This is a script that processes linguistic data from a PDF file. The PDF file is set to a default that anticipates that utterances will be in three lines like the following:
this is an example DEM.PROX COP INDEF example 'This is an example."
This is adjustable, but all additional headers and footnotes must be removed for data to be processed correctly.
Data is saved in a .csv format that is formatted as follows:
this DEM.PROX this is an example "This is an example." is COP this is an example "This is an example." an INDEF this is an example "This is an example." example example this is an example "This is an example."
It does not matter if there are line breaks between utterances or between an utterance and its gloss or its gloss and its translation.
If you decide to process your PDF outside of R before importing it, I strongly recommend using Adobe Acrobat to properly format PDFs prior to processing. In my experience, other PDF to .doc converters ruin formatting, creating extra labor in order to get the script to work properly.
That being said, in my experience, it's been much easier to manually remove data from R than to search through a PDF or .doc file.