Nutch
  1. Nutch
  2. NUTCH-1308

Unnecessary truncate content configuration, and logging in parse-zip/ZipParser

    Details

    • Type: Bug Bug
    • Status: Open
    • Priority: Major Major
    • Resolution: Unresolved
    • Affects Version/s: 1.4, nutchgora
    • Fix Version/s: 2.4
    • Component/s: parser
    • Labels:
      None

      Description

      Two issues here...

      1) Recently ferdy committed NUTCH-965 which skips parsing of truncated documents. Parse zip has it's own implementation for the same when it should really draw on the aforementioned implementation.
      2) If (in the offending piece of code mentioned above) truncation occurs, we get an incorrect log message the "Parser can't handle incomplete pdf files"!!! This is incorrect, shouldn't be there, and should be removed.

      
      72      if (contentLen != null && contentInBytes.length != len) {
      73 	return new ParseStatus(ParseStatus.FAILED,
      74 	ParseStatus.FAILED_TRUNCATED, "Content truncated at "
      75 	+ contentInBytes.length
      76 	+ " bytes. Parser can't handle incomplete pdf file.")
      77 	.getEmptyParseResult(content.getUrl(), getConf());
      78 	}
      

      For clarity, the issue is present in both Nutchgora branch[1] and Nutch trunk[2]

      [1] https://svn.apache.org/viewvc/nutch/branches/nutchgora/src/plugin/parse-zip/src/java/org/apache/nutch/parse/zip/ZipParser.java?diff_format=h&view=markup
      [2] https://svn.apache.org/viewvc/nutch/trunk/src/plugin/parse-zip/src/java/org/apache/nutch/parse/zip/ZipParser.java?diff_format=h&view=markup
      [2]

        Issue Links

          Activity

            People

            • Assignee:
              Lewis John McGibbney
              Reporter:
              Lewis John McGibbney
            • Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

              Dates

              • Created:
                Updated:

                Development