在项目中，利用iText和flying saucer生成pdf文件，网上所说的中文不显示的问题倒是没有遇到，不过就是中文字体时，由于字符宽度是按字母计算的，同样字数会导致一行显示很长不换行，从而超过版面宽度显示不全的问题。经过分析和查找，终于的到解决方案，另外也解决了中文标点符号出现在行首的现象。
首先，我们要明白这几个东西的真实含义：
Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS ： 4E00-9FBF：CJK 统一表意符号
Character.UnicodeBlock.CJK_COMPATIBILITY_IDEOGRAPHS ：F900-FAFF：CJK 兼容象形文字Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A ：3400-4DBF：CJK 统一表意符号扩展 A
CJK的意思是“Chinese，Japanese，Korea”的简写 ，实际上就是指中日韩三国的象形文字的Unicode编码
Character.UnicodeBlock.GENERAL_PUNCTUATION ：2000-206F：常用标点Character.UnicodeBlock.CJK_SYMBOLS_AND_PUNCTUATION ：3000-303F：CJK 符号和标点Character.UnicodeBlock.HALFWIDTH_AND_FULLWIDTH_FORMS ：FF00-FFEF：半角及全角形式
在flying saucer生成pdf的过程中，源代码中是按组和行宽来换行的，具体逻辑如下：
以空格为分组符，任意两个相邻空格之间的字符都是不可换行的，当该行多组字符的总长到达最接近行宽的时候，也就是再加一个字符组就会超过行宽的时候，就会换行
这种方式对英文是很凑效的，但是对中文是不行的，因为中文不是以空格分隔的。yye_javaeye老兄对源码做了改进，判断是否是中文字符，如果是中文字符，则把每个中文字符做一组，然后再用组+行宽来换行。他添加了两个方法：isChinese(char c)和getStrRight(String s,int left)， 改过后的部分源码如下：
package org.xhtmlrenderer.layout;
import org.xhtmlrenderer.css.constants.IdentValue;
import org.xhtmlrenderer.css.style.CalculatedStyle;
import org.xhtmlrenderer.render.FSFont;
/**
* A utility class that scans the text of a single inline box, looking for the
* next break point.
* @author Torbjrn Gannholm
*/
public class Breaker {
。。。。。。
public static void breakText(LayoutContext c,
LineBreakContext context, int avail, CalculatedStyle style) {
。。。。。。
String currentString = context.getStartSubstring();
int left = 0;
// int right = currentString.indexOf(WhitespaceStripper.SPACE, left + 1);
int right = getStrRight(currentString,left);
int lastWrap = 0;
int graphicsLength = 0;
int lastGraphicsLength = 0;
while (right > 0 && graphicsLength <= avail) {
lastGraphicsLength = graphicsLength;
graphicsLength += c.getTextRenderer().getWidth(
c.getFontContext(), font, currentString.substring(left, right));
lastWrap = left;
left = right;
// right = currentString.indexOf(WhitespaceStripper.SPACE, left + 1);
right = getStrRight(currentString,left+1);
}
。。。。。。
}
private static boolean isChinese(char c) {
Character.UnicodeBlock ub = Character.UnicodeBlock.of(c);
if (ub == Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS
|| ub == Character.UnicodeBlock.CJK_COMPATIBILITY_IDEOGRAPHS
|| ub == Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A
|| ub == Character.UnicodeBlock.GENERAL_PUNCTUATION
|| ub == Character.UnicodeBlock.CJK_SYMBOLS_AND_PUNCTUATION
|| ub == Character.UnicodeBlock.HALFWIDTH_AND_FULLWIDTH_FORMS) {
return true;
}
return false;
}
private static int getStrRight(String s,int left){
if(left>=s.length())
return -1;
char[] ch = s.toCharArray();
for(int i = left;i<ch.length;i++){
if(isChinese(ch[i]) || ' ' == ch[i]){
return i==0?i+1:i;
}
}
return -1;
}
}
这种方法确实实现了中文汉字的换行，但是也带来了“标点符号在行首”的问题，因为他把标点符号也视作汉字了，这样标点符号也被分组被换行了，解决的方法就是把标点符号从isChinese方法中去掉。其源码如下：
/*
* Breaker.java
* Copyright (c) 2004, 2005 Torbj锟絩n Gannholm,
* Copyright (c) 2005 Wisconsin Court System
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public License
* as published by the Free Software Foundation; either version 2.1
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*
*/
package org.xhtmlrenderer.layout;
import org.xhtmlrenderer.css.constants.IdentValue;
import org.xhtmlrenderer.css.style.CalculatedStyle;
import org.xhtmlrenderer.render.FSFont;
/**
* A utility class that scans the text of a single inline box, looking for the
* next break point.
* @author Torbj锟絩n Gannholm
*/
public class Breaker {
public static void breakFirstLetter(LayoutContext c, LineBreakContext context,
int avail, CalculatedStyle style) {
FSFont font = style.getFSFont(c);
context.setEnd(getFirstLetterEnd(context.getMaster(), context.getStart()));
context.setWidth(c.getTextRenderer().getWidth(
c.getFontContext(), font, context.getCalculatedSubstring()));
if (context.getWidth() > avail) {
context.setNeedsNewLine(true);
context.setUnbreakable(true);
}
}
private static int getFirstLetterEnd(String text, int start) {
int i = start;
while (i < text.length()) {
char c = text.charAt(i);
int type = Character.getType(c);
if (type == Character.START_PUNCTUATION ||
type == Character.END_PUNCTUATION ||
type == Character.INITIAL_QUOTE_PUNCTUATION ||
type == Character.FINAL_QUOTE_PUNCTUATION ||
type == Character.OTHER_PUNCTUATION) {
i++;
} else {
break;
}
}
if (i < text.length()) {
i++;
}
return i;
}
public static void breakText(LayoutContext c,
LineBreakContext context, int avail, CalculatedStyle style) {
FSFont font = style.getFSFont(c);
IdentValue whitespace = style.getWhitespace();
// ====== handle nowrap
if (whitespace == IdentValue.NOWRAP) {
context.setEnd(context.getLast());
context.setWidth(c.getTextRenderer().getWidth(
c.getFontContext(), font, context.getCalculatedSubstring()));
return;
}
//check if we should break on the next newline
if (whitespace == IdentValue.PRE ||
whitespace == IdentValue.PRE_WRAP ||
whitespace == IdentValue.PRE_LINE) {
int n = context.getStartSubstring().indexOf(WhitespaceStripper.EOL);
if (n > -1) {
context.setEnd(context.getStart() + n + 1);
context.setWidth(c.getTextRenderer().getWidth(
c.getFontContext(), font, context.getCalculatedSubstring()));
context.setNeedsNewLine(true);
context.setEndsOnNL(true);
} else if (whitespace == IdentValue.PRE) {
context.setEnd(context.getLast());
context.setWidth(c.getTextRenderer().getWidth(
c.getFontContext(), font, context.getCalculatedSubstring()));
}
}
//check if we may wrap
if (whitespace == IdentValue.PRE ||
(context.isNeedsNewLine() && context.getWidth() <= avail)) {
return;
}
context.setEndsOnNL(false);
String currentString = context.getStartSubstring();
int left = 0;
// int right = currentString.indexOf(WhitespaceStripper.SPACE, left + 1);
int right = getStrRight(currentString,left);
int lastWrap = 0;
int graphicsLength = 0;
int lastGraphicsLength = 0;
while (right > 0 && graphicsLength <= avail) {
lastGraphicsLength = graphicsLength;
graphicsLength += c.getTextRenderer().getWidth(
c.getFontContext(), font, currentString.substring(left, right));
lastWrap = left;
left = right;
// right = currentString.indexOf(WhitespaceStripper.SPACE, left + 1);
right = getStrRight(currentString,left+1);
}
if (graphicsLength <= avail) {
//try for the last bit too!
lastWrap = left;
lastGraphicsLength = graphicsLength;
graphicsLength += c.getTextRenderer().getWidth(
c.getFontContext(), font, currentString.substring(left));
}
if (graphicsLength <= avail) {
context.setWidth(graphicsLength);
context.setEnd(context.getMaster().length());
//It fit!
return;
}
context.setNeedsNewLine(true);
if (lastWrap != 0) {//found a place to wrap
context.setEnd(context.getStart() + lastWrap);
context.setWidth(lastGraphicsLength);
} else {//unbreakable string
if (left == 0) {
left = currentString.length();
}
context.setEnd(context.getStart() + left);
context.setUnbreakable(true);
if (left == currentString.length()) {
context.setWidth(c.getTextRenderer().getWidth(
c.getFontContext(), font, context.getCalculatedSubstring()));
} else {
context.setWidth(graphicsLength);
}
}
return;
}
private static boolean isChinese(char c) {
Character.UnicodeBlock ub = Character.UnicodeBlock.of(c);
if (ub == Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS
|| ub == Character.UnicodeBlock.CJK_COMPATIBILITY_IDEOGRAPHS
|| ub == Character.UnicodeBlock.CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A) {
return true;
}
return false;
}
private static int getStrRight(String s,int left){
if(left>=s.length())
return -1;
char[] ch = s.toCharArray();
for(int i = left;i<ch.length;i++){
if(isChinese(ch[i]) || ' ' == ch[i]){
return i==0?i+1:i;
}
}
return -1;
}
}
由于我使用的是maven管理项目依赖，包名有所变化
依赖：
<dependency> <groupId>org.xhtmlrenderer</groupId> <artifactId>flying-saucer-pdf-itext5</artifactId> <version>9.0.1</version> </dependency>
上面会引入以下几个依赖项：
flying-saucer-parent-9.0.1.pom
flying-saucer-pdf-itext5-9.0.1.jar
flying-saucer-core-9.0.1.jar
core-renderer-R8.jar
上面对应的代码就在flying-saucer-core-9.0.1.jar包中，修改之即可。
另外需要注意的就是，目前flying-saucer支持的最高版本是CSS 2.1，如果用了CSS 3的内容，可能会不起作用而导致版面错乱。
http://bettereveryday.iteye.com/blog/611561