baseparser.java

非常有用的操作pdf文件的java源码

        COSBase retval = null;

        skipSpaces();
        int nextByte = pdfSource.peek();
        char c = (char)nextByte;
        switch(c)
        {
            case '<':
            {
                int leftBracket = pdfSource.read();//pull off first left bracket
                c = (char)pdfSource.peek(); //check for second left bracket
                pdfSource.unread( leftBracket );
                if(c == '<')
                {

                    retval = parseCOSDictionary();
                    skipSpaces();
                }
                else
                {
                    retval = parseCOSString();
                }
                break;
            }
            case '[': // array
            {
                retval = parseCOSArray();
                break;
            }
            case '(':
                retval = parseCOSString();
                break;
            case '/':   // name
                retval = parseCOSName();
                break;
            case 'n':   // null
            {
                String nullString = readString();
                if( !nullString.equals( "null") )
                {
                    throw new IOException("Expected='null' actual='" + nullString + "'");
                }
                retval = COSNull.NULL;
                break;
            }
            case 't':
            {
                byte[] trueBytes = new byte[4];
                int amountRead = pdfSource.read( trueBytes, 0, 4 );
                String trueString = new String( trueBytes, 0, amountRead );
                if( trueString.equals( "true" ) )
                {
                    retval = COSBoolean.TRUE;
                }
                else
                {
                    throw new IOException( "expected true actual='" + trueString + "' " + pdfSource );
                }
                break;
            }
            case 'f':
            {
                byte[] falseBytes = new byte[5];
                int amountRead = pdfSource.read( falseBytes, 0, 5 );
                String falseString = new String( falseBytes, 0, amountRead );
                if( falseString.equals( "false" ) )
                {
                    retval = COSBoolean.FALSE;
                }
                else
                {
                    throw new IOException( "expected false actual='" + falseString + "' " + pdfSource );
                }
                break;
            }
            case 'R':
                pdfSource.read();
                retval = new COSObject(null);
                break;
            case (char)-1:
                return null;
            default:
            {
                if( Character.isDigit(c) || c == '-' || c == '+' || c == '.')
                {
                    StringBuffer buf = new StringBuffer();
                    int ic = pdfSource.read();
                    c = (char)ic;
                    while( Character.isDigit( c )||
                           c == '-' ||
                           c == '+' ||
                           c == '.' ||
                           c == 'E' ||
                           c == 'e' )
                    {
                        buf.append( c );
                        ic = pdfSource.read();
                        c = (char)ic;
                    }
                    if( ic != -1 )
                    {
                        pdfSource.unread( ic );
                    }
                    retval = COSNumber.get( buf.toString() );
                }
                else
                {
                    //This is not suppose to happen, but we will allow for it
                    //so we are more compatible with POS writers that don't
                    //follow the spec
                    String badString = readString();
                    //throw new IOException( "Unknown dir object c='" + c +
                    //"' peek='" + (char)pdfSource.peek() + "' " + pdfSource );
                    if( badString == null || badString.length() == 0 )
                    {
                        int peek = pdfSource.peek();
                        // we can end up in an infinite loop otherwise
                        throw new IOException( "Unknown dir object c='" + c +
                           "' cInt=" + (int)c + " peek='" + (char)peek + "' peekInt=" + peek + " " + pdfSource );
                    }

                }
            }
        }
        return retval;
    }

    /**
     * This will read the next string from the stream.
     *
     * @return The string that was read from the stream.
     *
     * @throws IOException If there is an error reading from the stream.
     */
    protected String readString() throws IOException
    {
        skipSpaces();
        StringBuffer buffer = new StringBuffer();
        int c = pdfSource.read();
        while( !isEndOfName((char)c) && !isClosing(c) && c != -1 )
        {
            buffer.append( (char)c );
            c = pdfSource.read();
        }
        if (c != -1)
        {
            pdfSource.unread(c);
        }
        return buffer.toString();
    }

    /**
     * This will read bytes until the end of line marker occurs.
     *
     * @param theString The next expected string in the stream.
     *
     * @return The characters between the current position and the end of the line.
     *
     * @throws IOException If there is an error reading from the stream or theString does not match what was read.
     */
    protected String readExpectedString( String theString ) throws IOException
    {
        int c = pdfSource.read();
        while( isWhitespace(c) && c != -1)
        {
            c = pdfSource.read();
        }
        StringBuffer buffer = new StringBuffer( theString.length() );
        int charsRead = 0;
        while( !isEOL(c) && c != -1 && charsRead < theString.length() )
        {
            char next = (char)c;
            buffer.append( next );
            if( theString.charAt( charsRead ) == next )
            {
                charsRead++;
            }
            else
            {
                throw new IOException( "Error: Expected to read '" + theString +
                    "' instead started reading '" +buffer.toString() + "'" );
            }
            c = pdfSource.read();
        }
        while( isEOL(c) && c != -1 )
        {
            c = pdfSource.read();
        }
        if (c != -1)
        {
            pdfSource.unread(c);
        }
        return buffer.toString();
    }

    /**
     * This will read the next string from the stream up to a certain length.
     *
     * @param length The length to stop reading at.
     *
     * @return The string that was read from the stream of length 0 to length.
     *
     * @throws IOException If there is an error reading from the stream.
     */
    protected String readString( int length ) throws IOException
    {
        skipSpaces();

        int c = pdfSource.read();
        
        //average string size is around 2 and the normal string buffer size is
        //about 16 so lets save some space.
        StringBuffer buffer = new StringBuffer(length);
        while( !isWhitespace(c) && !isClosing(c) && c != -1 && buffer.length() < length &&
            c != '[' &&
            c != '<' &&
            c != '(' &&
            c != '/' )
        {
            buffer.append( (char)c );
            c = pdfSource.read();
        }
        if (c != -1)
        {
            pdfSource.unread(c);
        }
        return buffer.toString();
    }

    /**
     * This will tell if the next character is a closing brace( close of PDF array ).
     *
     * @return true if the next byte is ']', false otherwise.
     *
     * @throws IOException If an IO error occurs.
     */
    protected boolean isClosing() throws IOException
    {
        return isClosing(pdfSource.peek());
    }
    
    /**
     * This will tell if the next character is a closing brace( close of PDF array ).
     *
     * @param c The character to check against end of line
     * @return true if the next byte is ']', false otherwise.
     */
    protected boolean isClosing(int c) 
    {
        return c == ']';
    }

    /**
     * This will read bytes until the end of line marker occurs.
     *
     * @return The characters between the current position and the end of the line.
     *
     * @throws IOException If there is an error reading from the stream.
     */
    protected String readLine() throws IOException
    {
        int c = pdfSource.read();
        while(isWhitespace(c) && c != -1)
        {
            c = pdfSource.read();
        }
        StringBuffer buffer = new StringBuffer( 11 );
        
        while( !isEOL(c) && c != -1 )
        {
            buffer.append( (char)c );
            c = pdfSource.read();
        }
        while( isEOL(c) && c != -1 )
        {
            c = pdfSource.read();
        }
        if (c != -1)
        {
            pdfSource.unread(c);
        }
        return buffer.toString();
    }

    /**
     * This will tell if the next byte to be read is an end of line byte.
     *
     * @return true if the next byte is 0x0A or 0x0D.
     *
     * @throws IOException If there is an error reading from the stream.
     */
    protected boolean isEOL() throws IOException
    {
        return isEOL(pdfSource.peek());
    }
    
    /**
     * This will tell if the next byte to be read is an end of line byte.
     *
     * @param c The character to check against end of line
     * @return true if the next byte is 0x0A or 0x0D.
     */
    protected boolean isEOL(int c)
    {
        return c == 10 || c == 13;
    }

    /**
     * This will tell if the next byte is whitespace or not.
     *
     * @return true if the next byte in the stream is a whitespace character.
     *
     * @throws IOException If there is an error reading from the stream.
     */
    protected boolean isWhitespace() throws IOException
    {
        return isWhitespace( pdfSource.peek() );
    }

    /**
     * This will tell if the next byte is whitespace or not.
     *
     * @param c The character to check against whitespace
     *
     * @return true if the next byte in the stream is a whitespace character.
     */
    protected boolean isWhitespace( int c )
    {
        return c == 0 || c == 9 || c == 12  || c == 10
        || c == 13 || c == 32;
    }

    /**
     * This will skip all spaces and comments that are present.
     *
     * @throws IOException If there is an error reading from the stream.
     */
    protected void skipSpaces() throws IOException
    {
        //log( "skipSpaces() " + pdfSource );
        int c = pdfSource.read();
        // identical to, but faster as: isWhiteSpace(c) || c == 37
        while(c == 0 || c == 9 || c == 12  || c == 10
                || c == 13 || c == 32 || c == 37)//37 is the % character, a comment
        {
            if ( c == 37 )
            {
                // skip past the comment section
                c = pdfSource.read();
                while(!isEOL(c) && c != -1)
                {
                    c = pdfSource.read();
                }
            }
            else 
            {
                c = pdfSource.read();
            }
        }
        if (c != -1)
        {
            pdfSource.unread(c);
        }
        //log( "skipSpaces() done peek='" + (char)pdfSource.peek() + "'" );
    }

    /**
     * This will read an integer from the stream.
     *
     * @return The integer that was read from the stream.
     *
     * @throws IOException If there is an error reading from the stream.
     */
    protected int readInt() throws IOException
    {
        skipSpaces();
        int retval = 0;

        int lastByte = 0;
        StringBuffer intBuffer = new StringBuffer();
        while( (lastByte = pdfSource.read() ) != 32 &&
        lastByte != 10 &&
        lastByte != 13 &&
        lastByte != 0 && //See sourceforge bug 853328
        lastByte != -1 )
        {
            intBuffer.append( (char)lastByte );
        }
        try
        {
            retval = Integer.parseInt( intBuffer.toString() );
        }
        catch( NumberFormatException e )
        {
            throw new IOException( "Error: Expected an integer type, actual='" + intBuffer + "'" );
        }
        return retval;
    }

    /**
     * This will add an xref.
     *
     * @param xref The xref to add.
     */
    public void addXref( PDFXref xref )
    {
        xrefs.add(xref);
    }

    /**
     * This will get all of the xrefs.
     *
     * @return A list of all xrefs.
     */
    public List getXrefs()
    {
        return xrefs;
    }
}