NAME

       dwww-cache - manage the dwww cache of converted documents

SYNOPSIS

       dwww-cache --lookup type location
       dwww-cache --store type location
       dwww-cache --list type location
       dwww-cache --list-all
       dwww-cache --clean

DESCRIPTION

       dwww-cache  manages  the  cache  of converted documents; it is part of dwww(7).  An option
       (the first argument) specifies the operation.  Some operations operate on the whole cache,
       and  need  no  other  arguments.  Others need the second and third argument to specify the
       type and pathname of the original document.  (See  dwww-convert(8)  for  more  information
       about the arguments.)

       The operations are:

       --lookup
              If  the  document  is  in  the cache, output it to the standard output.  Otherwise,
              output nothing and return a non-zero status.

       --store
              Read the document from the standard input, output it to  the  standard  output  and
              store  it  into  the  cache.   If  there was an older version already in the cache,
              remove it.

       --list Output information about one document in the cache to  the  standard  output.   The
              information  is  one  line,  with  four  space  delimited fields: type, pathname of
              original, pathname of cached, and permanent flag (y or n).

       --list-all
              Like --list, but for all documents.

       --clean
              Forget all cached documents that have been deleted.

       dwww-cache maintains a database with information of the cached documents.   The  documents
       themselves  are  stored  in separate files in the cache directory.  When old documents are
       removed from the cache by a suitable crontab entry, they will still exist in the database.
       This  operation  removes  all  entries  from the database where either the original or the
       converted file is now missing.

       To stop the cache from growing too large, an entry in root's crontab should remove  cached
       files that have not been accessed for a while.  For example, the following commands remove
       all documents that have not been accessed for ten days:

              find /var/cache/dwww -atime +10 | xargs rm -f
              dwww-cache --clean

       The idea is to first delete the old cached files and then  clean  up  the  database.   The
       policy  of  cleaning  the  cache  has  been kept outside of dwww-cache to keep the program
       simple, and to allow  maximum  flexibility.   The  default  dwww  installation  creates  a
       /etc/cron.daily/dwww, which automatically cleans the cache each day.

FILES

       /var/cache/dwww/db
              The cache directory.

       /var/cache/dwww/db/.cache_db
              The  database  with information about all cached documents.  This is a binary file,
              specially encoded for small size and high speed,  and  should  not  be  touched  by
              anyone but dwww-cache.

SEE ALSO

       dwww(7), dwww-convert(8).

AUTHOR

       Lars Wirzenius <liw@iki.fi>.  Modified by Robert Luberda <robert@debian.org>.
       See dwww(7) for copyrights and stuff.