XMLPULL源码解读记录 方法解读parseEndTag

a、解析过程中遇到</字符，进入endTag解析。

b、代码详解

    public int parseEndTag() throws XmlPullParserException, IOException {
        //ASSUMPTION ch is past "</"
        // [42] ETag ::=  '</' Name S? '>'
        //紧接在</后面的字符需要检查是否符合isNameStartChar()条件
        char ch = more();
        if(!isNameStartChar(ch)) {
            throw new XmlPullParserException(
                "expected name start and not "+printable(ch), this, null);
        }
        //类似parseStartTag，将posStart置于<字符
        posStart = pos - 3;
        final int nameStart = pos - 1 + bufAbsoluteStart;
        do {
            ch = more();
        } while(isNameChar(ch));

        // now we go one level down -- do checks
        //--depth;  //FIXME

        // check that end tag name is the same as start tag
        //String name = new String(buf, nameStart - bufAbsoluteStart,
        //                           (pos - 1) - (nameStart - bufAbsoluteStart));
        //int last = pos - 1;
        int off = nameStart - bufAbsoluteStart;
        //final int len = last - off;
        final int len = (pos - 1) - off;
        final char[] cbuf = elRawName[depth];
        //如果与startTag的元素名称长度不一致，抛异常
        if(elRawNameEnd[depth] != len) {
            // construct strings for exception
            final String startname = new String(cbuf, 0, elRawNameEnd[depth]);
            final String endname = new String(buf, off, len);
            throw new XmlPullParserException(
                "end tag name </"+endname+"> must match start tag name <"+startname+">"
                    +" from line "+elRawNameLine[depth], this, null);
        }
        //如果与startTag的元素名称不完全相同，抛异常
        for (int i = 0; i < len; i++)
        {
            if(buf[off++] != cbuf[i]) {
                // construct strings for exception
                final String startname = new String(cbuf, 0, len);
                final String endname = new String(buf, off - i - 1, len);
                throw new XmlPullParserException(
                    "end tag name </"+endname+"> must be the same as start tag <"+startname+">"
                        +" from line "+elRawNameLine[depth], this, null);
            }
        }

        //忽略元素名称结尾部分空字符
        while(isS(ch)) { ch = more(); } // skip additional white spaces
        //没有>紧随则抛出异常
        if(ch != '>') {
            throw new XmlPullParserException(
                "expected > to finish end tag not "+printable(ch)
                    +" from line "+elRawNameLine[depth], this, null);
        }

        //namespaceEnd = elNamespaceCount[ depth ]; //FIXME
        posEnd = pos;
        pastEndTag = true;
        return eventType = END_TAG;
    }

 

c、代码要点

c.1、第一个字符必须符合isNameStartChar()条件。xml规范中大多有这个要求，名称后面出现空格等空白字符是允许的。

c.2、解析元素名称，检查与startTag对应元素名称是否一致。

c.3、紧接元素名称后，是否有>字符收尾。

c.4、返回END_TAG事件。