chatparser.java

编辑视频文件

            String timeValue = Long.toString(((long) timeSlot[1]));

            resultSlots.put(tsId, timeValue);
        }

        return resultSlots;
    }

    /* (non-Javadoc)
     * @see mpi.eudico.server.corpora.clomimpl.abstr.Parser#getTierNames(java.lang.String)
     */
    public ArrayList getTierNames(String fileName) {
        // tierNames in ELAN are either the main tier '*PAR' labels, or
        // the combination of tier label plus participant, like '%mor@PAR'
        parse(fileName);

        return tierNames;
    }

    /* (non-Javadoc)
     * @see mpi.eudico.server.corpora.clomimpl.abstr.Parser#getParticipantOf(java.lang.String, java.lang.String)
     */
    public String getParticipantOf(String tierName, String fileName) {
        String participant = "";

        if (tierName.startsWith("*")) {
            participant = tierName.substring(1); // main tier label without *
        } else {
            int i = tierName.indexOf(TIER_NAME_SEPARATOR); // part of tier name after @

            if ((i > 0) && (tierName.length() > (i + 2))) {
                participant = tierName.substring(i + 1);
            }
        }

        return participant;
    }

    /* (non-Javadoc)
     * @see mpi.eudico.server.corpora.clomimpl.abstr.Parser#getLinguisticTypeOf(java.lang.String, java.lang.String)
     */
    public String getLinguisticTypeIDOf(String tierName, String fileName) {
        String lingTypeId = "";

        if (tierName.startsWith("*")) {
            lingTypeId = MAIN_TYPE; // main tier label without *
        } else {
            int i = tierName.indexOf(TIER_NAME_SEPARATOR); // part of tier name after @

            if (i > 0) {
                lingTypeId = tierName.substring(0, i);
            }
        }

        return lingTypeId;
    }

    /* (non-Javadoc)
     * @see mpi.eudico.server.corpora.clomimpl.abstr.Parser#getParentNameOf(java.lang.String, java.lang.String)
     */
    public String getParentNameOf(String tierName, String fileName) {
        parse(fileName);

        return (String) parentHash.get(tierName);
    }

    /* (non-Javadoc)
     * @see mpi.eudico.server.corpora.clomimpl.abstr.Parser#getAnnotationsOf(java.lang.String, java.lang.String)
     */
    public ArrayList getAnnotationsOf(String tierName, String fileName) {
        parse(fileName);

        ArrayList resultAnnotRecords = new ArrayList();

        Iterator it = annotRecordToTierMap.keySet().iterator();

        while (it.hasNext()) {
            AnnotationRecord annRec = (AnnotationRecord) it.next();

            if (annotRecordToTierMap.get(annRec).equals(tierName)) {
                resultAnnotRecords.add(annRec);
            }
        }

        return resultAnnotRecords;
    }

    private void parse(String fileName) {
        if (lastParsed.equals(fileName)) {
            return;
        }

        // (re)set everything to null for each parse
        participantLine = null;
        mediaFileName = null;
        chatBlocks.clear();
        lingTypeRecords.clear();
        tierNames.clear();
        parentHash.clear();
        timeOrder.clear();
        timeSlots.clear();
        annotationRecords.clear();
        annotRecordToTierMap.clear();

        br = null;

        // parse the file
        lastParsed = fileName;

        // do actual parsing
        try {
            br = new BufferedReader(new FileReader(fileName));
        } catch (Exception fnf) {
            fnf.printStackTrace();
        }

        String line = null;

        try {
            if ((line = br.readLine()) != null) {
                if (line.startsWith("@UTF8")) { // CHAT UTF-8
                    br.close();
                    br = new BufferedReader(new InputStreamReader(
                                new FileInputStream(fileName), "UTF-8"));
                }
            }
        } catch (IOException iox) {
            iox.printStackTrace();
        }

        parseLines();
        processBlocks();

        try {
            br.close();
        } catch (IOException io) {
            io.printStackTrace();
        }
    }

    private void parseLines() {
        String line = null;
        String outputLine = "";
        boolean recordingParticipant = false;

        ArrayList chatBlock = null;

        try {
            while ((line = br.readLine()) != null) {
                // Participant lines from header
                if (line.startsWith("@Participants:")) {
                    recordingParticipant = true;
                    participantLine = line;
                } else if (recordingParticipant == true) {
                    if (!(line.startsWith("@") || line.startsWith("*") ||
                            line.startsWith("%"))) {
                        // continuation of participants line
                        participantLine += line;
                    } else { // new header line or block line, end recording
                        recordingParticipant = false;
                    }
                }

                // CHAT "blocks"
                if (line.startsWith("*")) { // new block

                    // finish last line of previous block
                    if (!outputLine.equals("") && (chatBlock != null)) {
                        addLineToBlock(outputLine, chatBlock);
                    }

                    // output block
                    if (chatBlock != null) {
                        chatBlocks.add(chatBlock);
                    }

                    // start new recording
                    chatBlock = new ArrayList();

                    // add line to new recording
                    outputLine = line;
                } else if (line.startsWith("%") ||
                        ((line.length() > 1) &&
                        line.substring(1).startsWith("%"))) { // other lines

                    // finish last line
                    if (!outputLine.equals("") && (chatBlock != null)) {
                        addLineToBlock(outputLine, chatBlock);
                    }

                    outputLine = line;

                    if ((mediaFileName == null) &&
                            (startsWithMediaLabel(line) ||
                            ((line.length() > 1) &&
                            startsWithMediaLabel(line.substring(1))))) { // bullet in chat-utf8

                        // parse this line, second token is media file name.
                        StringTokenizer st = new StringTokenizer(line);

                        if (st.hasMoreTokens()) { // 'eat' %snd label
                            st.nextToken();
                        }

                        if (st.hasMoreTokens()) {
                            mediaFileName = st.nextToken();
                        }

                        // strip off possible double quotes
                        if (mediaFileName.startsWith("\"")) {
                            mediaFileName = mediaFileName.substring(1);
                        }

                        if (mediaFileName.endsWith("\"")) {
                            mediaFileName = mediaFileName.substring(0,
                                    mediaFileName.length() - 1);
                        }
                    }
                } else if (!line.startsWith("@")) { // no label, continuation of previous line
                    outputLine += line;
                }
            }

            // finish last line
            if (!outputLine.equals("") && (chatBlock != null)) {
                addLineToBlock(outputLine, chatBlock);
            }

            // output last block
            if (chatBlock != null) {
                chatBlocks.add(chatBlock);
            }
        } catch (FileNotFoundException fex) {
            fex.printStackTrace();
        } catch (IOException iex) {
            iex.printStackTrace();
        }
    }

    /**
     * Helper method to avoid copy and paste
     *
     * @param file DOCUMENT ME!
     *
     * @return DOCUMENT ME!
     *
     * @throws IOException DOCUMENT ME!
     */

    //	private final BufferedReader file2br(File file) throws IOException {

    /*
       A file is opened from the operating system.
       This stream of bytes could be a UTF-8 encoded unicode stream.
       If a file interpreted as UTF-8 contains isolatin-1, the file
       cannot be read. An Exception is thrown.
       Therefore, special care has to be taken when reading in UTF-8.
       As a first measure, the filename is used to decide if to read as UTF-8.
       This has to be changend in a future version.
       This is just done in order to include Unicode characters into Eudico.
     */
    /*        Reader filereader;

            if (-1 != file.getName().lastIndexOf(".utf8.")) { // this means 'contains'
                filereader = new InputStreamReader(new FileInputStream(file),
                        "UTF-8");
            } else {
                // use the locale encoding.
                filereader = new FileReader(file);
            }

            BufferedReader br = new BufferedReader(filereader);

            return br;
        }
    */
    private void addLineToBlock(String theLine, ArrayList theBlock) {
        String label = null;
        String value = null;

        label = getLabelPart(theLine);
        value = getValuePart(theLine);

        if ((label != null) && (value != null)) {
            String[] line = { label, value };
            theBlock.add(line);
        } else if ((label != null) && (value == null)) {
            // maybe a valid tierlabel with empty annotation content
            String[] line = { label, "" };
            theBlock.add(line);
        }
    }

    private String getLabelPart(String theLine) {
        String label = null;

        int index = theLine.indexOf(':');

        if (index > 0) {
            label = theLine.substring(0, index);
        }

        return label;
    }

    private String getValuePart(String theLine) {
        String value = null;

        int index = theLine.indexOf(':');

        if (index < (theLine.length() - 2)) {
            value = theLine.substring(index + 1).trim();
        }