Re: Possibility of personalizing OpenOffice documents automatically?

From: Gregory Seidman (gsslist+debian_at_anthropohedron.net)
Date: 05/31/05

  • Next message: Aurélien Campéas: "Re: fortune bug ?" 
    Date: Tue, 31 May 2005 10:03:30 -0400
To: debian-user@lists.debian.org

    On Tue, May 31, 2005 at 12:36:51PM +0200, Herv? Piedvache wrote:
    } On Tuesday 31 May 2005 12:08, Lee Braiden wrote:
    } > On Tuesday 31 May 2005 11:06, herve@elma.fr wrote:
    } > > No I want the operators to not be able to modify the original document
    } > > ... And to have only a main administrator of the documents ...
    } > > It's a kind of contact center, where people manage phone calls and for
    } > > each call they will attribute to the customer a status ... and this
    } > > status will produce a postal mail ... to be generated immediatly or by
    } > > batch at the end of the day ... all the mail will be printed and put in
    } > > wrap to be posted !
    } > >
    } > > It's clearer ? ;o)
    } >
    } > The usual way to do this sort of thing would be with a webserver running on
    } > your intranet, which is linked to a database, and keeps a list of your
    } > customers. Staff can then pull up a webpage, click a button, and have the
    } > server generate reports.
    }
    } Yes ... marvelous ... but with what kind of tool do you manage the production
    } of the mail ?
    } Yes It'll be a web application .. no problem for this ... but I want to manage
    } about 100 mail style *** ... and be able to do the redaction of thoses mail
    } easily ... and then have a system to produce them in file and print them ...
    }
    } So how to do that ?

    An OOo file is a zip file containing XML files. I don't have deep insight
    into the XML schema used, but it is well-documented and with a little
    examination of a sample file I find that it is pretty simple. Importantly,
    simple (same-styled) spans of text in a document show up as actual text
    strings in the XML. It is therefore possible to put in something like
    XXXXX_1_XXXXX in the running text and decompress/find/replace/recompress to
    produce a new file. Note that it is content.xml that you want to modify.

    The following script works for at least a simple case, though I recommend
    rewriting this for performance reasons (note that in this implementation
    replacements should be listed one per line in the replacements file, in
    order, and must not contain the # character):

    #!/bin/sh

    if test $# -ne 3
    then
            echo "Usage: $0 <input SXW file> <output SXW file> <replacements file>"
            exit 1
    fi

    PROCDIR=/tmp/$$
    EDFILE=$PROCDIR/edscr
    export PROCDIR

    mkdir $PROCDIR
    awk '{ printf("1,$s#XXXXX_%d_XXXXX#%s#g\n", NR, $0) }' "$3" > $EDFILE
    echo w >> $EDFILE
    echo q >> $EDFILE
    cp "$1" $PROCDIR/file.sxw
    (
    cd $PROCDIR
    unzip file.sxw content.xml
    ed -s content.xml < edscr
    zip -r file.sxw content.xml
    )
    cp $PROCDIR/file.sxw "$2"
    rm -rf $PROCDIR

    } regards,
    } Herv? Piedvache
    --Greg 

    -- 
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
  • Next message: Aurélien Campéas: "Re: fortune bug ?"