|
|
|
* ?
Using wildcards |
If a search does not work . . .
Finding strings in returns
Exporting returns as a HTML file
The internal editor
As you will have seen in the section How it works, Corpus Presenter has a variety of options for searching texts. The most basic of these is the Simple find option which can be activated on the main level of the program once a corpus or some text files have been loaded.
The Simple find option can be used to search in the current loaded text or it can span a branch of a tree or comb through texts starting from the current one to the end of the tree. To search in more than one text, click on the tree on the right (or list of files if you do not have a tree display).
After a search run has been completed and some returns have been reported you can click on the button Storage which will open the text box for holding returns temporarily. The returns have been deposited in the clipboard and you can paste the contents via Ctrl-V or Paste in the right click mouse menu. As the returns are in the Windows clipboard you can also retrieve them into any other program you may have loaded, such as your own word processor.
The Quick search function allows you to search for a string in one or more texts which are contained in a data set. Obviously, only searches through text files are permissible (if your corpus contains images or sound files, then these are ignored, databases can be searched through separately). This option initiates a dialogue in which you specify what string you wish to look for. This window also allows you to specify the range of a search, to gather finds in a list and store these to disk.
In addition to a straight search, this option allows you to store the finds in a list which can then be consulted after a search (it is retained in memory until a fresh search is made). Returns can also be stored to disk, the default name Quick_Search.lst is suggested, though you can use your own of course. You can re-load this list at some later point and, assuming that you are use the same corpus or set of files with which you generated the list in the first place, you can jump to the location in a text where the find was made by just double-clicking on a row in the list of finds.
You can enter a search string directly or use an input list just as with word lists and complex retrieval tasks. Returns can be collected if required and can be then transferred to the line grid on the retrieval level if necessary. You can select all items or only a subset in the list of finds. With the latter you decide what items to select by either pressing the Shift key and then either the Up or Down Arrow key (to mark contiguous items) or by pressing the Ctrl key and then either the Up or Down Arrow key (to mark non-contiguous items).
Pressing the Escape key will cancel the location procedure and show you the number of finds which have been made up to the point of interruption.
The wildcards * and ?, which you will know from other software, can be used here to increase the flexibility of searches. The question mark stands for a single unspecified character while the asterisk stands for one or more than one of these.
A space, tab stop, page break, carriage return or line feed cannot be an unspecified character, and hence cannot be captured by an asterisk. The entry ‘h*r’ will return ‘harder’ (if present in the text(s) scanned) but not ‘had their’ as there is a space between ‘h’ and ‘r’. You can, of course, search for just ‘had their’.
Alternatively, on the advanced search level, you could search for ‘had’ and ‘their’, or better still, search for a series of verb forms (one of which would be ‘had’) and a series of possessive pronouns (one of which would be ‘their’).