Searching within a document allows you to quickly find tokens that match certain criteria. Search queries can currently only by executed within an opened document; there is no functionality to search across multiple documents yet.
To open the search dialog, click on “Search” in the toolbar of the “Edit” tab while you’ve opened a document in CorA. Alternatively, for any token in the editor, open the dropdown menu () and choose “Search for similar…” to open the dialog pre-filled with the annotations of the respective token.
Defining a search query
In the search dialog, you can define conditions that must apply to a token for it to match the search query.
Specify if a token must match all or any (i.e., at least one) of these conditions via the dropdown box at the top.
Add or remove conditions from the list by clicking on or , respectively.
Click the “Reset” button to quickly clear all conditions.
Each condition is made up of a search field, a matching criterion, and (optionally) a value.
Search fields mainly correspond to annotation layers or flags within the document, and should be self-explanatory. Two fields deserve some further explanation:
Token: Searches for modernized tokens in all representations, i.e., “trans”, “utf”, and “ascii”. Alternatively, you can restrict the search to the “trans” representation by choosing “Token (Transcription)” instead.
POS: Searches for the combined part-of-speech and morphology tag; e.g., to find a token with POS tag “VVIMP” and morphology “Sg”, you need to enter “VVIMP.Sg” as the search value.
Matching criteria define the relation between the search field and its value, and should also be mostly self-explanatory. The criterion “matches regex” can be used if you are familiar with regular expressions. CorA relies on the MySQL implementation of regular expressions for this feature, which uses POSIX Extended Regular Expression (ERE) syntax.
The value of the search condition is always case-insensitive. It can be left empty to specifically search for fields with empty or non-empty values.
All search conditions always refer to a single token; there is currently no way to define criteria based on a token’s context, such as properties of a preceding token.
Browsing search results
When you perform a search, a new tab “Search” becomes available, which contains a list of all search results.
At the top of this page, you’ll find a summary of the search, including the number of matching tokens and a list of the search criteria for reference. In the toolbar, click the button “Modify search” to bring up the search dialog again.
Search results are presented in a table which looks and behaves almost exactly like the editor table. That means you can modify annotations directly within this list, just like you would in the editor. The only differences to the actual editor are:
You can click on any token or its line number to jump directly to this token in the main editor. This can be used, for example, to see the context of a token in the text.
Rows are more visually separated (to indicate that you’re not viewing a continuous flow of text) and the dropdown menu () is not available.
You can also navigate search results within the editor, without the need to constantly switch back to the “Search” tab, by using the buttons and in the editor’s toolbar. This will jump to the previous or next search result in the list, starting from either the first result, or the one you last clicked on in the “Search” tab.