Demo
Open your browser and navigate to http://localhost:8080/kitodo
after OCR-D modules and Kitodo are started.
Enter the username testadmin
and the password test
in the login dialog.
(Unless, of course, you are using an external Kitodo instance,
or have customized the data generated
from make prepare-examples
.)
Note: If it is the first launch of Kitodo.Production, then the
Indexing
tab of the system page is displayed, because indexing still needs to be done. To perform the indexing, click on the buttonCreate ElasticSearch mapping
. After the mapping is created, click on the buttonStart indexing
next to theWhole index
label. After a few seconds, the index is created and you can navigate to the dashboard by clicking on the Kitodo.Production logo.
Kitodo extensions
After the steps for installation of extra resources into Kitodo,
specifically the subtarget make prepare-data
, and starting up Kitodo.Production,
a number of extensions will become available:
- a new script task for OCR processing from the process directory
- a new script task for OCR processing from the export directory
- a new example workflow which includes both of them
You can find the latter under Projects
-> Workflows
-> OCR_Workflow
.
In this Kitodo workflow, the script tasks OCR from Process Dir
(before manual Structure Editor)
and OCR from Export Dir
(after DMS Export) have been added to trigger automatic OCR processing with OCR-D.
Moreover, if during installation the subtarget make prepare-examples
has been run, then
an example user, project and process are present in the database. (That process uses the new workflow,
but is stopped just prior to the first OCR.)
Execute script task "OCR from Process Dir"
This script task executes the script script_ocr_process_dir.sh
from the scripts
directory
and passes the selected process id
and the current task id
as parameters.
To execute this script task manually, navigate from the dashboard to All processes
by clicking
on the button in the Processes widget, or use the URL http://localhost:8080/kitodo/pages/processes.jsf?tabIndex=0.
Select a process to execute the script task, and click on Possible actions
and then on Execute KitodoScript
.
Type the following text into the form field:
action:runscript "stepname:OCR from Process Dir" "script:OCR Process Dir"
Then click on Execute KitodoScript
.
(This will run the simplistic Tesseract-based default workflow
asynchronously. The process status will change as soon as the job
is finished.)
Watch docker logs
, or look under the hood with the Monitor.
Execute script task "OCR from Export Dir"
This script task executes the script script_ocr_export_dir.sh
from the scripts
directory and passes the selected process title
and the current task id
as parameters.
The automatic script task is bound to the manual task Export DMS
in the workflow, because it needs the METS from the export directory that is created via the DMS export.
To export the METS of a process manually, navigate from the dashboard to All processes
by clicking on the button in the Processes widget, or use the URL http://localhost:8080/kitodo/pages/processes.jsf?tabIndex=0.
Select a process to export, click on Possible actions
and then on Export DMS
. After a successful export,
the METS is located in the export directory /usr/local/kitodo/dms-export/
in a subdirectory named by
the selected process.
Note: The export directory can be changed under the project settings and can only be found in our sample project under this path
/usr/local/kitodo/dms-export/
. After that the status of the task can be set toCompleted
and our script taskOCR from Export Dir
will be executed automatically.
Equivalently, to execute the script task manually after the METS has been exported (in some other workflow),
enter an Execute KitodoScript
action in the same way as above, but with the text:
action:runscript "stepname:OCR from Export Dir" "script:OCR Export Dir"
The
script_ocr_export_dir.sh
is still work in process, and does not perform any actual processing of the METS at the moment. (It only prints a message which parameters it was called with.)