Question

Hi,

I have some scripts which generate some data and update a page. Instead of going through the filesystem (which gives me NO history/RCS protection) I was wondering if a script already exists that would checkout through the web the page and give you a file in your local workspace, and then you can store it back ?

vision:

cd /tmp
TwikiStore -get http://server.company.com/bin/twiki/Main/WebLeftBar > tmp1
echo "news " >> tmp1
cat tmp1 | TwikiStore -put http://server.company.com/bin/twiki/Main/WebLeftBar

Environment

-- GillesEricDescamps - 28 Oct 2004

Answer

Gilles Eric - take a look at building a command set for the TWikiShellContrib - this is where such functionality should go.

-- MartinCleaver - 29 Oct 2004

see also mechanise-test.pl on TWikiTestInfrastructure for another example of how this could be done (here's a snippet of the core part of what would be required:)

    $mech->follow_link( text => 'Edit' );
    $mech->form_name( 'main' );

    my $topicText = $mech->value( 'text' );
    $topicText .= "\n\nadd more text to the topic\n\n-- %MAINWEB%.TWikiTester\n";
    $mech->field( text => $topicText );

    $mech->click_button( value => 'Save' );

-- WillNorris - 29 Oct 2004

Indeed, you could even wrap such CGI access in a TWikiShellContrib Command set, so that you get modularised command line access to parts of TWiki that don't have formal APIs.

-- MartinCleaver - 29 Oct 2004

Yet another way is via WebDAVPlugin.

-- PeterThoeny - 04 Dec 2004