RE: bibrank error

["Kam-ming Ku" <kmku@xxxxxx>]

I used bibrank -verbose=9  and here are the message:
How can I delete the word ``nih\xe5\x8dn''


2007-04-29 10:59:21 --> ......... unchanged hitlist for ``collin''
2007-04-29 10:59:21 --> ......... unchanged hitlist for ``rosalind''
2007-04-29 10:59:21 --> ......... inserting hitlist for ``nih\xe5\x8dn''
2007-04-29 10:59:21 --> Exception caught: 'ascii' codec can't encode characters in position 3-4: ordinal not in range(128)
  File "/usr/lib/python2.4/site-packages/invenio/bibrank_word_indexer.py", line 1022, in word_index
    wordTable.repair()
  File "/usr/lib/python2.4/site-packages/invenio/bibrank_word_indexer.py", line 835, in repair
    self.put_into_db("emergency")
  File "/usr/lib/python2.4/site-packages/invenio/bibrank_word_indexer.py", line 327, in put_into_db
    self.put_word_into_db(word, self.value[word])
  File "/usr/lib/python2.4/site-packages/invenio/bibrank_word_indexer.py", line 437, in put_word_into_db
    run_sql("INSERT INTO %s (term, hitlist) VALUES ('%s', '%s')" % (self.tablename, escape_string(word), serialize_via_marshal(set)))

 

-----Original Message-----
From: Kam-ming Ku [mailto:kmku@xxxxxx]
Sent: Thursday, April 26, 2007 11:42 PM
To: 'Tibor Simko'
Subject: RE: bibrank error


Bibrabk -R gives:

2007-04-26 23:40:59 --> rnkWORD01F for 93539-93618 is in consistent state
2007-04-26 23:40:59 --> rnkWORD01F adding records #93539-#93618 started
2007-04-26 23:40:59 --> rnkWORD01F adding records #93539-#93618 ended
2007-04-26 23:40:59 --> rnkWORD01F normal wordtable flush started
2007-04-26 23:40:59 --> ...updating 19615 words into rnkWORD01R started
2007-04-26 23:41:03 --> ......processed 1961/19615 words
2007-04-26 23:41:04 --> Exception caught: 'ascii' codec can't encode characters in position 3-4: ordinal not in range(128)


-----Original Message-----
From: Tibor Simko [mailto:tibor.simko@xxxxxxx]
Sent: Thursday, April 26, 2007 10:30 PM
To: kmku@xxxxxxxxxxxxx
Cc: project-cdsware-users@xxxxxxx
Subject: Re: bibrank error

Hello:

On Thu, 26 Apr 2007, Kam-ming Ku wrote:

> I run 'bibrank --check' & 'bibrank --repair' ...  a number of errors!!

Have you run the rank weight recalculation process (bibrank -R)?  This is needed if you add many new records to your existing document corpus; see some musings on the necessity of running this process at the HOWTO Run guide:

   <http://cdsware.cern.ch:8000/admin/howto/run.html>

Best regards
--
Tibor Simko ** CERN Document Server ** <http://cds.cern.ch/>