xmlentityscanner.java

来自「JAVA的一些源码 JAVA2 STANDARD EDITION DEVELO」· Java 代码 · 共 1,660 行 · 第 1/5 页

     * <strong>Note:</strong> This method does not guarantee to return
     * the longest run of parsed character data. This method may return
     * before markup due to reaching the end of the input buffer or any
     * other reason.
     * <p>
     * <strong>Note:</strong> The fields contained in the XMLString
     * structure are not guaranteed to remain valid upon subsequent calls
     * to the entity scanner. Therefore, the caller is responsible for
     * immediately using the returned character data or making a copy of
     * the character data.
     *
     * @param content The content structure to fill.
     *
     * @return Returns the next character on the input, if known. This
     *         value may be -1 but this does <em>note</em> designate
     *         end of file.
     *
     * @throws IOException  Thrown if i/o error occurs.
     * @throws EOFException Thrown on end of file.
     */
    public int scanContent(XMLString content) throws IOException {
        if (DEBUG_BUFFER) {
            System.out.print("(scanContent: ");
            XMLEntityManager.print(fCurrentEntity);
            System.out.println();
        }

        // load more characters, if needed
        if (fCurrentEntity.position == fCurrentEntity.count) {
            load(0, true);
        }
        else if (fCurrentEntity.position == fCurrentEntity.count - 1) {
            fCurrentEntity.ch[0] = fCurrentEntity.ch[fCurrentEntity.count - 1];
            load(1, false);
            fCurrentEntity.position = 0;
        }

        // normalize newlines
        int offset = fCurrentEntity.position;
        int c = fCurrentEntity.ch[offset];
        int newlines = 0;
        boolean external = fCurrentEntity.isExternal();
        if (c == '\n' || (c == '\r' && external)) {
            if (DEBUG_BUFFER) {
                System.out.print("[newline, "+offset+", "+fCurrentEntity.position+": ");
                XMLEntityManager.print(fCurrentEntity);
                System.out.println();
            }
            do {
                c = fCurrentEntity.ch[fCurrentEntity.position++];
                if (c == '\r' && external) {
                    newlines++;
                    fCurrentEntity.lineNumber++;
                    fCurrentEntity.columnNumber = 1;
                    if (fCurrentEntity.position == fCurrentEntity.count) {
                        offset = 0;
                        fCurrentEntity.position = newlines;
                        if (load(newlines, false)) {
                            break;
                        }
                    }
                    if (fCurrentEntity.ch[fCurrentEntity.position] == '\n') {
                        fCurrentEntity.position++;
                        offset++;
                    }
                    /*** NEWLINE NORMALIZATION ***/
                    else {
                        newlines++;
                    }
                }
                else if (c == '\n') {
                    newlines++;
                    fCurrentEntity.lineNumber++;
                    fCurrentEntity.columnNumber = 1;
                    if (fCurrentEntity.position == fCurrentEntity.count) {
                        offset = 0;
                        fCurrentEntity.position = newlines;
                        if (load(newlines, false)) {
                            break;
                        }
                    }
                }
                else {
                    fCurrentEntity.position--;
                    break;
                }
            } while (fCurrentEntity.position < fCurrentEntity.count - 1);
            for (int i = offset; i < fCurrentEntity.position; i++) {
                fCurrentEntity.ch[i] = '\n';
            }
            int length = fCurrentEntity.position - offset;
            if (fCurrentEntity.position == fCurrentEntity.count - 1) {
                content.setValues(fCurrentEntity.ch, offset, length);
                if (DEBUG_BUFFER) {
                    System.out.print("]newline, "+offset+", "+fCurrentEntity.position+": ");
                    XMLEntityManager.print(fCurrentEntity);
                    System.out.println();
                }
                return -1;
            }
            if (DEBUG_BUFFER) {
                System.out.print("]newline, "+offset+", "+fCurrentEntity.position+": ");
                XMLEntityManager.print(fCurrentEntity);
                System.out.println();
            }
        }

        // inner loop, scanning for content
        while (fCurrentEntity.position < fCurrentEntity.count) {
            c = fCurrentEntity.ch[fCurrentEntity.position++];
            if (!XMLChar.isContent(c)) {
                fCurrentEntity.position--;
                break;
            }
        }
        int length = fCurrentEntity.position - offset;
        fCurrentEntity.columnNumber += length - newlines;
        content.setValues(fCurrentEntity.ch, offset, length);

        // return next character
        if (fCurrentEntity.position != fCurrentEntity.count) {
            c = fCurrentEntity.ch[fCurrentEntity.position];
            // REVISIT: Does this need to be updated to fix the
            //          #x0D ^#x0A newline normalization problem? -Ac
            if (c == '\r' && external) {
                c = '\n';
            }
        }
        else {
            c = -1;
        }
        if (DEBUG_BUFFER) {
            System.out.print(")scanContent: ");
            XMLEntityManager.print(fCurrentEntity);
            System.out.println(" -> '"+(char)c+"'");
        }
        return c;

    } // scanContent(XMLString):int

    /**
     * Scans a range of attribute value data, setting the fields of the
     * XMLString structure, appropriately.
     * <p>
     * <strong>Note:</strong> The characters are consumed.
     * <p>
     * <strong>Note:</strong> This method does not guarantee to return
     * the longest run of attribute value data. This method may return
     * before the quote character due to reaching the end of the input
     * buffer or any other reason.
     * <p>
     * <strong>Note:</strong> The fields contained in the XMLString
     * structure are not guaranteed to remain valid upon subsequent calls
     * to the entity scanner. Therefore, the caller is responsible for
     * immediately using the returned character data or making a copy of
     * the character data.
     *
     * @param quote   The quote character that signifies the end of the
     *                attribute value data.
     * @param content The content structure to fill.
     *
     * @return Returns the next character on the input, if known. This
     *         value may be -1 but this does <em>note</em> designate
     *         end of file.
     *
     * @throws IOException  Thrown if i/o error occurs.
     * @throws EOFException Thrown on end of file.
     */
    public int scanLiteral(int quote, XMLString content)
        throws IOException {
        if (DEBUG_BUFFER) {
            System.out.print("(scanLiteral, '"+(char)quote+"': ");
            XMLEntityManager.print(fCurrentEntity);
            System.out.println();
        }

        // load more characters, if needed
        if (fCurrentEntity.position == fCurrentEntity.count) {
            load(0, true);
        }
        else if (fCurrentEntity.position == fCurrentEntity.count - 1) {
            fCurrentEntity.ch[0] = fCurrentEntity.ch[fCurrentEntity.count - 1];
            load(1, false);
            fCurrentEntity.position = 0;
        }

        // normalize newlines
        int offset = fCurrentEntity.position;
        int c = fCurrentEntity.ch[offset];
        int newlines = 0;
        boolean external = fCurrentEntity.isExternal();
        if (c == '\n' || (c == '\r' && external)) {
            if (DEBUG_BUFFER) {
                System.out.print("[newline, "+offset+", "+fCurrentEntity.position+": ");
                XMLEntityManager.print(fCurrentEntity);
                System.out.println();
            }
            do {
                c = fCurrentEntity.ch[fCurrentEntity.position++];
                if (c == '\r' && external) {
                    newlines++;
                    fCurrentEntity.lineNumber++;
                    fCurrentEntity.columnNumber = 1;
                    if (fCurrentEntity.position == fCurrentEntity.count) {
                        offset = 0;
                        fCurrentEntity.position = newlines;
                        if (load(newlines, false)) {
                            break;
                        }
                    }
                    if (fCurrentEntity.ch[fCurrentEntity.position] == '\n') {
                        fCurrentEntity.position++;
                        offset++;
                    }
                    /*** NEWLINE NORMALIZATION ***/
                    else {
                        newlines++;
                    }
                    /***/
                }
                else if (c == '\n') {
                    newlines++;
                    fCurrentEntity.lineNumber++;
                    fCurrentEntity.columnNumber = 1;
                    if (fCurrentEntity.position == fCurrentEntity.count) {
                        offset = 0;
                        fCurrentEntity.position = newlines;
                        if (load(newlines, false)) {
                            break;
                        }
                    }
                }
                else {
                    fCurrentEntity.position--;
                    break;
                }
            } while (fCurrentEntity.position < fCurrentEntity.count - 1);
            for (int i = offset; i < fCurrentEntity.position; i++) {
                fCurrentEntity.ch[i] = '\n';
            }
            int length = fCurrentEntity.position - offset;
            if (fCurrentEntity.position == fCurrentEntity.count - 1) {
                content.setValues(fCurrentEntity.ch, offset, length);
                if (DEBUG_BUFFER) {
                    System.out.print("]newline, "+offset+", "+fCurrentEntity.position+": ");
                    XMLEntityManager.print(fCurrentEntity);
                    System.out.println();
                }
                return -1;
            }
            if (DEBUG_BUFFER) {
                System.out.print("]newline, "+offset+", "+fCurrentEntity.position+": ");
                XMLEntityManager.print(fCurrentEntity);
                System.out.println();
            }
        }

        // scan literal value
        while (fCurrentEntity.position < fCurrentEntity.count) {
            c = fCurrentEntity.ch[fCurrentEntity.position++];
            if ((c == quote &&
                 (!fCurrentEntity.literal || external))
                || c == '%' || !XMLChar.isContent(c)) {
                fCurrentEntity.position--;
                break;
            }
        }
        int length = fCurrentEntity.position - offset;
        fCurrentEntity.columnNumber += length - newlines;
        content.setValues(fCurrentEntity.ch, offset, length);

        // return next character
        if (fCurrentEntity.position != fCurrentEntity.count) {
            c = fCurrentEntity.ch[fCurrentEntity.position];
            // NOTE: We don't want to accidentally signal the
            //       end of the literal if we're expanding an
            //       entity appearing in the literal. -Ac
            if (c == quote && fCurrentEntity.literal) {
                c = -1;
            }
        }
        else {
            c = -1;
        }
        if (DEBUG_BUFFER) {
            System.out.print(")scanLiteral, '"+(char)quote+"': ");
            XMLEntityManager.print(fCurrentEntity);
            System.out.println(" -> '"+(char)c+"'");
        }
        return c;

    } // scanLiteral(int,XMLString):int

    /**
     * Scans a range of character data up to the specified delimiter,
     * setting the fields of the XMLString structure, appropriately.
     * <p>
     * <strong>Note:</strong> The characters are consumed.
     * <p>
     * <strong>Note:</strong> This assumes that the internal buffer is
     * at least the same size, or bigger, than the length of the delimiter
     * and that the delimiter contains at least one character.
     * <p>
     * <strong>Note:</strong> This method does not guarantee to return
     * the longest run of character data. This method may return before
     * the delimiter due to reaching the end of the input buffer or any
     * other reason.
     * <p>
     * <strong>Note:</strong> The fields contained in the XMLString
     * structure are not guaranteed to remain valid upon subsequent calls
     * to the entity scanner. Therefore, the caller is responsible for
     * immediately using the returned character data or making a copy of
     * the character data.
     *
     * @param delimiter The string that signifies the end of the character
     *                  data to be scanned.
     * @param data      The data structure to fill.
     *
     * @return Returns true if there is more data to scan, false otherwise.
     *
     * @throws IOException  Thrown if i/o error occurs.
     * @throws EOFException Thrown on end of file.
     */
    public boolean scanData(String delimiter, XMLStringBuffer buffer)
        throws IOException {

        // REVISIT: This method does not need to use a string buffer.
        //          The change would avoid the array copies and increase
        //          performance. -Ac
        //
        //          Currently, this method is called for scanning CDATA
        //          sections, comments,  and processing instruction data.