More bugs fixed, updated readme

Readme.md

@@ -2,21 +2,22 @@
 
 ## Purpose
 
-Generate regular expressions from natural language. Currently WIP, but should look something like this:
+Generate regular expressions from natural language.
 
-Instead of a convoluted mess of symbols why not
+Instead of a convoluted mess of symbols like `/([\w\.=\-]*\w+)/g` why not
 
     using global matching
-    create a group called "capture_me" 
+    create a group called capture_me
-        match 0+ words or "." or "=" or "-"
+        match 0+ characters or "." or "=" or "-"
         match 1+ words
 
+Is the former not much easier to read and bug fix than the latter?
+
 Running the program should result in the following output:
 
-    Your regex = /\$([\w\.=\-]*[\w]+)/g
+    Your regex = /(?<capture_me>[\w\.\=\-]*\w++)/g
-    "capture_me" is group id 1
 
-Is the former not much easier to read and bug fix than the latter?
+You can then use your regex in your language of choice, with Human2Regex validating your regex for you.
 
 Another example
 
@@ -61,16 +62,24 @@ Another example
 
 Running the program should result in the following output:
 
-    Your regex = /^(https?:\/\/)?((\w\.)*)(:\d+)?([\w_\-]\.\w)((/[\w_\-]))?(\?([\w_\-]=[\w_\-]))?(#.*)$/g
+    Your regex = /^(?<protocol>https?\:\/\/)?(?<subdomain>(\w+\.)*)?(?<domain>(?:\w+|_|\-)+\.\w+)\:?\d*(?<path>(\/(?:\w+|_|\-)*)*)?(\?(?<query>((?:\w+|_|\-)+\=(?:\w+|_|\-)+)*))?(#.*)?$/g
-    "protocol" is group id 1
+
-    "subdomain" is group id 2
+Which one would you rather debug?
-    "domain" is group id 4
-    "path" is group id 5
-    "query" is group id 5 or 6 if "path" exists
 
 ## Usage
-Configure config.ts
+Build
-Run
 
     npm run build
 
+Run
+    
+    point web browser to: docs/index.html
+
+Test
+
+    npm t
+
+
+## Todo
+- Seperate website and source code. Move to yarn/npm
+- Add more regex options such as back references, subroutines, lookahead/behind, and more character classes (eg,  `[:alpha:]`)

package.json

@@ -1,6 +1,6 @@
 {
   "name": "human2regex",
-  "version": "0.9.0",
+  "version": "0.9.5",
   "description": "Humanized Regular Expressions",
   "main": "bundle.min.js",
   "devDependencies": {
@@ -8,20 +8,20 @@
     "@types/html-minifier": "^3.5.3",
     "@types/jest": "^26.0.15",
     "@types/mustache": "^4.0.1",
-    "@typescript-eslint/eslint-plugin": "^4.4.0",
+    "@typescript-eslint/eslint-plugin": "^4.6.1",
-    "@typescript-eslint/parser": "^4.4.0",
+    "@typescript-eslint/parser": "^4.6.1",
     "before-build-webpack": "^0.2.9",
-    "copy-webpack-plugin": "^6.2.1",
+    "copy-webpack-plugin": "^6.3.0",
     "css-loader": "^4.3.0",
     "eslint": "^7.11.0",
     "glob": "^7.1.6",
     "html-minifier": "^4.0.0",
-    "jest": "^26.6.1",
+    "jest": "^26.6.3",
     "mini-css-extract-plugin": "^1.0.0",
     "mustache": "^4.0.1",
     "optimize-css-assets-webpack-plugin": "^5.0.4",
     "ts-jest": "^26.4.3",
-    "ts-loader": "^8.0.4",
+    "ts-loader": "^8.0.9",
     "ts-node": "^9.0.0",
     "typescript": "^4.0.5",
     "webpack": "^4.44.2",
@@ -37,7 +37,7 @@
   "author": "Patrick Demian",
   "license": "MIT",
   "dependencies": {
-    "chevrotain": "^7.0.2",
+    "chevrotain": "^7.0.3",
     "codemirror": "^5.58.2"
   },
   "repository": {

src/generator.ts

@@ -334,9 +334,13 @@ export class MatchSubStatementCST extends H2RCST {
         let ret = "";
 
         let require_grouping = false;
+        let dont_clobber_plus = false;
 
         if (str.length === 1) {
             ret = str[0];
+            if (ret.endsWith("+")) {
+                dont_clobber_plus = true;
+            }
         }
         // we can use regex's [] for single chars, otherwise we need a group
         else if (str.every(isSingleRegexCharacter)) {
@@ -349,11 +353,37 @@ export class MatchSubStatementCST extends H2RCST {
         }
 
         if (this.count) {
+            if (dont_clobber_plus) {
+                const clobber = this.count.toRegex(language);
+
+                // + can be ignored as well as a count as long as that count is > 0
+                switch (clobber) {
+                    case "*":
+                    case "?":
+                        ret = "(?:" + ret + ")" + clobber;
+                        break;
+                    case "+":
+                        // ignore
+                        break;
+                    default:
+                        if (clobber.startsWith("{0")) {
+                            ret = "(?:" + ret + ")" + clobber;
+                        }
+                        else {
+                            // remove + and replace with count
+                            ret.substring(0, ret.length - 1) + clobber;
+                        }
+                        break;
+                }
+            }
+            else {
                 if (require_grouping) {
                     ret = "(?:" + ret + ")";
                 }
+
                 ret += this.count.toRegex(language);
             }
+        }
 
         return ret;
     }

src/tokens.ts

@@ -27,8 +27,8 @@ import { createToken, Lexer } from "chevrotain";
 /** @internal */ export const And = createToken({name: "And", pattern: /and|,/i});
 /** @internal */ export const Word = createToken({name: "WordSpecifier", pattern: /word(s)?/i});
 /** @internal */ export const Digit = createToken({name: "DigitSpecifier", pattern: /digit(s)?/i});
-/** @internal */ export const Character = createToken({name: "CharacterSpecifier", pattern: /character(s)?/i});
+/** @internal */ export const Character = createToken({name: "CharacterSpecifier", pattern: /(character|letter)s?/i});
-/** @internal */ export const Whitespace = createToken({name: "WhitespaceSpecifier", pattern: /(white space|whitespace)(s)?/i});
+/** @internal */ export const Whitespace = createToken({name: "WhitespaceSpecifier", pattern: /(white space|whitespace)s?/i});
 /** @internal */ export const Boundary = createToken({name: "BoundarySpecifier", pattern: /(word )boundary/i});
 /** @internal */ export const Number = createToken({name: "NumberSpecifier", pattern: /number(s)?/i});
 /** @internal */ export const Unicode = createToken({name: "UnicodeSpecifier", pattern: /unicode( class)?/i});

tests/generator.spec.ts

@@ -32,6 +32,11 @@ describe("Generator functionality", function() {
         const reg1 = parser.parse();
         expect(reg1.validate(RegexDialect.JS).length).toBe(0);
         expect(reg1.toRegex(RegexDialect.JS)).toBe("/https?/");
+
+        parser.input = lexer.tokenize("match 1+ words").tokens;
+        const reg2 = parser.parse();
+        expect(reg2.validate(RegexDialect.JS).length).toBe(0);
+        expect(reg2.toRegex(RegexDialect.JS)).toBe("/\\w+/"); // used to generate w++. make sure not to regress
     });
 
     it("validates invalid regexes", function() {

webpack.config.js

@@ -1,3 +1,5 @@
+/* eslint-disable @typescript-eslint/explicit-function-return-type */
+/* eslint-disable @typescript-eslint/naming-convention */
 /* eslint-disable @typescript-eslint/no-var-requires */
 /* eslint-disable no-undef */
 const path = require("path");
@@ -29,7 +31,6 @@ const config = {
 
 
 function build_mustache() {
-
 	if (!existsSync(config.dst)){
 		mkdirSync(config.dst);
 	}