"There have to be at least 1 columns containing DocumentCells !"

[timvancann]

Hi,

I keep getting this warning which does not allow me to continue the process. I get this warning when trying to do the following:

I have a directory which contains several .txt files. I want to read them in seperately via a loop, process each file seperately with a dictionary tagger, sentence extractor and much more and then write a new output file for each iteration.

This is my setup (I tried both):

Flat file Document Parser (out) -> (in) TableRow To Variable Loop Start (var out) -> (var in) Variable to TableRow (out) -> (in) Sentence Extractor (out) **.

OR

File List (out) -> (in) TableRow To Variable Loop Start (var out) -> (var in) Variable Based File Reader (out) -> (in) Sentence Extractor (out) **.

** It doesn't matter which node I use in the TextProcessing part, sentence extracter, dictionary tagger, POS tagger, BAG of words, whatever... It ALWAYS gives the warning: "There have to be at least 1 columns containing DocumentCells !"...

Any thoughts or solutions? Can't finish my project this way :-(.

[gabriel]

I guess, you just need to execute the loop start node in order to configure all downstream nodes. After this step, you should be able to execute all nodes at once using the "Execute All" button. Hope this helps?

[andyg]

Im pretty sure you need to use the "Strings to Document" node to convert the text files into documents.

 

Thanks

[kilian.thiel]

andyg is right, what You need is the "Strings to Document" converter node. The sentence extractor as well as many other nodes of the text processing plugin work on DocumentCells not on StringCell. Simply convert Your strings into documents and You can use all the nodes on them.

Nodes which require columns with DocumentsCells will give the following warnig if there aren't any: "There have to be at least 1 column containing DocumentCells !"