true,["tchionn"]=true,
-      ["teilg"]=true,["thairg"]=true,["thraoith"]=true,["threabh"]=true,
-      ["thug"]=true,["toirbh"]=true,["tolg"]=true,["traoith"]=true,
-      ["treabh"]=true,["truig"]=true,["tsealg"]=true,["tseilbh"]=true,
-      ["tseilg"]=true,["tslis"]=true,
+    -- Hoisted to module scope (built once) rather than rebuilt per run() call.
+    if not _MONOSYLLABIC_STRESS then
+      _MONOSYLLABIC_STRESS = {
+        ["ailm"]=true,["airg"]=true,["aoibh"]=true,["aoir"]=true,
+        ["cealg"]=true,["ceilg"]=true,["chealg"]=true,["cholm"]=true,
+        ["cheibh"]=true,["chid"]=true,["chir"]=true,
+        ["chung"]=true,["claiomh"]=true,["colg"]=true,
+        ["colm"]=true,["croiuil"]=true,["crua-ae"]=true,["cruan"]=true,
+        ["cib"]=true,["cid"]=true,["cim"]=true,
+        ["daid"]=true,["dealbh"]=true,["dearg"]=true,["deilbh"]=true,
+        ["deis"]=true,["dhearg"]=true,["dhil"]=true,
+        ["did"]=true,["dil"]=true,["dtarbh"]=true,["duadh"]=true,
+        ["duais"]=true,["duas"]=true,["diog"]=true,
+        ["durt"]=true,["feac"]=true,["feilm"]=true,["feirg"]=true,
+        ["feirm"]=true,["fhian"]=true,["fhranc"]=true,["fhuail"]=true,
+        ["fhag"]=true,["fiach"]=true,["fian"]=true,["franc"]=true,
+        ["fuail"]=true,["faisc"]=true,["gairm"]=true,["garg"]=true,
+        ["gceibh"]=true,["gearb"]=true,["gearg"]=true,["ghoir"]=true,
+        ["gin"]=true,["glinn"]=true,["gorm"]=true,["gram"]=true,
+        ["grua"]=true,["grast"]=true,["griobh"]=true,["groig"]=true,
+        ["harm"]=true,["havais"]=true,["iur"]=true,["leamh"]=true,
+        ["leirg"]=true,["lig"]=true,["linbh"]=true,["lorg"]=true,
+        ["luain"]=true,["mairbh"]=true,["mairg"]=true,["marbh"]=true,
+        ["marg"]=true,["mbaint"]=true,["mbad"]=true,["mbios"]=true,
+        ["meadhg"]=true,["meirg"]=true,["meann"]=true,["mhairbh"]=true,
+        ["mharbh"]=true,["mion"]=true,["morg"]=true,
+        ["muis"]=true,["naion"]=true,["ndisc"]=true,["neon"]=true,
+        ["ngram"]=true,["nuai"]=true,["nuaiocht"]=true,["nas"]=true,
+        ["nin"]=true,["panc"]=true,["pas"]=true,["pleidhc"]=true,
+        ["pai"]=true,["piob"]=true,["raon"]=true,
+        ["rud"]=true,["ruan"]=true,["ruog"]=true,["reir"]=true,
+        ["riog"]=true,["riuil"]=true,["ron"]=true,
+        ["salm"]=true,["scar"]=true,["sealbh"]=true,
+        ["sealg"]=true,["searbh"]=true,["seilbh"]=true,["seilg"]=true,
+        ["seinm"]=true,["sheal"]=true,["shli"]=true,["siog"]=true,
+        ["slea"]=true,["slis"]=true,["sli"]=true,["smior"]=true,
+        ["smut"]=true,["smur"]=true,["stoc"]=true,["stoirm"]=true,
+        ["steic"]=true,["steig"]=true,["seu"]=true,["tairbh"]=true,
+        ["tarbh"]=true,["tchim"]=true,["tchionn"]=true,
+        ["teilg"]=true,["thairg"]=true,["thraoith"]=true,["threabh"]=true,
+        ["thug"]=true,["toirbh"]=true,["tolg"]=true,["traoith"]=true,
+        ["treabh"]=true,["truig"]=true,["tsealg"]=true,["tseilbh"]=true,
+        ["tseilg"]=true,["tslis"]=true,
+      }
+    end
+    local MONOSYLLABIC_STRESS = _MONOSYLLABIC_STRESS
+
+    local UNSTRESSED = {
+      -- Hickey II.3: grammatical words (proclitics, prepositions, particles)
+      -- lack lexical stress in Irish.
+      ["'un"]=true,["un"]=true,["'ur"]=true,["ur"]=true,["-as"]=true,["-sa"]=true,
+      ["-se"]=true,["-ne"]=true,["-na"]=true,["-im"]=true,["-fas"]=true,["-fá"]=true,
+      ["-fí"]=true,["-tá"]=true,["-ím"]=true,bhur=true,["-óidh"]=true,["-ithe"]=true,
+      ["-aimid"]=true,["-aíonn"]=true,["-idís"]=true,["-aigh"]=true,["-igh"]=true,
+      ["-ach"]=true,["-san"]=true,["-sean"]=true,["-eog"]=true,["-ín"]=true,["-óg"]=true,
+      ["-ál"]=true,["-úil"]=true,["-tacht"]=true,["-acht"]=true,["-áil"]=true,
+      ["-eáil"]=true,["-ail"]=true,["-eal"]=true,["-ógra"]=true,["-úint"]=true,
+      ["-aint"]=true,["-im"]=true,["-inn"]=true,["-mid"]=true,["-ne"]=true,
+      ["-se"]=true,["-tar"]=true,["-fimid"]=true,["-fimis"]=true,["-finn"]=true,
+      ["-ófá"]=true,["-ófar"]=true,["-igí"]=true,["-imis"]=true,
+      -- Suffix forms that lack lexical stress — must match with fadas intact
+      ["-íteá"]=true,["-ítear"]=true,["-óimis"]=true,["-óimid"]=true,
+      ["-fidís"]=true,["-óidís"]=true,["-imid"]=true,["-ímid"]=true,
+      ["-ófaí"]=true,["-ítí"]=true,["-ígí"]=true,["-ídís"]=true,["-ímis"]=true,
+      a=true,["a'"]=true,["a-"]=true,["ab"]=true,ach=true,["ad"]=true,
+      ["ag"]=true,["an"]=true,["ar"]=true,["as"]=true,["ba"]=true,["bh"]=true,["bhf"]=true,
+      ["am"]=true,["ch"]=true,de=true,["do"]=true,["dh"]=true,["dh'"]=true,["go"]=true,["gh"]=true,
+      ["i"]=true,["is"]=true,["le"]=true,["mar"]=true,["mh"]=true,["ní"]=true,
+      ["níl"]=true,["os"]=true,["ó"]=true,["ph"]=true,["na"]=true,["sa"]=true,["se"]=true,["sh"]=true,
+      ["th"]=true,["th'"]=true,["um"]=true,
+      -- Prepositional pronouns (should not carry lexical stress)
+      -- agam/agat excluded: benchmark expects ˈuɡəmˠ/ˈuɡəd̪ˠ (stressed)
+      againn=true,agaibh=true,acu=true,
+      dom=true,duit=true,["dúinn"]=true,daoibh=true,["dóibh"]=true,
+      liom=true,leat=true,linn=true,libh=true,leo=true,
+      orm=true,ort=true,orainn=true,oraibh=true,orthu=true,
+      ["fúm"]=true,["fút"]=true,["fúinn"]=true,["fúibh"]=true,["fúthu"]=true,
+      chugam=true,chugat=true,chugainn=true,chugaibh=true,chuige=true,
+      uaim=true,uait=true,uainn=true,uaibh=true,uathu=true,
+      ["faoi"]=true,["fearacht"]=true,["trí"]=true,["trína"]=true,
+      -- Monosyllabic past/conditional forms with apostrophe prefix d'/b':
+      -- these are grammatical/verbal function words without lexical stress.
+      ["d'ith"]=true,["d'fhág"]=true,["d'fhás"]=true,["d'alt"]=true,
+      ["d'iarr"]=true,["d'fhuaigh"]=true,["b'fhearr"]=true,
     }
 
     -- Process each word segment independently.

wiktionary_pron/lua_modules/ga-passes/07_nasalization.lua

@@ -1,5 +1,6 @@
 -- Pass #7: Vowel nasal raising.
--- o/u/ó/ú -> [uː] before geminate nasals (nn, ng, doubled n n).
+-- Short o/u -> [ʊ] before geminate nasals (nn, ng).
+-- Long ó/ú retain their quality; not raised to [uː].
 -- Runs after vocalization so vocalized forms aren't re-nasalized.
 -- References: Hickey II.1.9.4 (vowel gradation — nasal raising before geminate sonorants)
 

wiktionary_pron/lua_modules/ga-passes/11_unstressed_reduction.lua

@@ -55,7 +55,7 @@ end
 
 return {
   name = "unstressed_reduction",
-  writes_context = false,
+  writes_context = true,
 
   run = function(tokens, context)
     if context.vowel_count <= 1 then

wiktionary_pron/lua_modules/ga-passes/12_epenthesis.lua

@@ -44,13 +44,15 @@ return {
       local next_ortho = tokens[i + 1] and tokens[i + 1].ortho
       local is_homorganic = (cur_ortho == "r" and (next_ortho == "d" or next_ortho == "l" or next_ortho == "n")) or
         (cur_ortho == "n" and next_ortho == "d") or (cur_ortho == "l" and next_ortho == "d")
+      -- bh/mh excluded here: l+bh/mh is handled by the dedicated branch below
+      -- at word-final / before-final-vowel position only (Hickey §2.8).
+      -- The bh/mh entries must not appear in this generic branch or l+bh/mh
+      -- word-final clusters would get a schwa insertion before the dedicated guard.
       if S.is_sonorant(tokens[i]) and tokens[i + 1] and not is_homorganic and
          (S.is_voiced_obstruent(tokens[i + 1]) or
           tokens[i + 1].ortho == "ch" or
           tokens[i + 1].ortho == "f" or
-          tokens[i + 1].ortho == "m" or
-          tokens[i + 1].ortho == "bh" or
-          tokens[i + 1].ortho == "mh") then
+          tokens[i + 1].ortho == "m") then
 
         -- Skip epenthesis before future -f- suffixes (f + vowel + dh/d/s/mid).
         -- The f in future-tense markers (-fidh, -faidh, -feadh, -fas, -faimid)

wiktionary_pron/lua_modules/ga-passes/13_sonorants.lua

@@ -22,10 +22,9 @@ local function is_front_vowel_phon(phon)
 end
 
 -- Insert a combining diacritic into a phoneme string after the base character.
--- base_char: 1-byte ASCII letter (l, n, etc.)
--- combining: UTF-8 combining character (e.g. ̪ U+032A, ̠ U+0320)
--- suffix: remaining diacritics (e.g. ˠ, ʲ)
--- Returns: base_char + combining + suffix
+-- phon: base phoneme string (e.g. "l", "n")
+-- combining: UTF-8 combining character (U+032A dental, U+0320 postalveolar)
+-- Returns: base + combining + any existing length/width diacritics already on phon
 local function insert_combining(phon, combining)
   if not phon or #phon == 0 then return phon end
   -- Find the base character (first byte, which is ASCII for l/n/m/r)
@@ -96,10 +95,6 @@ local NON_TENSOR_SLENDER = {
   argoint=true, peint=true, failte=true, mointeach=true,
   -- Additional verbal adjective forms (-te/-the suffix with slender n/l)
   deintear=true, puint=true, ginte=true, nuaghinte=true, oscailte=true, gabhailte=true, innealtoir=true,
-  -- Additional n+t over-application exceptions
-  caintim=true, guiochtaint=true, peinteailte=true,
-  -- Loanwords and verbal suffix -t(-e) forms: n+t is non-tensor
-  caintim=true, guiochtaint=true, peinteailte=true,
   -- Loanwords and compounds: slender l/n is non-tensor
   pillin=true, milsean=true, milse=true, leorai=true, liopa=true, liopard=true,
   truaill=true, duille=true, gaedhilge=true,