Add multichar replacements in smartquotes

fix #115

README.md

@@ -111,7 +111,10 @@ var md = require('markdown-it')({
   typographer:  false,
 
   // Double + single quotes replacement pairs, when typographer enabled,
-  // and smartquotes on. Set doubles to '«»' for Russian, '„“' for German.
+  // and smartquotes on. Could be either a String or an Array.
+  //
+  // For example, you can use '«»„“' for Russian, '„“‚‘' for German,
+  // and ['«\xA0', '\xA0»', '‹\xA0', '\xA0›'] for French (including nbsp).
   quotes: '“”‘’',
 
   // Highlighter function. Should return escaped HTML,

lib/index.js

@@ -152,9 +152,10 @@ function normalizeLinkText(url) {
  * - __typographer__  - `false`. Set `true` to enable [some language-neutral
  *   replacement](https://github.com/markdown-it/markdown-it/blob/master/lib/rules_core/replacements.js) +
  *   quotes beautification (smartquotes).
- * - __quotes__ - `“”‘’`, string. Double + single quotes replacement pairs, when
+ * - __quotes__ - `“”‘’`, String or Array. Double + single quotes replacement
- *   typographer enabled and smartquotes on. Set doubles to '«»' for Russian,
+ *   pairs, when typographer enabled and smartquotes on. For example, you can
- *   '„“' for German.
+ *   use `'«»„“'` for Russian, `'„“‚‘'` for German, and
+ *   `['«\xA0', '\xA0»', '‹\xA0', '\xA0›']` for French (including nbsp).
  * - __highlight__ - `null`. Highlighter function for fenced code blocks.
  *   Highlighter `function (str, lang)` should return escaped HTML. It can also
  *   return empty string if the source was not changed and should be escaped externaly.

lib/presets/commonmark.js

@@ -15,7 +15,10 @@ module.exports = {
     typographer:  false,
 
     // Double + single quotes replacement pairs, when typographer enabled,
-    // and smartquotes on. Set doubles to '«»' for Russian, '„“' for German.
+    // and smartquotes on. Could be either a String or an Array.
+    //
+    // For example, you can use '«»„“' for Russian, '„“‚‘' for German,
+    // and ['«\xA0', '\xA0»', '‹\xA0', '\xA0›'] for French (including nbsp).
     quotes: '\u201c\u201d\u2018\u2019' /* “”‘’ */,
 
     // Highlighter function. Should return escaped HTML,

lib/presets/default.js

@@ -15,7 +15,10 @@ module.exports = {
     typographer:  false,
 
     // Double + single quotes replacement pairs, when typographer enabled,
-    // and smartquotes on. Set doubles to '«»' for Russian, '„“' for German.
+    // and smartquotes on. Could be either a String or an Array.
+    //
+    // For example, you can use '«»„“' for Russian, '„“‚‘' for German,
+    // and ['«\xA0', '\xA0»', '‹\xA0', '\xA0›'] for French (including nbsp).
     quotes: '\u201c\u201d\u2018\u2019' /* “”‘’ */,
 
     // Highlighter function. Should return escaped HTML,

lib/presets/zero.js

@@ -16,7 +16,10 @@ module.exports = {
     typographer:  false,
 
     // Double + single quotes replacement pairs, when typographer enabled,
-    // and smartquotes on. Set doubles to '«»' for Russian, '„“' for German.
+    // and smartquotes on. Could be either a String or an Array.
+    //
+    // For example, you can use '«»„“' for Russian, '„“‚‘' for German,
+    // and ['«\xA0', '\xA0»', '‹\xA0', '\xA0›'] for French (including nbsp).
     quotes: '\u201c\u201d\u2018\u2019' /* “”‘’ */,
 
     // Highlighter function. Should return escaped HTML,

lib/rules_core/replacements.js

@@ -1,7 +1,5 @@
 // Simple typographyc replacements
 //
-// '' → ‘’
-// "" → “”. Set '«»' for Russian, '„“' for German, empty to disable
 // (c) (C) → ©
 // (tm) (TM) → ™
 // (r) (R) → ®

lib/rules_core/smartquotes.js

@@ -19,7 +19,7 @@ function replaceAt(str, index, ch) {
 function process_inlines(tokens, state) {
   var i, token, text, t, pos, max, thisLevel, item, lastChar, nextChar,
       isLastPunctChar, isNextPunctChar, isLastWhiteSpace, isNextWhiteSpace,
-      canOpen, canClose, j, isSingle, stack;
+      canOpen, canClose, j, isSingle, stack, openQuote, closeQuote;
 
   stack = [];
 
@@ -104,16 +104,28 @@ function process_inlines(tokens, state) {
           if (stack[j].level < thisLevel) { break; }
           if (item.single === isSingle && stack[j].level === thisLevel) {
             item = stack[j];
+
             if (isSingle) {
-              tokens[item.token].content = replaceAt(
+              openQuote = state.md.options.quotes[2];
-                tokens[item.token].content, item.pos, state.md.options.quotes[2]);
+              closeQuote = state.md.options.quotes[3];
-              token.content = replaceAt(
-                token.content, t.index, state.md.options.quotes[3]);
             } else {
-              tokens[item.token].content = replaceAt(
+              openQuote = state.md.options.quotes[0];
-                tokens[item.token].content, item.pos, state.md.options.quotes[0]);
+              closeQuote = state.md.options.quotes[1];
-              token.content = replaceAt(token.content, t.index, state.md.options.quotes[1]);
             }
+
+            // replace token.content *before* tokens[item.token].content,
+            // because, if they are pointing at the same token, replaceAt
+            // could mess up indices when quote length != 1
+            token.content = replaceAt(token.content, t.index, closeQuote);
+            tokens[item.token].content = replaceAt(
+              tokens[item.token].content, item.pos, openQuote);
+
+            pos += closeQuote.length - 1;
+            if (item.token === i) { pos += openQuote.length - 1; }
+
+            text = token.content;
+            max = text.length;
+
             stack.length = j;
             continue OUTER;
           }

test/misc.js

@@ -275,3 +275,36 @@ describe('maxNesting', function () {
   });
 
 });
+
+
+describe('smartquotes', function () {
+  var md = markdownit({
+    typographer: true,
+
+    // all strings have different length to make sure
+    // we didn't accidentally count the wrong one
+    quotes: [ '[[[', ']]', '(((((', '))))' ]
+  });
+
+  it('Should support multi-character quotes', function () {
+    assert.strictEqual(
+      md.render('"foo" \'bar\''),
+      '<p>[[[foo]] (((((bar))))</p>\n'
+    );
+  });
+
+  it('Should support nested multi-character quotes', function () {
+    assert.strictEqual(
+      md.render('"foo \'bar\' baz"'),
+      '<p>[[[foo (((((bar)))) baz]]</p>\n'
+    );
+  });
+
+  it('Should support multi-character quotes in different tags', function () {
+    assert.strictEqual(
+      md.render('"a *b \'c *d* e\' f* g"'),
+      '<p>[[[a <em>b (((((c <em>d</em> e)))) f</em> g]]</p>\n'
+    );
+  });
+
+});