meh lexer

docs/js/custom-element/lib/lexer.js

@@ -0,0 +1,91 @@
+function splitSymbols(v) {
+  if (v.match(/\w/)) return v;
+  return v.split(``);
+}
+
+class Lexer {
+  constructor(code) {
+    this.scope = 0;
+    this.pos = 0;
+    this.tokens = code.split(/\b/).map(splitSymbols).flat();
+    this.scopes = [];
+    console.log(this.tokens);
+  }
+
+  parse() {
+    while (this.pos < this.tokens.length) {
+      let token = this.tokens[this.pos++];
+
+      if ([`const`, `let`, "var"].includes(token)) {
+        this.parseVariable(token);
+      }
+
+      // ...
+      else if ([`'`, `"`, "`"].includes(token)) {
+        this.parseString(token);
+      }
+
+      // ...
+      else if (token === `(`) {
+        let functor,
+          i = 2;
+        do {
+          functor = this.tokens[this.pos - i++];
+        } while (functor.match(/\s+/));
+
+        // TODO: maths is fun?
+
+        console.log(`[${this.scope}]: ${functor}(...`);
+      } else if (token === `)`) {
+      }
+
+      // ...
+      else if (token === `{`) {
+        this.scopes[this.pos] = ++this.scope;
+      }
+
+      // ...
+      else if (token === `}`) {
+        this.scopes[this.pos] = --this.scope;
+      }
+    }
+
+    console.log(this.scopes);
+  }
+
+  parseVariable(type) {
+    let name;
+    do {
+      name = this.tokens[this.pos++];
+    } while (name.match(/\s+/));
+    console.log(`[${this.scope}]: ${type} ${name}`);
+  }
+
+  parseString(symbol) {
+    // we technically don't really care about the contents
+    // of strings, as they don't introduce new variables
+    // or functions that we need to care about.
+    let token;
+    let buffer = [symbol];
+    let blen = 1;
+    do {
+      token = this.tokens[this.pos++];
+      buffer.push(token);
+      blen++;
+    } while (
+      token !== symbol &&
+      buffer[blen - 2] !== `\\` &&
+      this.pos < this.tokens.length
+    );
+    //    buffer = buffer.join(``);
+    //    if (symbol === "`") {
+    //      this.parseTemplateString(buffer);
+    //    }
+  }
+
+  //  parseTemplateString(buffer) {
+  //    // console.log(buffer);
+  //  }
+}
+
+export { Lexer };

docs/js/custom-element/lib/perform-code-surgery.js

@@ -1,9 +1,8 @@
 import { GraphicsAPI } from "../api/graphics-api.js";
 
 export default function performCodeSurgery(code) {
-  // 0. strip out block comments and whitespace
+  // 0. strip out superfluous whitespace
 
-  code = code.replace(/\\\*[\w\s\r\n]+?\*\\/, ``);
   code = code.replace(/\r?\n(\r?\n)+/, `\n`);
 
   // 1. ensure that anything that needs to run by first calling its super function, does so.

docs/js/custom-element/lib/split-code-sections.js

@@ -3,6 +3,10 @@
  * We're going to regexp our way to flawed victory here.
  */
 export default function splitCodeSections(code) {
+  // removs comments and superfluous white space.
+  code = code.replace(/\\\*[\w\s\r\n]+?\*\\/, ``);
+  code = code.replace(/\r?\n(\r?\n)+/, `\n`);
+
   const re = /\b[\w\W][^\s]*?\([^)]*\)[\r\n\s]*{/;
   const cuts = [];
   for (let result = code.match(re); result; result = code.match(re)) {