hullproof 0.0.0-stage → 0.5.0

package/dist/map/lex-more.js

@@ -0,0 +1,305 @@
+// Java, C#, Kotlin, Swift and Rust: line and block comments, "strings" with escapes, text blocks and raw strings, character literals.
+function maskC(text, lang) {
+    const n = text.length;
+    const out = [];
+    let i = 0;
+    const put = (from, to) => { out.push(blank(text.slice(from, to))); };
+    const keep = (from, to) => { out.push(text.slice(from, to)); };
+    const closed = (open, close, escapes, doubled = false) => {
+        // index of the closing delimiter at or after open, or -1
+        let j = open;
+        while (j < n) {
+            if (escapes && text[j] === '\\') {
+                j += 2;
+                continue;
+            }
+            if (text.startsWith(close, j)) {
+                if (doubled && text.startsWith(close + close, j)) {
+                    j += close.length * 2;
+                    continue;
+                }
+                return j;
+            }
+            j++;
+        }
+        return -1;
+    };
+    while (i < n) {
+        const c = text[i];
+        const c2 = text[i + 1];
+        if (c === '/' && c2 === '/') {
+            const e = text.indexOf('\n', i);
+            const end = e < 0 ? n : e;
+            put(i, end);
+            i = end;
+            continue;
+        }
+        if (c === '/' && c2 === '*') {
+            const e = text.indexOf('*/', i + 2);
+            const end = e < 0 ? n : e + 2;
+            put(i, end);
+            i = end;
+            continue;
+        }
+        // C#: $"...", @"...", $@"..." ; raw strings in every language: """..."""
+        let pre = 0;
+        if (lang === 'cs') {
+            while (text[i + pre] === '$' || text[i + pre] === '@')
+                pre++;
+        }
+        if (lang === 'rs' && (c === 'r' || (c === 'b' && c2 === 'r')) && /^b?r#*"/.test(text.slice(i, i + 12))) {
+            const m = /^b?r(#*)"/.exec(text.slice(i, i + 12));
+            const start = i + m[0].length;
+            const e = text.indexOf('"' + m[1], start);
+            const end = e < 0 ? n : e;
+            keep(i, start);
+            put(start, end);
+            if (e >= 0)
+                keep(e, e + 1 + m[1].length);
+            i = e < 0 ? n : e + 1 + m[1].length;
+            continue;
+        }
+        const q = i + pre;
+        if (text[q] === '"') {
+            if (text.startsWith('"""', q)) {
+                const e = text.indexOf('"""', q + 3);
+                const end = e < 0 ? n : e;
+                keep(i, q + 3);
+                put(q + 3, end);
+                if (e >= 0)
+                    keep(e, e + 3);
+                i = e < 0 ? n : e + 3;
+                continue;
+            }
+            const verbatim = lang === 'cs' && text.slice(i, q).includes('@');
+            const e = closed(q + 1, '"', !verbatim, verbatim);
+            const end = e < 0 ? n : e;
+            keep(i, q + 1);
+            put(q + 1, end);
+            if (e >= 0)
+                keep(e, e + 1);
+            i = e < 0 ? n : e + 1;
+            continue;
+        }
+        if (c === '\'' && lang !== 'rs' || (c === '\'' && lang === 'rs' && /^'(?:\\.[^']*|[^\\'])'/.test(text.slice(i, i + 12)))) {
+            const e = closed(i + 1, '\'', true);
+            if (e >= 0 && e - i <= 12) {
+                keep(i, i + 1);
+                put(i + 1, e);
+                keep(e, e + 1);
+                i = e + 1;
+                continue;
+            }
+        }
+        keep(i, i + 1);
+        i++;
+    }
+    return out.join('');
+}
+const blank = (s) => s.replace(/[^\n]/g, ' ');
+export function maskMore(text, lang) {
+    if (lang === 'java' || lang === 'cs' || lang === 'kt' || lang === 'swift' || lang === 'rs')
+        return maskC(text, lang);
+    const n = text.length;
+    const out = [];
+    const put = (from, to) => { out.push(blank(text.slice(from, to))); };
+    const keep = (from, to) => { out.push(text.slice(from, to)); };
+    let i = 0;
+    let prevSig = '';
+    let prevWord = '';
+    // Heredoc terminators waiting for the end of the current line (Ruby and PHP).
+    let pending = [];
+    const quoted = (q, escapes) => {
+        let j = i + 1;
+        while (j < n && text[j] !== q && (text[j] !== '\n' || lang !== 'go')) {
+            if (escapes && text[j] === '\\')
+                j++;
+            j++;
+        }
+        const close = j < n && text[j] === q ? j : -1;
+        keep(i, i + 1);
+        put(i + 1, close < 0 ? Math.min(j, n) : close);
+        if (close >= 0)
+            keep(close, close + 1);
+        i = close < 0 ? Math.min(j, n) : close + 1;
+        prevSig = q;
+        prevWord = '';
+    };
+    const flushHeredocs = () => {
+        // i is just after a newline. Blank lines until each pending terminator line.
+        for (const h of pending) {
+            for (;;) {
+                if (i >= n)
+                    break;
+                const e = text.indexOf('\n', i);
+                const end = e < 0 ? n : e;
+                const row = text.slice(i, end);
+                const isEnd = (lang === 'rb' ? row.trim() : row.replace(/^\s+/, '').replace(/[;,)\s]+$/, '')) === h.id;
+                if (isEnd) {
+                    keep(i, end);
+                    i = end;
+                    break;
+                }
+                put(i, end);
+                i = end;
+                if (i < n) {
+                    out.push('\n');
+                    i++;
+                }
+            }
+        }
+        pending = [];
+    };
+    while (i < n) {
+        const c = text[i];
+        const c2 = text[i + 1];
+        if (c === '\n') {
+            keep(i, i + 1);
+            i++;
+            if (pending.length)
+                flushHeredocs();
+            // Ruby block comments
+            if (lang === 'rb' && text.startsWith('=begin', i)) {
+                const e = text.indexOf('\n=end', i);
+                const end = e < 0 ? n : text.indexOf('\n', e + 1) < 0 ? n : text.indexOf('\n', e + 1);
+                put(i, end);
+                i = end;
+            }
+            continue;
+        }
+        if (lang === 'go') {
+            if (c === '/' && c2 === '/') {
+                const e = text.indexOf('\n', i);
+                const end = e < 0 ? n : e;
+                put(i, end);
+                i = end;
+                continue;
+            }
+            if (c === '/' && c2 === '*') {
+                const e = text.indexOf('*/', i + 2);
+                const end = e < 0 ? n : e + 2;
+                put(i, end);
+                i = end;
+                continue;
+            }
+            if (c === '"') {
+                quoted('"', true);
+                continue;
+            }
+            if (c === '\'') {
+                quoted('\'', true);
+                continue;
+            }
+            if (c === '`') {
+                const e = text.indexOf('`', i + 1);
+                const end = e < 0 ? n : e;
+                keep(i, i + 1);
+                put(i + 1, end);
+                if (e >= 0)
+                    keep(e, e + 1);
+                i = e < 0 ? n : e + 1;
+                prevSig = '`';
+                continue;
+            }
+        }
+        else if (lang === 'rb') {
+            if (i === 0 && text.startsWith('=begin')) {
+                const e = text.indexOf('\n=end');
+                const end = e < 0 ? n : e + 5;
+                put(0, end);
+                i = end;
+                continue;
+            }
+            if (c === '#') {
+                const e = text.indexOf('\n', i);
+                const end = e < 0 ? n : e;
+                put(i, end);
+                i = end;
+                continue;
+            }
+            const hd = c === '<' && c2 === '<' ? /^<<([~-]?)(['"`]?)([A-Za-z_]\w*)\2/.exec(text.slice(i, i + 60)) : null;
+            if (hd && !/[\w)\]]\s*$/.test(text.slice(Math.max(0, i - 2), i).replace(/ $/, '')) || (hd && /[=(,]\s*$/.test(text.slice(Math.max(0, i - 12), i)))) {
+                pending.push({ id: hd[3], squiggly: hd[1] === '~' });
+                keep(i, i + hd[0].length);
+                i += hd[0].length;
+                continue;
+            }
+            if (c === '"' || c === '\'' || c === '`') {
+                quoted(c, true);
+                continue;
+            }
+            if (c === '/' && /^[(,=~{|&!;]$/.test(prevSig) || (c === '/' && /^(?:when|if|unless|and|or|not|return|split|match|scan|gsub|sub)$/.test(prevWord))) {
+                let j = i + 1;
+                let cls = false;
+                while (j < n && text[j] !== '\n') {
+                    if (text[j] === '\\') {
+                        j += 2;
+                        continue;
+                    }
+                    if (text[j] === '[')
+                        cls = true;
+                    else if (text[j] === ']')
+                        cls = false;
+                    else if (text[j] === '/' && !cls)
+                        break;
+                    j++;
+                }
+                if (j < n && text[j] === '/') {
+                    keep(i, i + 1);
+                    put(i + 1, j);
+                    keep(j, j + 1);
+                    i = j + 1;
+                    prevSig = '/';
+                    prevWord = '';
+                    continue;
+                }
+            }
+        }
+        else {
+            if ((c === '/' && c2 === '/') || (c === '#' && c2 !== '[')) {
+                const e = text.indexOf('\n', i);
+                const end = e < 0 ? n : e;
+                put(i, end);
+                i = end;
+                continue;
+            }
+            if (c === '/' && c2 === '*') {
+                const e = text.indexOf('*/', i + 2);
+                const end = e < 0 ? n : e + 2;
+                put(i, end);
+                i = end;
+                continue;
+            }
+            const hd = c === '<' && c2 === '<' && text[i + 2] === '<' ? /^<<<\s*(['"]?)([A-Za-z_]\w*)\1/.exec(text.slice(i, i + 60)) : null;
+            if (hd) {
+                pending.push({ id: hd[2], squiggly: true });
+                keep(i, i + hd[0].length);
+                i += hd[0].length;
+                continue;
+            }
+            if (c === '"' || c === '\'') {
+                quoted(c, true);
+                continue;
+            }
+        }
+        keep(i, i + 1);
+        if (!/\s/.test(c)) {
+            prevSig = c;
+            if (/[A-Za-z_]/.test(c)) {
+                const m = /^[A-Za-z_]\w*/.exec(text.slice(i, i + 20));
+                if (m) {
+                    prevWord = m[0];
+                    out.pop();
+                    keep(i, i + m[0].length);
+                    i += m[0].length;
+                    prevSig = 'w';
+                    continue;
+                }
+            }
+            else
+                prevWord = '';
+        }
+        i++;
+    }
+    return out.join('');
+}