collective.filemeta 1.0rc1

Background information

When for example MS Office or PDF documents are stored in Plone as normal File content, document metadata embedded in the documents is not used. Such metadata is embedded inside the files in various type-specific formats, and might include information such as title, description and keywords, or EXIF/IPTC metadata for images. Nor does Plone provide a screenshot or cover page thumbnail of the document.

This packages aims to provide those missing features in a generic, pluggable manner.

Note that in addition to this package, you need additional packages to provide the actual file type - specific metadata extraction, such as:

• Products.OpenXml for MS Office document support

• collective.pdfdocument for PDF support

Features provided

• Generic pluggable mechanism to retrieve metadata from different file types, update content with it, and even round-trip the metadata back to the file from Plone content, if needed

• Behavior to toggle the mechanism by content type

• An example optional ‘Document File’ Dexterity content type that gets its metadata and cover image automatically copied over from the uploaded document file

• Optionally enhanced built-in File content type so that it gets it metadata updated from uploaded file (including cover image)

• A nicer default view (used both for the example type & enhanced File)

Features not provided

• viewing the document contents

• indexing of documents (would be a good fit though)

• asynchronous operation

Note on metadata and cover images

Usually people don’t bother with document metadata so prior to uploading, you should check the document properties.

For MS Office docs, make sure that the “store preview” option is selected, before saving the document. PDFs have no embedded cover image; for them, collective.pdfdocument converts the first page of the PDF into a PNG cover image.

Implementation notes

Metadata is processed by a subscriber that attemtps to retrieve the metadata using a ‘IFileMetaProvider’ utility whose name matches the primary file field mime type. If the metadata can be extracted from file, it is copied over by another ‘IContentMetaUpdater’ utility whose name matches the metadata label (such as “title”), or metadata standard or format (such as “dc”, for Dublin Core, or “exif”). There’s also a ‘IFileMetaUpdater’ interface for updating the file when someone updates the metadata of the Plone content object.

Installation

Install collective.filemeta by adding it to your buildout:

[buildout]

...

eggs =
    collective.filemeta

and then running bin/buildout. Remember that you will also need some additional packages that provide the file type -specific extraction of document file metadata. For MS Office OpenXML and PDF support, the buildout would have:

[buildout]

...

eggs =
    collective.filemeta
    Products.OpenXml
    collective.pdfdocument

Contribute

• Issue Tracker: https://github.com/collective/collective.filemeta/issues

• Source Code: https://github.com/collective/collective.filemeta

Support

If you are having issues, please submit them to tracker or contact the author.

License

The project is licensed under the GPLv2.

1.0rc0 (unreleased)

• Don’t break when behavior is set on content type without image field [petri]

• Other minor fixes [petri]

1.0a1 (unreleased)

• Initial release. [petri]