Adding Transcription Content

From UA Libraries Digital Services Planning and Documentation

Revision as of 18:54, 4 April 2012 (edit) Jlderidder (Talk | contribs) (New page: Over time, this will get simpler, and hopefully more automated... that is, unless it becomes a part of Acumen functionality first. But for now...) ← Previous diff		Current revision (20:37, 3 October 2012) (edit) (undo) Jlderidder (Talk | contribs)
(3 intermediate revisions not shown.)
Line 1:		Line 1:
-	Over time, this will get simpler, and hopefully more automated... that is, unless it becomes a part of Acumen functionality first.	+	'''1)''' For a new collection (one not yet in omeka), you need to create a collection in the [http://transcribe.lib.ua.edu/admin/ administrative interface], or in the transcribe.omeka_collections table. Verify the collection id number assigned (auto-incrementally) by the software in this database table. You'll need to enter it as the first variable in the loadDbase script.
-	But for now...	+
		+	For an existing collection, look in the transcribe.omeka_collections table to obtain the id for this collection. This is NOT the same as our collection identifier.
		+
		+
		+
		+	'''''On libcontent1.lib.ua.edu:'''''
		+
		+
		+
		+	'''2)''' change directory to /srv/scripts/transcripts/omeka/
		+
		+	'''3)''' empty the metadata and images directories
		+
		+	'''4)''' run: getContent {collnum} {boxnum}
		+
		+	for example: getContent u0003_0000581 687
		+	This will collect the _2048 sized images and item-level metadata from /srv/www/htdocs/content for the identified content.
		+
		+	Images go into the ./images directory, and metadata in the ./metadata directory.
		+
		+	Note that if the box number in the metadata is incorrect, that item will not be selected.
		+
		+	'''5)''' run: remove2048
		+
		+	This renames the images in the ./images directory to take the "_2048" out of the file name
		+
		+	'''6)''' run: makeDer
		+
		+	This creates derivatives and dumps them into the respective directories in /srv/www/transcribe/archive:
		+	# files/ (original image)
		+	# fullsize/ (800 pixels on the long side)
		+	# square_thumbnails/ (200 pixels on each side)
		+	# thumbnails/ (200 pixels on the long side)
		+
		+	'''7)''' run: `loadDbase num` where `num` is the collection id number assigned by the database in step 1. Later versions of this software should pull the collection names and IDs from the database and give you a selection from which to choose.
		+
		+	In short, loadDbase pulls metadata out of the MODS files and creates the entries in the Omeka database. More detailed information is available in /srv/scripts/transcripts/omeka/dothis
		+
		+	'''8)''' modify the header of documentIt to identify the collections and boxes added. Then run it: `documentIt` to add the record of what we're transcribing to the InfoTrack database.
		+
		+
		+	[[User:Jlderidder|Jlderidder]] 14:17, 4 April 2012 (CDT)

---

Current revision

1) For a new collection (one not yet in omeka), you need to create a collection in the administrative interface, or in the transcribe.omeka_collections table. Verify the collection id number assigned (auto-incrementally) by the software in this database table. You'll need to enter it as the first variable in the loadDbase script.

For an existing collection, look in the transcribe.omeka_collections table to obtain the id for this collection. This is NOT the same as our collection identifier.

On libcontent1.lib.ua.edu:

2) change directory to /srv/scripts/transcripts/omeka/

3) empty the metadata and images directories

4) run: getContent {collnum} {boxnum}

for example: getContent u0003_0000581 687 This will collect the _2048 sized images and item-level metadata from /srv/www/htdocs/content for the identified content.

Images go into the ./images directory, and metadata in the ./metadata directory.

Note that if the box number in the metadata is incorrect, that item will not be selected.

5) run: remove2048

This renames the images in the ./images directory to take the "_2048" out of the file name

6) run: makeDer

This creates derivatives and dumps them into the respective directories in /srv/www/transcribe/archive:

1. files/ (original image)
2. fullsize/ (800 pixels on the long side)
3. square_thumbnails/ (200 pixels on each side)
4. thumbnails/ (200 pixels on the long side)

7) run: `loadDbase num` where `num` is the collection id number assigned by the database in step 1. Later versions of this software should pull the collection names and IDs from the database and give you a selection from which to choose.

In short, loadDbase pulls metadata out of the MODS files and creates the entries in the Omeka database. More detailed information is available in /srv/scripts/transcripts/omeka/dothis

8) modify the header of documentIt to identify the collections and boxes added. Then run it: `documentIt` to add the record of what we're transcribing to the InfoTrack database.

Jlderidder 14:17, 4 April 2012 (CDT)