Closed ethanwhite closed 7 years ago
I suggest writing to the authors http://esapubs.org/archive/ecol/E095/178/#authors and informing them about the errors in the data.
The authors are using tab delimiters and end of line as "\t\t\t" making it hard to handle the data because they also use \n
to mark a continuing line as shown in the fig below.
When the retriever tries to detect for incomplete lines, it will fail in this case since the line has an \n
and has the correct required number of fields(2) yet the second field is incomplete
'Ref_34', 'Emmons L.H. & F. Feer. 1997. Neotropical Rain Forest Mammals, a Field Guide. 2da ed. The \n'
Ah, OK. So was this failing on Python 2 as well? I thought it was working there.
On Python 2, It does not fail but it install the wrong data. We need to add a check for such errors. Sample installed using able data
Installed rows / do not match the total number of rows in the raw data
Fixed by #740.