refactor: remove slow utils (#56)

medzuslovjansky · Jun 23, 2024 · 25e9f63 · 25e9f63
1 parent f4f682c
commit 25e9f63
Show file tree

Hide file tree

Showing 11 changed files with 41 additions and 225 deletions.
diff --git a/src/common/fluentVowels.ts b/src/common/fluentVowels.ts
@@ -1,7 +1,4 @@
-import { ALL_CHARACTERS, CONSONANT_CHARACTERS } from '../constants';
-
-const LETTERS = new Set(ALL_CHARACTERS);
-const CONSONANTS = new Set(CONSONANT_CHARACTERS);
+import { ALL_LETTERS, ALL_CONSONANTS } from '../substitutions';
 
 export function markFluentVowel(word: string, add: string): string {
   let i = 0;
@@ -26,7 +23,7 @@ export function inferFluentVowel(word: string): string {
 
   while (i > 0) {
     const char = word[i];
-    if (!LETTERS.has(char)) {
+    if (!ALL_LETTERS.has(char)) {
       end = i;
       replaced = false;
     }
@@ -53,7 +50,7 @@ function replaceFluentVowel(word: string, j: number): string {
 }
 
 function isLastSyllable(word: string, i: number, end: number): boolean {
-  if (i === end - 2) return CONSONANTS.has(word[i + 1]);
+  if (i === end - 2) return ALL_CONSONANTS.has(word[i + 1]);
   if (i === end - 3) return word[i + 1] === 'n' && word[i + 2] === 'j';
   return false;
 }
diff --git a/src/constants/alphabet.ts b/src/constants/alphabet.ts
diff --git a/src/constants/index.ts b/src/constants/index.ts
@@ -1,3 +1,2 @@
-export * from './alphabet';
 export * from './bcp47';
 export * from './glagolitic';
diff --git a/src/noun/declensionNoun.ts b/src/noun/declensionNoun.ts
@@ -5,10 +5,11 @@
 import { declensionAdjective } from '../adjective';
 import { inferFluentVowel, markFluentVowel } from '../common';
 import type { Noun } from '../partOfSpeech';
-import { matchEnd, removeBrackets, replaceStringAt } from '../utils';
+import { removeBrackets, replaceStringAt } from '../utils';
 import { establishGender } from './establishGender';
 
-const AEEO = ['a', 'e', 'ę', 'o'];
+// endings like -i, -u are not declinable usually
+const AEEO$ = /[aeęo]$/;
 
 export function declensionNounFlat(
   rawNoun: string,
@@ -189,7 +190,7 @@ function establish_root(noun: string, gender: string) {
     noun.lastIndexOf('ȯ'),
   );
 
-  const hasVowelEnding = matchEnd(noun, [AEEO]);
+  const hasVowelEnding = AEEO$.test(noun);
 
   if (noun == 'lėv' || noun == 'lev') {
     result = 'ljv';
@@ -283,7 +284,7 @@ function nominative_sg(noun: string, root: string, gender: string) {
   if (gender == 'f2') {
     result = root;
   }
-  if (gender == 'f3' && root.lastIndexOf('v') == root.length - 1) {
+  if (gender == 'f3' && root.endsWith('v')) {
     result = root.substring(0, root.length - 1) + 'ȯv';
   } else if (gender == 'f3') {
     result = noun;
@@ -361,7 +362,7 @@ function instrumental_sg(root: string, gender: string) {
     result = root + 'ojų';
   } else if (gender == 'f2') {
     result = root + 'jų';
-  } else if (gender == 'f3' && root.lastIndexOf('v') == root.length - 1) {
+  } else if (gender == 'f3' && root.endsWith('v')) {
     result = root.substring(0, root.length - 1) + 'ȯvjų';
   } else if (gender == 'f3') {
     result = root + 'jų';

diff --git a/src/substitutions.ts b/src/substitutions.ts
@@ -1,19 +1,16 @@
-export const ALL_CHARACTERS =
-  'aáàăâåąāæbcćçčdďđḓeéèĕêěëėęēǝfghiíìĭîīıjĵklĺľļłŀǉmnńňñņǌoóòŏôöȯǫœpqrŕṙřsśšŠtťṱuúùŭûůũųūvwxyýzźżž'.split(
-    '',
-  );
+export const ALL_LETTERS = new Set(
+  'aáàăâåąāæbcćçčdďđḓeéèĕêěëėęēǝfghiíìĭîīıjĵklĺľļłŀǉmnńňñņǌoóòŏôöȯǫœpqrŕṙřsśšŠtťṱuúùŭûůũųūvwxyýzźżž',
+);
 
-export const ANY = ALL_CHARACTERS;
+export const ALL_CONSONANTS = new Set(
+  'bcćçčdďđḓfghklĺľļłŀǉmnńňñņǌpqrŕṙřsśštťṱvwxzźżž',
+);
 
-export const CONSONANT_CHARACTERS =
-  'bcćçčdďđḓfghklĺľļłŀǉmnńňñņǌpqrŕṙřsśštťṱvwxzźżž'.split('');
+export const ALL_VOWELS = new Set(
+  'aáàăâåąāæeéèĕêěëėęēǝiíìĭîīıoóòŏôöȯǫœuúùŭûůũųūyý',
+);
 
-export const CONSONANT = CONSONANT_CHARACTERS;
-
-export const VOWEL_CHARACTERS =
-  'aáàăâåąāæeéèĕêěëėęēǝiíìĭîīıoóòŏôöȯǫœuúùŭûůũųūyý'.split('');
-
-export const VOWEL = VOWEL_CHARACTERS;
+export const VOWELS = new Set('aåeęěėioȯuųy');
 
 export const LJ_NJ = ['lj', 'nj'];
 export const LJj_NJj = ['lj', 'ĺj', 'ľj', 'ǉ', 'nj', 'ńj', 'ňj', 'ñj', 'ǌ'];

diff --git a/src/utils/index.ts b/src/utils/index.ts
@@ -1,7 +1,5 @@
 export * from './areArraysEqual';
 export * from './compactArray';
-export * from './matchStart';
-export * from './matchEnd';
 export * from './memoizeLastCall';
 export * from './removeBrackets';
 export * from './replaceStringAt';

diff --git a/src/utils/matchEnd.test.ts b/src/utils/matchEnd.test.ts
diff --git a/src/utils/matchEnd.ts b/src/utils/matchEnd.ts
diff --git a/src/utils/matchStart.test.ts b/src/utils/matchStart.test.ts
diff --git a/src/utils/matchStart.ts b/src/utils/matchStart.ts
diff --git a/src/verb/conjugationVerb.ts b/src/verb/conjugationVerb.ts
@@ -2,17 +2,18 @@
  * @source http://steen.free.fr/interslavic/conjugator.html
  */
 
-import { compactArray, matchEnd } from '../utils';
+import { compactArray } from '../utils';
 import { parsePos, Verb } from '../partOfSpeech';
-import {
-  BIG_YUS,
-  BIG_YUS_LOOSE,
-  IOTATED_SMALL_YUS,
-  SMALL_YUS,
-} from '../substitutions';
+import { BIG_YUS, IOTATED_SMALL_YUS, SMALL_YUS } from '../substitutions';
 
-const _SE = [' se', ' se'];
-const SE_ = ['sę ', 'se '];
+const _SE = /. s[eę]$/;
+const SE_ = /s[eę] $/;
+const EVA_OVA = /[eo]va$/;
+const NUU = /..n[uų]$/;
+const OUEE = /^..?[eěou]$/;
+const BDSZE = /[bdsz]ję$/;
+const AEE = /[aeě]$/;
+const MEUU = /[meuų-]$/;
 
 const PREFIXES = [
   'prědpo',
@@ -48,16 +49,7 @@ const PREFIXES = [
   'v',
 ];
 
-const NON_REGULAR_VERBS = [
-  'věděti',
-  'vedeti',
-  'jesti',
-  'jěsti',
-  'dati',
-  'dųti',
-  'byti',
-  'žegti',
-];
+const NON_REGULAR_VERBS = /(v[eě]d[eě]ti|j[eě]sti|d[aų]ti|byti|žegti)$/;
 
 const irregular_stems = { da: 1, je: 1, jě: 1, ja: 1, vě: 1 };
 
@@ -253,9 +245,9 @@ function splitReflexive(inf: string) {
 
 function prefix(inf: string) {
   // get prefixes for some non-regular verbs
-  const irregular = matchEnd(inf, [NON_REGULAR_VERBS]);
-  if (irregular) {
-    const maybePrefix = inf.slice(0, -irregular.length);
+  const match = inf.match(NON_REGULAR_VERBS);
+  if (match) {
+    const maybePrefix = inf.slice(0, -match[1].length);
     if (PREFIXES.includes(maybePrefix)) {
       return maybePrefix;
     }
@@ -370,13 +362,13 @@ function derive_present_tense_stem(infinitive_stem_string: string): string {
     result = 'uměĵ';
   } else if (result === 'hova') {
     result = 'hovaĵ';
-  } else if (matchEnd(result, [['o', 'e'], 'va'])) {
+  } else if (EVA_OVA.test(result)) {
     result = result.slice(0, -3) + 'uj';
-  } else if (result.length > 3 && matchEnd(result, ['n', BIG_YUS_LOOSE])) {
+  } else if (NUU.test(result)) {
     result = result.slice(0, -1);
-  } else if (result.length < 4 && matchEnd(result, [['o', 'u', 'e', 'ě']])) {
+  } else if (OUEE.test(result)) {
     result = result + 'j';
-  } else if (matchEnd(result, [['b', 'd', 's', 'z'], IOTATED_SMALL_YUS])) {
+  } else if (BDSZE.test(result)) {
     result = result.slice(0, -2) + 'ȯjm';
   } else if (result.endsWith(IOTATED_SMALL_YUS)) {
     result = result.slice(0, -1) + 'm';
@@ -386,7 +378,7 @@ function derive_present_tense_stem(infinitive_stem_string: string): string {
     result = result.slice(0, -1) /*+ 'm'*/;
   } else if (result.endsWith('y')) {
     result = result + 'j';
-  } else if (matchEnd(result, [['a', 'e', 'ě']])) {
+  } else if (AEE.test(result)) {
     result = result + 'ĵ';
   }
   return result;
@@ -398,9 +390,9 @@ function present_tense_stem(pref: string, pts: string, is: string) {
   if (pts.length == 0) {
     result = derive_present_tense_stem(is);
   } else {
-    if (matchEnd(pts, _SE) && pts.length > 3) {
+    if (_SE.test(pts)) {
       pts = pts.slice(0, -3);
-    } else if (matchEnd(pts, SE_)) {
+    } else if (SE_.test(pts)) {
       pts = pts.slice(3);
     }
 
@@ -412,7 +404,7 @@ function present_tense_stem(pref: string, pts: string, is: string) {
       }
     }
 
-    if (matchEnd(pts, [['-', 'm', 'e', 'ų', 'u']])) {
+    if (MEUU.test(pts)) {
       result = pts.slice(0, -1);
     } else {
       result = pts;