better RTF encoding detection; don't fail on RTF files with .doc extension; fixed XML serialisation

ChangeLog

@@ -6,11 +6,13 @@
 * (planned) Fixed authors list/tags list editing
 * (planned) CSS selectors priority
 
-===== 2.3 (Feb ??, 2015) =====
+===== 2.3 (Feb 14, 2015) =====
 * File format filter option (can be changed from the bookshelf plugin)
 * Samsung multi-window support
 * Bookshelf view item in the library
 * Updated Persian localisation by biomanw (https://github.com/biomanw)
+* Better RTF encoding detection
+* Don't fail on RTF files with .doc extension
 
 ===== 2.2.7 (Jan 30, 2015) =====
 * Fixed behaviour for pages with multiple images (fb2.zip only)

assets/resources/application/neutral.xml

@@ -7,7 +7,7 @@
 		<node name="txt" value="txt"/>
 		<node name="msdoc" value="Microsoft Word"/>
 		<node name="HTML" value="HTML"/>
-		<node name="rtf" value="RTF"/>
+		<node name="RTF" value="RTF"/>
 		<node name="PDF" value="PDF"/>
 		<node name="DjVu" value="DjVu"/>
 		<node name="CBZ" value="Comic book archive (cbz, cbr)"/>

jni/NativeFormats/fbreader/src/formats/doc/DocPlugin.cpp

@@ -27,6 +27,7 @@
 #include "DocMetaInfoReader.h"
 #include "DocBookReader.h"
 #include "DocStreams.h"
+#include "../rtf/RtfPlugin.h"
 #include "../../bookmodel/BookModel.h"
 #include "../../library/Book.h"
 
@@ -36,6 +37,17 @@ DocPlugin::DocPlugin() {
 DocPlugin::~DocPlugin() {
 }
 
+static bool isRtf(const Book &book) {
+	shared_ptr<ZLInputStream> stream = book.file().inputStream();
+	if (stream.isNull() || !stream->open()) {
+		return false;
+	}
+	char buffer[6] = { 0 };
+	stream->read(buffer, 5);
+	static const std::string RTF_BYTES = "{\\rtf";
+	return RTF_BYTES == buffer;
+}
+
 bool DocPlugin::providesMetainfo() const {
 	return true;
 }
@@ -49,6 +61,10 @@ bool DocPlugin::acceptsFile(const ZLFile &file) const {
 }
 
 bool DocPlugin::readMetainfo(Book &book) const {
+	if (::isRtf(book)) {
+		return RtfPlugin().readMetainfo(book);
+	}
+
 	if (!DocMetaInfoReader(book).readMetainfo()) {
 		return false;
 	}
@@ -71,5 +87,9 @@ bool DocPlugin::readLanguageAndEncoding(Book &/*book*/) const {
 }
 
 bool DocPlugin::readModel(BookModel &model) const {
+	if (::isRtf(*model.book())) {
+		return RtfPlugin().readModel(model);
+	}
+
 	return DocBookReader(model, model.book()->encoding()).readBook();
 }

jni/NativeFormats/fbreader/src/formats/rtf/RtfPlugin.cpp

@@ -34,23 +34,16 @@ bool RtfPlugin::providesMetainfo() const {
 }
 
 const std::string RtfPlugin::supportedFileType() const {
-	return "rtf";
+	return "RTF";
 }
 
 bool RtfPlugin::readMetainfo(Book &book) const {
+	readLanguageAndEncoding(book);
+
 	if (!RtfDescriptionReader(book).readDocument(book.file())) {
 		return false;
 	}
 
-	if (book.encoding().empty()) {
-		book.setEncoding(ZLEncodingConverter::UTF8);
-	} else if (book.language().empty()) {
-		shared_ptr<ZLInputStream> stream = new RtfReaderStream(book.file(), 50000);
-		if (!stream.isNull()) {
-			detectLanguage(book, *stream, book.encoding());
-		}
-	}
-
 	return true;
 }
 
@@ -64,5 +57,19 @@ bool RtfPlugin::readModel(BookModel &model) const {
 }
 
 bool RtfPlugin::readLanguageAndEncoding(Book &book) const {
+	if (book.encoding().empty()) {
+		shared_ptr<ZLInputStream> stream = new RtfReaderStream(book.file(), 50000);
+		if (!stream.isNull()) {
+			detectEncodingAndLanguage(book, *stream);
+		}
+		if (book.encoding().empty()) {
+			book.setEncoding(ZLEncodingConverter::UTF8);
+		}
+	} else if (book.language().empty()) {
+		shared_ptr<ZLInputStream> stream = new RtfReaderStream(book.file(), 50000);
+		if (!stream.isNull()) {
+			detectLanguage(book, *stream, book.encoding());
+		}
+	}
 	return true;
 }

src/org/geometerplus/fbreader/book/XMLSerializer.java

@@ -35,9 +35,13 @@ import org.geometerplus.zlibrary.core.util.ZLColor;
 import org.geometerplus.zlibrary.text.view.ZLTextPosition;
 
 class XMLSerializer extends AbstractSerializer {
+	private StringBuilder builder() {
+		return new StringBuilder("<?xml version='1.1' encoding='UTF-8'?>");
+	}
+
 	@Override
 	public String serialize(BookQuery query) {
-		final StringBuilder buffer = new StringBuilder();
+		final StringBuilder buffer = builder();
 		appendTag(buffer, "query", false,
 			"limit", String.valueOf(query.Limit),
 			"page", String.valueOf(query.Page)
@@ -136,7 +140,7 @@ class XMLSerializer extends AbstractSerializer {
 
 	@Override
 	public String serialize(BookmarkQuery query) {
-		final StringBuilder buffer = new StringBuilder();
+		final StringBuilder buffer = builder();
 		appendTag(buffer, "query", false,
 			"visible", String.valueOf(query.Visible),
 			"limit", String.valueOf(query.Limit),
@@ -164,7 +168,7 @@ class XMLSerializer extends AbstractSerializer {
 
 	@Override
 	public String serialize(Book book) {
-		final StringBuilder buffer = new StringBuilder();
+		final StringBuilder buffer = builder();
 		serialize(buffer, book);
 		return buffer.toString();
 	}
@@ -262,7 +266,7 @@ class XMLSerializer extends AbstractSerializer {
 
 	@Override
 	public String serialize(Bookmark bookmark) {
-		final StringBuilder buffer = new StringBuilder();
+		final StringBuilder buffer = builder();
 		appendTag(
 			buffer, "bookmark", false,
 			"id", String.valueOf(bookmark.getId()),
@@ -325,7 +329,7 @@ class XMLSerializer extends AbstractSerializer {
 
 	@Override
 	public String serialize(HighlightingStyle style) {
-		final StringBuilder buffer = new StringBuilder();
+		final StringBuilder buffer = builder();
 		final ZLColor bgColor = style.getBackgroundColor();
 		final ZLColor fgColor = style.getForegroundColor();
 		appendTag(buffer, "style", true,
@@ -441,23 +445,41 @@ class XMLSerializer extends AbstractSerializer {
 		}
 	}
 
-	private static String escapeForXml(String data) {
-		if (data.indexOf('&') != -1) {
-			data = data.replaceAll("&", "&amp;");
+	private static CharSequence escapeForXml(String data) {
+		final StringBuilder buffer = new StringBuilder();
+
+		final int len = data.length();
+		for (int i = 0; i < len; ++i) {
+			final char ch = data.charAt(i);
+			switch (ch) {
+				case '\u0009':
+					buffer.append(ch);
+					break;
+				default:
+					if ((ch >= '\u0020' && ch <= '\uD7FF') ||
+						(ch >= '\u0E00' && ch <= '\uFFFD')) {
+						buffer.append(ch);
+					}
+					break;
+				case '&':
+					buffer.append("&amp;");
+					break;
+				case '<':
+					buffer.append("&lt;");
+					break;
+				case '>':
+					buffer.append("&gt;");
+					break;
+				case '"':
+					buffer.append("&quot;");
+					break;
+				case '\'':
+					buffer.append("&apos;");
+					break;
+			}
 		}
-		if (data.indexOf('<') != -1) {
-			data = data.replaceAll("<", "&lt;");
-		}
-		if (data.indexOf('>') != -1) {
-			data = data.replaceAll(">", "&gt;");
-		}
-		if (data.indexOf('\'') != -1) {
-			data = data.replaceAll("'", "&apos;");
-		}
-		if (data.indexOf('"') != -1) {
-			data = data.replaceAll("\"", "&quot;");
-		}
-		return data;
+
+		return buffer;
 	}
 
 	private static void clear(StringBuilder buffer) {