@jterrazz/intelligence 4.2.0 → 6.0.0

package/dist/formatting.js

@@ -0,0 +1,2819 @@
+import { createRequire } from "node:module";
+import { readFileSync } from "node:fs";
+import path from "node:path";
+import nspell from "nspell";
+//#region src/formatting/clean-ai-text.ts
+const INVISIBLE_CHARS_RE = /[\u00AD\u180E\u200B-\u200C\u200E-\u200F\u202A-\u202E\u2060-\u2064\u2066-\u2069\uFEFF]/g;
+const ASCII_CTRL_RE = /[\u0000-\u0008\u000B\u000C\u000E-\u001F\u007F]/g;
+const SPACE_LIKE_RE = /[\u00A0\u1680\u2000-\u200A\u202F\u205F\u3000]/g;
+const MULTIPLE_SPACES_RE = / {2,}/g;
+const CR_RE = /\r\n?/g;
+const CITATION_RE = / *\(oaicite:\d+\)\{index=\d+\}/g;
+const EM_DASH_SEPARATOR_RE = /\s*[—–―‒]\s*/g;
+const TYPOGRAPHY_REPLACEMENTS = [
+	{
+		pattern: /[\u2018\u2019\u201A]/g,
+		replacement: "'"
+	},
+	{
+		pattern: /[\u201C\u201D\u201E]/g,
+		replacement: "\""
+	},
+	{
+		pattern: /\u2026/g,
+		replacement: "..."
+	},
+	{
+		pattern: /[\u2022\u25AA-\u25AB\u25B8-\u25B9\u25CF]/g,
+		replacement: "-"
+	}
+];
+/**
+* Cleans AI-generated text by removing AI artifacts and normalizing typography.
+*
+* @param text - The text to clean
+* @param options - Cleaning options
+* @returns The cleaned text
+*/
+function cleanAiText(text, options = {}) {
+	const { normalizeEmDashesToCommas = true, collapseSpaces = true } = options;
+	if (!text) return "";
+	let result = text;
+	result = result.replace(CR_RE, "\n");
+	result = result.replace(CITATION_RE, "");
+	result = result.normalize("NFKC");
+	result = result.replace(INVISIBLE_CHARS_RE, "");
+	result = result.replace(ASCII_CTRL_RE, "");
+	if (normalizeEmDashesToCommas) result = result.replace(EM_DASH_SEPARATOR_RE, ", ");
+	result = result.replace(SPACE_LIKE_RE, " ");
+	for (const { pattern, replacement } of TYPOGRAPHY_REPLACEMENTS) result = result.replace(pattern, replacement);
+	if (collapseSpaces) result = result.replace(MULTIPLE_SPACES_RE, " ").trim();
+	return result;
+}
+//#endregion
+//#region src/formatting/preserved-terms.generated.ts
+/**
+* GENERATED FILE — do not edit, run `npm run generate:terms`.
+*
+* Source: @cspell/dict-companies + @cspell/dict-software-terms, filtered
+* through the `dictionary-en` ambiguity check (see
+* scripts/generate-preserved-terms.ts for the full pipeline).
+*
+* Candidates before ambiguity filter: 3053
+* Excluded as ambiguous (lowercase form is a common English word): 516
+* Kept: 2512
+*
+* Examples of ambiguity exclusions:
+*   ADMIN -> "admin" (computing-acronyms.txt)
+*   DB -> "db" (computing-acronyms.txt)
+*   DRAM -> "dram" (computing-acronyms.txt)
+*   RAM -> "ram" (computing-acronyms.txt)
+*   REF -> "ref" (computing-acronyms.txt)
+*   UHF -> "uhf" (computing-acronyms.txt)
+*   WAN -> "wan" (computing-acronyms.txt)
+*   ZIP -> "zip" (computing-acronyms.txt)
+*   A -> "a" (companies.txt)
+*   AD -> "ad" (companies.txt)
+*   AT -> "at" (companies.txt)
+*   Adobe -> "adobe" (companies.txt)
+*/
+const GENERATED_PRESERVED_TERMS = [
+	"2GIS",
+	"3M",
+	"AAA",
+	"AAAI",
+	"Aachen",
+	"Aalto",
+	"AAPL",
+	"Aarhus",
+	"ABB",
+	"Abbott",
+	"ABBV",
+	"AbbVie",
+	"ABBYY",
+	"Abdulaziz",
+	"Abdullah",
+	"Aberdeen",
+	"ABIOMED",
+	"Abios",
+	"AbiosGaming",
+	"ABMD",
+	"ABNF",
+	"Accenture",
+	"ACFS",
+	"Ackermann",
+	"ACM",
+	"ACS",
+	"Activision",
+	"ADBE",
+	"Adblock",
+	"Adecco",
+	"Adelaide",
+	"AdGuard",
+	"ADI",
+	"Adidas",
+	"ADOT",
+	"ADSK",
+	"Adsterra",
+	"Advanced",
+	"AECOM",
+	"Aegon",
+	"AEON",
+	"AES",
+	"AES128",
+	"Aetna",
+	"Aeva",
+	"Affero",
+	"AFLAC",
+	"Agari",
+	"AGCO",
+	"agentLess",
+	"Agilent",
+	"Agricole",
+	"Ahold",
+	"AICPA",
+	"Airbnb",
+	"Aires",
+	"Airgas",
+	"Airlines",
+	"Airways",
+	"Aisin",
+	"AK",
+	"AKAM",
+	"Akamai",
+	"Alaska",
+	"Albemarle",
+	"Alberta",
+	"Alcoa",
+	"Aldi",
+	"Alexandria",
+	"Alfresa",
+	"ALGN",
+	"Alipay",
+	"AlixPartners",
+	"ALLE",
+	"Alleghany",
+	"Allegion",
+	"Allen",
+	"Allergan",
+	"Alliant",
+	"Allianz",
+	"Allstate",
+	"Alstom",
+	"Altri",
+	"Altria",
+	"AMAT",
+	"Ambientia",
+	"Amcor",
+	"AMCR",
+	"Ameren",
+	"America",
+	"American",
+	"Ameriprise",
+	"AmerisourceBergen",
+	"AMETEK",
+	"Amgen",
+	"AMGN",
+	"Amherst",
+	"Amphenol",
+	"AMR",
+	"Amsterdam",
+	"Amtrack",
+	"AMZN",
+	"Anadarko",
+	"Analytics",
+	"Andersons",
+	"Andes",
+	"Andrews",
+	"ANET",
+	"Anglo",
+	"Anheuser-Busch",
+	"Anixter",
+	"ANSS",
+	"Ansteel",
+	"ANSYS",
+	"ANTM",
+	"Antwerp",
+	"Aon",
+	"Apache",
+	"Apartments",
+	"Appetize",
+	"Applied",
+	"Appointedd",
+	"Aprilia",
+	"Aptiv",
+	"APTV",
+	"Aramark",
+	"ArcelorMittal",
+	"Archer-Daniels-Midland",
+	"Argentaria",
+	"Arista",
+	"Arize",
+	"Arizona",
+	"Armour",
+	"ARNES",
+	"Arthur",
+	"Arts",
+	"arXiv",
+	"Asana",
+	"Asbury",
+	"Ashland",
+	"ASLR",
+	"Assicurazioni",
+	"Associates",
+	"Assurances",
+	"Assurant",
+	"Aston",
+	"AstraZeneca",
+	"ATAPI",
+	"Atlassian",
+	"Atmos",
+	"ATVI",
+	"Auchan",
+	"Auckland",
+	"Audi",
+	"Austin",
+	"Australia",
+	"Australian",
+	"Auto-Owners",
+	"autoClose",
+	"Autodesk",
+	"Autoliv",
+	"autoLoader",
+	"Automattic",
+	"autoMock",
+	"autoMocking",
+	"autoMocks",
+	"AutoNation",
+	"autoScroll",
+	"autoSpy",
+	"AutoZone",
+	"AvalonBay",
+	"Avery",
+	"AVGO",
+	"Avis",
+	"Avito",
+	"Aviv",
+	"Aviva",
+	"Avnet",
+	"Avon",
+	"AXA",
+	"azureCR",
+	"azureRM",
+	"azureTools",
+	"azureWebsites",
+	"Baden-Wurttemberg",
+	"BAE",
+	"Bahls",
+	"Bahn",
+	"Bakeries",
+	"Baltimore",
+	"Banco",
+	"Bancorp",
+	"Bancshares",
+	"Bandung",
+	"Bangalore",
+	"Bankengruppe",
+	"Bankin",
+	"Banks",
+	"Baosteel",
+	"Barbara",
+	"Barcelona",
+	"Barclays",
+	"Barnes",
+	"baseAddress",
+	"Basecamp",
+	"Basel",
+	"baseNode",
+	"baseObj",
+	"basePath",
+	"basePtr",
+	"BASF",
+	"Bauman",
+	"Baxter",
+	"Bayer",
+	"Bazaarvoice",
+	"BB",
+	"Becton",
+	"Beihang",
+	"Beijing",
+	"Beirut",
+	"Belarusian",
+	"Belfast",
+	"Bemis",
+	"Ben",
+	"Benefits",
+	"Bentley",
+	"Bergen",
+	"Berkeley",
+	"Berkley",
+	"Berksfile",
+	"Berkshire",
+	"Berlin",
+	"Bern",
+	"Bertelsmann",
+	"Bharat",
+	"BHP",
+	"BIIB",
+	"Bilbao",
+	"Billiton",
+	"Bing",
+	"Bio-Rad",
+	"Biogen",
+	"Biomet",
+	"Birmingham",
+	"bitCount",
+	"Bitly",
+	"Bitnami",
+	"Bitrate",
+	"bitSize",
+	"Bjango",
+	"BKNG",
+	"BlackRock",
+	"Blackstone",
+	"blahBlah",
+	"Bloomberg",
+	"Bloomreach",
+	"Bluesky",
+	"Blueway",
+	"BMW",
+	"BNP",
+	"Boeing",
+	"BOM",
+	"Bombay",
+	"Bonn",
+	"Boots",
+	"Booz",
+	"BorgWarner",
+	"Bosch",
+	"Boston",
+	"bottomLeft",
+	"bottomRight",
+	"Bouygues",
+	"BP",
+	"BPCE",
+	"Bradesco",
+	"Brands",
+	"Brandwatch",
+	"Brasil",
+	"Brazilian",
+	"Brewblox",
+	"Bridgestone",
+	"Bridgewater",
+	"Bristol",
+	"Bristol-Myers",
+	"British",
+	"BRK",
+	"Broadcom",
+	"Broadridge",
+	"Brown-Forman",
+	"browserConfig",
+	"Bruijn",
+	"Brunei",
+	"Brunswick",
+	"Brussels",
+	"bSize",
+	"BSON",
+	"BT",
+	"BTN",
+	"BTNS",
+	"Buenos",
+	"bufIndex",
+	"bufSize",
+	"Bugatti",
+	"buildFolder",
+	"Bunge",
+	"Bunnings",
+	"BYOD",
+	"bzFile",
+	"bzipFile",
+	"Cablevision",
+	"Cabot",
+	"Cadillac",
+	"Caesars",
+	"Calendly",
+	"Calgary",
+	"California",
+	"Calpine",
+	"Caltech",
+	"Caltex",
+	"Cambridge",
+	"Camerfirma",
+	"Cameron",
+	"Campbell",
+	"Camunda",
+	"Canada",
+	"Canterbury",
+	"Capgemini",
+	"Capnproto",
+	"Cardiff",
+	"Caremark",
+	"Caribbean",
+	"CarMax",
+	"Carnegie",
+	"Carolina",
+	"CARR",
+	"Carrefour",
+	"Casey's",
+	"Caskroom",
+	"Castparts",
+	"Catalent",
+	"categoryName",
+	"Cathay",
+	"Causely",
+	"CBOE",
+	"CBRE",
+	"CBS",
+	"CDATA",
+	"CDATASection",
+	"CDN",
+	"CDNS",
+	"CDW",
+	"Cegid",
+	"CEIP",
+	"Celanese",
+	"Celgene",
+	"Centene",
+	"CenterPoint",
+	"Centers",
+	"Centrica",
+	"CenturyLink",
+	"Cepton",
+	"CERN",
+	"Cerner",
+	"certId",
+	"certStore",
+	"Certum",
+	"CEST",
+	"CEVA",
+	"CFE",
+	"CFONB",
+	"CGA",
+	"CH2M",
+	"Chalmers",
+	"Chambersign",
+	"changeDir",
+	"changeList",
+	"changeLists",
+	"Charles",
+	"Chartered",
+	"ChefDK",
+	"ChemChina",
+	"Chemicals",
+	"Cheng",
+	"CHEP",
+	"Chesapeake",
+	"Chevrolet",
+	"Chicago",
+	"childNode",
+	"Chile",
+	"Chinese",
+	"Chipotle",
+	"Chocolately",
+	"Cholesky",
+	"Chronosphere",
+	"CHRW",
+	"Chrysler",
+	"CHS",
+	"CHTR",
+	"Chubb",
+	"Chubu",
+	"Chulalongkorn",
+	"CIGNA",
+	"Cincinnati",
+	"Cinemark",
+	"CINF",
+	"Cintas",
+	"Cisco",
+	"Citibank",
+	"CITIC",
+	"Citigroup",
+	"Citizens",
+	"Citrix",
+	"Citylink",
+	"classPath",
+	"Cliffs",
+	"Clojure",
+	"Clorox",
+	"Cloudflare",
+	"Cloudinary",
+	"CMCSA",
+	"CME",
+	"CMOS",
+	"CMS",
+	"CN",
+	"CNAME",
+	"CNCF",
+	"CNP",
+	"Coca-Cola",
+	"Codecentric",
+	"codeId",
+	"codeMirror",
+	"Codento",
+	"codePage",
+	"COFCO",
+	"CoffeeScript",
+	"Colgate-Palmolive",
+	"Collibra",
+	"Colombia",
+	"COLOPL",
+	"colorRef",
+	"Columbia",
+	"Comcast",
+	"Comerica",
+	"Commerzbank",
+	"Communications",
+	"Communities",
+	"Comodo",
+	"Compal",
+	"Companies",
+	"Complutense",
+	"componentManager",
+	"COMPUTERNAME",
+	"Con-way",
+	"ConAgra",
+	"Concho",
+	"ConocoPhillips",
+	"CONSOL",
+	"containerName",
+	"Controls",
+	"Coors",
+	"Copart",
+	"Copenhagen",
+	"Coralogix",
+	"Core-Mark",
+	"CorePack",
+	"Cornell",
+	"Corning",
+	"Corteva",
+	"Cosmo",
+	"Costco",
+	"Coursera",
+	"CPC",
+	"CPRT",
+	"CQRS",
+	"CRD",
+	"Creations",
+	"Credly",
+	"CRH",
+	"Criteo",
+	"crossSite",
+	"Crowdin",
+	"CSAIL",
+	"CSAT",
+	"CSCO",
+	"CSRF",
+	"CST",
+	"csvLine",
+	"CSX",
+	"CTAS",
+	"CTL",
+	"CTLT",
+	"CTSH",
+	"CTVA",
+	"CTXS",
+	"Cummins",
+	"curLoc",
+	"Curtin",
+	"customDomain",
+	"cValue",
+	"cVar",
+	"cVars",
+	"CVE",
+	"CVS",
+	"CWE",
+	"CyberTrust",
+	"Cybozu",
+	"Dachser",
+	"Dai-ichi",
+	"Daimler",
+	"Daiwa",
+	"DAL",
+	"Dana",
+	"Danaher",
+	"Daniels",
+	"Danone",
+	"Darden",
+	"Darmstadt",
+	"Dartmouth",
+	"Darussalam",
+	"DAST",
+	"Databricks",
+	"dataCenter",
+	"dataCenters",
+	"Datadog",
+	"dataFile",
+	"Datanami",
+	"Datang",
+	"dateInterval",
+	"dateObject",
+	"dateTime",
+	"dateTimezone",
+	"datFile",
+	"Datorama",
+	"Davis",
+	"DaVita",
+	"DAWG",
+	"Dayjs",
+	"Dazs",
+	"DBaaS",
+	"DBAL",
+	"dbLog",
+	"DBs",
+	"DCs",
+	"DDMMYY",
+	"DDOG",
+	"DDoS",
+	"DDR",
+	"DDTHH",
+	"Deakin",
+	"debugBreak",
+	"Decker",
+	"DeepScale",
+	"Deere",
+	"Delhaize",
+	"Delhi",
+	"Deloitte",
+	"Denki",
+	"Denmark",
+	"Dennison",
+	"Denso",
+	"Dentsply",
+	"Depot",
+	"Deutsche",
+	"Deutschland",
+	"DEV",
+	"devContainers",
+	"deviceId",
+	"deviceInfo",
+	"deviceIp",
+	"deviceName",
+	"Devices",
+	"Devon",
+	"DEVs",
+	"DexCom",
+	"Dexia",
+	"Dezzai",
+	"DFKI",
+	"Dick's",
+	"Dickinson",
+	"Diego",
+	"Digia",
+	"DigiCert",
+	"Dillard's",
+	"Dior",
+	"DirecTV",
+	"dirId",
+	"dirLen",
+	"dirPath",
+	"DISCA",
+	"DISCK",
+	"Disney",
+	"DLTR",
+	"DocBlock",
+	"docId",
+	"docSearch",
+	"Docusaurus",
+	"docUtil",
+	"docUtils",
+	"DOM",
+	"Domino's",
+	"Domtar",
+	"Dongfeng",
+	"Donnelley",
+	"DOS",
+	"dotNET",
+	"Douyin",
+	"Dover",
+	"Dow",
+	"downloadFile",
+	"downloadFolder",
+	"downloadId",
+	"dPad",
+	"Dr",
+	"Dresden",
+	"driveInfo",
+	"driveName",
+	"DRM",
+	"dSize",
+	"DSNU",
+	"dstFile",
+	"dstStream",
+	"DTE",
+	"Dublin",
+	"Ducati",
+	"DuPont",
+	"Durham",
+	"dValue",
+	"Dwight",
+	"DXC",
+	"DXCM",
+	"Dynatrace",
+	"Dyno",
+	"EADS",
+	"eAPI",
+	"Eastman",
+	"Eaton",
+	"eBPF",
+	"ECDHE",
+	"ECDSA",
+	"ECMA",
+	"Ecolab",
+	"Ecole",
+	"Ecopetrol",
+	"Edeka",
+	"Edinburgh",
+	"Edison",
+	"Edwards",
+	"EET",
+	"EGA",
+	"Egnyte",
+	"Eindhoven",
+	"Electricite",
+	"Eleuther",
+	"Elfsight",
+	"Eli",
+	"EMC",
+	"EMCOR",
+	"Emerson",
+	"Emojipedia",
+	"Emory",
+	"endScroll",
+	"endUnless",
+	"Enel",
+	"Energie",
+	"ENI",
+	"Entergy",
+	"Enterprises",
+	"entryName",
+	"Environnement",
+	"EOG",
+	"EOL",
+	"EOLs",
+	"EPFL",
+	"EPYC",
+	"EQIX",
+	"Equifax",
+	"Equinix",
+	"Equities",
+	"Erasmus",
+	"Ericsson",
+	"Erie",
+	"errorCode",
+	"errorLog",
+	"errorLogs",
+	"ESA",
+	"eSignature",
+	"ESMTPS",
+	"ESNEXT",
+	"Esri",
+	"Essex",
+	"ETag",
+	"ETags",
+	"ETH",
+	"Ethereum",
+	"Etsy",
+	"Ettercap",
+	"Eurasian",
+	"Euris",
+	"eventLog",
+	"Everest",
+	"Evergy",
+	"Eversource",
+	"EVRG",
+	"Exelon",
+	"Exeter",
+	"Exherbo",
+	"EXOR",
+	"EXPD",
+	"EXPE",
+	"Expedia",
+	"Expeditors",
+	"Expensify",
+	"Experian",
+	"Exps",
+	"externalProject",
+	"Exts",
+	"Exxon",
+	"EY",
+	"F5",
+	"FaaS",
+	"Fabra",
+	"Facebook",
+	"Fannie",
+	"Fargo",
+	"Fastenal",
+	"FAW",
+	"FBHS",
+	"FCStone",
+	"featureId",
+	"featureStore",
+	"FedEx",
+	"Fenosa",
+	"Ferrari",
+	"FFIV",
+	"Figma",
+	"fileData",
+	"fileDir",
+	"fileExt",
+	"fileFolder",
+	"fileId",
+	"fileIdx",
+	"fileIndex",
+	"fileInfo",
+	"fileLen",
+	"fileLog",
+	"filePath",
+	"fileUtil",
+	"fileUtils",
+	"Finmeccanica",
+	"Fireblocks",
+	"FirstEnergy",
+	"Fiserv",
+	"FISV",
+	"FITB",
+	"FIXURL",
+	"Flattr",
+	"Flavors",
+	"FleetCor",
+	"Flextronics",
+	"Flickr",
+	"Flipkart",
+	"FLIR",
+	"Florida",
+	"Flowserve",
+	"Fluor",
+	"FMC",
+	"folderInfo",
+	"folderName",
+	"Foncire",
+	"Fonticons",
+	"Foods",
+	"Formosa",
+	"Fortinet",
+	"Fortive",
+	"FOXA",
+	"FPM",
+	"FQDNS",
+	"Fragrances",
+	"frameBuffer",
+	"France",
+	"France-KLM",
+	"Franklin",
+	"Franz",
+	"Freddie",
+	"Freeport-McMoRan",
+	"freeSpace",
+	"Freiburg",
+	"Freightlinx",
+	"Frenet",
+	"Fresenius",
+	"Freshworks",
+	"Fried",
+	"FTNT",
+	"Fudan",
+	"Fujifilm",
+	"Fujitsu",
+	"fullScreen",
+	"Futurice",
+	"fValue",
+	"Gadjah",
+	"Gallagher",
+	"Galway",
+	"gamePad",
+	"GameStop",
+	"Gannett",
+	"Garmin",
+	"Gartner",
+	"GasTerra",
+	"Gazprom",
+	"GCE",
+	"GCESD",
+	"GCM",
+	"GCP",
+	"GCS",
+	"GDF",
+	"GDPR",
+	"Gebruder",
+	"Geely",
+	"Gemfile",
+	"GEMM",
+	"Generale",
+	"Generali",
+	"Geneva",
+	"Genworth",
+	"George",
+	"Georgetown",
+	"Georgia",
+	"GeoTrust",
+	"getFont",
+	"getSel",
+	"Ghent",
+	"GICS",
+	"Gilead",
+	"Giphy",
+	"GitHub",
+	"GitLab",
+	"GLAPI",
+	"Glasgow",
+	"GlaxoSmithKline",
+	"Glencore",
+	"GlobalSign",
+	"GmbH",
+	"GNSS",
+	"gnuLocaleDir",
+	"Gofore",
+	"Goldman",
+	"Goodyear",
+	"GOOG",
+	"GOOGL",
+	"Gothenburg",
+	"GP",
+	"GPIO",
+	"GPKG",
+	"Gradio",
+	"Grafana",
+	"Grainger",
+	"Grammarly",
+	"Graphcool",
+	"Graybar",
+	"Graz",
+	"Greenland",
+	"GRMN",
+	"Groningen",
+	"groupBox",
+	"Groupe",
+	"gRPC",
+	"Grumman",
+	"GS",
+	"GSM",
+	"GUI",
+	"GUID",
+	"GUIs",
+	"Guodian",
+	"Gwangju",
+	"gzFile",
+	"gzipFile",
+	"Hacktoberfest",
+	"Hai",
+	"Hallak",
+	"Halliburton",
+	"Hamilton",
+	"Hanesbrands",
+	"Haniel",
+	"Hanyang",
+	"Hapag-Lloyd",
+	"Harbin",
+	"Harley-Davidson",
+	"Harris",
+	"Hartford",
+	"Harvard",
+	"Hasbro",
+	"HashiCorp",
+	"Hathaway",
+	"HBAN",
+	"hBrush",
+	"HCA",
+	"hCursor",
+	"HD",
+	"HDD",
+	"HDMI",
+	"Healthpeak",
+	"HeBei",
+	"Hebrew",
+	"Hee",
+	"Heidelberg",
+	"Heineken",
+	"Heinz",
+	"helloWorld",
+	"Helsinki",
+	"Henan",
+	"Henry",
+	"Heptio",
+	"Heraeus",
+	"Heriot-Watt",
+	"Hershey",
+	"Hesai",
+	"Hess",
+	"Hetzner",
+	"Hewlett",
+	"Hewlett-Packard",
+	"hFile",
+	"HHMM",
+	"hIcon",
+	"Hilton",
+	"Hindustan",
+	"Hipercard",
+	"Hitachi",
+	"hKey",
+	"hModule",
+	"Hokkaido",
+	"Holcim",
+	"Holdings",
+	"HollyFrontier",
+	"Hologic",
+	"HOLX",
+	"HOMAG",
+	"HOMEDRIVE",
+	"HOMEPATH",
+	"Honda",
+	"Honeywell",
+	"Hong",
+	"Hopkins",
+	"Hormel",
+	"Horton",
+	"Hotels",
+	"Howmet",
+	"howTo",
+	"HPBW",
+	"hrTime",
+	"HSBC",
+	"hScroll",
+	"HSIC",
+	"HSLA",
+	"HSVA",
+	"HTML",
+	"HTTP",
+	"HTTPS",
+	"HTTPSD",
+	"httpSocket",
+	"httpUrl",
+	"Hua",
+	"Huadian",
+	"Huaneng",
+	"Huawei",
+	"Huazhong",
+	"HubSpot",
+	"Hughes",
+	"Hulu",
+	"Humana",
+	"Humboldt",
+	"Hungarian",
+	"Huntington",
+	"Hutchison",
+	"hWin",
+	"hWnd",
+	"Hyundai",
+	"IANA",
+	"Iberdrola",
+	"Icahn",
+	"ICCC",
+	"iCloud",
+	"ICLR",
+	"iData",
+	"Idealo",
+	"Idean",
+	"Idec",
+	"Idemitsu",
+	"IDEX",
+	"Idexx",
+	"IDXX",
+	"IEEE",
+	"IETF",
+	"IHS",
+	"IKEA",
+	"Illinois",
+	"Illumina",
+	"ILMN",
+	"imageSize",
+	"Imgix",
+	"Imgur",
+	"Imperva",
+	"importLib",
+	"INAP",
+	"InBev",
+	"inBuffer",
+	"Income",
+	"INCY",
+	"Incyte",
+	"indexFile",
+	"India",
+	"Indian",
+	"Indonesia",
+	"Industries",
+	"Ineos",
+	"Infisical",
+	"Infomaniak",
+	"Infosys",
+	"ING",
+	"Ingalls",
+	"Ingersoll",
+	"Ingram",
+	"Ingredion",
+	"Initializers",
+	"Innovations",
+	"Innoviz",
+	"Ins",
+	"Instagram",
+	"instanceDir",
+	"Instruments",
+	"Int'l",
+	"INTC",
+	"Integrated",
+	"Integrys",
+	"Interpublic",
+	"intervalObj",
+	"Intesa",
+	"INTL",
+	"INTU",
+	"Invesco",
+	"IOUtil",
+	"IP",
+	"IPG",
+	"IPGP",
+	"ipNet",
+	"iPos",
+	"IQVIA",
+	"Ireland",
+	"IRSA",
+	"Irvine",
+	"isArray",
+	"isCallable",
+	"isDisabled",
+	"ISEA",
+	"isEmpty",
+	"isObject",
+	"Isovalent",
+	"ISRG",
+	"isStr",
+	"isString",
+	"IssueHunt",
+	"issueId",
+	"isUnDef",
+	"Italia",
+	"Italiane",
+	"Itanium",
+	"Itausa-Investimentos",
+	"itemId",
+	"Itochu",
+	"Iubenda",
+	"iVal",
+	"Jabil",
+	"Jacobs",
+	"Jagiellonian",
+	"James",
+	"Jarden",
+	"Jardine",
+	"JavaScript",
+	"JavaScriptReact",
+	"JBHT",
+	"JBoss",
+	"JBS",
+	"JDBC",
+	"Jekylling",
+	"Jekyllrb",
+	"Jemmic",
+	"Jerry's",
+	"JetBlue",
+	"Jetifier",
+	"JFE",
+	"JFrog",
+	"Jiangsu",
+	"Jiaotong",
+	"Jizhong",
+	"JKHY",
+	"JM",
+	"JNPR",
+	"Johns",
+	"Johnson",
+	"Jones",
+	"Joyent",
+	"JPDC",
+	"JPMorgan",
+	"jPos",
+	"JSON",
+	"JSONLD",
+	"JSTOR",
+	"Jude",
+	"Jupyter",
+	"jVal",
+	"JWK",
+	"JWKS",
+	"JWT",
+	"JX",
+	"Kabbage",
+	"Kabushiki",
+	"Kailuan",
+	"Kaisha",
+	"Kalgin",
+	"Kanpur",
+	"Kansai",
+	"Kansas",
+	"Karlsruhe",
+	"Kawasaki",
+	"KBC",
+	"KBR",
+	"KDDI",
+	"KEDA",
+	"Keio",
+	"Kellogg",
+	"Kelly",
+	"Kentucky",
+	"Kerberos",
+	"KeyCloak",
+	"KeyCorp",
+	"keyDown",
+	"keyMenu",
+	"keyOf",
+	"KEYS",
+	"Keysight",
+	"keyVault",
+	"KFC",
+	"KFUPM",
+	"KFW",
+	"Khalifa",
+	"Kia",
+	"Kiewit",
+	"Kimberly-Clark",
+	"Kimco",
+	"Kinder",
+	"King's",
+	"KKR",
+	"KLA",
+	"KLAC",
+	"Knuth",
+	"Ko-fi",
+	"Kobe",
+	"Kohl's",
+	"Komatsu",
+	"Konami",
+	"Kong",
+	"Koo",
+	"Korea",
+	"Kosan",
+	"Kotlin",
+	"KotlinScript",
+	"KPMG",
+	"kPos",
+	"Kraft",
+	"Kragpur",
+	"Kramdown",
+	"Kroger",
+	"KTM",
+	"KU",
+	"Kuehne",
+	"KUKA",
+	"Kung",
+	"KupiVIP",
+	"Kuula",
+	"kVal",
+	"kvCopy",
+	"kvPair",
+	"Kyoto",
+	"Kyung",
+	"Kyushu",
+	"L-3",
+	"L'Oreal",
+	"L3Harris",
+	"Laboratories",
+	"Labs",
+	"Lafarge",
+	"Lamborghini",
+	"Lamoda",
+	"Lancaster",
+	"Lancia",
+	"Landesbank",
+	"Langfuse",
+	"Langtrace",
+	"Las",
+	"Lastfm",
+	"Lauder",
+	"Lauren",
+	"Lausanne",
+	"lButton",
+	"LDOS",
+	"Lear",
+	"Leaseweb",
+	"LEDDAR",
+	"Leeds",
+	"Leggett",
+	"Lego",
+	"Leica",
+	"Leicester",
+	"Leiden",
+	"Leidos",
+	"Leishen",
+	"Lennar",
+	"Lenovo",
+	"Letterboxd",
+	"Leucadia",
+	"Leuven",
+	"Levenshtein",
+	"Levis",
+	"LexisNexis",
+	"Lexus",
+	"LHS",
+	"Liatrio",
+	"libATA",
+	"libDir",
+	"libDirs",
+	"Liberapay",
+	"Librato",
+	"Lidl",
+	"Lifesciences",
+	"Lightstep",
+	"Lilly",
+	"Lincoln",
+	"Linde",
+	"lineDown",
+	"Lines",
+	"LinkCode",
+	"LinkedIn",
+	"linkName",
+	"linkRef",
+	"Liverpool",
+	"Livox",
+	"LKQ",
+	"LLC",
+	"Lloyds",
+	"LOCALEDIR",
+	"localeInfo",
+	"localStorage",
+	"Lockheed",
+	"Loews",
+	"Logitech",
+	"Logz",
+	"Lomonosov",
+	"London",
+	"LONGLONG",
+	"Lonza",
+	"LoongArch",
+	"Loongson",
+	"Lorillard",
+	"Los",
+	"LOSCAM",
+	"Lots",
+	"Loughborough",
+	"Louis",
+	"Lovatt",
+	"Lowe's",
+	"LP",
+	"lParam",
+	"lpBuffer",
+	"lpctStr",
+	"lpDWord",
+	"LPSTR",
+	"lptStr",
+	"lpVoid",
+	"lpWord",
+	"LRCX",
+	"lResult",
+	"lSize",
+	"LSM",
+	"LTS",
+	"LTSV",
+	"Lufthansa",
+	"Lukoil",
+	"Lumibird",
+	"Luminar",
+	"Lumitics",
+	"Lund",
+	"Lunr",
+	"Lutherans",
+	"lVar",
+	"lVars",
+	"LXC",
+	"Lyft",
+	"Lyon",
+	"LyondellBasell",
+	"Maastricht",
+	"Machines",
+	"macOS",
+	"Macquarie",
+	"Macy's",
+	"Mada",
+	"Madrid",
+	"Mae",
+	"Maersk",
+	"Magna",
+	"Mahalanobis",
+	"Mahidol",
+	"Mailchimp",
+	"Malaya",
+	"Malaysia",
+	"Manchester",
+	"ManpowerGroup",
+	"Manulife",
+	"Mapfre",
+	"Marietta",
+	"MarketAxess",
+	"Markets",
+	"Markit",
+	"Marquard",
+	"Marriott",
+	"Marubeni",
+	"Maruhan",
+	"Mary",
+	"Maryland",
+	"Masco",
+	"Maserati",
+	"Massachusetts",
+	"Massey",
+	"MasterCard",
+	"Materials",
+	"Matheson",
+	"Matomo",
+	"Mattel",
+	"Mattermost",
+	"maxCPUs",
+	"Mazda",
+	"mButton",
+	"McCormick",
+	"McDonald's",
+	"McGill",
+	"McGraw",
+	"MCHP",
+	"McKesson",
+	"McLaren",
+	"McLennan",
+	"McMaster",
+	"MDLZ",
+	"MeadWestvaco",
+	"Medco",
+	"MediaWiki",
+	"Medipal",
+	"Medtronic",
+	"Meiji",
+	"Melanox",
+	"Melbourne",
+	"Mellon",
+	"MemDB",
+	"Mendeley",
+	"Mercari",
+	"Mercedes",
+	"Mercedes-Benz",
+	"MERCHANTABILITY",
+	"Merchants",
+	"Merck",
+	"messageId",
+	"messageIds",
+	"messageList",
+	"Metals",
+	"MetLife",
+	"Mettler",
+	"Mexican",
+	"Mexico",
+	"MGM",
+	"Miami",
+	"Michelin",
+	"Michigan",
+	"Microsoft",
+	"Mid-America",
+	"Midjourney",
+	"Migros",
+	"Milano",
+	"Mills",
+	"Minecraft",
+	"Ming",
+	"minMaxInfo",
+	"Minmetals",
+	"Minnesota",
+	"Mirantis",
+	"MITM",
+	"Mitsubishi",
+	"Mitsui",
+	"Mixpanel",
+	"Mizuho",
+	"MKTX",
+	"MLP",
+	"MMDDYY",
+	"MMDDYYYY",
+	"MNST",
+	"Mobil",
+	"Mobis",
+	"mockFile",
+	"mockFn",
+	"mockFunction",
+	"Mockito",
+	"mockPath",
+	"Mohawk",
+	"Mol",
+	"Molina",
+	"Moller-Maersk",
+	"Molson",
+	"Monash",
+	"Mondelez",
+	"Mondiale",
+	"Mondoo",
+	"Mongo",
+	"Monsanto",
+	"Monterey",
+	"Montreal",
+	"Moody's",
+	"Moore's",
+	"Moovit",
+	"Morgan",
+	"Morris",
+	"Morrison",
+	"Moscow",
+	"Motorola",
+	"Motors",
+	"Mouseflow",
+	"mouseWheel",
+	"Movil",
+	"Mozilla",
+	"MRC",
+	"MS",
+	"MSCI",
+	"MSFT",
+	"msgList",
+	"MSSQLSERVER",
+	"multiLog",
+	"multiPath",
+	"Munich",
+	"Murphy",
+	"Musixmatch",
+	"MXIM",
+	"Myer",
+	"Mylan",
+	"Nagel",
+	"Nagoya",
+	"Nanjing",
+	"Nanyang",
+	"Napatech",
+	"Narrowable",
+	"Narrowables",
+	"Nasdaq",
+	"Nava",
+	"Navarra",
+	"Naver",
+	"Navistar",
+	"NCLH",
+	"NCR",
+	"NDA",
+	"NDAQ",
+	"NDJSON",
+	"NEC",
+	"Nederlanden",
+	"Nemours",
+	"NetApp",
+	"Netflix",
+	"Netlify",
+	"Networks",
+	"Neuralink",
+	"Newcastle",
+	"NEWDATE",
+	"NEWDECIMAL",
+	"newDir",
+	"Newell",
+	"newFile",
+	"newId",
+	"Newmont",
+	"Nextel",
+	"NextEra",
+	"nextIdx",
+	"nFile",
+	"NFLX",
+	"Nginx",
+	"Nginxinc",
+	"Nielsen",
+	"NII",
+	"Nijmegen",
+	"Nike",
+	"Nikto",
+	"Nintendo",
+	"Nippon",
+	"NiSource",
+	"Nissan",
+	"NIST",
+	"Nitor",
+	"nKey",
+	"NKSJ",
+	"NLOK",
+	"NLSN",
+	"noCookie",
+	"nodeMap",
+	"NodeNext",
+	"nodeObj",
+	"nodeP",
+	"nodePtr",
+	"Nokia",
+	"Nomura",
+	"nonStatic",
+	"Nordcloud",
+	"Nordea",
+	"Nordstrom",
+	"NordVPN",
+	"NOREF",
+	"NoReply",
+	"Norfolk",
+	"Normale",
+	"Northrop",
+	"NortonLifeLock",
+	"Norwegian",
+	"noSlash",
+	"NOSQL",
+	"notArray",
+	"NOTFOUND",
+	"notObject",
+	"Notre",
+	"Nottingham",
+	"Novartis",
+	"Novell",
+	"Novosibirsk",
+	"NRG",
+	"NTAP",
+	"NTRS",
+	"Nubity",
+	"Nucor",
+	"Nullable",
+	"Nullables",
+	"Nullsoft",
+	"NVDA",
+	"Nvidia",
+	"NVR",
+	"NWSA",
+	"NYSE",
+	"O'Lakes",
+	"O'Reilly",
+	"Oaktree",
+	"OAuthLib",
+	"objectId",
+	"ObservIQ",
+	"ODBC",
+	"ODFL",
+	"Ohio",
+	"OIDC",
+	"Oilwell",
+	"okCancel",
+	"Okta",
+	"OLAP",
+	"oldFile",
+	"OleDLL",
+	"Olly",
+	"OLM",
+	"OLTP",
+	"Omaha",
+	"Omnicare",
+	"Omnicom",
+	"Omnition",
+	"OMV",
+	"oneOf",
+	"ONEOK",
+	"Onex",
+	"Ontario",
+	"OOBE",
+	"OOCL",
+	"OOTB",
+	"Opel",
+	"OpenAI",
+	"Openbase",
+	"OpenDNS",
+	"OpenID",
+	"OpenTelemetry",
+	"Opscode",
+	"ORCL",
+	"orderId",
+	"ORLEN",
+	"ORLY",
+	"ORM",
+	"OSA",
+	"Osaka",
+	"OSDI",
+	"Oshkosh",
+	"OSINT",
+	"Oslo",
+	"OSS",
+	"OSTIF",
+	"Otago",
+	"Otechie",
+	"OTIS",
+	"OTL",
+	"OTLP",
+	"Ottawa",
+	"OU",
+	"outBuffer",
+	"outDir",
+	"Outreachy",
+	"OWASP",
+	"Owens",
+	"Owens-Illinois",
+	"Oy",
+	"Ozforex",
+	"Ozon",
+	"PACCAR",
+	"Packard",
+	"Padova",
+	"pageData",
+	"PagerDuty",
+	"pageView",
+	"pageViews",
+	"Palantir",
+	"Paludis",
+	"Panasonic",
+	"Panner",
+	"Papaki",
+	"PARCO",
+	"parentNode",
+	"Paribas",
+	"Paris",
+	"Parker-Hannifin",
+	"Partners",
+	"Parts",
+	"Patreon",
+	"Paulo",
+	"Paxos",
+	"PAYC",
+	"Paychex",
+	"Paycom",
+	"Paylocity",
+	"Payments",
+	"PayPal",
+	"PAYX",
+	"PBCT",
+	"PBF",
+	"pBuf",
+	"pBuffer",
+	"PCAP",
+	"PCAPNG",
+	"PCAR",
+	"pcFileDir",
+	"pConnection",
+	"PCTL",
+	"PDVSA",
+	"Peabody",
+	"Pemex",
+	"Penh",
+	"Penn",
+	"Penney",
+	"Pennsylvania",
+	"Penske",
+	"Pentair",
+	"Penteston",
+	"People's",
+	"Pepsi",
+	"PepsiCo",
+	"Percona",
+	"PerkinElmer",
+	"Perrigo",
+	"Petersburg",
+	"Petrobras",
+	"Petronas",
+	"PetSmart",
+	"Peugeot",
+	"pFile",
+	"Pfizer",
+	"Pharmaceuticals",
+	"Pharmahandel",
+	"Philip",
+	"Philips",
+	"Phillips",
+	"Phnom",
+	"Photonics",
+	"PHPs",
+	"Pingdom",
+	"Pinterest",
+	"Pioneering",
+	"Pittsburgh",
+	"Pixar",
+	"PKCS",
+	"PKGDATADIR",
+	"PKN",
+	"Plains",
+	"Plans",
+	"Platt",
+	"PLC",
+	"Plex",
+	"PLSQL",
+	"PNC",
+	"Po",
+	"Pohang",
+	"POJO",
+	"POJOs",
+	"Politecnico",
+	"Pompeu",
+	"Pont",
+	"Porsche",
+	"Portainer",
+	"Porto",
+	"POSCO",
+	"POSIX",
+	"postCondition",
+	"Poste",
+	"Postgres",
+	"PPG",
+	"PPL",
+	"Prague",
+	"Praxair",
+	"Precommit",
+	"Preformat",
+	"prevIdx",
+	"PRGO",
+	"Priceline",
+	"Princeton",
+	"printName",
+	"Prisma",
+	"PRNU",
+	"processId",
+	"Processing",
+	"procId",
+	"Procter",
+	"Products",
+	"projectId",
+	"Prologis",
+	"PromQL",
+	"Properties",
+	"Protobuf",
+	"Protobufs",
+	"Proxmox",
+	"PSEG",
+	"PSP",
+	"PSTR",
+	"PTT",
+	"Pty",
+	"Publix",
+	"PulteGroup",
+	"Purdue",
+	"Putra",
+	"pVar",
+	"PVH",
+	"PwC",
+	"PYPL",
+	"Qatar",
+	"QCOM",
+	"Qorvo",
+	"QRVO",
+	"QUALCOMM",
+	"Quanergy",
+	"Queen's",
+	"Queensland",
+	"QuesmaOrg",
+	"Quintiles",
+	"QuoVadis",
+	"Rabobank",
+	"Rackspace",
+	"Ralph",
+	"Randstad",
+	"Raymond",
+	"Raytheon",
+	"rayTrace",
+	"rayTracing",
+	"Razorpay",
+	"RBS",
+	"rButton",
+	"RDP",
+	"Reachability",
+	"readDir",
+	"readError",
+	"readErrorLog",
+	"readErrorLogs",
+	"readErrors",
+	"readFile",
+	"Reaktor",
+	"Realogy",
+	"Rebrandly",
+	"Recv",
+	"Redbud",
+	"Reddit",
+	"Reddit's",
+	"RedHat",
+	"refCount",
+	"refCounting",
+	"referenceCounting",
+	"Refining",
+	"REFLEX",
+	"refNum",
+	"REFs",
+	"refVal",
+	"refValue",
+	"Regeneron",
+	"Regions",
+	"REGN",
+	"regRead",
+	"relDate",
+	"relDateTime",
+	"Relex",
+	"relTime",
+	"Renault",
+	"Rentals",
+	"repoRoot",
+	"Repsol",
+	"Research",
+	"Reserve",
+	"ResMed",
+	"ResNet",
+	"Resorts",
+	"Resources",
+	"Responsys",
+	"Restaurants",
+	"retryCancel",
+	"Revolut",
+	"Reynolds",
+	"RGBIR",
+	"RHS",
+	"Ricoh",
+	"RIEGL",
+	"RIMAS",
+	"Rio",
+	"RMIT",
+	"Robert",
+	"Robinson",
+	"Roblox",
+	"Roche",
+	"Rochester",
+	"Rock-Tenn",
+	"Rockwell",
+	"Rohlig",
+	"rollDown",
+	"Rollins",
+	"Rolls",
+	"Rome",
+	"Rosneft",
+	"Ross",
+	"ROST",
+	"Rotterdam",
+	"Rowe",
+	"rowId",
+	"Royce",
+	"RPATH",
+	"RPC",
+	"RPCs",
+	"RSA",
+	"rSize",
+	"Rubbermaid",
+	"Russia",
+	"Rutgers",
+	"Ruuvi",
+	"RWE",
+	"RWTH",
+	"Ryder",
+	"S-Oil",
+	"Saab",
+	"SaaS",
+	"Sabic",
+	"Sachs",
+	"Safeway",
+	"SAIC",
+	"Sainsbury",
+	"Saint-Gobain",
+	"Sales",
+	"Salesforce",
+	"sameSite",
+	"Samsung",
+	"San",
+	"SanDisk",
+	"Sands",
+	"Sanmina",
+	"Sanofi",
+	"Sanpaolo",
+	"Santa",
+	"Santander",
+	"Sao",
+	"SAST",
+	"SATA",
+	"Saud",
+	"SBA",
+	"SBAC",
+	"Sberbank",
+	"SBOM",
+	"SBUX",
+	"SCADA",
+	"Schein",
+	"schemeId",
+	"Schenker",
+	"Schlumberger",
+	"Schneider",
+	"SCHW",
+	"Schwab",
+	"Sciences",
+	"Scotia",
+	"Scotland",
+	"screenShare",
+	"Scripts",
+	"scrollInfo",
+	"SDRAM",
+	"SE",
+	"Seagate",
+	"Sealed",
+	"Sears",
+	"securityLevel",
+	"Seiki",
+	"Sekiyu",
+	"Semilux",
+	"Semmle",
+	"Sempra",
+	"Sencha",
+	"Sensedia",
+	"Seoul",
+	"Serret",
+	"ServiceNow",
+	"Services",
+	"setAffinity",
+	"setEnv",
+	"setFont",
+	"setRedraw",
+	"setSel",
+	"setText",
+	"SHA256",
+	"Shagang",
+	"Shandong",
+	"Shanxi",
+	"Sheffield",
+	"Shenhua",
+	"Sherwin-Williams",
+	"Shopify",
+	"Shougang",
+	"Showa",
+	"SHV",
+	"sideEffect",
+	"Siemens",
+	"SIGHUP",
+	"SIGINT",
+	"SIGKDD",
+	"SIGKILL",
+	"signOff",
+	"SIGQUIT",
+	"SIGTERM",
+	"SigV4",
+	"SigV4A",
+	"Silae",
+	"Silicom",
+	"SIMD",
+	"Simon",
+	"Singapore",
+	"SinnerSchrader",
+	"Sinochem",
+	"Sinomach",
+	"Sinopec",
+	"Sirona",
+	"Sismology",
+	"Sistema",
+	"SIVB",
+	"SK",
+	"Skaffold",
+	"Skoda",
+	"Skyworks",
+	"SL",
+	"SLA",
+	"sLen",
+	"Slevomat",
+	"SLI",
+	"Slimpay",
+	"SLM",
+	"SLO",
+	"slotId",
+	"Smania",
+	"Smarkets",
+	"SmartyStreets",
+	"Smithfield",
+	"Smucker",
+	"Snap-on",
+	"Snapchat",
+	"Snapple",
+	"SNCF",
+	"SNPS",
+	"Snyk",
+	"Societe",
+	"socketId",
+	"Sodexo",
+	"Softbank",
+	"SoftLayer",
+	"Solaris",
+	"SolarWinds",
+	"Solita",
+	"Solr",
+	"Solutions",
+	"someHost",
+	"Sonera",
+	"Sons",
+	"Sony",
+	"Sorbonne",
+	"sourceIdx",
+	"sourceMap",
+	"sourceMaps",
+	"Southampton",
+	"SpaceX",
+	"Sparc",
+	"SPDX",
+	"SPGI",
+	"Spirent",
+	"Spotify",
+	"SQL",
+	"SQRL",
+	"SquashFS",
+	"Squibb",
+	"SRAM",
+	"srcDir",
+	"srcFile",
+	"srcFolder",
+	"srcStream",
+	"SSD",
+	"SSE",
+	"SSID",
+	"SSP",
+	"SSRF",
+	"Staat",
+	"Stanford",
+	"Stanley",
+	"Staples",
+	"Starbucks",
+	"Starfield",
+	"Starwood",
+	"States",
+	"staticFile",
+	"Staticman",
+	"Staticman's",
+	"Stationers",
+	"Statoil",
+	"stdCall",
+	"stdWin",
+	"Steelers",
+	"Stellantis",
+	"STERIS",
+	"Stimulsoft",
+	"Stockholm",
+	"Stores",
+	"strLen",
+	"Stryker",
+	"Subaru",
+	"subDirective",
+	"subDirectives",
+	"Suez",
+	"Suisse",
+	"Sumitomo",
+	"Suncor",
+	"Sungkyunkwan",
+	"Sunoco",
+	"SunTrust",
+	"Supercomputing",
+	"Superfeedr",
+	"Supermarkets",
+	"Supervalu",
+	"Surgutneftegas",
+	"SUSE",
+	"sUser",
+	"Susser",
+	"Sussex",
+	"Suzuken",
+	"Suzuki",
+	"SVB",
+	"Svc",
+	"SVGA",
+	"SwedenCentral",
+	"Swiftype",
+	"Swinburne",
+	"Swiss",
+	"Swisscom",
+	"SWKS",
+	"Swyftx",
+	"Sydney",
+	"Symantec",
+	"Synnex",
+	"Synopsys",
+	"Sysco",
+	"sysError",
+	"sysKeyDown",
+	"sysRoot",
+	"sysRoots",
+	"Systems",
+	"T-Mobile",
+	"Taiwan",
+	"Take-Two",
+	"tarFile",
+	"Targa",
+	"Tartu",
+	"Tata",
+	"Taylor's",
+	"tBody",
+	"tChar",
+	"tCol",
+	"TCP",
+	"tDiff",
+	"TE",
+	"TechnipFMC",
+	"Technologies",
+	"TechRadar",
+	"Telecom",
+	"Teledyne",
+	"Teleflex",
+	"Telefonica",
+	"Telekom",
+	"Telia",
+	"Telstra",
+	"Tencent",
+	"Tenneco",
+	"Teradyne",
+	"Terex",
+	"Tesco",
+	"Tesla",
+	"Tesoro",
+	"testEnv",
+	"testFile",
+	"testFiles",
+	"testId",
+	"testLog",
+	"testLogs",
+	"testMessage",
+	"testMsg",
+	"testParam",
+	"testParams",
+	"testSuite",
+	"testSuites",
+	"testVal",
+	"testValue",
+	"testVar",
+	"Tewoo",
+	"Texas",
+	"texCoord",
+	"texCoords",
+	"textFile",
+	"textFiles",
+	"Textron",
+	"tFoot",
+	"Thanos",
+	"tHead",
+	"Theming",
+	"Thermo",
+	"Thrivent",
+	"ThyssenKrupp",
+	"TIAA-CREF",
+	"Ticino",
+	"Tidelift",
+	"Tieto",
+	"Tiffany",
+	"TikTok",
+	"timeFrame",
+	"timeFrames",
+	"timeInterval",
+	"timeStep",
+	"timeSteps",
+	"Tinto",
+	"TJX",
+	"tmpDir",
+	"tmpFile",
+	"tmpStr",
+	"TMUS",
+	"TNK-BP",
+	"Todoist",
+	"Tokio",
+	"Tokyo",
+	"Toledo",
+	"Tomsk",
+	"Tongji",
+	"topLeft",
+	"topRight",
+	"Torchmark",
+	"Toronto",
+	"Toronto-Dominion",
+	"Toshiba",
+	"Towers",
+	"Toyota",
+	"Toys",
+	"TPDS",
+	"tPtr",
+	"Traceloop",
+	"Trakt",
+	"Trane",
+	"TransDigm",
+	"Translogix",
+	"TravelCenters",
+	"Travelers",
+	"Trimble",
+	"Tripadvisor",
+	"Truist",
+	"TRW",
+	"TSBuild",
+	"TSBuildInfo",
+	"TSCO",
+	"TSDB",
+	"Tsing",
+	"Tsinghua",
+	"tSize",
+	"TSLA",
+	"TSQL",
+	"TTWO",
+	"TUI",
+	"Tumblr",
+	"Turku",
+	"tValue",
+	"Twente",
+	"Twenty-First",
+	"Twilio",
+	"TWTR",
+	"Tyler",
+	"TypeCheck",
+	"Typedef",
+	"TypeScriptReact",
+	"Tyson",
+	"tzId",
+	"tzObj",
+	"UAB",
+	"UAE",
+	"Uber",
+	"UBS",
+	"UC",
+	"UCIM",
+	"UCLA",
+	"uCoord",
+	"uCoords",
+	"UCSI",
+	"UCUM",
+	"UDR",
+	"UEFA",
+	"UFJ",
+	"UGent",
+	"UGI",
+	"UI",
+	"ULTA",
+	"Ultrapar",
+	"unDef",
+	"UniCredit",
+	"Unilever",
+	"uniqueId",
+	"uniqueIdentifier",
+	"UNIST",
+	"United",
+	"UnitedHealth",
+	"Universiteit",
+	"unixRoot",
+	"unSetter",
+	"Unsplash",
+	"UNSW",
+	"Unum",
+	"Upcloud",
+	"updateTimestamp",
+	"uPos",
+	"Uppsala",
+	"Urbana-Champaign",
+	"URS",
+	"US",
+	"USA",
+	"USENIX",
+	"userId",
+	"Utilities",
+	"UTP",
+	"Utrecht",
+	"UUID",
+	"UUIDs",
+	"uVal",
+	"Vagrantfile",
+	"Valarian",
+	"Valeo",
+	"Valero",
+	"Vanderbilt",
+	"Varco",
+	"Variadic",
+	"Varian",
+	"Vasona",
+	"Vattenfall",
+	"vCursor",
+	"vDSO",
+	"Vegas",
+	"Veikkaus",
+	"Velodyne",
+	"Ventas",
+	"Veolia",
+	"Verdaccio",
+	"Verisec",
+	"Verisign",
+	"Verisk",
+	"Verizon",
+	"VF",
+	"vFile",
+	"VGA",
+	"VGL",
+	"VHS",
+	"VIAC",
+	"Viacom",
+	"ViacomCBS",
+	"Viber",
+	"Victoria",
+	"VictoriaMetrics",
+	"Vienna",
+	"Vimeo",
+	"Vinci",
+	"Vincit",
+	"Virginia",
+	"VirtualBox",
+	"VirusTotal",
+	"Visteon",
+	"Visy",
+	"Vivendi",
+	"Vizcaya",
+	"VLAN",
+	"VLANs",
+	"VLDB",
+	"VMDK",
+	"VMOMI",
+	"VMware",
+	"VMXNET",
+	"VNC",
+	"Vodafone",
+	"Volkswagen",
+	"Volvo",
+	"Vornado",
+	"VPN",
+	"VPS",
+	"VRAM",
+	"VRSK",
+	"VRSN",
+	"VRTX",
+	"VSAN",
+	"vScroll",
+	"vSphere",
+	"vSwitch",
+	"VTEX",
+	"Vulcan",
+	"Vulkan",
+	"Vultr",
+	"Vyatta",
+	"W3C",
+	"Wabtec",
+	"Wacom",
+	"Wageningen",
+	"WAgent",
+	"waitUntil",
+	"Wal-Mart",
+	"Walgreen",
+	"Walgreens",
+	"Walmart",
+	"Walt",
+	"WAP",
+	"Warner",
+	"Warsaw",
+	"Warwick",
+	"Waseda",
+	"Washington",
+	"Watchr",
+	"Waterloo",
+	"Watson",
+	"Waze",
+	"Weaviate",
+	"Webex",
+	"webIndex",
+	"webIndexFile",
+	"WebJS",
+	"WEC",
+	"Wedos",
+	"Weiqiao",
+	"Weiss",
+	"WellCare",
+	"WellPoint",
+	"Wells",
+	"Welltower",
+	"WESCO",
+	"Wesfarmers",
+	"Westinghouse",
+	"Westlake",
+	"Weston",
+	"Westpac",
+	"WestRock",
+	"Weyerhaeuser",
+	"Whampoa",
+	"WhatsApp",
+	"widgetManager",
+	"Wikidata",
+	"Wikimedia",
+	"Wikipedia",
+	"Wiktionary",
+	"Wildberries",
+	"Williams",
+	"Willis",
+	"Wilmar",
+	"WinDLL",
+	"WindowsLogin",
+	"Windstream",
+	"WinNT",
+	"WinOS",
+	"WinRM",
+	"WinRS",
+	"WIPO",
+	"Wisconsin-Madison",
+	"WiseTech",
+	"Wishart",
+	"Wistron",
+	"Wix",
+	"WLTW",
+	"Wm",
+	"WMIC",
+	"Wollongong",
+	"Woolworths",
+	"Woori",
+	"WordPress",
+	"workArea",
+	"workDir",
+	"workFolder",
+	"worldGen",
+	"WPAR",
+	"wParam",
+	"WPARs",
+	"writeDir",
+	"writeFile",
+	"WSMan",
+	"WSUS",
+	"Wuhan",
+	"WWAN",
+	"WWW",
+	"Wyndham",
+	"WYNN",
+	"WYSIWYG",
+	"Xbox",
+	"Xcel",
+	"Xcode",
+	"xCoord",
+	"xCoords",
+	"Xenial",
+	"XFS",
+	"Xi'an",
+	"Xiaohongshu",
+	"Xiaomi",
+	"Xilinx",
+	"XING",
+	"Xinxing",
+	"xKey",
+	"XLNX",
+	"xMax",
+	"xMin",
+	"XML",
+	"XMLRPC",
+	"xPos",
+	"Xproto",
+	"XRAY",
+	"XSD",
+	"XSRF",
+	"Xstrata",
+	"xVal",
+	"Xware",
+	"YAGNI",
+	"Yale",
+	"Yamada",
+	"Yamaha",
+	"Yandex",
+	"Yassir",
+	"Yasuda",
+	"Yat-Sen",
+	"yCoord",
+	"yCoords",
+	"yesNo",
+	"Yext",
+	"yKey",
+	"Yleisradio",
+	"yMax",
+	"yMin",
+	"Yonsei",
+	"York",
+	"YouTube",
+	"YPF",
+	"yPos",
+	"Yusen",
+	"yVal",
+	"Zalando",
+	"Zapier",
+	"Zappos",
+	"ZBRA",
+	"zCoord",
+	"zCoords",
+	"Zealand",
+	"ZEIT",
+	"Zend",
+	"Zentrale",
+	"Zerodha",
+	"ZFS",
+	"Zhejiang",
+	"Zhihu",
+	"Ziggo",
+	"Zillow",
+	"Zimmer",
+	"ZION",
+	"Zions",
+	"zipFile",
+	"zKey",
+	"zLog",
+	"zMax",
+	"zMin",
+	"Zoetis",
+	"Zoho",
+	"ZPool",
+	"ZPools",
+	"zPos",
+	"Zscaler",
+	"Zuora",
+	"Zurich",
+	"zVal"
+];
+//#endregion
+//#region src/formatting/to-sentence-case.ts
+/**
+* Normalize Title Case overuse (a common AI/marketing tic) back to sentence
+* case while preserving acronyms, mixed-case proper nouns, and the first word
+* of each sentence.
+*
+* The function is conservative: it only rewrites strings whose ratio of
+* capitalized words is high enough that the input is unambiguously in Title
+* Case. Normal prose with one or two proper nouns is left untouched.
+*
+* Philosophy (important, and the inverse of the naive approach): we only
+* lowercase a capitalized/ALL-CAPS token when we can positively confirm its
+* lowercase form is a real word — via the generated preserved-terms list, or
+* via an actual EN/FR spelling dictionary lookup (through `nspell`, which
+* expands Hunspell affixes, so inflected forms like "Published" or
+* "Reported" resolve correctly even though only their stems are dictionary
+* headwords). Anything we don't recognize is assumed to be a proper noun and
+* its original casing is preserved. This is deliberately asymmetric:
+* over-capitalizing a rare common word is a mild cosmetic miss, but
+* decapitalizing an unrecognized proper noun (a person, a place, a brand) is
+* an outright factual error. The old approach did the opposite — lowercase
+* by default, preserve only what's on an explicit allowlist — which
+* systematically mangled any proper noun the allowlist didn't happen to
+* cover.
+*
+* Known limitation: proper nouns whose lowercase homograph is a real
+* dictionary word get wrongly lowercased mid-sentence — e.g. "Macron"
+* ("macron" is an English word), "Musk" ("musk" is a fragrance note), "API"
+* ("api" is a dictionary-fr headword). Same class of ambiguity as "Apple"
+* the fruit vs. "Apple" the company; see the test suite for more cases.
+* Callers who need a specific term protected can pass it via
+* `preservedTerms`.
+*
+* Examples:
+*   "Your Next AI Skill Is Worldbuilding" -> "Your next AI skill is Worldbuilding"
+*   "How To Use The API For HTTP Calls"   -> "How to use the API for HTTP calls"
+*   "The OpenAI Revolution"               -> "The OpenAI revolution"
+*   "Macron Meets Zelensky In Kyiv"       -> "Macron meets Zelensky in Kyiv" (see caveat above: "macron" collides with a real word, so it *does* get lowercased — see tests)
+*   "Cursor: the compression of mechanical work" -> unchanged
+*   "AI is making us smarter"             -> unchanged
+*/
+/**
+* Terms `npm run generate:terms` (see scripts/generate-preserved-terms.ts)
+* cannot derive from the cspell dictionaries, kept here by hand. Stays
+* short and deliberate on purpose:
+*
+*  - 'I' — the pronoun is always capitalized in English. It's excluded from
+*    the generated list because lowercase "i" is a genuine dictionary
+*    headword (the math/code variable) in both EN and FR, so a plain
+*    spelling-dictionary lookup would incorrectly lowercase it; this one
+*    hardcoded linguistic rule overrides that.
+*  - 'AI' — a real gap: dictionary-en does list uppercase "AI" as a valid
+*    headword these days, but lowercase "ai" is *also* a genuine French verb
+*    form ("j'ai" — I have). With French enabled by default, a plain
+*    dictionary lookup on the lowercased token would find "ai" valid via
+*    French and wrongly lowercase the acronym. The explicit preserved-term
+*    entry (checked before any dictionary access) sidesteps the collision
+*    entirely.
+*  - 'OAuth' — not a dictionary headword in either language, so on its own
+*    it would already survive mid-title via the "unrecognized -> preserve"
+*    default. The entry earns its keep for input that doesn't already use
+*    the canonical casing (e.g. "Oauth"), restoring it to "OAuth".
+*
+* Deliberately NOT re-added, even though they were in the old hardcoded
+* list and cspell doesn't source them either:
+*  - 'Apple', 'Meta', 'Linear', 'Operator', 'Codex', 'Google', 'Slack',
+*    'Notion', 'Cursor', 'Amazon', 'Grok', 'TypeScript', 'Python', 'React'
+*    — each collides with a genuine lowercase English dictionary word
+*    (apple, meta, linear, operator, codex, google, slack, notion, cursor,
+*    amazon, grok, typescript, python, react). Preserving them
+*    unconditionally would wrongly force-capitalize the common word too;
+*    the dictionary-lookup fallback applies the same ambiguity rule
+*    automatically for terms that aren't hand- or cspell-listed.
+*  - 'Node.js', 'Next.js' — can't be represented as a single preserved
+*    term anyway: the tokenizer below splits on '.', so these would only
+*    ever match a lone "Node"/"Next" token. Both "node" and "next" are
+*    common dictionary words, so adding them would incorrectly preserve
+*    ordinary title-case capitalization of unrelated sentences.
+*  - 'iOS', 'macOS' — no longer needed. Both have an internal lower→upper
+*    transition ("iOS", "macOS"), so the mixed-case heuristic (checked
+*    before any dictionary access) already preserves them without an
+*    explicit entry.
+*/
+const MANUAL_PRESERVED_TERMS = [
+	"I",
+	"AI",
+	"OAuth"
+];
+const WORD_PATTERN = /(?<word>[\p{L}0-9']+)/u;
+const HAS_LOWER_THEN_UPPER = /\p{Ll}\p{Lu}/u;
+const HAS_UPPER = /\p{Lu}/u;
+const HAS_LOWER = /\p{Ll}/u;
+const HAS_LETTER = /\p{L}/u;
+const SENTENCE_END_PATTERN = /[.!?]/;
+const DICTIONARY_PACKAGE = {
+	en: "dictionary-en",
+	fr: "dictionary-fr"
+};
+/**
+* Lazily-built singleton map (lowercase -> canonical casing) for the
+* combined default preserved terms. Built once on first use and reused for
+* every call that doesn't ask to replace the defaults entirely.
+*/
+let defaultPreservedMap = null;
+function getDefaultPreservedMap() {
+	if (!defaultPreservedMap) {
+		const map = /* @__PURE__ */ new Map();
+		for (const term of GENERATED_PRESERVED_TERMS) map.set(term.toLowerCase(), term);
+		for (const term of MANUAL_PRESERVED_TERMS) map.set(term.toLowerCase(), term);
+		defaultPreservedMap = map;
+	}
+	return defaultPreservedMap;
+}
+/**
+* Lazily-constructed, per-language `nspell` singletons. Parsing a Hunspell
+* affix + dictionary file costs roughly 100-300ms, so an instance is only
+* ever built the first time a token *actually* requires a dictionary lookup
+* (never at module import, never at the first `toSentenceCase` call if that
+* call's tokens are all resolved by the cheaper fast paths first). Each
+* language is independent: a call that only uses `languages: ['en']` never
+* pays the French parsing cost, and vice versa.
+*/
+const spellCheckers = /* @__PURE__ */ new Map();
+/**
+* `dictionary-en`/`dictionary-fr` ship an async default export (they read
+* their `.aff`/`.dic` files with `fs.promises.readFile` at import time), but
+* `toSentenceCase` must stay synchronous. We bypass their package entry
+* point entirely and read the underlying Hunspell data files ourselves with
+* a synchronous `readFileSync`, resolving the package directory via
+* `createRequire` (works from both the ESM and CJS build outputs).
+*/
+const nodeRequire = createRequire(import.meta.url);
+function loadSpellChecker(language) {
+	const packageEntry = nodeRequire.resolve(DICTIONARY_PACKAGE[language]);
+	const packageDir = path.dirname(packageEntry);
+	return nspell({
+		aff: readFileSync(path.join(packageDir, "index.aff")),
+		dic: readFileSync(path.join(packageDir, "index.dic"))
+	});
+}
+function getSpellChecker(language) {
+	let checker = spellCheckers.get(language);
+	if (!checker) {
+		checker = loadSpellChecker(language);
+		spellCheckers.set(language, checker);
+	}
+	return checker;
+}
+/**
+* Memoizes dictionary verdicts (`"<lang>\0<word>"` -> `correct(word)`) since
+* headline-style text repeats the same handful of words constantly. Bounded
+* to avoid unbounded growth across a long-running process; eviction is a
+* simple FIFO (oldest-inserted entry dropped first) rather than true LRU —
+* good enough for this access pattern and far cheaper to maintain.
+*/
+const MAX_VERDICT_CACHE_ENTRIES = 1e4;
+const spellVerdictCache = /* @__PURE__ */ new Map();
+function isKnownSpelling(word, language) {
+	const key = `${language}${word}`;
+	const cached = spellVerdictCache.get(key);
+	if (cached !== void 0) return cached;
+	const verdict = getSpellChecker(language).correct(word);
+	if (spellVerdictCache.size >= MAX_VERDICT_CACHE_ENTRIES) {
+		const oldestKey = spellVerdictCache.keys().next().value;
+		if (oldestKey !== void 0) spellVerdictCache.delete(oldestKey);
+	}
+	spellVerdictCache.set(key, verdict);
+	return verdict;
+}
+/** Whether `word` is recognized (case-sensitively) by any of `languages`, checked in order. */
+function isKnownInAnyLanguage(word, languages) {
+	for (const language of languages) if (isKnownSpelling(word, language)) return true;
+	return false;
+}
+/**
+* Normalizes Title Case overuse back to sentence case.
+*
+* @param text - The text to normalize
+* @param options - Normalization options
+* @returns The normalized text, or the original text if it doesn't look like Title Case
+*/
+function toSentenceCase(text, options = {}) {
+	if (!text) return text;
+	const localPreservedMap = /* @__PURE__ */ new Map();
+	for (const term of options.preservedTerms ?? []) localPreservedMap.set(term.toLowerCase(), term);
+	const defaultMap = options.replacePreserved ? null : getDefaultPreservedMap();
+	const lookupPreserved = (lower) => localPreservedMap.get(lower) ?? defaultMap?.get(lower);
+	const threshold = options.titleCaseThreshold ?? .6;
+	const minWords = options.minWords ?? 3;
+	const languages = options.languages ?? ["en", "fr"];
+	const tokens = text.split(WORD_PATTERN);
+	const wordTokens = tokens.filter((t) => WORD_PATTERN.test(t) && HAS_LETTER.test(t));
+	if (wordTokens.length < minWords) return text;
+	if (wordTokens.filter((t) => HAS_UPPER.test(t[0] ?? "")).length / wordTokens.length < threshold) return text;
+	let isFirstWord = true;
+	let nextStartsSentence = false;
+	return tokens.map((tok) => {
+		if (!WORD_PATTERN.test(tok) || !HAS_LETTER.test(tok)) {
+			if (SENTENCE_END_PATTERN.test(tok)) nextStartsSentence = true;
+			return tok;
+		}
+		const startsNewSentence = isFirstWord || nextStartsSentence;
+		isFirstWord = false;
+		nextStartsSentence = false;
+		const preserved = lookupPreserved(tok.toLowerCase());
+		if (preserved) return preserved;
+		if (HAS_LOWER_THEN_UPPER.test(tok)) return tok;
+		if (startsNewSentence) return tok[0].toUpperCase() + tok.slice(1).toLowerCase();
+		if (!HAS_UPPER.test(tok)) return tok;
+		if (!HAS_LOWER.test(tok) && tok.length >= 2) {
+			const lower = tok.toLowerCase();
+			if (isKnownInAnyLanguage(lower, languages)) return lower;
+			const titleCased = tok[0] + tok.slice(1).toLowerCase();
+			if (isKnownInAnyLanguage(titleCased, languages)) return titleCased;
+			return tok;
+		}
+		const lower = tok.toLowerCase();
+		if (isKnownInAnyLanguage(lower, languages)) return lower;
+		return tok;
+	}).join("");
+}
+//#endregion
+export { cleanAiText, toSentenceCase };
+
+//# sourceMappingURL=formatting.js.map