Using the Admin panel
The Admin panel allows you to:
- index your files;
- harvest RDF from them;
- access Solr and Sesame's interface through the Components tab;
- use introspection on Cocoon's processes.
Indexing here refers to the process through which Solr collects information from your EpiDoc xml files. It is needed in order for you to be able to perform searching, create epigraphic indices, and populate facets for faceted search. The first time you add your files to EFES, you should Index all of them through the "Index all (search)" button.
This may take a while, depending on how many files you have placed in your webapps/ROOT/content/xml/epidoc/ folder. Once the process is complete, you will get a Solr index result for all files, notifying you that adding new data from "epidoc" and "indices" was successful.
Each time you add new xml files or change something in the markup of the files that have already been indexed, you must Index again so that the changes are reflected in data that is being collected.
If you have a bigger corpus, it makes sense to only index the files which you know have been changed/added. This can be done by clicking on the "Index" link next to the file in the list of epidoc files:
Harvesting is the process of collecting data from the EpiDoc xml files and the authority lists and storing it into RDF - a representation of knowledge in the form of triple statements. These RDF can be exported and shared between projects, resulting in shared ontologies, authorities etc.
On a smaller scale, harvesting in EFES is needed because it is closely tied up to the values in your facets and some of the indices pages. In order to harvest rdf from your files, you need to create an RDF repository. This only needs to be done once in the beginning of your project, after which you can start harvesting and populating your facets and indices.
Click on the "Harvest all (RDF)" button to harvest the RDF from all files from your project. Just like with indexing, it is possible to harvest individual files as well by clicking on the link next to any given file).
The harvesting fails if one or more xml files contain spaces in their names (e.g. 'doc 1.xml'): remember to avoid spaces inside the filenames or to replace them with underscores or hyphens ('doc1.xml', or 'doc_1.xml', or 'doc-1.xml').
The introspection feature of the Admin panel is extremely useful for visualising certain processes without having to look for them in the code of the sitemaps. It is mainly used for developing and testing new functionality and debugging.
Since Cocoon operates on sitemaps consisting of multiple mapmatches which are referenced on multiple places, tracking down these processes across multiple files can be difficult. The Match for URL functionality allows you to enter any root relevant url and find the mapmatch responsible for it. Let's say we want to see which mapmatch generates the page that has a list of all the inscriptions in our project. We can navigate to this page from the menu bar - we'll be taken to http://127.0.0.1:9999/en/inscriptions/. We need to copy everything that is after the string representing the local host and paste it in the search box.
When we hit return we will receive the Kiln match introspection for the URL /en/inscriptions/, which shows the mapmatch giving the instructions for the creation of the page that lives under this url. If we hover over the value of the id, we can see which sitemap this mapmatch is a part of - in our case, this mapmatch has an id="local-epidoc-index-display" and is within the main.xmap sitemap. We can also see all the components of the mapmatch which builds the page as we see it in the browser, such as templates and xsl files, and track down any potential problems.
Another possible scenario is that we know the id of the mapmatch and we want to quickly look at it in the browser without having to open the sitemap that contains it in the xml editor. This is where the Match by ID function comes handy, as it provides a list of all defined mapmatch id's, which can be displayed upon clicking on them.
You can also see a list of all existing templates in alphabetical orders, with a possibility to display their xslt in the browser.
