pdfencodings.java

来自「有关对pdf操作的代码」· Java 代码 · 共 798 行 · 第 1/3 页

            String text = extra.byteToChar(bytes, encoding);
            if (text != null)
                return text;
        }
        char ch[] = null;
        if (encoding.equals(BaseFont.WINANSI))
            ch = winansiByteToChar;
        else if (encoding.equals(PdfObject.TEXT_PDFDOCENCODING))
            ch = pdfEncodingByteToChar;
        if (ch != null) {
            int len = bytes.length;
            char c[] = new char[len];
            for (int k = 0; k < len; ++k) {
                c[k] = ch[bytes[k] & 0xff];
            }
            return new String(c);
        }
        try {
            return new String(bytes, encoding);
        }
        catch (UnsupportedEncodingException e) {
            throw new ExceptionConverter(e);
        }
    }
    
    /** Checks is <CODE>text</CODE> only has PdfDocEncoding characters.
     * @param text the <CODE>String</CODE> to test
     * @return <CODE>true</CODE> if only PdfDocEncoding characters are present
     */    
    public static boolean isPdfDocEncoding(String text) {
        if (text == null)
            return true;
        int len = text.length();
        for (int k = 0; k < len; ++k) {
            char char1 = text.charAt(k);
            if (char1 < 128 || (char1 > 160 && char1 <= 255))
                continue;
            if (!pdfEncoding.containsKey(char1))
                return false;
        }
        return true;
    }
    
    static final HashMap cmaps = new HashMap();
    /** Assumes that '\\n' and '\\r\\n' are the newline sequences. It may not work for
     * all CJK encodings. To be used with loadCmap().
     */    
    public static final byte CRLF_CID_NEWLINE[][] = new byte[][]{{(byte)'\n'}, {(byte)'\r', (byte)'\n'}};

    /** Clears the CJK cmaps from the cache. If <CODE>name</CODE> is the
     * empty string then all the cache is cleared. Calling this method
     * has no consequences other than the need to reload the cmap
     * if needed.
     * @param name the name of the cmap to clear or all the cmaps if the empty string
     */    
    public static void clearCmap(String name) {
        synchronized (cmaps) {
            if (name.length() == 0)
                cmaps.clear();
            else
                cmaps.remove(name);
        }
    }
    
    /** Loads a CJK cmap to the cache with the option of associating
     * sequences to the newline.
     * @param name the CJK cmap name
     * @param newline the sequences to be replaced by a newline in the resulting CID. See <CODE>CRLF_CID_NEWLINE</CODE>
     */    
    public static void loadCmap(String name, byte newline[][]) {
        try {
            char planes[][] = null;
            synchronized (cmaps) {
                planes = (char[][])cmaps.get(name);
            }
            if (planes == null) {
                planes = readCmap(name, newline);
                synchronized (cmaps) {
                    cmaps.put(name, planes);
                }
            }
        }
        catch (IOException e) {
            throw new ExceptionConverter(e);
        }        
    }
    
    /** Converts a <CODE>byte</CODE> array encoded as <CODE>name</CODE>
     * to a CID string. This is needed to reach some CJK characters
     * that don't exist in 16 bit Unicode.</p>
     * The font to use this result must use the encoding "Identity-H"
     * or "Identity-V".</p>
     * See ftp://ftp.oreilly.com/pub/examples/nutshell/cjkv/adobe/.
     * @param name the CJK encoding name
     * @param seq the <CODE>byte</CODE> array to be decoded
     * @return the CID string
     */    
    public static String convertCmap(String name, byte seq[]) {
        return convertCmap(name, seq, 0, seq.length);
    }
    
    /** Converts a <CODE>byte</CODE> array encoded as <CODE>name</CODE>
     * to a CID string. This is needed to reach some CJK characters
     * that don't exist in 16 bit Unicode.</p>
     * The font to use this result must use the encoding "Identity-H"
     * or "Identity-V".</p>
     * See ftp://ftp.oreilly.com/pub/examples/nutshell/cjkv/adobe/.
     * @param name the CJK encoding name
     * @param start the start offset in the data
     * @param length the number of bytes to convert
     * @param seq the <CODE>byte</CODE> array to be decoded
     * @return the CID string
     */    
    public static String convertCmap(String name, byte seq[], int start, int length) {
        try {
            char planes[][] = null;
            synchronized (cmaps) {
                planes = (char[][])cmaps.get(name);
            }
            if (planes == null) {
                planes = readCmap(name, (byte[][])null);
                synchronized (cmaps) {
                    cmaps.put(name, planes);
                }
            }
            return decodeSequence(seq, start, length, planes);
        }
        catch (IOException e) {
            throw new ExceptionConverter(e);
        }        
    }
    
    static String decodeSequence(byte seq[], int start, int length, char planes[][]) {
        StringBuffer buf = new StringBuffer();
        int end = start + length;
        int currentPlane = 0;
        for (int k = start; k < end; ++k) {
            int one = (int)seq[k] & 0xff;
            char plane[] = planes[currentPlane];
            int cid = plane[one];
            if ((cid & 0x8000) == 0) {
                buf.append((char)cid);
                currentPlane = 0;
            }
            else
                currentPlane = cid & 0x7fff;
        }
        return buf.toString();
    }

    static char[][] readCmap(String name, byte newline[][]) throws IOException {
        ArrayList planes = new ArrayList();
        planes.add(new char[256]);
        readCmap(name, planes);
        if (newline != null) {
            for (int k = 0; k < newline.length; ++k)
                encodeSequence(newline[k].length, newline[k], BaseFont.CID_NEWLINE, planes);
        }
        char ret[][] = new char[planes.size()][];
        return (char[][])planes.toArray(ret);
    }
    
    static void readCmap(String name, ArrayList planes) throws IOException {
        String fullName = BaseFont.RESOURCE_PATH + "cmaps/" + name;
        InputStream in = BaseFont.getResourceStream(fullName);
        if (in == null)
            throw new IOException("The Cmap " + name + " was not found.");
        encodeStream(in, planes);
        in.close();
    }
    
    static void encodeStream(InputStream in, ArrayList planes) throws IOException {
        BufferedReader rd = new BufferedReader(new InputStreamReader(in, "iso-8859-1"));
        String line = null;
        int state = CIDNONE;
        byte seqs[] = new byte[7];
        while ((line = rd.readLine()) != null) {
            if (line.length() < 6)
                continue;
            switch (state) {
                case CIDNONE: {
                    if (line.indexOf("begincidrange") >= 0)
                        state = CIDRANGE;
                    else if (line.indexOf("begincidchar") >= 0)
                        state = CIDCHAR;
                    else if (line.indexOf("usecmap") >= 0) {
                        StringTokenizer tk = new StringTokenizer(line);
                        String t = tk.nextToken();
                        readCmap(t.substring(1), planes);
                    }
                    break;
                }
                case CIDRANGE: {
                    if (line.indexOf("endcidrange") >= 0) {
                        state = CIDNONE;
                        break;
                    }
                    StringTokenizer tk = new StringTokenizer(line);
                    String t = tk.nextToken();
                    int size = t.length() / 2 - 1;
                    long start = Long.parseLong(t.substring(1, t.length() - 1), 16);
                    t = tk.nextToken();
                    long end = Long.parseLong(t.substring(1, t.length() - 1), 16);
                    t = tk.nextToken();
                    int cid = Integer.parseInt(t);
                    for (long k = start; k <= end; ++k) {
                        breakLong(k, size, seqs);
                        encodeSequence(size, seqs, (char)cid, planes);
                        ++cid;
                    }
                    break;
                }
                case CIDCHAR: {
                    if (line.indexOf("endcidchar") >= 0) {
                        state = CIDNONE;
                        break;
                    }
                    StringTokenizer tk = new StringTokenizer(line);
                    String t = tk.nextToken();
                    int size = t.length() / 2 - 1;
                    long start = Long.parseLong(t.substring(1, t.length() - 1), 16);
                    t = tk.nextToken();
                    int cid = Integer.parseInt(t);
                    breakLong(start, size, seqs);
                    encodeSequence(size, seqs, (char)cid, planes);
                    break;
                }
            }
        }
    }
    
    static void breakLong(long n, int size, byte seqs[]) {
        for (int k = 0; k < size; ++k) {
            seqs[k] = (byte)(n >> ((size - 1 - k) * 8));
        }
    }

    static void encodeSequence(int size, byte seqs[], char cid, ArrayList planes) {
        --size;
        int nextPlane = 0;
        for (int idx = 0; idx < size; ++idx) {
            char plane[] = (char[])planes.get(nextPlane);
            int one = (int)seqs[idx] & 0xff;
            char c = plane[one];
            if (c != 0 && (c & 0x8000) == 0)
                throw new RuntimeException("Inconsistent mapping.");
            if (c == 0) {
                planes.add(new char[256]);
                c = (char)((planes.size() - 1) | 0x8000);
                plane[one] = c;
            }
            nextPlane = c & 0x7fff;
        }
        char plane[] = (char[])planes.get(nextPlane);
        int one = (int)seqs[size] & 0xff;
        char c = plane[one];
        if ((c & 0x8000) != 0)
            throw new RuntimeException("Inconsistent mapping.");
        plane[one] = cid;
    }

    /** Adds an extra encoding.
     * @param name the name of the encoding. The encoding recognition is case insensitive
     * @param enc the conversion class
     */    
    public static void addExtraEncoding(String name, ExtraEncoding enc) {