bug fixes and more tests

docs/index.html

@@ -29,4 +29,4 @@ create an optional group
 create an optional group
 	# fragment, again, we don't care, so ignore everything afterwards
 	match "#"
-	match 0+ any thing</textarea><h4>Errors:</h4><textarea readonly="readonly" class="form-control" id="errors" rows="5"></textarea></div><br><div class="col-lg-4 tenpx-margin-bottom"><div class="cheatsheet"><h2>Cheat Sheet:</h2><p>Full documentation available <a href="/tutorial.html">here</a></p><p class="font-weight-bold">Matching</p><p><code class="cm-s-idea">match "hello world"</code> matches "hello world" exactly</p><p></p><p><code class="cm-s-idea">match "hello" then optionally " world"</code> matches "hello" or "hello world"</p><p><code class="cm-s-idea">match "hello" or "world"</code> matches "hello" or "world</p><p><code class="cm-s-idea">match a word</code> matches any word</p><p class="font-weight-bold">Repetition</p><p><code class="cm-s-idea">match 0+ "hello"</code> matches 0 or more "hello"s</p><p><code class="cm-s-idea">match 3 "hello"</code> matches exactly "hellohellohello"</p><p><code class="cm-s-idea">match 1 to 5 "hello"</code> matches between 1 to 5 "hello"s</p><p><code class="cm-s-idea">repeat 0 or more</code> repeats the intended text 0 or more times (default)</p><p><code class="cm-s-idea">optionally repeat between 3 to 5</code> optionally repeats the indented text 3 to 5 times</p><p class="font-weight-bold">Grouping</p><p><code class="cm-s-idea">create a group called mygroup</code> creates a group called "mygroup"</p><p><code class="cm-s-idea">create an optional group</code> creates an unnamed optional group</p><p class="font-weight-bold">Using</p><p><code class="cm-s-idea">using global and case insensitive</code> uses the 'g' and 'i' flags</p><p class="font-weight-bold">Misc</p><p><code class="cm-s-idea">// comment</code> is a single line comment</p><p><code class="cm-s-idea">/* comment */</code> is a multi line comment</p></div></div></div></div><footer><div class="container"><div class="row"><div class="col-lg-8 col-md-10 mx-auto"><p class="copyright">Copyright &copy; 2021 Patrick Demian. This page's source code is available at <a rel="noopener noreferrer" href="https://github.com/pdemian/human2regex">github.com/pdemian/human2regex</a></p></div></div></div></footer></div><script defer="defer" src="/bundle.min.js"></script></body></html>
+	match 0+ any thing</textarea><h4>Errors:</h4><textarea readonly="readonly" class="form-control" id="errors" rows="5"></textarea></div><br><div class="col-lg-4 tenpx-margin-bottom"><div class="cheatsheet"><h2>Cheat Sheet:</h2><p>Full documentation available <a href="/tutorial.html">here</a></p><p class="font-weight-bold">Matching</p><p><code class="cm-s-idea">match "hello world"</code> matches "hello world" exactly</p><p></p><p><code class="cm-s-idea">match "hello" then optionally " world"</code> matches "hello" or "hello world"</p><p><code class="cm-s-idea">match "hello" or "world"</code> matches "hello" or "world"</p><p><code class="cm-s-idea">match a word</code> matches any word</p><p class="font-weight-bold">Repetition</p><p><code class="cm-s-idea">match 0+ "hello"</code> matches 0 or more "hello"s</p><p><code class="cm-s-idea">match 3 "hello"</code> matches exactly "hellohellohello"</p><p><code class="cm-s-idea">match 1 to 5 "hello"</code> matches between 1 to 5 "hello"s</p><p><code class="cm-s-idea">repeat 0 or more</code> repeats the intended text 0 or more times (default)</p><p><code class="cm-s-idea">optionally repeat between 3 to 5</code> optionally repeats the indented text 3 to 5 times</p><p class="font-weight-bold">Grouping</p><p><code class="cm-s-idea">create a group called mygroup</code> creates a group called "mygroup"</p><p><code class="cm-s-idea">create an optional group</code> creates an unnamed optional group</p><p class="font-weight-bold">Using</p><p><code class="cm-s-idea">using global and case insensitive</code> uses the 'g' and 'i' flags</p><p class="font-weight-bold">Misc</p><p><code class="cm-s-idea">// comment</code> is a single line comment</p><p><code class="cm-s-idea">/* comment */</code> is a multi line comment</p></div></div></div></div><footer><div class="container"><div class="row"><div class="col-lg-8 col-md-10 mx-auto"><p class="copyright">Copyright &copy; 2021 Patrick Demian. This page's source code is available at <a rel="noopener noreferrer" href="https://github.com/pdemian/human2regex">github.com/pdemian/human2regex</a></p></div></div></div></footer></div><script defer="defer" src="/bundle.min.js"></script></body></html>

lib/generator_helper.js

@@ -44,9 +44,10 @@ function minMatchString(arr, depth = 0) {
     if (arr.length === 1) {
         return utilities_1.first(arr);
     }
-    // base case: arr is all single letters
+    // base case: arr is all single letters or ranges
-    if (arr.every(utilities_1.isSingleRegexCharacter)) {
+    if (arr.every((value) => utilities_1.isSingleRegexCharacter(value) || utilities_1.isRangeRegex(value))) {
-        return "[" + arr.join("") + "]";
+        // if range, don't forget to remove '[' and ']'
+        return "[" + arr.map((x) => utilities_1.isSingleRegexCharacter(x) ? x : x.substring(1, x.length - 1)).join("") + "]";
     }
     // now the real magic begins
     // You are not expected to understand this

lib/utilities.d.ts

@@ -64,6 +64,16 @@ export declare function combineFlags(value: number, flag: number): number;
  * @internal
  */
 export declare function isSingleRegexCharacter(char: string): boolean;
+/**
+ * Checks to see if the character is a range
+ *
+ * @remarks a range is in the format of [X-Y] where X and Y are valid single regex characters
+ *
+ * @param str the string to check
+ * @returns if the value is a regex range
+ * @internal
+ */
+export declare function isRangeRegex(str: string): boolean;
 /**
  * Gets the first element of an array
  * @remarks does not validate if array has any elements

lib/utilities.js

@@ -1,7 +1,7 @@
 "use strict";
 /*! Copyright (c) 2021 Patrick Demian; Licensed under MIT */
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.CommonError = exports.append = exports.regexEscape = exports.removeQuotes = exports.findLastIndex = exports.last = exports.first = exports.isSingleRegexCharacter = exports.combineFlags = exports.hasFlag = exports.makeFlag = exports.usefulConditional = exports.unusedParameter = void 0;
+exports.CommonError = exports.append = exports.regexEscape = exports.removeQuotes = exports.findLastIndex = exports.last = exports.first = exports.isRangeRegex = exports.isSingleRegexCharacter = exports.combineFlags = exports.hasFlag = exports.makeFlag = exports.usefulConditional = exports.unusedParameter = void 0;
 /**
  * The following section is used because the linter is set up to warn about certain operations
  * and for good reason! I'd much rather have these functions than accidently use bitwise operations, or
@@ -86,6 +86,31 @@ function isSingleRegexCharacter(char) {
         char.length === 1;
 }
 exports.isSingleRegexCharacter = isSingleRegexCharacter;
+/**
+ * Checks to see if the character is a range
+ *
+ * @remarks a range is in the format of [X-Y] where X and Y are valid single regex characters
+ *
+ * @param str the string to check
+ * @returns if the value is a regex range
+ * @internal
+ */
+function isRangeRegex(str) {
+    if (!str.startsWith("[") && !str.endsWith("]")) {
+        return false;
+    }
+    const split = str.substring(1, str.length - 1).split("-");
+    if (split.length !== 2) {
+        return false;
+    }
+    //hack: check to ensure that we aren't escaped
+    if (split[0].endsWith("\\") && split[0] !== "\\\\") {
+        return false;
+    }
+    // ensure the "-" wasn't escaped
+    return isSingleRegexCharacter(split[0]) && isSingleRegexCharacter(split[1]);
+}
+exports.isRangeRegex = isRangeRegex;
 /**
  * Gets the first element of an array
  * @remarks does not validate if array has any elements

package-lock.json

@@ -1,6 +1,6 @@
 {
   "name": "human2regex",
-  "version": "1.1.3",
+  "version": "1.1.4",
   "lockfileVersion": 1,
   "requires": true,
   "dependencies": {

package.json

@@ -1,6 +1,6 @@
 {
   "name": "human2regex",
-  "version": "1.1.3",
+  "version": "1.1.4",
   "description": "Humanized Regular Expressions",
   "main": "./lib/index.js",
   "typings": "./lib/index.d.ts",

src/docs/index.hbs

@@ -41,7 +41,7 @@
 						<p class="font-weight-bold">Matching</p>
 						<p>{{i-code}}match "hello world"{{end-i-code}} matches "hello world" exactly<p>
 						<p>{{i-code}}match "hello" then optionally " world"{{end-i-code}} matches "hello" or "hello world"</p>
-						<p>{{i-code}}match "hello" or "world"{{end-i-code}} matches "hello" or "world</p>
+						<p>{{i-code}}match "hello" or "world"{{end-i-code}} matches "hello" or "world"</p>
 						<p>{{i-code}}match a word{{end-i-code}} matches any word
 						<p class="font-weight-bold">Repetition</p>
 						<p>{{i-code}}match 0+ "hello"{{end-i-code}} matches 0 or more "hello"s</p>

src/generator_helper.ts

@@ -5,7 +5,7 @@
  * @packageDocumentation
  */
 
-import { first, isSingleRegexCharacter } from "./utilities";
+import { first, isRangeRegex, isSingleRegexCharacter } from "./utilities";
 
 /**
  * Minimizes the match string by finding duplicates or substrings in the array
@@ -49,9 +49,10 @@ function minMatchString(arr: string[], depth: number = 0): string {
         return first(arr);
     }
 
-    // base case: arr is all single letters
+    // base case: arr is all single letters or ranges
-    if (arr.every(isSingleRegexCharacter)) {
+    if (arr.every((value) => isSingleRegexCharacter(value) || isRangeRegex(value))) {
-        return "[" + arr.join("") + "]";
+        // if range, don't forget to remove '[' and ']'
+        return "[" + arr.map((x) => isSingleRegexCharacter(x) ? x :x.substring(1, x.length-1)).join("") + "]";
     }
 
     // now the real magic begins

src/utilities.ts

@@ -95,6 +95,35 @@ export function isSingleRegexCharacter(char: string): boolean {
            char.length === 1;
 }
 
+/**
+ * Checks to see if the character is a range
+ * 
+ * @remarks a range is in the format of [X-Y] where X and Y are valid single regex characters
+ * 
+ * @param str the string to check
+ * @returns if the value is a regex range
+ * @internal
+ */
+export function isRangeRegex(str: string): boolean {
+    if (!str.startsWith("[") && !str.endsWith("]")) {
+        return false;
+    }
+
+    const split = str.substring(1, str.length-1).split("-");
+
+    if (split.length !== 2) {
+        return false;
+    }
+    
+    //hack: check to ensure that we aren't escaped
+    if (split[0].endsWith("\\") && split[0] !== "\\\\") {
+        return false;
+    }
+
+    // ensure the "-" wasn't escaped
+    return isSingleRegexCharacter(split[0]) && isSingleRegexCharacter(split[1]);
+}
+
 /**
  * Gets the first element of an array
  * @remarks does not validate if array has any elements

tests/generator.spec.ts

@@ -99,6 +99,16 @@ describe("Generator functionality", function() {
         const reg1 = parser.parse(toks1);
         expect(reg1.validate(RegexDialect.JS).length).toBe(0);
         expect(reg1.toRegex(RegexDialect.JS)).toBe("/[\\u0061-\\u007A]/");
+
+        const toks2 = lexer.tokenize('match "0"..."9" or "a".."z" or "A".."Z"').tokens;
+        const reg2 = parser.parse(toks2);
+        expect(reg2.validate(RegexDialect.JS).length).toBe(0);
+        expect(reg2.toRegex(RegexDialect.JS)).toBe("/[0-9a-zA-Z]/");
+
+        const toks3 = lexer.tokenize('match "0" or "a".."z" or "A".."Z"').tokens;
+        const reg3 = parser.parse(toks3);
+        expect(reg3.validate(RegexDialect.JS).length).toBe(0);
+        expect(reg3.toRegex(RegexDialect.JS)).toBe("/[0a-zA-Z]/");
     });
 
     it("handles specifiers", function() {
@@ -193,6 +203,16 @@ describe("Generator functionality", function() {
         const reg7 = parser.parse(toks7);
         expect(reg7.validate(RegexDialect.JS).length).toBe(0);
         expect(reg7.toRegex(RegexDialect.JS)).toBe("/[^>]*/");
+
+        const toks8 = lexer.tokenize('match "a" or "a"').tokens;
+        const reg8 = parser.parse(toks8);
+        expect(reg8.validate(RegexDialect.JS).length).toBe(0);
+        expect(reg8.toRegex(RegexDialect.JS)).toBe("/a/");
+
+        const toks9 = lexer.tokenize('match "a".."z" or "a".."z"').tokens;
+        const reg9 = parser.parse(toks9);
+        expect(reg9.validate(RegexDialect.JS).length).toBe(0);
+        expect(reg9.toRegex(RegexDialect.JS)).toBe("/[a-z]/");
     });
 
     it("can generate backreferences", function() {

tests/generator_helper.spec.ts

@@ -12,7 +12,9 @@ describe("Generator helper functionality", function() {
             { from: [ "a1x1z", "a2y2z", "a3z3z" ], to: "a(?:1x1|2y2|3z3)z" },
             { from: [ "ab", "cd" ], to: "ab|cd" },
             { from: [ "abc", "bc" ], to: "a?bc" },
-            { from: [ "abc", "xb" ], to: "abc|xb" }
+            { from: [ "abc", "xb" ], to: "abc|xb" },
+            { from: [ "a", "a" ], to: "a" },
+            { from: [ "a-z", "a-z" ], to: "a-z" }
         ];
 
         for (const c of test_cases) {

tests/utilities.spec.ts

@@ -1,7 +1,7 @@
 /*! Copyright (c) 2021 Patrick Demian; Licensed under MIT */
 
 import "../src/utilities";
-import { isSingleRegexCharacter, findLastIndex, removeQuotes, regexEscape, hasFlag, combineFlags, makeFlag, first, last, CommonError, append } from "../src/utilities";
+import { isSingleRegexCharacter, findLastIndex, removeQuotes, regexEscape, hasFlag, combineFlags, makeFlag, first, last, CommonError, append, isRangeRegex } from "../src/utilities";
 import { UsingFlags, ISemanticError } from "../src/generator";
 import { IRecognitionException, ILexingError, createTokenInstance } from "chevrotain";
 import { Indent } from "../src/tokens";
@@ -65,6 +65,18 @@ describe("Utility functions", function() {
         expect(isSingleRegexCharacter("💩")).toBe(false);
     });
 
+    it("can determine if something is a range", function() {
+        expect(isRangeRegex("")).toBe(false);
+        expect(isRangeRegex("-3")).toBe(false);
+        expect(isRangeRegex("[]")).toBe(false);
+        expect(isRangeRegex("[-3]")).toBe(false);
+        expect(isRangeRegex("[a-z]")).toBe(true);
+        expect(isRangeRegex("[\\u1234-\\u1234]")).toBe(true);
+        expect(isRangeRegex("[௹-௹]")).toBe(true);
+        expect(isRangeRegex("[\\-3]")).toBe(false);
+        expect(isRangeRegex("[\\\\-3]")).toBe(true);
+    });
+
     it("should remove quotes correctly", function() {
         expect(removeQuotes('""')).toEqual("");
         expect(removeQuotes('"hello world"')).toEqual("hello world");