- ${{result.section ? `
${{result.section}}
` : ''}}
+ ${{result.section ? `
${{result.section}}${{phoneticIndicator}}
` : phoneticIndicator}}
${{titleHtml}}
${{result.matchedHeading ? `
${{highlightText(result.matchedHeading, query)}}
` : ''}}
${{previewHtml}}
diff --git a/docs.html b/docs.html
index c5f1523..1044b2a 100644
--- a/docs.html
+++ b/docs.html
@@ -531,6 +531,19 @@
font-size: 0.9rem;
}
+ .search-phonetic-badge {
+ display: inline-block;
+ margin-left: 0.5rem;
+ padding: 0.15rem 0.4rem;
+ background: var(--sage);
+ color: var(--pine);
+ border-radius: 4px;
+ font-size: 0.7rem;
+ font-style: italic;
+ text-transform: none;
+ letter-spacing: normal;
+ }
+
.search-shortcut {
display: none;
margin-left: 0.5rem;
@@ -1217,6 +1230,197 @@
Hemlock Language Manual
let selectedIndex = -1;
let currentResults = [];
+ // Metaphone phonetic encoding algorithm
+ // Converts words to phonetic codes for fuzzy matching
+ function metaphone(word) {
+ if (!word) return '';
+ word = word.toUpperCase();
+ let result = '';
+ let i = 0;
+
+ // Skip initial silent letters
+ if (word.length >= 2) {
+ const start = word.substring(0, 2);
+ if (['KN', 'GN', 'PN', 'AE', 'WR'].includes(start)) {
+ i = 1;
+ }
+ }
+
+ while (i < word.length && result.length < 6) {
+ const c = word[i];
+ const prev = i > 0 ? word[i - 1] : '';
+ const next = i < word.length - 1 ? word[i + 1] : '';
+ const next2 = i < word.length - 2 ? word[i + 2] : '';
+
+ // Skip duplicates except CC
+ if (c === prev && c !== 'C') {
+ i++;
+ continue;
+ }
+
+ switch (c) {
+ case 'A': case 'E': case 'I': case 'O': case 'U':
+ // Vowels only kept at start
+ if (i === 0) result += c;
+ break;
+
+ case 'B':
+ // B silent after M at end
+ if (!(prev === 'M' && i === word.length - 1)) result += 'B';
+ break;
+
+ case 'C':
+ if (next === 'I' && next2 === 'A') {
+ result += 'X'; // CIA -> X
+ } else if (['I', 'E', 'Y'].includes(next)) {
+ result += 'S'; // Soft C
+ } else if (next === 'H') {
+ result += 'X'; // CH -> X
+ i++;
+ } else {
+ result += 'K'; // Hard C
+ }
+ break;
+
+ case 'D':
+ if (next === 'G' && ['I', 'E', 'Y'].includes(next2)) {
+ result += 'J'; // DGE, DGI, DGY -> J
+ i++;
+ } else {
+ result += 'T';
+ }
+ break;
+
+ case 'F':
+ result += 'F';
+ break;
+
+ case 'G':
+ if (next === 'H') {
+ if (!['A', 'E', 'I', 'O', 'U'].includes(next2)) {
+ i++; // GH silent
+ } else {
+ result += 'K';
+ }
+ } else if (next === 'N' && (i === word.length - 2 || (next2 === 'E' && i === word.length - 3))) {
+ // GN at end or GNE at end is silent
+ } else if (['I', 'E', 'Y'].includes(next)) {
+ result += 'J'; // Soft G
+ } else {
+ result += 'K'; // Hard G
+ }
+ break;
+
+ case 'H':
+ // H only if followed by vowel and not after vowel
+ if (['A', 'E', 'I', 'O', 'U'].includes(next) && !['A', 'E', 'I', 'O', 'U'].includes(prev)) {
+ result += 'H';
+ }
+ break;
+
+ case 'J':
+ result += 'J';
+ break;
+
+ case 'K':
+ if (prev !== 'C') result += 'K';
+ break;
+
+ case 'L':
+ result += 'L';
+ break;
+
+ case 'M':
+ result += 'M';
+ break;
+
+ case 'N':
+ result += 'N';
+ break;
+
+ case 'P':
+ if (next === 'H') {
+ result += 'F';
+ i++;
+ } else {
+ result += 'P';
+ }
+ break;
+
+ case 'Q':
+ result += 'K';
+ break;
+
+ case 'R':
+ result += 'R';
+ break;
+
+ case 'S':
+ if (next === 'H') {
+ result += 'X';
+ i++;
+ } else if (next === 'I' && ['O', 'A'].includes(next2)) {
+ result += 'X'; // SIO, SIA -> X
+ } else {
+ result += 'S';
+ }
+ break;
+
+ case 'T':
+ if (next === 'I' && ['O', 'A'].includes(next2)) {
+ result += 'X'; // TIO, TIA -> X
+ } else if (next === 'H') {
+ result += '0'; // TH -> 0 (theta)
+ i++;
+ } else if (next !== 'C' || next2 !== 'H') {
+ result += 'T';
+ }
+ break;
+
+ case 'V':
+ result += 'F';
+ break;
+
+ case 'W':
+ if (['A', 'E', 'I', 'O', 'U'].includes(next)) result += 'W';
+ break;
+
+ case 'X':
+ result += 'KS';
+ break;
+
+ case 'Y':
+ if (['A', 'E', 'I', 'O', 'U'].includes(next)) result += 'Y';
+ break;
+
+ case 'Z':
+ result += 'S';
+ break;
+ }
+ i++;
+ }
+ return result;
+ }
+
+ // Get phonetic codes for all words in a string
+ function getPhoneticCodes(text) {
+ if (!text) return [];
+ const words = text.toLowerCase().replace(/[^a-z\s]/g, '').split(/\s+/);
+ return words.filter(w => w.length > 2).map(w => ({ word: w, code: metaphone(w) }));
+ }
+
+ // Check if query matches any word phonetically
+ function phoneticMatch(queryCodes, targetCodes) {
+ for (const qc of queryCodes) {
+ for (const tc of targetCodes) {
+ if (qc.code && tc.code && qc.code === tc.code && qc.word !== tc.word) {
+ return tc.word; // Return the matched word for highlighting
+ }
+ }
+ }
+ return null;
+ }
+
// Build search index from PAGES
function buildSearchIndex() {
const index = [];
@@ -1248,13 +1452,19 @@
Hemlock Language Manual
.trim()
.substring(0, 200);
+ // Compute phonetic codes for title and headings
+ const titleCodes = getPhoneticCodes(title);
+ const headingCodes = headings.flatMap(h => getPhoneticCodes(h));
+
index.push({
title: title,
pageId: page.id,
section: section,
headings: headings,
content: page.content.toLowerCase(),
- preview: preview
+ preview: preview,
+ titleCodes: titleCodes,
+ headingCodes: headingCodes
});
}
return index;
@@ -1267,12 +1477,14 @@
Hemlock Language Manual
if (!query || query.length < 2) return [];
const q = query.toLowerCase().trim();
+ const queryCodes = getPhoneticCodes(q);
const results = [];
for (const item of searchIndex) {
let score = 0;
let matchedHeading = null;
let matchContext = '';
+ let phoneticMatchWord = null;
// Check title (highest priority)
const titleLower = item.title.toLowerCase();
@@ -1309,6 +1521,32 @@
Hemlock Language Manual
if (end < item.content.length) matchContext = matchContext + '...';
}
+ // Phonetic matching (finds similar-sounding words even with typos)
+ if (score === 0 && queryCodes.length > 0) {
+ // Check title phonetically
+ const titleMatch = phoneticMatch(queryCodes, item.titleCodes);
+ if (titleMatch) {
+ score = 35; // Phonetic title match
+ phoneticMatchWord = titleMatch;
+ }
+
+ // Check headings phonetically
+ if (!titleMatch) {
+ const headingMatch = phoneticMatch(queryCodes, item.headingCodes);
+ if (headingMatch) {
+ score = Math.max(score, 25); // Phonetic heading match
+ phoneticMatchWord = headingMatch;
+ // Find the heading containing this match
+ for (const heading of item.headings) {
+ if (heading.toLowerCase().includes(headingMatch)) {
+ matchedHeading = heading;
+ break;
+ }
+ }
+ }
+ }
+ }
+
if (score > 0) {
results.push({
title: item.title,
@@ -1317,7 +1555,8 @@
Hemlock Language Manual
score: score,
matchedHeading: matchedHeading,
preview: matchContext || item.preview,
- query: q
+ query: q,
+ phoneticMatch: phoneticMatchWord
});
}
}
@@ -1343,13 +1582,20 @@
Hemlock Language Manual
}
const html = results.map((result, index) => {
- const titleHtml = highlightText(result.title, query);
- const previewHtml = highlightText(result.preview, query);
+ // Highlight both query and phonetic match word
+ let titleHtml = highlightText(result.title, query);
+ let previewHtml = highlightText(result.preview, query);
+ if (result.phoneticMatch) {
+ titleHtml = highlightText(titleHtml, result.phoneticMatch);
+ previewHtml = highlightText(previewHtml, result.phoneticMatch);
+ }
const selectedClass = index === selectedIndex ? ' selected' : '';
+ const phoneticIndicator = result.phoneticMatch ?
+ `
sounds like "${result.phoneticMatch}"` : '';
return `
- ${result.section ? `
${result.section}
` : ''}
+ ${result.section ? `
${result.section}${phoneticIndicator}
` : phoneticIndicator}
${titleHtml}
${result.matchedHeading ? `
${highlightText(result.matchedHeading, query)}
` : ''}
${previewHtml}