Command-line tool

Command-line interface to the odea toolkit.

This tool takes a single filename (path) as its input, representing a source file within a BagIt archival collection managed by odea, and will update the Bag structure according to the options given.

usage: odea [-h] [--new DIR] [--update] [--derive] [--publish]
            [--filename FILENAME] [--index] [--archive ARCHIVE]
            [--baseurl BASEURL] [--license LICENSE]

The --filename argument is required for --update, --derive, and --publish commands.

Available options are:

--new DIR

initialize a new collection in <DIR>

--update

tag the file and create or update the file and item metadata

--derive

create derivatives

--publish

create an html description page for the corresponding item

--filename FILENAME

file to be processed by update/derive/publish

--index

update the collection html index with information about the corresponding item

--archive ARCHIVE

the name of the archive, for html and bag-info.json

--baseurl BASEURL

the base URL for the archive, for html output

--license LICENSE

license or copyright text for html output

--new

The --new command creates a new directory structure in the given directory. The command will fail if the given path is not a valid directory path. If the --archive flag is given, the archive name will be added to the file bag-info.json on creation.

$ odea --new .
$ tree
.
├── bag-info.json
├── bagit.txt
├── data
│   └── deriv
├── file_metadata
├── html
└── item_metadata

5 directories, 2 files

--update

The --update command will perform the following operations.

  1. “Tag” a new file by adding a uuid string to the filename and the string “SRC” as its format, such that the file is renamed to the format <basename>.SRC.<uuid>.ext. If any of the parent directories contain a uuid in the directory name, the file is considered to be part of a multi- file item (such as a web page with associated media), and will not be renamed.

  2. Create a new json metadata document for the file, if one does not yet exist.

  3. Obtain the sha256 hash, modification time, and size of the file and add these to the json metadata. If no title is present, use the basename of the input filename as the title, replacing underscores with spaces.

  4. Create thumbnail images for the file.

The command requires an input file set by --filename, representing a source item in the payload directory.

--derive

The --derive command will generate derivatives based on a mapping of input filename extension to derive functions. The following list indicates the derivatives that will be generated for each recognized input file type.

list of urls (.url)

  • pf-webarc (.webarc directory)

plain text markup (.md, .txt)

  • df-pandoc-html

ReStructuredText (.rst)

  • df-docutils-html

raster image (.bmp, .gif, .jpg, .jpeg, .png, .psd, .tif, .tiff)

  • df-med-img (.png)

  • df-lg-img (.png)

audio file (.mp3, .wav, .wma, .ogg)

  • pf-wav

  • df-mp3

office document (.odt, .odp, .doc, .docx, .ppt, .pptx)

  • df-pdf-doc

vector image (.eps, .svg)

  • pf-vector (.svg)

  • df-pdf-vector

video (.avi, .flv, .mov, .mpeg, .mp4, .webm, .ogv)

  • df-360p-vp9-400k (.webm)

  • df-video-still (.jpg)

  • df-h264 (.mp4)

The above rules should cover the majority of input file types and use cases. For advanced file processing, custom scripts can be built using the odea python library.

The command requires an input file set by --filename, representing a source item in the payload directory.

--publish

The --publish command will generate an html description page for an item in the archive. The page will be stored in the html directory in the bag root, with the name <uuid>.html corresponding to the item identifier.

The command requires an input file set by --filename, representing a source item in the payload directory.

--index

The --index command will generate an html index for the collection as a whole, using metadata from the file bag-info.json in the collection root.