<div dir="ltr">Hi all,<br><br>We are developing an OwnCloud application for use in a University setting (<a href="https://github.com/uws-eresearch/apps/tree/master/crate_it">https://github.com/uws-eresearch/apps/tree/master/crate_it</a> and <a href="https://github.com/uws-eresearch/apps/tree/master/file_previewer">https://github.com/uws-eresearch/apps/tree/master/file_previewer</a>).<br>
<br>One of the functions of our app is being able to show file-previews for any kind of file, not just images. At the moment we are doing this via a back-end service which creates web-ready previews for a variety of document types including office docs (word, powerpoint, spreadsheets) for both the Microsoft and open document formats as well as a variety of other file formats.<br>
<br>We are wondering about the best way to integrate this functionality, and what plans there might be for OC6.<br>* Are there events in the system which are triggered whenever there are new files via the web UI AND via the sync client? We want to be able to generate web-ready views of files as they are added, probably via a message queue<br>
* Is there a standard interface for previews/thumbnails, like the current image viewer that can be registered somewhere so other plugins and the files app can access them? Currently we have registered the actions to view previews with fileactions registry.<br>
* What are the plans for the Lucene search service? It should be possible to combine some services. For example, if you're rendering a Word document into HTML then the rendering process can extract a full-text version of the doc, and also some metadata such as title and author, that could be put in the Lucene index and possibly also the main database.<br>
<br>We'd love to contribute in this area and are wondering how best to get involved.<br><br>Our current application is using this approach:<br><br>* A crude file-watching service crawls the OC data directory directly (TODO: tie this to OC events or use the main DB)<br>
* Each file is placed on a message queue and dealt with by a conversion service which generates HTML and/or image thumbails, down-samples video etc (this uses the Integrated Content Environment (ICE) <a href="https://code.google.com/p/integrated-content-environment/">https://code.google.com/p/integrated-content-environment/</a>)<br>
* The converted assets are kept in a repository - this is based on The Fascinator, a repository platfom used in a number of Australian universities for managing research data: <a href="https://sites.google.com/site/fascinatorhome/">https://sites.google.com/site/fascinatorhome/</a><br>
* Our OC app hooks in to the Files app, so when a user clicks on a file it looks for an HTML version of it in the repository.<br><br>The Fascinator also includes an Apache Solr index, so could provide both full-text and faceted search, but we need to evaluate whether this is necessary, given that OC has Lucene.<div>
<br></div>Dr Peter Sefton<div style>Lloyd Harischandra</div></div>