EPrints Documentation - User contributions [en-gb]

Include Javascript in the workflow

2015-06-17T20:58:43Z

Pm705@zepler.net:

Sometimes you need to augment an EPrints workflow in a specific way. One of the ways you can do this is using javascript but it is not immediately clear how to do that. There are several options available which may suit what you need to do.

==Pure Javascript==

This should be the prefered method although it is not always straight forward to use depending what your trying to do. EPrints stores javascript files in <code><eprints_root>/archives/<archive_id>/cfg/static/javascript/auto/</code> and it EPrints ships with PrototypeJS. If your workflow problem is specific enough you can simply use
<source lang="javascript">
$$('.any-css-selectors').each( function( matching_element ) {
// your code here
});
</source>

==Custom input renderer==

You can overwrite the input renderer for the field you wish to modify. This gives you access to information used to render the field.

<source lang="perl">
{
'name' => 'courses',
'type' => 'text',
'multiple' => 1,
'text_index' => 1,
'render_input' => sub {
( $field, $repository, $current_value, $dataset, $staff, $hidden_fields, $object, $basename ) = @_;

my $dom = $field->render_input_field_actual( $repository, $current_value, $dataset, $staff, $hidden_fields, $object, $basename );
my $script = $repository->xml->create_element("script", type=>"text/javascript");
$script->appendChild($repository->create_text_node('
// your javascript here '." which can have $perl interlaced "
));
}
},
</source>

==XHTML Component==

You if you are in hurry or prototyping then you can use a HTML Component

<source lang='xml'>

<component type='XHTML'>
<script type='text/javascript'>
//your javascript goes here
</script>
</component>

</source>

The downside of this approach is it leaves a tiny bit of workflow component floating in your workflow so you probably wouldn't use it in production.

[[Category:Howto]]

Include Javascript in the workflow

2015-06-17T20:54:02Z

Pm705@zepler.net: /* Custom input renderer */

Sometimes you need to augment an EPrints workflow in a specific way. One of the ways you can do this is using javascript but it is not immediately clear how to do that. There are several options available which may suit what you need to do.

==Pure Javascript==

This should be the prefered method although it is not always straight forward to use depending what your trying to do. EPrints stores javascript files in <eprints_root>/archives/<archive_id>/cfg/static/javascript/auto/ and it EPrints ships with PrototypeJS. If your workflow problem is specific enough you can simply us $$('any-css-selectors').whateveryouneedtodo()

==Custom input renderer==

You can overwrite the input renderer for the field you wish to modify. This gives you access to information used to render the field.

<source lang="perl">
{
'name' => 'courses',
'type' => 'text',
'multiple' => 1,
'text_index' => 1,
'render_input' => sub {
( $field, $repository, $current_value, $dataset, $staff, $hidden_fields, $object, $basename ) = @_;

my $dom = $field->render_input_field_actual( $repository, $current_value, $dataset, $staff, $hidden_fields, $object, $basename );
my $script = $repository->xml->create_element("script", type=>"text/javascript");
$script->appendChild($repository->create_text_node('
// your javascript here '." which can have $perl interlaced "
));
}
},
</source>

==XHTML Component==

You if you are in hurry or prototyping then you can use a HTML Component

<source lang='xml'>

<component type='XHTML'>
<script type='text/javascript'>
//your javascript goes here
</script>
</component>

</source>

The downside of this approach is it leaves a tiny bit of workflow component floating in your workflow so you probably wouldn't use it in production.

[[Category:Howto]]

Include Javascript in the workflow

2015-06-17T20:24:33Z

Pm705@zepler.net: /* Custom input renderer */

Include Javascript in the workflow

2015-06-17T20:20:03Z

Pm705@zepler.net: /* Custom input renderer */

Include Javascript in the workflow

2015-06-17T20:08:05Z

Pm705@zepler.net: /* Pure Javascript */

Include Javascript in the workflow

2015-06-17T19:54:02Z

Pm705@zepler.net:

Sometimes you need to augment an EPrints workflow in a specific way. One of the ways you can do this is using javascript but it is not immediately clear how to do that. There are several options available which may suit what you need to do.

==Pure Javascript==

EPrints stores javascript files in <eprints_root>/archives/<archive_id>/cfg/static/javascript/auto/ and it EPrints ships with PrototypeJS.

[[Category:Howto]]

Include Javascript in the workflow

2015-06-17T19:48:43Z

Pm705@zepler.net: Created page with 'Category:Howto'

[[Category:Howto]]

ReCollect

2013-03-24T22:45:51Z

Pm705@zepler.net: /* Customizing ReCollect */

[[Category:EPrints_3_Plugins]]
ReCollect is a Bazaar package for EPrints which adds extensive data workflow support to EPrints. Fields are compatible with Datacite and INSPIRE. This plugin was created by the University of Essex (http://researchdataessex.posterous.com/) and packaged by the University of Southampton as part of the DataPool project (http://datapool.soton.ac.uk/).

==Customizing ReCollect==
Recollect has a large set of default fields. The aim is to create consistent field names accross the eprints community. You can easilly add or remove fields from the default ReCollect workflow by editing the dataset stage in workflows/eprints/default.xml. If you have suggestions for fields you would like added to ReCollect please email Patrick McSweeney ([mailto:pm5@ecs.soton.ac.uk pm5@ecs.soton.ac.uk])

ReCollect

2013-03-24T22:45:37Z

Pm705@zepler.net: /* Customizing ReCollect */

ReCollect

2013-03-24T22:45:21Z

Pm705@zepler.net:

ReCollect

2013-03-24T22:28:08Z

Pm705@zepler.net: Created page with 'Category:EPrints_3_Plugins'

[[Category:EPrints_3_Plugins]]

EdShareCollectionsInstallation

2010-05-25T15:17:35Z

Pm705@zepler.net: /* Install CGI Scripts */

[[Category:EdShare]]
==Introduction==
The collections plugin is part of [[EdShare]] suite. It is based on code developed in EdShare and LanguageBox. The goal of this project is to produce a package that allows the easy installation of the collections functionality on a vanilla EPrints installation.

'''For more details about this plugin see [[EdShareCollections]].'''

==Requirements==
The EdShare Collections plugin has been tested and confirmed to run on EPrints 3.1.3.

==Installation==

'''As with all package installations it is important to remember to backup your installation of EPrints before continuing.'''

===Extract Package===
The package should be extracted in the appropriate archive directory of your EPrints installation. Run the following commands replacing ''ARCHIVEID'' with the name of the archive you want the collections on, and replace ''VERSION'' with the appropriate version number based on the package you have downloaded;

cd /opt/eprints3/archives/ARCHIVEID
tar xzf collections-VERSION.tar.gz

These commands assume that you installed EPrints to ''/opt/eprints3'' substitute this for something else if you installed it elsewhere. Users of the Ubuntu version of EPrints will find that it is installed by default at ''/usr/share/eprints3''.

===Install CGI Scripts===
Link the collections CGI scripts to the EPrints CGI directory:

ln -s /opt/eprints3/archives/ARCHIVEID/cgi/users/collectionselect/ /opt/eprints3/cgi/users
ln -s /opt/eprints3/archives/ARCHIVEID/cgi/users/bookmarks/ /opt/eprints3/cgi/users
ln -s /opt/eprints3/archives/ARCHIVEID/cgi/bookmarks/ /opt/eprints3/cgi/

===Patch Default Workflow===
Assuming that you have patch installed on you system then you need to run;

patch -p0 < default_collection_workflow.patch

'''This patch will only work if you have not yet modified your default workflow.''' If you have then you will need to add the following to your code inside the ''<flow>'' tags, this modification is what an unmodified default workflow should look like once changed;

<flow>
<stage ref="type"/>
<epc:choose>
<epc:when test="type = 'collection'">
<stage ref="collection"/>
</epc:when>
<epc:otherwise>
<stage ref="files"/>
<stage ref="core"/>
<stage ref="subjects"/>
</epc:otherwise>
</epc:choose>
</flow>

Then at the work flow file still inside the ''<workflow>'' tags you need to add the following;

<stage name="collection">
<component type="Field::Multi">
<title><epc:phrase ref="Plugin/Collection:details_title"/></title>
<field ref="title" required="yes" input_lookup_url="{$config{perl_url}}/users/lookup/title_duplicates" input_lookup_params="eprintid={eprintid}"/>
<field ref="abstract"/>
</component>
<component type="Field::CollectionSelect">
<field ref="relation"/>
</component>
</stage>

===Edit Configuration Files===

====archive/ARCHIVEID/cfg/namedsets/eprint====
Add <code>collection</code> to the ''types for eprint'' list. The position it is in this list will be the position it will appear when choosing a type in the new item workflow.

====archive/ARCHIVEID/cfg/cfg.d/eprint_validate.pl====
In the ''validate_eprint'' function look for the line <code>my @problems = ();</code> and after it add;

return( @problems ) if( $eprint->get_type eq 'collection' );

====archive/ARCHIVEID/cfg/cfg.d/eprint_warnings.pl====
You will need to add an exception here to stop warnings appearing for users when they deposit a collection. A warning will appear because no files have been loaded for the collection but this is fine. If you have not edited this file before look for the following line;

if( @docs == 0 )

Replace it with the following;

if( $eprint->get_type ne 'collection' && @docs == 0 )

This should stop the warning from appearing.

====archives/ARCHIVEID/cfg/citations/eprint/default.xml====
Assuming you haven't modified the default citation you want to look for the first ''<choose>'' block at the top of the citation. And add the following;

<choose>

<when test="type = 'collection'">
<print expr="userid"/>
</when>

</choose>

This will put the user citation for the user that created this collection at the front of the default EPrint citation.

====archives/ARCHIVEID/cfg/cfg.d/collection.pl====
'''OPTIONAL'''

The collections extension provides you with two control boxes that appear on your abstract pages, a bookmark controls box and collection membership box. The collection membership box will not render on the collections page. It is possible to disable these boxes or change where they appear on the page.

To disable either of the boxes look in this file for either of the following lines;

$c->{plugins}->{"Screen::EPrint::Box::BookmarkTools"}->{params}->{disable} = 0;
$c->{plugins}->{"Screen::EPrint::Box::CollectionMembership"}->{params}->{disable} = 0;

If you want to disable either of the boxes then you just need to set the value of this setting to 1 and then regenerate your abstract pages and now the boxes will appear or not according to your preferences.

You can also choose where on the page the boxes appear. Replace the 1000 with undef and the undef with 1000 for the position you want the box to appear. To hide the box use the method above to do so rather then declare all of the positions as undef.

$c->{plugins}->{"Screen::EPrint::Box::BookmarkTools"}->{appears}->{summary_top} = undef;
$c->{plugins}->{"Screen::EPrint::Box::BookmarkTools"}->{appears}->{summary_right} = 1000;
$c->{plugins}->{"Screen::EPrint::Box::BookmarkTools"}->{appears}->{summary_bottom} = undef;
$c->{plugins}->{"Screen::EPrint::Box::BookmarkTools"}->{appears}->{summary_left} = undef;

Again any changes will need to be followed by a regeneration of your abstract pages for them to be visible.

====archives/ARCHIVEID/cfg/cfg.d/z_bookmarks.pl====
'''OPTIONAL'''

This step is not actually required as part of the installation but it is important that you take note of it since there may be some undesired results when using the plugin.

The bookmarks plugin defines <code>$c->{get_users_owned_eprints}</code> in the ''z_bookmarks.pl'' file. If you are installing the plugin on a vanilla install of EPrints then there should not be a problem. But if you have installed other plugins, or implemented this function yourself then you will need to modify the function and comment this function out in ''z_bookmarks.pl''.

====archives/ARCHIVEID/cfg/cfg.d/collection_render.pl====
'''OPTIONAL'''

It is possible to customize the layout of the collection pages by modifying the ''collection_render'' function located in ''archives/ARCHIVEID/cfg/cfg.d/collection_render.pl''. This function is very similar to the ''eprint_render'' function except that certain fields in the EPrint object will not be set. Refer to the default function for more details.

===Restart Web Server===
Once you have completed all of these previous steps, you will need to restart your web server in the usual fashion.

==Bugs==
Please report all bugs to the [http://lslvm-mmr4.ecs.soton.ac.uk/ OneShare Bugtracker]. There is a project setup for OneShare collections called ''OneShare Packages >> Collections''.

MePrintsInstall

2010-05-24T10:28:04Z

Pm705@zepler.net: /* Getting Started */

[[Category:MePrints]]
This page provides technical instructions for installing the MePrints extension. To find out what MePrints is, and for an overview of MePrints features, see [[MePrintsOverview]].

==Installation (EPrints 3.1.3+)==

Download the latest release to your '''local repository directory''' (eg. /opt/eprints3/archives/ARCHIVEID/).

Extract files:

tar xzvf meprints_xx.tgz

===Install bin scripts===

Edit the bin/generate_meprints file and check the include path on the first line. For example if you have installed EPrints in /var/lib/eprints3 change the line from:

#!/usr/bin/perl -w -I/opt/eprints3/perl_lib

to:

#!/usr/bin/perl -w -I/var/lib/eprints3/perl_lib

===Install cgi scripts===

Link the MePrints cgi scripts into the EPrints cgi directory:

ln -s /opt/eprints3/archives/ARCHIVEID/cgi/meprints/ /opt/eprints3/cgi/
ln -s /opt/eprints3/archives/ARCHIVEID/cgi/users/meprints/ /opt/eprints3/cgi/users/

===Update database===

Add the new MePrints user fields (defined in cfg/cfg.d/z_meprints.pl) to your repository database:

cd /opt/eprints3/
bin/epadmin update_database_structure ARCHIVEID --verbose

===Apply MePrints patches===

Apply the MePrints patch files:

cd /opt/eprints3
patch -R -p0 < archives/ARCHIVE_ID/meprints_1.0.patch

DO NOT APPLY THIS FOLLOWING PATCH ON EPrints 3.2, it is not required;

patch -R -p0 < archives/ARCHIVE_ID/meprints_dataset.patch #YOU DONT NEED THIS ON 3.2 trying it might break your install

These patches change two core EPrints modules:

# Update/Views.pm - allow view pages to be generated for any dataset (MePrints adds views over the public_profile_users dataset)
# DataSet.pm - add a new dataset called public_profile_users which is the set of all users who have opted to make their public profile page visible.

==Getting Started==

To activate MePrints, you will need to make some changes to your repository setup.

===cfg/cfg.d/plugins.pl===
You can skip this step on MePrints 1.3 and later.

Add the following lines to your cfg/cfg.d/plugins.pl configuration file: to use the MePrints homepage instead instead of the default EPrints "Profile" page:

$c->{plugins}->{"Screen::User::View"}->{appears}->{key_tools} = undef;
$c->{plugin_alias_map}->{"Screen::User::View"} = "Screen::User::Homepage";

===cfg/cfg.d/user_fields_automatic.pl===

Add the following lines to your cfg/cfg.d/user_fields_automatic.pl configuration file to update the MePrints public profile page automatically when a user changes his or her profile information,

<pre>
$c->{user_fields_automatic} = sub {

my $user = @_;

...

if( defined $user->{changed} )
{
$user->remove_static();
}

};
</pre>

===cfg/cfg.d/eprint_fields_automatic.pl===

Add the following line to your cfg/cfg.d/eprint_fields_automatic.pl configuration file so that the MePrints public profile page automatically changes when an EPrint they have deposited is updated.

$eprint->get_user->remove_static;

===cfg/cfg.d/urls.pl===

To activate the canonical profile URLs, add "/profile/" to the rewrite_exceptions list in your cfg/cfg.d/urls.pl configuration file:

<pre>
$c->{rewrite_exceptions} = [ '/cgi/', '/archive/', '/profile/' ];
</pre>

===cfg/apachevhost.conf===

Add the following directives to your cfg/apachevhost.conf configuration file to enable canonical profile URLs (for example http://myrepository.com/profile/jsmith):

<pre>
<Location "/profile">
SetHandler perl-script
PerlResponseHandler EPrints::Plugin::MePrints::MePrintsHandler
</Location>
</pre>

===cfg/workflows/user/default.xml===

Add the following lines to your cfg/workflows/user/default.xml configuration file:

<stage name="default">
...

<component type="Field::Multi">
<title><epc:phrase ref="user_section_personal" /></title>
...
</component>

<component type="Field::Multi">
<title><epc:phrase ref="user_section_meprints" /></title>
<field ref="profile_visibility" required="yes"/>
<field ref="jobtitle"/>
<field ref="expertise"/>
<field ref="biography"/>
<field ref="qualifications"/>
</component>

===cfg/citations/user/default.xml===

Replace the default EPrints user citation with the MePrints citation (which includes a thumbnail of the user's profile picture) run:

cd /opt/eprints3/archives/ARCHIVEID
cp cfg/citations/user/default_with_thumbnail.xml cfg/citations/user/default.xml

'''Note: if you have already made changes to the default citation, you should examine both files and merge the MePrints changes into default.xml

===Authorship in MePrints===
Meprints 1.1 and later change the definition of ownership in the respository. This means that users who have the same email address as that listed in an items creators field belong to that user once the item enters the archive. This allows users to see statistics about papers they have authored without have to upload their own copy of the paper. You can return this to the eprints default by removing the cfg/cfg.d/authorship_meprints.pl file

===Confirm Permissions===
The last thing to do before using MePrints is to ensure that the permissions are correct on the ''meprints'' directory. To do this run the following commands, there probably will not be any output from running this command;
cd archives/ARCHIVEID
chmod g+wx meprints

If the permissions are not correct then you may encounter errors when using MePrints.

==Testing==

After applying all the changes above, restart Apache to activate MePrints.

===Homepage===

Log into your repository and should see your MePrints Homepage with default widget layout and profile picture.

====Update profile====

Click the Modify Profile button and add some information to the new MePrints fields (Job Title, Expertise, Biography, Qualifications). Set your Profile Visibility to '''Public'''. Click the Save button to return to your Homepage where the information you entered should now be displayed.

====Upload photo====

Click the Profile Picture button, choose a file and click Upload. You will be returned to your Homepage where your picture will now be displayed.

====MePrints Widgets====

By default your homepage displays several widgets (see user_homepage_defaults setting in cfg/cfg.d/z_meprints.pl). Try adding and removing widgets using the widget controls at the bottom of the screen. You can also rearrange widgets by dragging them up/down or between columns. To reset to the default widget layout click the Reset button in the widget control bar.

===Public Profile===

Enter the following URL to view your public profile page:

http://myrepository.com/profile/USERNAME

where USERNAME is your EPrints username. The layout of the MePrints widgets on your profile page is controlled by the user_profile_defaults setting in cfg/cfg.d/z_meprints.pl.

Go back to your MePrints homepage and click Modify Profile. Change some of your profile information (eg. Job Title, Expertise, Biography, Qualifications) and click Save. Return to your public profile page and check that your profile information has been updated.

===MePrints Views===

Run:

bin/generate_views ARCHIVEID --generate menus

Check http://myrepository.com/view/ - you should see 2 new views (you may want to link these into your site template).

Explore the view pages and check that you are listed. Make sure that you can click through from the browse listing to your public profile page.

===MePrints search===

Go to: http://myrepository.com/cgi/meprints/search

Check that you can find your profile using keywords from your user profile information (eg. expertise, qualifications).

Make sure that you can click through from the search result list to your public profile page.

===MePrints Embeddable===

If you want to use the default minimal style supplied for MePrints Embeddable then add the following code to the ''<head>'' section of your page:
<link rel="stylesheet" type="text/css" href="http://myrepository.com/style/meprints_embed.css"/>

====MePrints Embeddable Widgets====
It is possible to embed individual widgets that are viewable on your public profile on your own page. If you go to the public profile page you will see an embed links on the widgets that can be embedded. Click the link and instructions will appear telling you how to embed the widget on your page.

====MePrints Embeddable Profile====
Edit an HTML page and add the following code to the ''<head>'' section of the page:

<script type="text/javascript" src="http://myrepository.com/cgi/meprints/embed_js?username=USERNAME"></script>

Where ''USERNAME'' is your username on the repository.

Then inside the ''<body>'' section of the page you need to add the following code:

<div id="meprints_profile"></div>

This will be the location where your embeddable profile will go on your page.

===MePrints for admins===

Using an administrator account go to the Admin screen and choose '''Search users'''. Enter a username or other criteria to find at least one user account (not the account you are logged in with).

Make sure that you can click through from the results list to the user's homepage.

You should be able to modify the user's profile information by clicking Administer Account.

The widget layout should reflect the user's own preference, and you should not be able to make changes to their widget layout.

====Hide Profile====

Change the user's profile visibility to '''private'''.

Log out, and make sure that:

# the profile is no longer visible at /profile/USERNAME
# the profile no longer appears in search results
# the profile no longer appears in view pages
# the embeddable widget no longer displays the profile

You can also change the default visibility of MePrints profiles in cfg/cfg.d/z_meprints.pl
by default profiles are public unless a user specifically makes there profile private. change
$c->{default_profile_behavior} = 1;
to
$c->{default_profile_behavior} = 0;
to make the default private profiles

==Troubleshooting==

===Database errors===

When dragging and dropping Widgets around, MePrints shows database errors in the error log:

DBD::mysql::st execute failed: Duplicate entry '1-0' for key 1 at /opt/eprints3/perl_lib/EPrints/Database.pm line 1192.
at /opt/eprints3/perl_lib/EPrints/Database.pm line 1192
EPrints::Database::insert('EPrints::Database::mysql=HASH(0x2b593e11ebd0)', 'user_items_fields', 'ARRAY(0x2b593bd00510)', ...

This is a known problem which does not seem to affect widget layout.

===generate_views errors===

====Warnings====

Use of uninitialized value $id in hash element at /usr/share/eprints3/perl_lib/EPrints/Database.pm line 2979.

See http://trac.eprints.org/trac/ticket/3537

====EPrint citations====

After hiding all previously public profiles, eprint citations were rendered in the view list instead of users'. (need to work out how to reproduce this)

====Empty views====

After hiding all previously public profiles and regenerating views, view pages still show expertise/department values.

MePrintsInstall

2010-05-24T09:23:25Z

2009-12-04T10:50:18Z

Pm705@zepler.net: /* Cleaning up */

=Installation =

Begin by downloading PreviewPlus from : SOMEWHERE

== Prerequesit software ==
As prerequisits you will also require FFMPEG, LAME and OpenOffice.
On Ubuntu/Debian you can install these from the package manager.
On RedHat/Fedora you may have to install them by hand depending what repositories you have available.

FFMPEG:[http://ffmpeg.org/index.html http://ffmpeg.org/index.html]

LAME:[http://lame.sourceforge.net/ http://lame.sourceforge.net/]

OpenOffice:[http://www.openoffice.org/ http://www.openoffice.org/]

== Plugin Installation ==
Instructions assume eprints is installed at /opt/eprints3

Extract the tarball into your local archive directory:

cd /opt/eprints3/archives/ARCHIVEID
tar xvzf PreviewPlus.VERSION.tgz

== Configure office conversions ==
Depending what distribution of linux you are using you will find that open office installs in different places. You may need to edit <code>cfg/cfg.d/office_conversions.pl</code> so that <code>$c->{office_path}</code> points to you open office install.

== Symlink CGI scripts ==

ln -s /opt/eprints3/archives/ARCHIVEID/cgi/get_preview_plus /opt/eprints3/cgi/

== Adding the preview to eprint render ==
In eprint render find where you want the preview to appear on the page and add:
if($session->get_repository()->call("previews_exist", $eprint)){
$page->appendChild($session->get_repository()->call("make_preview_plus", $session, $eprint, "horizontal"));
}

== Cleaning up ==

You must restart your webserver.

You will probably also want to run /opt/eprints/bin/epadmin redo_thumbnails ARCHIVEID
but be warned this is a very expensive process if you have a large archive and should definately be run overnight.

== Setting up cron ==
You will probably want to run the /opt/eprints3/archives/ARCHIVEID/bin/generate_video_previews and /opt/eprints3/archives/ARCHIVEID/bin/generate_audio_previews nightly. The easiest way to do this is to add them to the eprints users contab.

Currently there is no mechanism to stop these scripts snowballing so make sure you leave long enough in the crontab for one job to finish before the next starts.

PreviewPlusInstallation

2009-12-04T10:49:58Z

Pm705@zepler.net:

=Installation =

Begin by downloading PreviewPlus from : SOMEWHERE

== Prerequesit software ==
As prerequisits you will also require FFMPEG, LAME and OpenOffice.
On Ubuntu/Debian you can install these from the package manager.
On RedHat/Fedora you may have to install them by hand depending what repositories you have available.

FFMPEG:[http://ffmpeg.org/index.html http://ffmpeg.org/index.html]

LAME:[http://lame.sourceforge.net/ http://lame.sourceforge.net/]

OpenOffice:[http://www.openoffice.org/ http://www.openoffice.org/]

== Plugin Installation ==
Instructions assume eprints is installed at /opt/eprints3

Extract the tarball into your local archive directory:

cd /opt/eprints3/archives/ARCHIVEID
tar xvzf PreviewPlus.VERSION.tgz

== Configure office conversions ==
Depending what distribution of linux you are using you will find that open office installs in different places. You may need to edit <code>cfg/cfg.d/office_conversions.pl</code> so that <code>$c->{office_path}</code> points to you open office install.

== Symlink CGI scripts ==

ln -s /opt/eprints3/archives/ARCHIVEID/cgi/get_preview_plus /opt/eprints3/cgi/

== Adding the preview to eprint render ==
In eprint render find where you want the preview to appear on the page and add:
if($session->get_repository()->call("previews_exist", $eprint)){
$page->appendChild($session->get_repository()->call("make_preview_plus", $session, $eprint, "horizontal"));
}

== Cleaning up ==

You must restart your webserver.

You will probably also want to run /opt/eprints/bin/epadmin redo_thumbnails ARCHIVEID
But be warned this is a very expensive process if you have a large archive and should definately be run overnight.

== Setting up cron ==
You will probably want to run the /opt/eprints3/archives/ARCHIVEID/bin/generate_video_previews and /opt/eprints3/archives/ARCHIVEID/bin/generate_audio_previews nightly. The easiest way to do this is to add them to the eprints users contab.

Currently there is no mechanism to stop these scripts snowballing so make sure you leave long enough in the crontab for one job to finish before the next starts.

PreviewPlusInstallation

2009-12-04T10:45:44Z

Pm705@zepler.net: /* Adding the preview to eprint render */

=Installation =

Begin by downloading PreviewPlus from : SOMEWHERE

== Prerequesit software ==
As prerequisits you will also require FFMPEG, LAME and OpenOffice.
On Ubuntu/Debian you can install these from the package manager.
On RedHat/Fedora you may have to install them by hand depending what repositories you have available.

FFMPEG:[http://ffmpeg.org/index.html http://ffmpeg.org/index.html]

LAME:[http://lame.sourceforge.net/ http://lame.sourceforge.net/]

OpenOffice:[http://www.openoffice.org/ http://www.openoffice.org/]

== Plugin Installation ==
Instructions assume eprints is installed at /opt/eprints3

Extract the tarball into your local archive directory:

cd /opt/eprints3/archives/ARCHIVEID
tar xvzf PreviewPlus.VERSION.tgz

== Configure office conversions ==
Depending what distribution of linux you are using you will find that open office installs in different places. You may need to edit <code>cfg/cfg.d/office_conversions.pl</code> so that <code>$c->{office_path}</code> points to you open office install.

== Symlink CGI scripts ==

ln -s /opt/eprints3/archives/ARCHIVEID/cgi/get_preview_plus /opt/eprints3/cgi/

== Adding the preview to eprint render ==
In eprint render find where you want the preview to appear on the page and add:
if($session->get_repository()->call("previews_exist", $eprint)){
$page->appendChild($session->get_repository()->call("make_preview_plus", $session, $eprint, "horizontal"));
}

== Setting up cron ==
You will probably want to run the /opt/eprints3/archives/ARCHIVEID/bin/generate_video_previews and /opt/eprints3/archives/ARCHIVEID/bin/generate_audio_previews nightly. The easiest way to do this is to add them to the eprints users contab.

Currently there is no mechanism to stop these scripts snowballing so make sure you leave long enough in the crontab for one job to finish before the next starts.