jhandl / hounder Goto Github PK

Read a file with Config that contains an #include statement. If the
included file is missing, the exception thrown by Config is misleading,
stating that the file that could not be read is the main file, not the
included file.

http://pastebin.com/SER8JMqj
this is what i get when i try to ./install.sh

mkdir: cannot create directory `logs': File exists
Exception in thread "main" java.lang.NoClassDefFoundError: 
com/flaptor/hounder/installer/InstallationWizard
Caused by: java.lang.ClassNotFoundException: 
com.flaptor.hounder.installer.InstallationWizard
        at java.net.URLClassLoader$1.run(URLClassLoader.java:217)
        at java.security.AccessController.doPrivileged(Native Method)
        at java.net.URLClassLoader.findClass(URLClassLoader.java:205)
        at java.lang.ClassLoader.loadClass(ClassLoader.java:321)
        at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:294)
        at java.lang.ClassLoader.loadClass(ClassLoader.java:266)
Could not find the main class: 
com.flaptor.hounder.installer.InstallationWizard. Program will exit.

i am using ubuntu .

When doing a crawler install, selecting manual seeds, the installation
turns out to use a different, default seeds file.

go to websearch and press submitquery, a parsing error is displayed

We are currently using a trick to speed up the copy of index directories
that involves making a hard link copy of all the files. This works _only_
because lucene does never modify an index segment, but creates a new one
and deletes the original, if necessary. We have added metadata
(index.properties) that does not follow this convention, and is modified.
This breaks the indexer copy mechanism, making the properties of all
indexes those of the current one.

once installed, tools like listPageDb.sh fail because the logs dir is missing.

Currently, Hounder does not check for duplicated fields. This generates
some unpredictable behaviour in combination with tokenization. Besides,
indexing fields like "AddId?" makes the indexer crash.

The html parser should output '\n' when a <br> is found.

What steps will reproduce the problem?

1. Install from gui or command line and there is java class not found 
exception on PageDB.

The problem is caused by library naming conflict.

The problem is that the install scripts are looking for the following jars

hounder-trunk.jar, hounder-trunk-deps.jar

The distribution however contains hounder-2.0.jar, hounder-deps-2.0.jar

What steps will reproduce the problem?
1. Use a client that enforces OpenSearch specs, such as perl's WWW::OpenSearch.

What is the expected output? What do you see instead?

Error while parsing Description XML: Node should be OpenSearchDescription: rss 
at /usr/share/perl5/WWW/OpenSearch.pm line 139
 at /usr/share/perl5/WWW/OpenSearch.pm line 139

What version of the product are you using? On what operating system?

hounder-2.0.1-binary_installer on Linux, Java 1.6.

Please provide any additional information below.

See the specification at 
http://opensearch.a9.com/spec/opensearchdescription/1.0/  The first element, 
OpenSearchDescription, does not appear in Hounder's output, which by default 
is: 

<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:opensearch="http://a9.com/-/spec/opensearchrss/1.0/" 
xmlns:hounder="http://www.flaptor.com/opensearchrss/1.0/" version="2.0">
  <channel>
    <title>Hounder:</title>
    <link>http://wwwdev:47012/opensearch/search.jsp?query=&amp;start=0&amp;hitsPerPage=10</link>
    <description>Hounder search results for:</description>
    <opensearch:totalResults>0</opensearch:totalResults>

    <opensearch:startIndex>0</opensearch:startIndex>
    <opensearch:itemsPerPage>10</opensearch:itemsPerPage>
    <hounder:query></hounder:query>
    <hounder:status>0</hounder:status>
    <hounder:statusDesc>OK</hounder:statusDesc>
  </channel>
</rss>

...It does, however, appear to be valid RSS at least.

cannot do a version release of "hounder-1.0.jar" and maintain those scripts
need a generic way to fix this

add extra fields that arent in opensearch, especially the indicator of what
result index to ask for in the next page