@keymanapp/kmc-model 17.0.85-alpha

package/build/test/test-default-apply-case.js

@@ -0,0 +1,105 @@
+import 'mocha';
+import { assert } from 'chai';
+import { defaultApplyCasing } from '../src/model-defaults.js';
+describe('The default applyCasing() function', function () {
+    // // --------
+    // // Definitions toward SMP testing.
+    // // Ref: https://unicode.org/charts/nameslist/n_1D400.html
+    // // Useful for tests related to strings with supplementary pairs.
+    // let u = function(code: number): string {
+    //   var H = Math.floor((code - 0x10000) / 0x400) + 0xD800;
+    //   var L = (code - 0x10000) % 0x400 + 0xDC00;
+    //   return String.fromCharCode(H, L);
+    // }
+    // Are there any known default-handled SMP cases?
+    // If not... we could make the defaultApplyCase function simpler by not worrying about SMP.
+    //
+    // let smp_a = u(0x1d5ba);  // MATHEMATICAL SANS-SERIF SMALL A
+    // let smp_p = u(0x1d5c9);
+    // let smp_l = u(0x1d5c5);
+    // let smp_e = u(0x1d5be);
+    // let smp_A = u(0x1d5a0);  // MATHEMATICAL SANS_SERIF CAPITAL A
+    // let smp_P = u(0x1d5af);
+    // let smp_L = u(0x1d5ab);
+    // let smp_E = u(0x1d5a4);
+    // // Unfortunately... the default JS .toUpperCase() implementation doesn't actually
+    // // map the 'SMALL' versions to the 'CAPITAL' versions.
+    // // ---------
+    describe('case:  \'lower\'', function () {
+        const testCases = [
+            // Note:  not written the Turkish way.  Turns out 'İ'.toLowerCase() decomposes the result,
+            // which would have made for a fairly yucky test.
+            ['Istanbul', 'istanbul'],
+            // The DEFAULT function is NOT responsible for understanding the Turkish
+            // case regarding the lowercasing of:
+            // 'I' U+0048 LATIN CAPITAL LETTER I to 'ı' U+0131 LATIN SMALL LETTER DOTLESS I
+            // For Turkic languages, the recommendation is to make a
+            // custom applyCasing function:
+            ['DİYARBAKIR', 'di̇yarbakir'],
+            // full-width romaji has corresponding capitalized versions:
+            ['ＡＥＳＴＨＥＴＩＣ', 'ａｅｓｔｈｅｔｉｃ'],
+            // "skýlos" is Greek for dog 🇬🇷🐶
+            // starts with an 's' and ends with an 's'
+            // which are DIFFERENT CHARACTERS in lowercased Greek!
+            ['ΣΚΥΛΟΣ', 'σκυλος'],
+            // Uncased syntax and numbers should pass through unscathed:
+            ['1234.?!', '1234.?!']
+        ];
+        for (let [input, expected] of testCases) {
+            it(`should lowercase '${input}' as '${expected}'`, function () {
+                assert.equal(defaultApplyCasing('lower', input), expected);
+            });
+        }
+    });
+    describe('case:  \'upper\'', function () {
+        const testCases = [
+            // Inverse of the corresponding 'lower' test.
+            ['istanbul', 'ISTANBUL'],
+            // The DEFAULT function is NOT responsible for understanding the Turkish
+            // case regarding the uppercasing of:
+            // 'ı' U+0131 LATIN SMALL LETTER DOTLESS I to 'I' U+0048 LATIN CAPITAL LETTER I
+            // For Turkic languages, the recommendation is to make a
+            // custom applyCasing function:
+            ['diyarbakır', 'DIYARBAKIR'],
+            // full-width romaji has corresponding capitalized versions:
+            ['ａｅｓｔｈｅｔｉｃ', 'ＡＥＳＴＨＥＴＩＣ'],
+            // "skýlos" is Greek for dog 🇬🇷🐶
+            // starts with an 's' and ends with an 's'
+            // which are DIFFERENT CHARACTERS in lowercased Greek!
+            ['σκυλος', 'ΣΚΥΛΟΣ'],
+            // Uncased syntax and numbers should pass through unscathed:
+            ['1234.?!', '1234.?!']
+        ];
+        for (let [input, expected] of testCases) {
+            it(`should uppercase '${input}' as '${expected}'`, function () {
+                assert.equal(defaultApplyCasing('upper', input), expected);
+            });
+        }
+    });
+    describe('case:  \'initial\'', function () {
+        const testCases = [
+            // Inverse of the corresponding 'lower' test.
+            ['istanbul', 'Istanbul'],
+            // The DEFAULT function is NOT responsible for understanding the Turkish
+            // case regarding the uppercasing of:
+            // 'ı' U+0131 LATIN SMALL LETTER DOTLESS I to 'I' U+0048 LATIN CAPITAL LETTER I
+            // For Turkic languages, the recommendation is to make a
+            // custom applyCasing function:
+            ['diyarbakır', 'Diyarbakır'],
+            // full-width romaji has corresponding capitalized versions:
+            ['ａｅｓｔｈｅｔｉｃ', 'Ａｅｓｔｈｅｔｉｃ'],
+            // "skýlos" is Greek for dog 🇬🇷🐶
+            // starts with an 's' and ends with an 's'
+            // which are DIFFERENT CHARACTERS in lowercased Greek!
+            ['σκυλος', 'Σκυλος'],
+            // Uncased syntax and numbers should pass through unscathed:
+            ['1234.?!', '1234.?!']
+        ];
+        for (let [input, expected] of testCases) {
+            it(`should initial-case '${input}' as '${expected}'`, function () {
+                assert.equal(defaultApplyCasing('initial', input), expected);
+            });
+        }
+    });
+});
+//# sourceMappingURL=test-default-apply-case.js.map

package/build/test/test-default-apply-case.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-default-apply-case.js","sourceRoot":"","sources":["../../test/test-default-apply-case.ts"],"names":[],"mappings":"AACA,OAAO,OAAO,CAAC;AACf,OAAO,EAAE,MAAM,EAAE,MAAM,MAAM,CAAC;AAC9B,OAAO,EAAE,kBAAkB,EAAE,MAAM,0BAA0B,CAAC;AAE9D,QAAQ,CAAC,oCAAoC,EAAE;IAC7C,cAAc;IACd,qCAAqC;IACrC,4DAA4D;IAC5D,mEAAmE;IACnE,2CAA2C;IAC3C,2DAA2D;IAC3D,+CAA+C;IAE/C,sCAAsC;IACtC,IAAI;IAEJ,iDAAiD;IACjD,2FAA2F;IAC3F,EAAE;IACF,8DAA8D;IAC9D,0BAA0B;IAC1B,0BAA0B;IAC1B,0BAA0B;IAE1B,gEAAgE;IAChE,0BAA0B;IAC1B,0BAA0B;IAC1B,0BAA0B;IAE1B,oFAAoF;IACpF,yDAAyD;IACzD,eAAe;IAEf,QAAQ,CAAC,kBAAkB,EAAE;QAC3B,MAAM,SAAS,GAAuB;YACpC,0FAA0F;YAC1F,iDAAiD;YACjD,CAAC,UAAU,EAAE,UAAU,CAAC;YAExB,wEAAwE;YACxE,qCAAqC;YACrC,+EAA+E;YAC/E,wDAAwD;YACxD,+BAA+B;YAC/B,CAAC,YAAY,EAAE,aAAa,CAAC;YAE7B,4DAA4D;YAC5D,CAAC,WAAW,EAAE,WAAW,CAAC;YAE1B,mCAAmC;YACnC,0CAA0C;YAC1C,sDAAsD;YACtD,CAAC,QAAQ,EAAE,QAAQ,CAAC;YAEpB,4DAA4D;YAC5D,CAAC,SAAS,EAAE,SAAS,CAAC;SACvB,CAAC;QAEF,KAAK,IAAI,CAAC,KAAK,EAAE,QAAQ,CAAC,IAAI,SAAS,EAAE;YACvC,EAAE,CAAC,qBAAqB,KAAK,SAAS,QAAQ,GAAG,EAAE;gBACjD,MAAM,CAAC,KAAK,CAAC,kBAAkB,CAAC,OAAO,EAAE,KAAK,CAAC,EAAE,QAAQ,CAAC,CAAC;YAC7D,CAAC,CAAC,CAAC;SACJ;IACH,CAAC,CAAC,CAAC;IAEH,QAAQ,CAAC,kBAAkB,EAAE;QAC3B,MAAM,SAAS,GAAuB;YACpC,6CAA6C;YAC7C,CAAC,UAAU,EAAE,UAAU,CAAC;YAExB,wEAAwE;YACxE,qCAAqC;YACrC,+EAA+E;YAC/E,wDAAwD;YACxD,+BAA+B;YAC/B,CAAC,YAAY,EAAE,YAAY,CAAC;YAE5B,4DAA4D;YAC5D,CAAC,WAAW,EAAE,WAAW,CAAC;YAE1B,mCAAmC;YACnC,0CAA0C;YAC1C,sDAAsD;YACtD,CAAC,QAAQ,EAAE,QAAQ,CAAC;YAEpB,4DAA4D;YAC5D,CAAC,SAAS,EAAE,SAAS,CAAC;SACvB,CAAC;QAEF,KAAK,IAAI,CAAC,KAAK,EAAE,QAAQ,CAAC,IAAI,SAAS,EAAE;YACvC,EAAE,CAAC,qBAAqB,KAAK,SAAS,QAAQ,GAAG,EAAE;gBACjD,MAAM,CAAC,KAAK,CAAC,kBAAkB,CAAC,OAAO,EAAE,KAAK,CAAC,EAAE,QAAQ,CAAC,CAAC;YAC7D,CAAC,CAAC,CAAC;SACJ;IACH,CAAC,CAAC,CAAC;IAEH,QAAQ,CAAC,oBAAoB,EAAE;QAC7B,MAAM,SAAS,GAAuB;YACpC,6CAA6C;YAC7C,CAAC,UAAU,EAAE,UAAU,CAAC;YAExB,wEAAwE;YACxE,qCAAqC;YACrC,+EAA+E;YAC/E,wDAAwD;YACxD,+BAA+B;YAC/B,CAAC,YAAY,EAAE,YAAY,CAAC;YAE5B,4DAA4D;YAC5D,CAAC,WAAW,EAAE,WAAW,CAAC;YAE1B,mCAAmC;YACnC,0CAA0C;YAC1C,sDAAsD;YACtD,CAAC,QAAQ,EAAE,QAAQ,CAAC;YAEpB,4DAA4D;YAC5D,CAAC,SAAS,EAAE,SAAS,CAAC;SACvB,CAAC;QAEF,KAAK,IAAI,CAAC,KAAK,EAAE,QAAQ,CAAC,IAAI,SAAS,EAAE;YACvC,EAAE,CAAC,wBAAwB,KAAK,SAAS,QAAQ,GAAG,EAAE;gBACpD,MAAM,CAAC,KAAK,CAAC,kBAAkB,CAAC,SAAS,EAAE,KAAK,CAAC,EAAE,QAAQ,CAAC,CAAC;YAC/D,CAAC,CAAC,CAAC;SACJ;IACH,CAAC,CAAC,CAAC;AACL,CAAC,CAAC,CAAC"}

package/build/test/test-default-search-term-to-key.d.ts

@@ -0,0 +1,2 @@
+import 'mocha';
+//# sourceMappingURL=test-default-search-term-to-key.d.ts.map

package/build/test/test-default-search-term-to-key.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-default-search-term-to-key.d.ts","sourceRoot":"","sources":["../../test/test-default-search-term-to-key.ts"],"names":[],"mappings":"AACA,OAAO,OAAO,CAAC"}

package/build/test/test-default-search-term-to-key.js

@@ -0,0 +1,148 @@
+import 'mocha';
+import { assert } from 'chai';
+import { defaultSearchTermToKey, defaultCasedSearchTermToKey, defaultApplyCasing } from '../src/model-defaults.js';
+describe('The default searchTermToKey() function', function () {
+    describe('languageUsesCasing: false', function () {
+        const testCases = [
+            // "İstanbul" has a U+0130 LATIN CAPITAL LETTER I WITH DOT ABOVE.
+            // Without any casing operations, only the I should be altered.
+            ['İstanbul', 'Istanbul'],
+            // Similarly...
+            ['DİYARBAKIR', 'DIYARBAKIR'],
+            // "skýlos" is Greek for dog 🇬🇷🐶
+            // starts with an 's' and ends with an 's'
+            // which are DIFFERENT CHARACTERS in lowercased Greek!
+            ['σκύλος', 'σκυλος'],
+            ['ΣΚΥΛΟΣ', 'ΣΚΥΛΟΣ'],
+            // full-width romaji is compatibility-canonical with ASCII characters:
+            ['ａｅｓｔｈｅｔｉｃ', 'aesthetic'],
+            // U+212B ANGSTROM SIGN (Å)
+            // U+00C5 LATIN CAPITAL LETTER A WITH RING ABOVE (Å)
+            // and should both normalize to 'A'
+            ['\u212B', 'A'],
+            ['\u00C5', 'A'],
+            // We should not fall for U+037E GREEK QUESTION MARK's trolling:
+            ['\u037e', ';'],
+            // Test presentational forms of Arabic:
+            // U+FE8D ARABIC LETTER ALEF ISOLATED FORM -> U+0627 ARABIC LETTER ALEF
+            // U+FEDF ARABIC LETTER LAM INITIAL FORM -> U+0644 ARABIC LETTER LAM
+            // U+FED8 ARABIC LETTER QAF MEDIAL FORM -> U+0642 ARABIC LETTER QAF
+            // U+FEEC ARABIC LETTER HEH MEDIAL FORM -> U+0647 ARABIC LETTER HEH
+            // U+FEEE ARABIC LETTER WAW FINAL FORM -> U+0648 ARABIC LETTER WAW
+            // U+FE93 ARABIC LETTER TEH MARBUTA ISOLATED FORM -> U+0629 ARABIC LETTER TEH MARBUTA
+            ['\uFE8D\uFEDF\uFED8\uFEEC\uFEEE\uFE93', '\u0627\u0644\u0642\u0647\u0648\u0629'],
+            // Combine both NFKD **AND** knocking off diacritics:
+            // U+01C4 LATIN CAPITAL LETTER DZ WITH CARON (DŽ) -> <U+0064, U+007A> (dz)
+            ['DŽ', 'DZ'],
+        ];
+        for (let [input, expected] of testCases) {
+            it(`should normalize '${input}' to '${expected}'`, function () {
+                assert.equal(defaultSearchTermToKey(input), expected);
+            });
+        }
+    });
+    describe('languageUsesCasing:  true (custom applyCasing, inverts lower- & upper- casing)', function () {
+        const testCases = [
+            // The DEFAULT function is NOT responsible for understanding the Turkish
+            // distinction between U+0130 LATIN CAPITAL LETTER I WITH DOT ABOVE and
+            // 'I' U+0048 LATIN CAPITAL LETTER I.
+            // For Turkic languages, the recommendation is thus to make a
+            // custom searchTermToKey function.
+            ['İstanbul', 'ISTANBUL'],
+            ['DİYARBAKIR', 'DIYARBAKIR'],
+            // "skýlos" is Greek for dog 🇬🇷🐶
+            // starts with an 's' and ends with an 's'
+            // which are DIFFERENT CHARACTERS in lowercased Greek!
+            ['σκύλος', 'ΣΚΥΛΟΣ'],
+            ['σκυλοσ', 'ΣΚΥΛΟΣ'],
+            // full-width romaji is compatibility-canonical with ASCII characters:
+            ['ａｅｓｔｈｅｔｉｃ', 'AESTHETIC'],
+            // U+212B ANGSTROM SIGN (Å)
+            // U+00C5 LATIN CAPITAL LETTER A WITH RING ABOVE (Å)
+            // and should both normalize to 'a'
+            ['\u212B', 'A'],
+            ['\u00C5', 'A'],
+            // We should not fall for U+037E GREEK QUESTION MARK's trolling:
+            ['\u037e', ';'],
+            // Test presentational forms of Arabic:
+            // U+FE8D ARABIC LETTER ALEF ISOLATED FORM -> U+0627 ARABIC LETTER ALEF
+            // U+FEDF ARABIC LETTER LAM INITIAL FORM -> U+0644 ARABIC LETTER LAM
+            // U+FED8 ARABIC LETTER QAF MEDIAL FORM -> U+0642 ARABIC LETTER QAF
+            // U+FEEC ARABIC LETTER HEH MEDIAL FORM -> U+0647 ARABIC LETTER HEH
+            // U+FEEE ARABIC LETTER WAW FINAL FORM -> U+0648 ARABIC LETTER WAW
+            // U+FE93 ARABIC LETTER TEH MARBUTA ISOLATED FORM -> U+0629 ARABIC LETTER TEH MARBUTA
+            ['\uFE8D\uFEDF\uFED8\uFEEC\uFEEE\uFE93', '\u0627\u0644\u0642\u0647\u0648\u0629'],
+            // Combine both NFKD **AND** knocking off diacritics:
+            // U+01C4 LATIN CAPITAL LETTER DZ WITH CARON (DŽ) -> <U+0064, U+007A> (dz)
+            ['DŽ', 'DZ'],
+        ];
+        // While a Turkish-based test would be nice, Turkish needs custom keying,
+        // as U+0130's default handling is... not ideal in Turkish.
+        //
+        // Instead, we can get a simple-enough test with inverted casing.
+        let customCasing = function (caseToApply, text, defaultApplyCasing) {
+            switch (caseToApply) {
+                case 'lower':
+                    return text.toUpperCase();
+                case 'upper':
+                    return text.toLowerCase();
+                case 'initial':
+                    return customCasing('upper', text.charAt(0), defaultApplyCasing) + text.substr(1);
+                default:
+                    return text;
+            }
+        };
+        let customCasingClosure = function (caseToApply, text) {
+            return customCasing(caseToApply, text, defaultApplyCasing);
+        };
+        for (let [input, expected] of testCases) {
+            it(`should normalize '${input}' to '${expected}'`, function () {
+                assert.equal(defaultCasedSearchTermToKey(input, customCasingClosure), expected);
+            });
+        }
+    });
+    describe('languageUsesCasing:  undefined (legacy 12.0 / 13.0 mode, uses `defaultApplyCasing`)', function () {
+        const testCases = [
+            // "İstanbul" has a U+0130 LATIN CAPITAL LETTER I WITH DOT ABOVE.
+            // This should be lowercased.
+            ['İstanbul', 'istanbul'],
+            // The DEFAULT function is NOT responsible for understanding the Turkish
+            // case regarding the lowercasing of:
+            // 'I' U+0048 LATIN CAPITAL LETTER I to 'ı' U+0131 LATIN SMALL LETTER DOTLESS I
+            // For Turkic languages, the recommendation is to make a
+            // custom searchTermToKey function:
+            ['DİYARBAKIR', 'diyarbakir'],
+            // "skýlos" is Greek for dog 🇬🇷🐶
+            // starts with an 's' and ends with an 's'
+            // which are DIFFERENT CHARACTERS in lowercased Greek!
+            ['σκύλος', 'σκυλος'],
+            ['ΣΚΥΛΟΣ', 'σκυλοσ'],
+            // full-width romaji is compatibility-canonical with ASCII characters:
+            ['ａｅｓｔｈｅｔｉｃ', 'aesthetic'],
+            // U+212B ANGSTROM SIGN (Å)
+            // U+00C5 LATIN CAPITAL LETTER A WITH RING ABOVE (Å)
+            // and should both normalize to 'a'
+            ['\u212B', 'a'],
+            ['\u00C5', 'a'],
+            // We should not fall for U+037E GREEK QUESTION MARK's trolling:
+            ['\u037e', ';'],
+            // Test presentational forms of Arabic:
+            // U+FE8D ARABIC LETTER ALEF ISOLATED FORM -> U+0627 ARABIC LETTER ALEF
+            // U+FEDF ARABIC LETTER LAM INITIAL FORM -> U+0644 ARABIC LETTER LAM
+            // U+FED8 ARABIC LETTER QAF MEDIAL FORM -> U+0642 ARABIC LETTER QAF
+            // U+FEEC ARABIC LETTER HEH MEDIAL FORM -> U+0647 ARABIC LETTER HEH
+            // U+FEEE ARABIC LETTER WAW FINAL FORM -> U+0648 ARABIC LETTER WAW
+            // U+FE93 ARABIC LETTER TEH MARBUTA ISOLATED FORM -> U+0629 ARABIC LETTER TEH MARBUTA
+            ['\uFE8D\uFEDF\uFED8\uFEEC\uFEEE\uFE93', '\u0627\u0644\u0642\u0647\u0648\u0629'],
+            // Combine both NFKD **AND** knocking off diacritics:
+            // U+01C4 LATIN CAPITAL LETTER DZ WITH CARON (DŽ) -> <U+0064, U+007A> (dz)
+            ['DŽ', 'dz'],
+        ];
+        for (let [input, expected] of testCases) {
+            it(`should normalize '${input}' to '${expected}'`, function () {
+                assert.equal(defaultCasedSearchTermToKey(input, defaultApplyCasing), expected);
+            });
+        }
+    });
+});
+//# sourceMappingURL=test-default-search-term-to-key.js.map

package/build/test/test-default-search-term-to-key.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-default-search-term-to-key.js","sourceRoot":"","sources":["../../test/test-default-search-term-to-key.ts"],"names":[],"mappings":"AACA,OAAO,OAAO,CAAC;AACf,OAAO,EAAC,MAAM,EAAC,MAAM,MAAM,CAAC;AAC5B,OAAO,EAAE,sBAAsB,EACtB,2BAA2B,EAC3B,kBAAkB,EAAE,MAAM,0BAA0B,CAAC;AAG9D,QAAQ,CAAC,wCAAwC,EAAE;IACjD,QAAQ,CAAC,2BAA2B,EAAE;QACpC,MAAM,SAAS,GAAuB;YACpC,iEAAiE;YACjE,+DAA+D;YAC/D,CAAC,UAAU,EAAE,UAAU,CAAC;YAExB,eAAe;YACf,CAAC,YAAY,EAAE,YAAY,CAAC;YAE5B,mCAAmC;YACnC,0CAA0C;YAC1C,sDAAsD;YACtD,CAAC,QAAQ,EAAE,QAAQ,CAAC;YACpB,CAAC,QAAQ,EAAE,QAAQ,CAAC;YAEpB,sEAAsE;YACtE,CAAC,WAAW,EAAE,WAAW,CAAC;YAE1B,2BAA2B;YAC3B,oDAAoD;YACpD,mCAAmC;YACnC,CAAC,QAAQ,EAAE,GAAG,CAAC;YACf,CAAC,QAAQ,EAAE,GAAG,CAAC;YAEf,gEAAgE;YAChE,CAAC,QAAQ,EAAE,GAAG,CAAC;YAEf,uCAAuC;YACvC,uEAAuE;YACvE,oEAAoE;YACpE,mEAAmE;YACnE,mEAAmE;YACnE,kEAAkE;YAClE,qFAAqF;YACrF,CAAC,sCAAsC,EAAE,sCAAsC,CAAC;YAEhF,qDAAqD;YACrD,yEAAyE;YACzE,CAAC,GAAG,EAAE,IAAI,CAAC;SACZ,CAAC;QAEF,KAAK,IAAI,CAAC,KAAK,EAAE,QAAQ,CAAC,IAAI,SAAS,EAAE;YACvC,EAAE,CAAC,qBAAqB,KAAK,SAAS,QAAQ,GAAG,EAAE;gBACjD,MAAM,CAAC,KAAK,CAAC,sBAAsB,CAAC,KAAK,CAAC,EAAE,QAAQ,CAAC,CAAC;YACxD,CAAC,CAAC,CAAC;SACJ;IACH,CAAC,CAAC,CAAC;IAEH,QAAQ,CAAC,gFAAgF,EAAE;QACzF,MAAM,SAAS,GAAuB;YACpC,wEAAwE;YACxE,uEAAuE;YACvE,qCAAqC;YACrC,6DAA6D;YAC7D,mCAAmC;YACnC,CAAC,UAAU,EAAE,UAAU,CAAC;YACxB,CAAC,YAAY,EAAE,YAAY,CAAC;YAE5B,mCAAmC;YACnC,0CAA0C;YAC1C,sDAAsD;YACtD,CAAC,QAAQ,EAAE,QAAQ,CAAC;YACpB,CAAC,QAAQ,EAAE,QAAQ,CAAC;YAEpB,sEAAsE;YACtE,CAAC,WAAW,EAAE,WAAW,CAAC;YAE1B,2BAA2B;YAC3B,oDAAoD;YACpD,mCAAmC;YACnC,CAAC,QAAQ,EAAE,GAAG,CAAC;YACf,CAAC,QAAQ,EAAE,GAAG,CAAC;YAEf,gEAAgE;YAChE,CAAC,QAAQ,EAAE,GAAG,CAAC;YAEf,uCAAuC;YACvC,uEAAuE;YACvE,oEAAoE;YACpE,mEAAmE;YACnE,mEAAmE;YACnE,kEAAkE;YAClE,qFAAqF;YACrF,CAAC,sCAAsC,EAAE,sCAAsC,CAAC;YAEhF,qDAAqD;YACrD,yEAAyE;YACzE,CAAC,GAAG,EAAE,IAAI,CAAC;SACZ,CAAC;QAEF,yEAAyE;QACzE,2DAA2D;QAC3D,EAAE;QACF,iEAAiE;QACjE,IAAI,YAAY,GAAG,UAAS,WAAuB,EACvB,IAAY,EACZ,kBAAkC;YAC5D,QAAO,WAAW,EAAE;gBAClB,KAAK,OAAO;oBACR,OAAO,IAAI,CAAC,WAAW,EAAE,CAAC;gBAC9B,KAAK,OAAO;oBACR,OAAO,IAAI,CAAC,WAAW,EAAE,CAAC;gBAC9B,KAAK,SAAS;oBACV,OAAO,YAAY,CAAC,OAAO,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,kBAAkB,CAAC,GAAG,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;gBACtF;oBACI,OAAO,IAAI,CAAC;aACf;QACL,CAAC,CAAA;QAED,IAAI,mBAAmB,GAAG,UAAS,WAAuB,EAAE,IAAY;YACtE,OAAO,YAAY,CAAC,WAAW,EAAE,IAAI,EAAE,kBAAkB,CAAC,CAAC;QAC7D,CAAC,CAAA;QAED,KAAK,IAAI,CAAC,KAAK,EAAE,QAAQ,CAAC,IAAI,SAAS,EAAE;YACvC,EAAE,CAAC,qBAAqB,KAAK,SAAS,QAAQ,GAAG,EAAE;gBACjD,MAAM,CAAC,KAAK,CAAC,2BAA2B,CAAC,KAAK,EAAE,mBAAqC,CAAC,EAAE,QAAQ,CAAC,CAAC;YACpG,CAAC,CAAC,CAAC;SACJ;IACH,CAAC,CAAC,CAAC;IAEH,QAAQ,CAAC,qFAAqF,EAAE;QAC9F,MAAM,SAAS,GAAuB;YACpC,iEAAiE;YACjE,6BAA6B;YAC7B,CAAC,UAAU,EAAE,UAAU,CAAC;YAExB,wEAAwE;YACxE,qCAAqC;YACrC,+EAA+E;YAC/E,wDAAwD;YACxD,mCAAmC;YACnC,CAAC,YAAY,EAAE,YAAY,CAAC;YAE5B,mCAAmC;YACnC,0CAA0C;YAC1C,sDAAsD;YACtD,CAAC,QAAQ,EAAE,QAAQ,CAAC;YACpB,CAAC,QAAQ,EAAE,QAAQ,CAAC;YAEpB,sEAAsE;YACtE,CAAC,WAAW,EAAE,WAAW,CAAC;YAE1B,2BAA2B;YAC3B,oDAAoD;YACpD,mCAAmC;YACnC,CAAC,QAAQ,EAAE,GAAG,CAAC;YACf,CAAC,QAAQ,EAAE,GAAG,CAAC;YAEf,gEAAgE;YAChE,CAAC,QAAQ,EAAE,GAAG,CAAC;YAEf,uCAAuC;YACvC,uEAAuE;YACvE,oEAAoE;YACpE,mEAAmE;YACnE,mEAAmE;YACnE,kEAAkE;YAClE,qFAAqF;YACrF,CAAC,sCAAsC,EAAE,sCAAsC,CAAC;YAEhF,qDAAqD;YACrD,yEAAyE;YACzE,CAAC,GAAG,EAAE,IAAI,CAAC;SACZ,CAAC;QAEF,KAAK,IAAI,CAAC,KAAK,EAAE,QAAQ,CAAC,IAAI,SAAS,EAAE;YACvC,EAAE,CAAC,qBAAqB,KAAK,SAAS,QAAQ,GAAG,EAAE;gBACjD,MAAM,CAAC,KAAK,CAAC,2BAA2B,CAAC,KAAK,EAAE,kBAAkB,CAAC,EAAE,QAAQ,CAAC,CAAC;YACjF,CAAC,CAAC,CAAC;SACJ;IACH,CAAC,CAAC,CAAC;AACL,CAAC,CAAC,CAAC"}

package/build/test/test-error-logger.d.ts

@@ -0,0 +1,2 @@
+export {};
+//# sourceMappingURL=test-error-logger.d.ts.map

package/build/test/test-error-logger.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-error-logger.d.ts","sourceRoot":"","sources":["../../test/test-error-logger.ts"],"names":[],"mappings":""}

package/build/test/test-error-logger.js

@@ -0,0 +1,26 @@
+import { MAX_MESSAGES, KeymanCompilerError, log } from "../src/model-compiler-errors.js";
+import { LogHoarder } from "./helpers/index.js";
+import { assert } from "chai";
+describe('error logger', function () {
+    beforeEach(function () {
+        this.logHoarder = (new LogHoarder).install();
+    });
+    afterEach(function () {
+        this.logHoarder.uninstall();
+        delete this.logHoarder;
+    });
+    it('should stop logging messages **after** a maximum', function () {
+        for (let i = 0; i < MAX_MESSAGES; i++) {
+            log(KeymanCompilerError.CWARN_DuplicateWordInSameFile, "fake error");
+        }
+        // We've logged *just enough messages. This error should not be found:
+        assert.isFalse(this.logHoarder.hasSeenCode(KeymanCompilerError.CWARN_TooManyErrorsOrWarnings));
+        // Log just one too many:
+        log(KeymanCompilerError.CWARN_DuplicateWordInSameFile, "fake error");
+        assert.isTrue(this.logHoarder.hasSeenCode(KeymanCompilerError.CWARN_TooManyErrorsOrWarnings));
+        // Log a DIFFERENT error -- it should not appear in the log
+        log(KeymanCompilerError.CWARN_MixedNormalizationForms, "fake error");
+        assert.isFalse(this.logHoarder.hasSeenCode(KeymanCompilerError.CWARN_MixedNormalizationForms));
+    });
+});
+//# sourceMappingURL=test-error-logger.js.map

package/build/test/test-error-logger.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-error-logger.js","sourceRoot":"","sources":["../../test/test-error-logger.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,YAAY,EAAE,mBAAmB,EAAE,GAAG,EAAE,MAAM,iCAAiC,CAAC;AACzF,OAAO,EAAE,UAAU,EAAE,MAAM,oBAAoB,CAAC;AAChD,OAAO,EAAE,MAAM,EAAE,MAAM,MAAM,CAAC;AAE9B,QAAQ,CAAC,cAAc,EAAE;IACvB,UAAU,CAAC;QACT,IAAI,CAAC,UAAU,GAAG,CAAC,IAAI,UAAU,CAAC,CAAC,OAAO,EAAE,CAAA;IAC9C,CAAC,CAAC,CAAA;IAEF,SAAS,CAAC;QACR,IAAI,CAAC,UAAU,CAAC,SAAS,EAAE,CAAC;QAC5B,OAAO,IAAI,CAAC,UAAU,CAAC;IACzB,CAAC,CAAC,CAAA;IAEF,EAAE,CAAC,kDAAkD,EAAE;QACrD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,YAAY,EAAE,CAAC,EAAE,EAAE;YACrC,GAAG,CAAC,mBAAmB,CAAC,6BAA6B,EAAE,YAAY,CAAC,CAAC;SACtE;QAED,sEAAsE;QACtE,MAAM,CAAC,OAAO,CAAC,IAAI,CAAC,UAAU,CAAC,WAAW,CACxC,mBAAmB,CAAC,6BAA6B,CAClD,CAAC,CAAC;QAEH,yBAAyB;QACzB,GAAG,CAAC,mBAAmB,CAAC,6BAA6B,EAAE,YAAY,CAAC,CAAC;QAErE,MAAM,CAAC,MAAM,CAAC,IAAI,CAAC,UAAU,CAAC,WAAW,CACvC,mBAAmB,CAAC,6BAA6B,CAClD,CAAC,CAAC;QAEH,2DAA2D;QAC3D,GAAG,CAAC,mBAAmB,CAAC,6BAA6B,EAAE,YAAY,CAAC,CAAC;QACrE,MAAM,CAAC,OAAO,CAAC,IAAI,CAAC,UAAU,CAAC,WAAW,CACxC,mBAAmB,CAAC,6BAA6B,CAClD,CAAC,CAAC;IACL,CAAC,CAAC,CAAA;AACJ,CAAC,CAAC,CAAA"}

package/build/test/test-join-word-breaker.d.ts

@@ -0,0 +1,2 @@
+export {};
+//# sourceMappingURL=test-join-word-breaker.d.ts.map

package/build/test/test-join-word-breaker.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-join-word-breaker.d.ts","sourceRoot":"","sources":["../../test/test-join-word-breaker.ts"],"names":[],"mappings":""}

package/build/test/test-join-word-breaker.js

@@ -0,0 +1,84 @@
+import { assert } from "chai";
+import defaultWordBreaker from './wordbreakers/default-wordbreaker-esm.js';
+import { decorateWithJoin } from '../src/join-word-breaker-decorator.js';
+describe('The join word breaker decorator', function () {
+    it('should decorate an existing word breaker', function () {
+        let breakWords = decorateWithJoin(defaultWordBreaker, ['-']);
+        assert.isFunction(breakWords);
+    });
+    const TEST_CASES = [
+        /* input,      joiners,  default breaks,        breaks with joins */
+        // Original test case from https://github.com/keymanapp/keyman/issues/2753
+        ['khui-chhùi', ['-'], ["khui", "-", "chhùi"], ["khui-chhùi"]],
+        // Plains Cree SRO:
+        ['ê-kotiskâwêyâhk', ['-'], ['ê', '-', 'kotiskâwêyâhk'], ['ê-kotiskâwêyâhk']],
+        // Edge cases:
+        // Joiner alone:
+        ['-', ['-'], ['-'], ['-']],
+        // Joiner at the end:
+        ['ni-', ['-'], ['ni', '-'], ['ni-']],
+        // Joiner at the end:
+        ['-ân', ['-'], ['-', 'ân'], ['-ân']],
+        // This was my guiding test case:
+        [
+            "-yâhk ê-nitawi-kotiskâwêyâhk ni-",
+            ["-"],
+            ["-", "yâhk", "ê", "-", "nitawi", "-", "kotiskâwêyâhk", "ni", "-"],
+            ["-yâhk", "ê-nitawi-kotiskâwêyâhk", "ni-"]
+        ],
+        // Do not perform any joins:
+        ["hello world", ["-"], ["hello", "world"], ["hello", "world"]],
+        // Joining using multiple joiners
+        [
+            "Email: no-body@example.com",
+            ["@", "-"],
+            ["Email", ":", "no", "-", "body", "@", "example.com"],
+            ["Email", ":", "no-body@example.com"]
+        ],
+        // Joining with two or more joiners in a row
+        [
+            "nobody@@example.com",
+            ["@"],
+            ["nobody", "@", "@", "example.com"],
+            ["nobody@@example.com"]
+        ],
+        // it should NOT join non-contiguous spans:
+        [
+            "this- is -bad",
+            ["-"],
+            ["this", "-", "is", "-", "bad"],
+            ["this-", "is", "-bad"]
+        ],
+        // different but adjacent joiners
+        [
+            "I made the kawé:-conjugator.",
+            ["-", ":"],
+            ["I", "made", "the", "kawé", ":", "-", "conjugator", "."],
+            ["I", "made", "the", "kawé:-conjugator", "."]
+        ],
+        // 3+ joiners in a row
+        [
+            // NB: – is U+2001 EN DASH
+            "This language is nut–=💠¤~ty!",
+            ["~", "–", "¤", "=", "💠"],
+            ["This", "language", "is", "nut", "–", "=", "💠", "¤", "~", "ty", "!"],
+            ["This", "language", "is", "nut–=💠¤~ty", "!"],
+        ],
+    ];
+    for (let [phrase, joiners, unjoined, expected] of TEST_CASES) {
+        it(`should break «${[phrase]}» as [${expected.join(' ;; ')}]`, function () {
+            let breakWords = decorateWithJoin(defaultWordBreaker, joiners);
+            let unjoinedResult = defaultWordBreaker(phrase).map(onlyText);
+            assert.deepEqual(unjoinedResult, unjoined);
+            let actualResult = breakWords(phrase).map(onlyText);
+            assert.deepEqual(actualResult, expected);
+        });
+    }
+    /**
+     * Get just the text from a span.
+     */
+    function onlyText(span) {
+        return span.text;
+    }
+});
+//# sourceMappingURL=test-join-word-breaker.js.map

package/build/test/test-join-word-breaker.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-join-word-breaker.js","sourceRoot":"","sources":["../../test/test-join-word-breaker.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,MAAM,EAAE,MAAM,MAAM,CAAC;AAC9B,OAAO,kBAAkB,MAAM,2CAA2C,CAAC;AAC3E,OAAO,EAAC,gBAAgB,EAAC,MAAM,uCAAuC,CAAC;AAEvE,QAAQ,CAAC,iCAAiC,EAAE;IAC1C,EAAE,CAAC,0CAA0C,EAAE;QAC7C,IAAI,UAAU,GAAG,gBAAgB,CAAC,kBAAkB,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC;QAC7D,MAAM,CAAC,UAAU,CAAC,UAAU,CAAC,CAAC;IAChC,CAAC,CAAC,CAAC;IAEH,MAAM,UAAU,GAA6C;QAC3D,oEAAoE;QACpE,0EAA0E;QAC1E,CAAC,YAAY,EAAE,CAAC,GAAG,CAAC,EAAI,CAAC,MAAM,EAAE,GAAG,EAAE,OAAO,CAAC,EAAE,CAAC,YAAY,CAAC,CAAC;QAE/D,mBAAmB;QACnB,CAAC,iBAAiB,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,EAAE,GAAG,EAAE,eAAe,CAAC,EAAE,CAAC,iBAAiB,CAAC,CAAC;QAE5E,cAAc;QAEd,gBAAgB;QAChB,CAAC,GAAG,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,CAAC,CAAC;QAC1B,qBAAqB;QACrB,CAAC,KAAK,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,IAAI,EAAE,GAAG,CAAC,EAAE,CAAC,KAAK,CAAC,CAAC;QACpC,qBAAqB;QACrB,CAAC,KAAK,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,EAAE,IAAI,CAAC,EAAE,CAAC,KAAK,CAAC,CAAC;QAEpC,iCAAiC;QACjC;YACE,kCAAkC;YAClC,CAAC,GAAG,CAAC;YACL,CAAC,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,GAAG,EAAE,QAAQ,EAAE,GAAG,EAAE,eAAe,EAAE,IAAI,EAAE,GAAG,CAAC;YAClE,CAAC,OAAO,EAAE,wBAAwB,EAAE,KAAK,CAAC;SAC3C;QAED,4BAA4B;QAC5B,CAAC,aAAa,EAAE,CAAC,GAAG,CAAC,EAAE,CAAC,OAAO,EAAE,OAAO,CAAC,EAAE,CAAC,OAAO,EAAE,OAAO,CAAC,CAAC;QAE9D,iCAAiC;QACjC;YACE,4BAA4B;YAC5B,CAAC,GAAG,EAAE,GAAG,CAAC;YACV,CAAC,OAAO,EAAE,GAAG,EAAE,IAAI,EAAE,GAAG,EAAE,MAAM,EAAE,GAAG,EAAE,aAAa,CAAC;YACrD,CAAC,OAAO,EAAE,GAAG,EAAE,qBAAqB,CAAC;SACtC;QAED,4CAA4C;QAC5C;YACE,qBAAqB;YACrB,CAAC,GAAG,CAAC;YACL,CAAC,QAAQ,EAAE,GAAG,EAAE,GAAG,EAAE,aAAa,CAAC;YACnC,CAAC,qBAAqB,CAAC;SACxB;QAED,2CAA2C;QAC3C;YACE,eAAe;YACf,CAAC,GAAG,CAAC;YACL,CAAC,MAAM,EAAE,GAAG,EAAE,IAAI,EAAE,GAAG,EAAE,KAAK,CAAC;YAC/B,CAAC,OAAO,EAAE,IAAI,EAAE,MAAM,CAAC;SACxB;QAED,iCAAiC;QACjC;YACE,8BAA8B;YAC9B,CAAC,GAAG,EAAE,GAAG,CAAC;YACV,CAAC,GAAG,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,EAAE,GAAG,EAAE,GAAG,EAAE,YAAY,EAAE,GAAG,CAAC;YACzD,CAAC,GAAG,EAAE,MAAM,EAAE,KAAK,EAAE,kBAAkB,EAAE,GAAG,CAAC;SAC9C;QAED,sBAAsB;QACtB;YACE,0BAA0B;YAC1B,+BAA+B;YAC/B,CAAC,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,GAAG,EAAE,IAAI,CAAC;YAC1B,CAAC,MAAM,EAAE,UAAU,EAAE,IAAI,EAAE,KAAK,EAAE,GAAG,EAAE,GAAG,EAAE,IAAI,EAAE,GAAG,EAAE,GAAG,EAAE,IAAI,EAAE,GAAG,CAAC;YACtE,CAAC,MAAM,EAAE,UAAU,EAAE,IAAI,EAAE,aAAa,EAAE,GAAG,CAAC;SAC/C;KACF,CAAA;IAED,KAAK,IAAI,CAAC,MAAM,EAAE,OAAO,EAAE,QAAQ,EAAE,QAAQ,CAAC,IAAI,UAAU,EAAE;QAC5D,EAAE,CAAC,iBAAiB,CAAC,MAAM,CAAC,SAAS,QAAQ,CAAC,IAAI,CAAC,MAAM,CAAC,GAAG,EAAE;YAC7D,IAAI,UAAU,GAAG,gBAAgB,CAAC,kBAAkB,EAAE,OAAO,CAAC,CAAC;YAC/D,IAAI,cAAc,GAAG,kBAAkB,CAAC,MAAM,CAAC,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;YAC9D,MAAM,CAAC,SAAS,CAAC,cAAc,EAAE,QAAQ,CAAC,CAAC;YAC3C,IAAI,YAAY,GAAG,UAAU,CAAC,MAAM,CAAC,CAAC,GAAG,CAAC,QAAQ,CAAC,CAAC;YACpD,MAAM,CAAC,SAAS,CAAC,YAAY,EAAE,QAAQ,CAAC,CAAC;QAC3C,CAAC,CAAC,CAAC;KACJ;IAED;;OAEG;IACH,SAAS,QAAQ,CAAC,IAAU;QAC1B,OAAO,IAAI,CAAC,IAAI,CAAC;IACnB,CAAC;AACH,CAAC,CAAC,CAAC"}

package/build/test/test-model-definitions.d.ts

@@ -0,0 +1,2 @@
+import 'mocha';
+//# sourceMappingURL=test-model-definitions.d.ts.map

package/build/test/test-model-definitions.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-model-definitions.d.ts","sourceRoot":"","sources":["../../test/test-model-definitions.ts"],"names":[],"mappings":"AAAA,OAAO,OAAO,CAAC"}

package/build/test/test-model-definitions.js

@@ -0,0 +1,165 @@
+import 'mocha';
+import { assert } from 'chai';
+import { ModelDefinitions } from '../src/model-definitions.js';
+describe('Model definition pseudoclosures', function () {
+    describe('14.0 defaults', function () {
+        describe('languageUsesCasing == true', function () {
+            // We don't need a complete spec for this, given the (currently) limited range of what
+            // the ModelPseudoclosure covers.
+            let modelSource = {
+                languageUsesCasing: true,
+                sources: [],
+                format: 'trie-1.0'
+            };
+            let pseudoclosure = new ModelDefinitions(modelSource);
+            const testCases = [
+                // Note:  not written the Turkish way.  Turns out 'İ'.toLowerCase() decomposes the result,
+                // which would have made for a fairly yucky test.
+                ['Istanbul', 'istanbul', 'istanbul'],
+                // The DEFAULT function is NOT responsible for understanding the Turkish
+                // case regarding the lowercasing of:
+                // 'I' U+0048 LATIN CAPITAL LETTER I to 'ı' U+0131 LATIN SMALL LETTER DOTLESS I
+                // For Turkic languages, the recommendation is to make a
+                // custom applyCasing function:
+                ['DİYARBAKIR', 'di̇yarbakir', 'diyarbakir'],
+                // full-width romaji has corresponding lowercased versions:
+                ['ＡＥＳＴＨＥＴＩＣ', 'ａｅｓｔｈｅｔｉｃ', 'aesthetic'],
+                // "skýlos" is Greek for dog 🇬🇷🐶
+                // starts with an 's' and ends with an 's'
+                // which are DIFFERENT CHARACTERS in lowercased Greek!
+                ['σκύλος', 'σκύλος', 'σκυλος'],
+                ['ΣΚΥΛΟΣ', 'σκυλος', 'σκυλοσ'],
+                // to make the distinction.  Both 'Σ's have the same char-code.
+                // Uncased syntax and numbers should pass through unscathed:
+                ['1234.?!', '1234.?!', '1234.?!'],
+                ['”', '”', '"'],
+                ["‘", "‘", "'"]
+            ];
+            for (let [input, cased, keyed] of testCases) {
+                it(`should case '${input}' as '${cased}'`, function () {
+                    assert.equal(pseudoclosure.applyCasing('lower', input), cased);
+                });
+                it(`should key  '${input}' as '${keyed}'`, function () {
+                    assert.equal(pseudoclosure.searchTermToKey(input), keyed);
+                });
+            }
+        });
+        describe('languageUsesCasing == false', function () {
+            // We don't need a complete spec for this, given the (currently) limited range of what
+            // the ModelPseudoclosure covers.
+            let modelSource = {
+                languageUsesCasing: false,
+                sources: [],
+                format: 'trie-1.0'
+            };
+            let pseudoclosure = new ModelDefinitions(modelSource);
+            const testCases = [
+                // Note:  not written the Turkish way.  Turns out 'İ'.toLowerCase() decomposes the result,
+                // which would have made for a fairly yucky test.
+                ['Istanbul', 'Istanbul'],
+                ['DİYARBAKIR', 'DIYARBAKIR'],
+                // full-width romaji has corresponding capitalized versions:
+                ['ＡＥＳＴＨＥＴＩＣ', 'AESTHETIC'],
+                // "skýlos" is Greek for dog 🇬🇷🐶
+                // starts with an 's' and ends with an 's'
+                // which are DIFFERENT CHARACTERS in lowercased Greek!
+                ['σκύλος', 'σκυλος'],
+                ['ΣΚΥΛΟΣ', 'ΣΚΥΛΟΣ'],
+                // to make the distinction.  Both 'Σ's have the same char-code.
+                // Uncased syntax and numbers should pass through unscathed:
+                ['1234.?!', '1234.?!'],
+                ['”', '"'],
+                ["‘", "'"]
+            ];
+            for (let [input, keyed] of testCases) {
+                it(`should key  '${input}' as '${keyed}'`, function () {
+                    assert.equal(pseudoclosure.searchTermToKey(input), keyed);
+                });
+            }
+        });
+    });
+    describe('Pre 14.0 defaults (languageUsesCasing == undefined)', function () {
+        // We don't need a complete spec for this, given the (currently) limited range of what
+        // the ModelPseudoclosure covers.
+        let modelSource = {
+            sources: [],
+            format: 'trie-1.0'
+        };
+        let pseudoclosure = new ModelDefinitions(modelSource);
+        const testCases = [
+            // Note:  not written the Turkish way.  Turns out 'İ'.toLowerCase() decomposes the result,
+            // which would have made for a fairly yucky test.
+            ['Istanbul', 'istanbul'],
+            ['DİYARBAKIR', 'diyarbakir'],
+            // full-width romaji has corresponding capitalized versions:
+            ['ＡＥＳＴＨＥＴＩＣ', 'aesthetic'],
+            // "skýlos" is Greek for dog 🇬🇷🐶
+            // starts with an 's' and ends with an 's'
+            // which are DIFFERENT CHARACTERS in lowercased Greek!
+            ['σκύλος', 'σκυλος'],
+            ['ΣΚΥΛΟΣ', 'σκυλοσ'],
+            // to make the distinction.  Both 'Σ's have the same char-code.
+            // Uncased syntax and numbers should pass through unscathed:
+            ['1234.?!', '1234.?!']
+        ];
+        for (let [input, keyed] of testCases) {
+            it(`should key  '${input}' as '${keyed}'`, function () {
+                assert.equal(pseudoclosure.searchTermToKey(input), keyed);
+            });
+        }
+    });
+    describe('Model-defined applyCasing + (dependent) searchTermToKey', function () {
+        // Note:  this test only implements enough Turkish-related stuff to facilitate
+        // a functional test.  Not guaranteed to be sufficient for actual Turkish use.
+        let turkishCasing = function (form, text, defaultApplyCasing) {
+            switch (form) {
+                case 'lower':
+                    return defaultApplyCasing(form, text
+                        .replace(/I/g, 'ı')
+                        .replace(/İ/g, 'i'));
+                case 'upper':
+                    return defaultApplyCasing(form, text
+                        .replace(/ı/g, 'I')
+                        .replace(/i/g, 'İ'));
+                case 'initial':
+                    return turkishCasing('upper', text.charAt(0), defaultApplyCasing) + text.substr(1);
+                default:
+                    return text;
+            }
+        };
+        let modelSource = {
+            languageUsesCasing: true,
+            applyCasing: turkishCasing,
+            searchTermToKey: function (wordform, applyCasing) {
+                return Array.from(wordform
+                    .normalize('NFC') // Mostly to avoid decomposing 'İ'
+                ) // end of `Array.from`
+                    .map(function (c) { return applyCasing('lower', c); }) // Will use custom `applyCasing` definition!
+                    .join('');
+            },
+            sources: [],
+            format: 'trie-1.0'
+        };
+        let pseudoclosure = new ModelDefinitions(modelSource);
+        const testCases = [
+            ['İstanbul', 'istanbul', 'istanbul'],
+            // The DEFAULT function is NOT responsible for understanding the Turkish
+            // case regarding the lowercasing of:
+            // 'I' U+0048 LATIN CAPITAL LETTER I to 'ı' U+0131 LATIN SMALL LETTER DOTLESS I
+            // For Turkic languages, the recommendation is to make a
+            // custom applyCasing function:
+            ['DİYARBAKIR', 'diyarbakır', 'diyarbakır'],
+            // Uncased syntax and numbers should pass through unscathed:
+            ['1234.?!', '1234.?!', '1234.?!']
+        ];
+        for (let [input, cased, keyed] of testCases) {
+            it(`should case '${input}' as '${cased}'`, function () {
+                assert.equal(pseudoclosure.applyCasing('lower', input), cased);
+            });
+            it(`should key  '${input}' as '${keyed}'`, function () {
+                assert.equal(pseudoclosure.searchTermToKey(input), keyed);
+            });
+        }
+    });
+});
+//# sourceMappingURL=test-model-definitions.js.map

package/build/test/test-model-definitions.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-model-definitions.js","sourceRoot":"","sources":["../../test/test-model-definitions.ts"],"names":[],"mappings":"AAAA,OAAO,OAAO,CAAC;AACf,OAAO,EAAE,MAAM,EAAE,MAAM,MAAM,CAAC;AAC9B,OAAO,EAAE,gBAAgB,EAAE,MAAM,6BAA6B,CAAC;AAG/D,QAAQ,CAAC,iCAAiC,EAAE;IAC1C,QAAQ,CAAC,eAAe,EAAE;QACxB,QAAQ,CAAC,4BAA4B,EAAE;YACrC,sFAAsF;YACtF,iCAAiC;YACjC,IAAI,WAAW,GAAuB;gBACpC,kBAAkB,EAAE,IAAI;gBACxB,OAAO,EAAE,EAAE;gBACX,MAAM,EAAE,UAAU;aACnB,CAAC;YAEF,IAAI,aAAa,GAAG,IAAI,gBAAgB,CAAC,WAAW,CAAC,CAAC;YAEtD,MAAM,SAAS,GAA+B;gBAC5C,0FAA0F;gBAC1F,iDAAiD;gBACjD,CAAC,UAAU,EAAE,UAAU,EAAE,UAAU,CAAC;gBAEpC,wEAAwE;gBACxE,qCAAqC;gBACrC,+EAA+E;gBAC/E,wDAAwD;gBACxD,+BAA+B;gBAC/B,CAAC,YAAY,EAAE,aAAa,EAAE,YAAY,CAAC;gBAE3C,2DAA2D;gBAC3D,CAAC,WAAW,EAAE,WAAW,EAAE,WAAW,CAAC;gBAEvC,mCAAmC;gBACnC,0CAA0C;gBAC1C,sDAAsD;gBACtD,CAAC,QAAQ,EAAE,QAAQ,EAAE,QAAQ,CAAC;gBAC9B,CAAC,QAAQ,EAAE,QAAQ,EAAE,QAAQ,CAAC;gBACG,+DAA+D;gBAEhG,4DAA4D;gBAC5D,CAAC,SAAS,EAAE,SAAS,EAAE,SAAS,CAAC;gBACjC,CAAC,GAAG,EAAE,GAAG,EAAE,GAAG,CAAC;gBACf,CAAC,GAAG,EAAE,GAAG,EAAE,GAAG,CAAC;aAChB,CAAC;YAEF,KAAK,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,IAAI,SAAS,EAAE;gBAC3C,EAAE,CAAC,gBAAgB,KAAK,SAAS,KAAK,GAAG,EAAE;oBACzC,MAAM,CAAC,KAAK,CAAC,aAAa,CAAC,WAAW,CAAC,OAAO,EAAE,KAAK,CAAC,EAAE,KAAK,CAAC,CAAC;gBACjE,CAAC,CAAC,CAAC;gBAEH,EAAE,CAAC,gBAAgB,KAAK,SAAS,KAAK,GAAG,EAAE;oBACzC,MAAM,CAAC,KAAK,CAAC,aAAa,CAAC,eAAe,CAAC,KAAK,CAAC,EAAE,KAAK,CAAC,CAAC;gBAC5D,CAAC,CAAC,CAAC;aACJ;QACH,CAAC,CAAC,CAAC;QAEH,QAAQ,CAAC,6BAA6B,EAAE;YACtC,sFAAsF;YACtF,iCAAiC;YACjC,IAAI,WAAW,GAAuB;gBACpC,kBAAkB,EAAE,KAAK;gBACzB,OAAO,EAAE,EAAE;gBACX,MAAM,EAAE,UAAU;aACnB,CAAC;YAEF,IAAI,aAAa,GAAG,IAAI,gBAAgB,CAAC,WAAW,CAAC,CAAC;YAEtD,MAAM,SAAS,GAAuB;gBACpC,0FAA0F;gBAC1F,iDAAiD;gBACjD,CAAC,UAAU,EAAE,UAAU,CAAC;gBAExB,CAAC,YAAY,EAAE,YAAY,CAAC;gBAE5B,4DAA4D;gBAC5D,CAAC,WAAW,EAAE,WAAW,CAAC;gBAE1B,mCAAmC;gBACnC,0CAA0C;gBAC1C,sDAAsD;gBACtD,CAAC,QAAQ,EAAE,QAAQ,CAAC;gBACpB,CAAC,QAAQ,EAAE,QAAQ,CAAC;gBACG,+DAA+D;gBAEtF,4DAA4D;gBAC5D,CAAC,SAAS,EAAE,SAAS,CAAC;gBAEtB,CAAC,GAAG,EAAE,GAAG,CAAC;gBACV,CAAC,GAAG,EAAE,GAAG,CAAC;aACX,CAAC;YAEF,KAAK,IAAI,CAAC,KAAK,EAAE,KAAK,CAAC,IAAI,SAAS,EAAE;gBACpC,EAAE,CAAC,gBAAgB,KAAK,SAAS,KAAK,GAAG,EAAE;oBACzC,MAAM,CAAC,KAAK,CAAC,aAAa,CAAC,eAAe,CAAC,KAAK,CAAC,EAAE,KAAK,CAAC,CAAC;gBAC5D,CAAC,CAAC,CAAC;aACJ;QACH,CAAC,CAAC,CAAC;IACL,CAAC,CAAC,CAAC;IACH,QAAQ,CAAC,qDAAqD,EAAE;QAC9D,sFAAsF;QACtF,iCAAiC;QACjC,IAAI,WAAW,GAAuB;YACpC,OAAO,EAAE,EAAE;YACX,MAAM,EAAE,UAAU;SACnB,CAAC;QAEF,IAAI,aAAa,GAAG,IAAI,gBAAgB,CAAC,WAAW,CAAC,CAAC;QAEtD,MAAM,SAAS,GAAuB;YACpC,0FAA0F;YAC1F,iDAAiD;YACjD,CAAC,UAAU,EAAE,UAAU,CAAC;YAExB,CAAC,YAAY,EAAE,YAAY,CAAC;YAE5B,4DAA4D;YAC5D,CAAC,WAAW,EAAE,WAAW,CAAC;YAE1B,mCAAmC;YACnC,0CAA0C;YAC1C,sDAAsD;YACtD,CAAC,QAAQ,EAAE,QAAQ,CAAC;YACpB,CAAC,QAAQ,EAAE,QAAQ,CAAC;YACG,+DAA+D;YAEtF,4DAA4D;YAC5D,CAAC,SAAS,EAAE,SAAS,CAAC;SACvB,CAAC;QAEF,KAAK,IAAI,CAAC,KAAK,EAAE,KAAK,CAAC,IAAI,SAAS,EAAE;YACpC,EAAE,CAAC,gBAAgB,KAAK,SAAS,KAAK,GAAG,EAAE;gBACzC,MAAM,CAAC,KAAK,CAAC,aAAa,CAAC,eAAe,CAAC,KAAK,CAAC,EAAE,KAAK,CAAC,CAAC;YAC5D,CAAC,CAAC,CAAC;SACJ;IACH,CAAC,CAAC,CAAC;IAEH,QAAQ,CAAC,yDAAyD,EAAE;QAClE,8EAA8E;QAC9E,8EAA8E;QAC9E,IAAI,aAAa,GAAG,UAAS,IAAgB,EAAE,IAAY,EAAE,kBAA8D;YACzH,QAAO,IAAI,EAAE;gBACb,KAAK,OAAO;oBACV,OAAO,kBAAkB,CAAC,IAAI,EAAE,IAAI;yBAC/B,OAAO,CAAC,IAAI,EAAE,GAAG,CAAC;yBAClB,OAAO,CAAC,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC;gBAC3B,KAAK,OAAO;oBACV,OAAO,kBAAkB,CAAC,IAAI,EAAE,IAAI;yBAC/B,OAAO,CAAC,IAAI,EAAE,GAAG,CAAC;yBAClB,OAAO,CAAC,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC;gBAC3B,KAAK,SAAS;oBACZ,OAAO,aAAa,CAAC,OAAO,EAAE,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,kBAAkB,CAAC,GAAG,IAAI,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;gBACrF;oBACE,OAAO,IAAI,CAAC;aACb;QACH,CAAC,CAAA;QAED,IAAI,WAAW,GAAuB;YACpC,kBAAkB,EAAE,IAAI;YACxB,WAAW,EAAE,aAAa;YAC1B,eAAe,EAAE,UAAS,QAAgB,EAAE,WAA2B;gBACrE,OAAO,KAAK,CAAC,IAAI,CAAC,QAAQ;qBACvB,SAAS,CAAC,KAAK,CAAC,CAAE,kCAAkC;iBACtD,CAAC,sBAAsB;qBACvB,GAAG,CAAC,UAAS,CAAC,IAAI,OAAO,WAAW,CAAC,OAAO,EAAE,CAAC,CAAC,CAAA,CAAA,CAAC,CAAC,CAAE,4CAA4C;qBAChG,IAAI,CAAC,EAAE,CAAC,CAAC;YACZ,CAAC;YACD,OAAO,EAAE,EAAE;YACX,MAAM,EAAE,UAAU;SACnB,CAAC;QAEF,IAAI,aAAa,GAAG,IAAI,gBAAgB,CAAC,WAAW,CAAC,CAAC;QAEtD,MAAM,SAAS,GAA+B;YAC5C,CAAC,UAAU,EAAE,UAAU,EAAE,UAAU,CAAC;YAEpC,wEAAwE;YACxE,qCAAqC;YACrC,+EAA+E;YAC/E,wDAAwD;YACxD,+BAA+B;YAC/B,CAAC,YAAY,EAAE,YAAY,EAAE,YAAY,CAAC;YAE1C,4DAA4D;YAC5D,CAAC,SAAS,EAAE,SAAS,EAAE,SAAS,CAAC;SAClC,CAAC;QAEF,KAAK,IAAI,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,IAAI,SAAS,EAAE;YAC3C,EAAE,CAAC,gBAAgB,KAAK,SAAS,KAAK,GAAG,EAAE;gBACzC,MAAM,CAAC,KAAK,CAAC,aAAa,CAAC,WAAW,CAAC,OAAO,EAAE,KAAK,CAAC,EAAE,KAAK,CAAC,CAAC;YACjE,CAAC,CAAC,CAAC;YAEH,EAAE,CAAC,gBAAgB,KAAK,SAAS,KAAK,GAAG,EAAE;gBACzC,MAAM,CAAC,KAAK,CAAC,aAAa,CAAC,eAAe,CAAC,KAAK,CAAC,EAAE,KAAK,CAAC,CAAC;YAC5D,CAAC,CAAC,CAAC;SACJ;IACH,CAAC,CAAC,CAAC;AACL,CAAC,CAAC,CAAC"}

package/build/test/test-override-script-defaults.d.ts

@@ -0,0 +1,2 @@
+export {};
+//# sourceMappingURL=test-override-script-defaults.d.ts.map

package/build/test/test-override-script-defaults.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"test-override-script-defaults.d.ts","sourceRoot":"","sources":["../../test/test-override-script-defaults.ts"],"names":[],"mappings":""}