Add option to disable collapsing of whitespaces (#905)

Author: marecar3

aztec/src/main/java/org/wordpress/aztec/Html.java

@@ -154,7 +154,7 @@ public static Spanned fromHtml(String source,
                                    Context context,
                                    List<IAztecPlugin> plugins,
                                    List<String> ignoredTags) {
-        return fromHtml(source, null, context, plugins, ignoredTags);
+        return fromHtml(source, null, context, plugins, ignoredTags, true);
     }
 
     /**
@@ -176,7 +176,8 @@ private static class HtmlParser {
      * <p>This uses TagSoup to handle real HTML, including all of the brokenness found in the wild.
      */
     public static Spanned fromHtml(String source, TagHandler tagHandler,
-                                   Context context, List<IAztecPlugin> plugins, List<String> ignoredTags) {
+                                   Context context, List<IAztecPlugin> plugins,
+                                   List<String> ignoredTags, boolean shouldIgnoreWhitespace) {
 
         Parser parser = new Parser();
         try {
@@ -195,7 +196,13 @@ public static Spanned fromHtml(String source, TagHandler tagHandler,
         source = preprocessSource(source, plugins);
 
         HtmlToSpannedConverter converter =
-                new HtmlToSpannedConverter(source, tagHandler, parser, context, plugins, ignoredTags);
+                new HtmlToSpannedConverter(source,
+                        tagHandler,
+                        parser,
+                        context,
+                        plugins,
+                        ignoredTags,
+                        shouldIgnoreWhitespace);
 
         return converter.convert();
     }
@@ -238,18 +245,21 @@ class HtmlToSpannedConverter implements org.xml.sax.ContentHandler, LexicalHandl
     private Html.TagHandler tagHandler;
     private Context context;
     private List<String> ignoredTags;
+    private boolean shouldIgnoreWhitespace;
 
     public HtmlToSpannedConverter(
             String source, Html.TagHandler tagHandler,
             Parser parser,
-            Context context, List<IAztecPlugin> plugins, List<String> ignoredTags) {
+            Context context, List<IAztecPlugin> plugins,
+            List<String> ignoredTags, boolean shouldIgnoreWhitespace) {
         this.source = source;
         this.plugins = plugins;
         this.spannableStringBuilder = new SpannableStringBuilder();
         this.tagHandler = tagHandler;
         this.reader = parser;
         this.context = context;
         this.ignoredTags = ignoredTags;
+        this.shouldIgnoreWhitespace = shouldIgnoreWhitespace;
     }
 
     public Spanned convert() {
@@ -718,6 +728,12 @@ public void characters(char ch[], int start, int length) throws SAXException {
             char c = ch[i + start];
 
             if (!insidePreTag && !insideCodeTag && c == ' ' || c == '\n') {
+
+                if (c == ' ' && !shouldIgnoreWhitespace) {
+                    sb.append(c);
+                    continue;
+                }
+
                 char pred;
                 int len = sb.length();
 

aztec/src/main/kotlin/org/wordpress/aztec/AztecParser.kt

@@ -68,18 +68,21 @@ class AztecParser @JvmOverloads constructor(val plugins: List<IAztecPlugin> = li
         val tidySource = tidy(source)
 
         val spanned = SpannableString(Html.fromHtml(tidySource,
-                AztecTagHandler(context, plugins), context, plugins, ignoredTags))
+                AztecTagHandler(context, plugins), context, plugins, ignoredTags, true))
 
         postprocessSpans(spanned)
 
         return spanned
     }
 
-    fun fromHtml(source: String, context: Context, shouldSkipTidying: Boolean = false): Spanned {
+    fun fromHtml(source: String,
+                 context: Context,
+                 shouldSkipTidying: Boolean = false,
+                 shouldIgnoreWhitespace: Boolean = true): Spanned {
         val tidySource = if (shouldSkipTidying) source else tidy(source)
 
         val spanned = SpannableStringBuilder(Html.fromHtml(tidySource,
-                AztecTagHandler(context, plugins), context, plugins, ignoredTags))
+                AztecTagHandler(context, plugins), context, plugins, ignoredTags, shouldIgnoreWhitespace))
 
         addVisualNewlinesToBlockElements(spanned)
         markBlockElementsAsParagraphs(spanned)

aztec/src/main/kotlin/org/wordpress/aztec/AztecText.kt

@@ -1118,6 +1118,10 @@ open class AztecText : AppCompatEditText, TextWatcher, UnknownHtmlSpan.OnUnknown
         return false
     }
 
+    open fun shouldIgnoreWhitespace(): Boolean {
+        return true
+    }
+
     override fun afterTextChanged(text: Editable) {
         if (isTextChangedListenerDisabled()) {
             subWatcherNestingLevel()
@@ -1168,7 +1172,7 @@ open class AztecText : AppCompatEditText, TextWatcher, UnknownHtmlSpan.OnUnknown
 
         var cleanSource = CleaningUtils.cleanNestedBoldTags(source)
         cleanSource = Format.removeSourceEditorFormatting(cleanSource, isInCalypsoMode, isInGutenbergMode)
-        builder.append(parser.fromHtml(cleanSource, context, shouldSkipTidying()))
+        builder.append(parser.fromHtml(cleanSource, context, shouldSkipTidying(), shouldIgnoreWhitespace()))
 
         Format.preProcessSpannedText(builder, isInCalypsoMode)
 

aztec/src/test/kotlin/org/wordpress/aztec/AztecParserTest.kt

@@ -1260,4 +1260,14 @@ class AztecParserTest : AndroidTestCase() {
         val output = mParser.toHtml(span)
         Assert.assertEquals(input, output)
     }
+
+    @Test
+    @Throws(Exception::class)
+    fun parseHtmlToSpanToHtmlParagraphWithMultipleWhitespace_isNotEqual() {
+        val input = "<p>   Hello There!</p>"
+        val inputAfterParser = "<p>Hello There!</p>"
+        val span = SpannableString(mParser.fromHtml(input, RuntimeEnvironment.application.applicationContext))
+        val output = mParser.toHtml(span)
+        Assert.assertEquals(output, inputAfterParser)
+    }
 }