What is Plush ?

Plush is PyLUcene SHell to play with a Lucene indexes interactively.

Main features:

• View store information.
• View indexes definition.
• Search using the Lucene Query Parser Syntax.
• Sort result list.
• Browse by document number.
• Top term occurences for a field, matching a regex.
• Support PyLucene 1.9.1, 2.0.0 and 2.1.0
• Interactive shell emacs like command history and editing features.
• Command line tool and thus scriptable.
• Easy installation, no java required.
• Can load NXLucene analyzers.
• Plush is free software distributed under the GNU GPL.
• Plush is written in python and can be easily customized.

Why using Plush instead of Luke ?

• Lucene indexes are on a remote server.
• You are using PyLucene no java available on the server.
• No X server available.
• You want to sort the result list.
• You want to have an idea of the search response time.
• You need to run command line scripts.

For other cases have a look to the great Luke tool.

Examples

$ plush
plush
Welcome to Plush 0.1.0, a Lucene interactive terminal.
Using PyLucene 2.0.0-3 and Lucene 2.0.0-453447.

  Type: \open [INDEX_PATH]    to open a lucene store
        \?                    for help
        \q or ^D              to quit.

plush> \open tmp/PyLucene-2.0.0-3/samples/index
Opening store: /home/ben/tmp/PyLucene-2.0.0-3/samples/index

plush> \d
Directory info
--------------
* Directory path             : /home/ben/tmp/PyLucene-2.0.0-3/samples/index
* Directory current version  : 1168939081493
* Number of docs             : 578 (max doc num: 580)
* Number of fields           : 3
* Index last modified        : 2007-01-30T11:04:46
* Index status               : unlocked
* Has deletions              : YES
* Directory implementation   : org.apache.lucene.store.FSDirectory

plush> \di
Index info
----------
Found 3 fields for doc #579:
* contents                     Stored Indexed Tokenized
* name                         Stored Indexed
* path                         Stored Indexed

plush> foo -bar
Searching: foo -bar
------------------------------------
  Default search index: contents
  Simple analyzer tokens: foo, bar
  Found 3 matching document(s) in 0.002s

* 00) score: 0.56, doc_num: 218
  name: migrating-from-nuxmetadirectories.txt
  path: /home...migrating-from-nuxmetadirectories.txt

* 01) score: 0.40, doc_num: 246
  name: creating-new-content-types.txt
  path: /home...creating-new-content-types.txt

* 02) score: 0.16, doc_num: 184
  name: guidelines-user_interface.txt
  path: /home...guidelines-user_interface.txt

  Found 3 matching document(s) in 0.002s

plush> \v 184
Found 3 fields for doc #184
contents                   S I T    : <snip>file content</snip>
name                       S I      : guidelines-user_interface.txt
path                       S I      : /home...guidelines-user_interface.txt

plush> \v 184 ^name
Found 3 fields for doc #184
name                       S I      : guidelines-user_interface.txt

plush> \set
  limit = 5
  select = name path
  default_field = contents
  sort_on =
  waterline =
  analyzer = Simple

plush> \set sort_on contents
 It is not possible to sort on a tokenized field


plush> \term contents
Top 5 term occurences for field 'contents' matching '*'.
occurences term
---------- ----
       455 http
       455 information
       452 more
       446 directory
       434 copy
---------- ----

plush> \set limit 7
  limit key setted.

plush> \term name \.txt$
Top 7 term occurences for field 'name' matching '\.txt$'.
occurences term
---------- ----
       449 README.txt
        13 DEPENDENCIES.txt
        13 refresh.txt
         6 TODO.txt
         5 INSTALL.txt
         5 LICENSE.txt
         3 COPYING.txt
---------- ----

plush> ^D
Bye

The command history is saved in the ~/.plush_history file.

Using NXLucene analyzers

You can use the NXLucene analyzers by setting an environnement variable NXLUCENE_HOME that point to the path where the analyzers python module can be loaded.

$ export NXLUCENE_HOME=/usr/local/nxlucene/src/nxlucene
$ plush
NXLucene analyzers Loaded.
...
plush> \set analyzer french
Error: Invalid analyzer.
   Possible values are: nxfrench, nxsort, Whitespace, Keyword, nxstandard, German, Stop, French, Standard, Simple, nxurl, nxkeyword
plush> \set analyzer nxfrench
plush> l'œuvre d'art
Searching: l'œuvre d'art
------------------------------------
Default search index: contents
nxfrench analyzer tokens: oeuvr, art
...

NXLucene analyzers are prefixed with nx.