How to Upload Raw Datasets

For the datafile, you can upload a .zip file, an xml-based file or a plain-text file. If you upload a single plain-text file instead, the system will assume that there is one annotation separated by a blank line each and process the file as such. If you upload a .zip file, the system will assume that it contains a number of plain-text files within, and process each of the text files within the .zip file as individual code-able units. You can, however, use macros or find and replace in Microsoft Word to insert a special delimiter ==--endcodeableunit--== in your raw data so that you can upload a zip archive and still code at the paragraph rather than whole document level. The delimiter has to be on a line all by itself - that is, you need to have:

  <the text><hard return>
  ==--endcodeableunit--==<hard return>
  <next text...>

If you upload an xml file (with a .xml extension), the system will verify that it conforms to the correct schema definition (which can be downloaded here) and process the file as such. A sample XML document and tips on using the XML upload functionality can be found here.

For the code file, you can choose to upload a plain-text file here with one code per line (up to a maxiumum of 255 codes) in the format of "code|description" (without the quotes) that will define one code per line. You can also add an additional optional "|keycode" parameter to the end to define the keystroke to use for that code (note that this should only be one character in length - if the keycode field is larger than one character, only the first character will be assigned). Also note that if you assign the same keycode to multiple items, only the first item assigned will retain that keycode. If you do not upload a code file, you will have to opportunity to define the codes on the next screen as well as assign the keyboard keys associated with each code.

Furthermore - if you are using the XML upload, any options set within the uploaded XML datafile will override any options set here. This option will also override any options for any codes you may have set either in the XML datafile or via the code file upload here, and will also override the flag for allowing multiple codes.

If you choose to allow users to define their own codes, you can upload or set your own suggested codes to use codes.

Also, if you do not assign a dataset name, it will take on the name of the data file uploaded (without any file extension).