From 92bef7daecfe583bda76726daf4cdaa7273b714b Mon Sep 17 00:00:00 2001 From: Simon Cambier Date: Wed, 30 Aug 2023 19:09:15 +0200 Subject: [PATCH] Bugfix/281 backticks (#282) * #281 - Added backticks as a separation character * #281 - Fixed case sensitive duplicates * Fixed highlighting bug, and improved perfs Also added logs --- src/globals.ts | 2 +- src/search/omnisearch.ts | 11 ++++++++--- src/search/query.ts | 10 ++++++---- src/tools/utils.ts | 23 ++++------------------- 4 files changed, 19 insertions(+), 27 deletions(-) diff --git a/src/globals.ts b/src/globals.ts index 2d54f44..bdb93cc 100644 --- a/src/globals.ts +++ b/src/globals.ts @@ -107,4 +107,4 @@ export function isCacheEnabled(): boolean { } export const SPACE_OR_PUNCTUATION = - /[|\t\n\r= -#%-*,.\/:;?@[-\]_{}\u00A0\u00A1\u00A7\u00AB\u00B6\u00B7\u00BB\u00BF\u037E\u0387\u055A-\u055F\u0589\u058A\u05BE\u05C0\u05C3\u05C6\u05F3\u05F4\u0609\u060A\u060C\u060D\u061B\u061E\u061F\u066A-\u066D\u06D4\u0700-\u070D\u07F7-\u07F9\u0830-\u083E\u085E\u0964\u0965\u0970\u09FD\u0A76\u0AF0\u0C77\u0C84\u0DF4\u0E4F\u0E5A\u0E5B\u0F04-\u0F12\u0F14\u0F3A-\u0F3D\u0F85\u0FD0-\u0FD4\u0FD9\u0FDA\u104A-\u104F\u10FB\u1360-\u1368\u1400\u166E\u1680\u169B\u169C\u16EB-\u16ED\u1735\u1736\u17D4-\u17D6\u17D8-\u17DA\u1800-\u180A\u1944\u1945\u1A1E\u1A1F\u1AA0-\u1AA6\u1AA8-\u1AAD\u1B5A-\u1B60\u1BFC-\u1BFF\u1C3B-\u1C3F\u1C7E\u1C7F\u1CC0-\u1CC7\u1CD3\u2000-\u200A\u2010-\u2029\u202F-\u2043\u2045-\u2051\u2053-\u205F\u207D\u207E\u208D\u208E\u2308-\u230B\u2329\u232A\u2768-\u2775\u27C5\u27C6\u27E6-\u27EF\u2983-\u2998\u29D8-\u29DB\u29FC\u29FD\u2CF9-\u2CFC\u2CFE\u2CFF\u2D70\u2E00-\u2E2E\u2E30-\u2E4F\u3000-\u3003\u3008-\u3011\u3014-\u301F\u3030\u303D\u30A0\u30FB\uA4FE\uA4FF\uA60D-\uA60F\uA673\uA67E\uA6F2-\uA6F7\uA874-\uA877\uA8CE\uA8CF\uA8F8-\uA8FA\uA8FC\uA92E\uA92F\uA95F\uA9C1-\uA9CD\uA9DE\uA9DF\uAA5C-\uAA5F\uAADE\uAADF\uAAF0\uAAF1\uABEB\uFD3E\uFD3F\uFE10-\uFE19\uFE30-\uFE52\uFE54-\uFE61\uFE63\uFE68\uFE6A\uFE6B\uFF01-\uFF03\uFF05-\uFF0A\uFF0C-\uFF0F\uFF1A\uFF1B\uFF1F\uFF20\uFF3B-\uFF3D\uFF3F\uFF5B\uFF5D\uFF5F-\uFF65]+/u + /[|\t\n\r= -`#%-*,.\/:;?@[-\]_{}\u00A0\u00A1\u00A7\u00AB\u00B6\u00B7\u00BB\u00BF\u037E\u0387\u055A-\u055F\u0589\u058A\u05BE\u05C0\u05C3\u05C6\u05F3\u05F4\u0609\u060A\u060C\u060D\u061B\u061E\u061F\u066A-\u066D\u06D4\u0700-\u070D\u07F7-\u07F9\u0830-\u083E\u085E\u0964\u0965\u0970\u09FD\u0A76\u0AF0\u0C77\u0C84\u0DF4\u0E4F\u0E5A\u0E5B\u0F04-\u0F12\u0F14\u0F3A-\u0F3D\u0F85\u0FD0-\u0FD4\u0FD9\u0FDA\u104A-\u104F\u10FB\u1360-\u1368\u1400\u166E\u1680\u169B\u169C\u16EB-\u16ED\u1735\u1736\u17D4-\u17D6\u17D8-\u17DA\u1800-\u180A\u1944\u1945\u1A1E\u1A1F\u1AA0-\u1AA6\u1AA8-\u1AAD\u1B5A-\u1B60\u1BFC-\u1BFF\u1C3B-\u1C3F\u1C7E\u1C7F\u1CC0-\u1CC7\u1CD3\u2000-\u200A\u2010-\u2029\u202F-\u2043\u2045-\u2051\u2053-\u205F\u207D\u207E\u208D\u208E\u2308-\u230B\u2329\u232A\u2768-\u2775\u27C5\u27C6\u27E6-\u27EF\u2983-\u2998\u29D8-\u29DB\u29FC\u29FD\u2CF9-\u2CFC\u2CFE\u2CFF\u2D70\u2E00-\u2E2E\u2E30-\u2E4F\u3000-\u3003\u3008-\u3011\u3014-\u301F\u3030\u303D\u30A0\u30FB\uA4FE\uA4FF\uA60D-\uA60F\uA673\uA67E\uA6F2-\uA6F7\uA874-\uA877\uA8CE\uA8CF\uA8F8-\uA8FA\uA8FC\uA92E\uA92F\uA95F\uA9C1-\uA9CD\uA9DE\uA9DF\uAA5C-\uAA5F\uAADE\uAADF\uAAF0\uAAF1\uABEB\uFD3E\uFD3F\uFE10-\uFE19\uFE30-\uFE52\uFE54-\uFE61\uFE63\uFE68\uFE6A\uFE6B\uFF01-\uFF03\uFF05-\uFF0A\uFF0C-\uFF0F\uFF1A\uFF1B\uFF1F\uFF20\uFF3B-\uFF3D\uFF3F\uFF5B\uFF5D\uFF5F-\uFF65]+/u diff --git a/src/search/omnisearch.ts b/src/search/omnisearch.ts index a91a684..1c94d66 100644 --- a/src/search/omnisearch.ts +++ b/src/search/omnisearch.ts @@ -188,6 +188,7 @@ export class Omnisearch { return [] } + logDebug('=== New search ===') logDebug('Starting search for', query) let fuzziness: number @@ -299,6 +300,9 @@ export class Omnisearch { // Sort results and keep the 50 best results = results.sort((a, b) => b.score - a.score).slice(0, 50) + if (results.length) + logDebug('First result:', results[0]) + const documents = await Promise.all( results.map(async result => await cacheManager.getDocument(result.id)) ) @@ -306,7 +310,7 @@ export class Omnisearch { // If the search query contains quotes, filter out results that don't have the exact match const exactTerms = query.getExactTerms() if (exactTerms.length) { - logDebug('Filtering with quoted terms') + logDebug('Filtering with quoted terms: ', exactTerms) results = results.filter(r => { const document = documents.find(d => d.path === r.id) const title = document?.path.toLowerCase() ?? '' @@ -353,8 +357,9 @@ export class Omnisearch { warnDebug('Stopped getMatches at', count, 'results') break } - const m = match[2] - if (m) matches.push({ match: m, offset: match.index + 1 }) + logDebug('match :', match) + const m = match[1] + if (m) matches.push({ match: m, offset: match.index }) } // If the query can be found "as is" in the text, put this match first diff --git a/src/search/query.ts b/src/search/query.ts index 5af9ec6..0032c5f 100644 --- a/src/search/query.ts +++ b/src/search/query.ts @@ -77,10 +77,12 @@ export class Query { public getExactTerms(): string[] { return [ - ...new Set([ - ...this.query.text.filter(o => o.split(' ').length > 1), - ...this.#inQuotes, - ]), + ...new Set( + [ + ...this.query.text.filter(o => o.split(' ').length > 1), + ...this.#inQuotes, + ].map(str => str.toLowerCase()) + ), ] } } diff --git a/src/tools/utils.ts b/src/tools/utils.ts index 7854847..a8eacfd 100644 --- a/src/tools/utils.ts +++ b/src/tools/utils.ts @@ -26,11 +26,8 @@ export function highlighter(str: string): string { return `${str}` } -export function highlighterGroups(_substring: string, ...args: any[]) { - // args[0] is the single char preceding args[1], which is the word we want to highlight - if (!!args[1].trim()) - return `${args[0]}${args[1]}` - return '<no content>' +export function highlighterGroups(substring: string, ...args: any[]): string { + return `${substring}` } export function escapeHTML(html: string): string { @@ -91,21 +88,9 @@ export function stringsToRegex(strings: string[]): RegExp { // sort strings by decreasing length, so that longer strings are matched first strings.sort((a, b) => b.length - a.length) - const joined = - '(' + - // Default word split is not applied if the user uses the cm-chs-patch plugin - (getChsSegmenter() - ? '' - : // Split on start of line, spaces, punctuation, or capital letters (for camelCase) - // We also add the hyphen to the list of characters that can split words - settings.splitCamelCase - ? `^|${SPACE_OR_PUNCTUATION.source}|\-|[A-Z]` - : `^|${SPACE_OR_PUNCTUATION.source}|\-`) + - ')' + - `(${strings.map(s => escapeRegex(s)).join('|')})` + const joined = `(${strings.map(s => escapeRegex(s)).join('|')})` - const reg = new RegExp(`${joined}`, 'giu') - return reg + return new RegExp(`${joined}`, 'giu') } export function extractHeadingsFromCache(