I have this Js function with hard coded filter parameters. It filter all the buckets sub objects when key start with a string from a given list. For now i havent found a way to put this list as an array...
function filter(buckets) {
return buckets.filter(({key}) => {
const _key = key.toLowerCase();
return (
!_key.startsWith("anciennes") && !_key.startsWith("anciens") && !_key.startsWith("arrondissements") && !_key.startsWith("autorites") && !_key.startsWith("cantons") && !_key.startsWith("capitales") &&
!_key.startsWith("chaines") && !_key.startsWith("chefs lieux") && !_key.startsWith("circonscriptions") && !_key.startsWith("communautes d") && !_key.startsWith("comtes d") && !_key.startsWith("constellations") && !_key.startsWith("continents") &&
!_key.startsWith("cours d") && !_key.startsWith("départements") && !_key.startsWith("districts d") && !_key.startsWith("établissements") && !_key.startsWith("états") && !_key.startsWith("ethnonymes") && !_key.startsWith("étoiles") &&
!_key.startsWith("europe") && !_key.startsWith("îles") && !_key.startsWith("lander-allemands") && !_key.startsWith("lieux-mythologiques") && !_key.startsWith("Localités d") && !_key.startsWith("mers") &&
!_key.startsWith("montagnes") && !_key.startsWith("municipalité") && !_key.startsWith("fromage") && !_key.startsWith("localités") && !_key.startsWith("noms de") && !_key.startsWith("numéros de") &&
!_key.startsWith("odonymes") && !_key.startsWith("organisations internationales") && !_key.startsWith("pays ") && !_key.startsWith("péninsules") && !_key.startsWith("préfectures") && !_key.startsWith("provinces") && !_key.startsWith("quartiers") &&
!_key.startsWith("régions") && !_key.startsWith("réserves indiennes") && !_key.startsWith("sous ") && !_key.startsWith("territoires") && !_key.startsWith("toponymes ") && !_key.startsWith("unions supranationales") && !_key.startsWith("villes du quebec") && !_key.startsWith("voivodies de pologne")
)
})
}
const buckets = [
{"key": "Aliments", "doc_count": 10}, {"key": "Adjectifs", "doc_count": 7}, {"key": "Vêtements", "doc_count": 6}, {"key": "Armures", "doc_count": 5}, {"key": "Anciennes divisions géographiques", "doc_count": 4}, {"key": "Super-règnes", "doc_count": 4},
{"key": "Eucaryotes", "doc_count": 3}, {"key": "Pays", "doc_count": 3}, {"key": "Antonomases", "doc_count": 2}, {"key": "Continents", "doc_count": 2}, {"key": "Europe", "doc_count": 2}, {"key": "France", "doc_count": 2},
{"key": "Localités", "doc_count": 2}, {"key": "Localités de France", "doc_count": 2}, {"key": "Plantes", "doc_count": 2}, {"key": "États", "doc_count": 2}, {"key": "Acaryotes", "doc_count": 1},
{"key": "Animaux", "doc_count": 1}, {"key": "Armes", "doc_count": 1}, {"key": "Aromates", "doc_count": 1}, {"key": "Chordés", "doc_count": 1}, {"key": "Couleurs", "doc_count": 1}, {"key": "Créatures mythologiques", "doc_count": 1},
{"key": "Fromages", "doc_count": 1}, {"key": "Fromages forts", "doc_count": 1}, {"key": "Fromages à pâte filée", "doc_count": 1}, {"key": "Fruits", "doc_count": 1}, {"key": "Langages informatiques", "doc_count": 1},
{"key": "Localités du département de la Corrèze", "doc_count": 1}, {"key": "Localités du département de la Dordogne", "doc_count": 1}, {"key": "Machines", "doc_count": 1}, {"key": "Oiseaux", "doc_count": 1}, {"key": "Tétrapodes", "doc_count": 1},
{"key": "Vertébrés", "doc_count": 1}, {"key": "Vie domestique", "doc_count": 1}, {"key": "Virus", "doc_count": 1}, {"key": "Volcans", "doc_count": 1}
];
jsbin here :
https://jsbin.com/xicejozafe/edit?js,console
how to transform this function to generate the !_key.startsWith( ... ) && from and array of string insted of this crazy hard coded long &&
const listOfTermToIgnore = ["anciennes", "anciens", "arrondissements", "autorites", "cantons", "capitales",
"chaines", "chefs lieux", "circonscriptions", "communautes d", "comtes d", "constellations", "continents",
"cours d", "départements", "districts d", "établissements", "états", "ethnonymes", "étoiles",
"europe", "îles", "lander-allemands", "lieux-mythologiques", "Localités d", "mers",
"montagnes", "municipalité", "fromage", "localités", "noms de", "numéros de",
"odonymes", "organisations internationales", "pays ", "péninsules", "préfectures", "provinces", "quartiers",
"régions", "réserves indiennes", "sous ", "territoires", "toponymes ", "unions supranationales", "villes du quebec", "voivodies de pologne" ]
thanks for reading
You can use the inverse of array.some. It will return false if _key starts with any item from the array, or true otherwise.
function filter(buckets) {
return buckets.filter(({key}) => {
const _key = key.toLowerCase();
return !listOfTermToIgnore.some(item => _key.startsWith(item))
})
}
Use array.every() to check all the elements of the array.
function filter(buckets) {
return buckets.filter(({key}) => {
const _key = key.toLowerCase();
return listOfTtermToIgnore.every(term => !_key.startsWith(term))
}
});
For the one-time cost of creating a regex, you get the following high performance solution:
function quotemeta(s) {
return String(s).replace(/\W/g, "\\$&");
}
const listOfTermToIgnore = [
"anciennes", "anciens", "arrondissements", "autorites", "cantons",
"capitales", "chaines", "chefs lieux", "circonscriptions",
"communautes d", "comtes d", "constellations", "continents",
"cours d", "départements", "districts d", "établissements",
"états", "ethnonymes", "étoiles", "europe", "îles",
"lander-allemands", "lieux-mythologiques", "mers", "montagnes",
"municipalité", "fromage", "localités", "noms de", "numéros de",
"odonymes", "organisations internationales", "pays ", "péninsules",
"préfectures", "provinces", "quartiers", "régions", "réserves indiennes",
"sous ", "territoires", "toponymes ", "unions supranationales",
"villes du quebec", "voivodies de pologne"
];
const filter_re = new RegExp(
"^(?:" + listOfTermToIgnore.map( _ => quotemeta(_) ).join("|") + ")"
);
function filter(buckets) {
return buckets.filter( _ => !filter_re.test( _.toLowerCase() ) );
}
console.log( filter( [ "anciennes", "xanciennes", "anciennesx", "banane" ] ) );It's common to use this approach in Perl, but Perl optimizes this kind of alternation into a trie, which is extremely fast. It just occurred to me that your favourite JavaScript might not do that, so it might not be as performant as I thought earlier. Still, moving the checking of all these strings to the regex engine should still be faster than checking them all in JavaScript code.
But if you had a truly long list of terms to ignore, I'd recommend using a trie.