Examples of TemporaryResources

Examples of org.apache.tika.io.TemporaryResources

144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
    if (!ExternalParser.check(checkCmd))
      return;

    XHTMLContentHandler xhtml = new XHTMLContentHandler(handler, metadata);

    TemporaryResources tmp = new TemporaryResources();
    File output = null;
    try {
      TikaInputStream tikaStream = TikaInputStream.get(stream, tmp);
      File input = tikaStream.getFile();
      long size = tikaStream.getLength();

      if (size >= config.getMinFileSizeToOcr() && size <= config.getMaxFileSizeToOcr()) {

        output = tmp.createTemporaryFile();
        doOCR(input, output, config);

        // Tesseract appends .txt to output file name
        output = new File(output.getAbsolutePath() + ".txt");

        if (output.exists())
          extractOutput(new FileInputStream(output), xhtml);

      }

    } finally {
      tmp.dispose();
      if (output != null)
        output.delete();

    }
  }
View Full Code Here

Examples of org.apache.tika.io.TemporaryResources

137
138
139
140
141
142
143
144
145
146
147
148
149
            Metadata metadata, ParseContext context)
            throws IOException, SAXException, TikaException {
        XHTMLContentHandler xhtml =
            new XHTMLContentHandler(handler, metadata);

        TemporaryResources tmp = new TemporaryResources();
        try {
            parse(TikaInputStream.get(stream, tmp),
                    xhtml, metadata, tmp);
        } finally {
            tmp.dispose();
        }
    }
View Full Code Here

Examples of org.apache.tika.io.TemporaryResources

66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
        // Check if we have access to the document
        if (input == null) {
            return MediaType.OCTET_STREAM;
        }

        TemporaryResources tmp = new TemporaryResources();
        try {
            TikaInputStream tis = TikaInputStream.get(input, tmp);

            byte[] prefix = new byte[1024]; // enough for all known formats
            int length = tis.peek(prefix);

            MediaType type = detectArchiveFormat(prefix, length);
            if (PackageParser.isZipArchive(type)
                    && TikaInputStream.isTikaInputStream(input)) {
                return detectZipFormat(tis);
            } else if (!type.equals(MediaType.OCTET_STREAM)) {
                return type;
            } else {
                return detectCompressorFormat(prefix, length);
            }
        } finally {
            try {
                tmp.dispose();
            } catch (TikaException e) {
                // ignore
            }
        }
    }
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.