Database

The Database is the analytical counterpart of the Catalog: one row per tablet, every recorded property available as a column and as a filter. It is designed to build a working corpus — all administrative texts from one scribal school, all tablets from a given findspot with photographs available — and to export it.

1. The table

Each row is a tablet, documented — where the information exists — through its physical characteristics (dimensions), palaeographic features (scribal school, scribe), textual classification (genre, typology, sub-type, language), archival metadata (inventory number, bibliographic reference, publication status) and archaeological context (collection, site, findspot). Two further columns record whether a transliteration and images are available.

An empty cell means the datum has not been recorded, which is not the same as saying it is absent from the object: the corpus is a work in progress and the density of description varies from archive to archive.

Clicking a siglum opens the text. Sigla of tablets without a transliteration are not links.

Sorting: click a column header. Sigla, inventory numbers and other alphanumeric identifiers sort naturally, so 2 precedes 10.

Reordering: click and drag column headers to reorder a table as it sees fit.

2. Simple filter

The database content can be filtered using the global search box in the top-right corner. This filter is case-insensitive and matches your search term against the entire table, including columns that aren't currently displayed. Only rows containing a match are shown.

3. Advanced filters

The Advanced filters button opens a panel on the right of the screen, holding one small box per filterable property: collection, site, language, scribal school, scribe, genre, typology, sub-type, bibliographic reference, duplicates, presence of images, presence of a transliteration, publication status, findspot.

Four things are worth knowing about them:

• Each box lists the values actually attested, with the number of texts beside each one. There are no empty options: if a value is offered, it has results.
• The boxes work as a cascade. Selecting a value in one box recalculates all the others over what is left. A value that disappears has not been hidden — in the combination you have built it has no results. This is what stops you from constructing a filter that returns nothing.
• Several values inside one box widen the result; values in different boxes narrow it. Choosing letter and administrative under Genre returns both kinds; adding Emar under Site returns only the ones from Emar.
• Multiselection. You can also select multiple rows at once by holding Shift or Cmd/Ctrl while clicking.
• The panel does not cover the table. It stays open alongside it, and the table remains readable and clickable while you filter — you can see the effect of each choice as you make it, instead of filtering blind.

The free-text box above the table searches all columns at once, and combines with the filters.

4. Columns and export

Columns shows and hides individual columns. Beyond tidying the screen, this is how you shape a table for export: hide what you do not need, and only the remaining columns are carried over.

Tools collects the operations on the current selection — that is, on the filtered set, not only on the page you are looking at:

5. Editorial columns

Two further columns exist for editorial accounts (Advanced users and Administrators) only, and are hidden by default — enable them from Columns:

• backup — direct access to the text file and to the annotations file kept for each tablet. A greyed-out icon means that file has not been generated.
• edit — the record editing form and the delete control.

They are not available to other accounts, in any form.