During a text mining workflow, I am facing the following problem: I close the loop cycles, start new ones and try to merge tables with a joiner node, but for every variation I get the error message “Can’t merge FlowVariable Stacks! (likely a loop problem.)”. This is the same phenomenon when I want to use the Concatenate node. No matter what Loop end I try to close, or if I use multiple loop end nodes, the process always gets stuck.
Hey there and welcome to the forum!
The issue stems from the attempt to concat tables that come from two parallel loops… that won’t work… which I’m afraid also means more information is needed to help you out.
Can you maybe explain what you are trying to do / share your workflow with dummy data?
I am processing text documents. I need to find ‘name + value’ pairs in them. I store the names in a reference table and look up the values for them. I manage the results per document. The “name+value” pairs belonging to the same document are identified by an iteration identifier (iteration identifier: 0 -123). Basically, I solved the problem: I have 124 tables, which contain the “name+value” pairs from each document. I write each table to separate worksheets in an Excel table. Since the names of the processed documents are too long (name+path), I cannot use this as the name of the Excel worksheets. The idea was to append the document names at the end of the process after the last non-empty row of the tables. At this point in the workflow, I have a table with 844 rows/2 columns and a table with 124 rows/1 column. The latter contains the names of the documents. Based on the iteration identifiers, I start a group loop on table 844/2, and a chunk loop on table 124/1, and combine the rows in a Concatenate node (in case of identical column names, the node inserts the name into the corresponding column after the last non-empty row). There may be problems with closing the loops, handling variable flow variables… and even what. I tried filtering the variables to handle the variables, I tried several types of loop end nodes, nothing has worked so far.
*: In the attached figure, after a simple row filtering, I pass a row from the rows of document names, so the process works without any further ado.
Can someone help???
hey there,
I read that you solved the problem that’s why I did not respond… what’s the challenge you are now facing? Can you share an example workflow rather than screenshots?
Hi @Zaharovka.
As @MartinDDDD said, you cannot concatenate two group start like that.
The workflow currently looks like this. I have attached a directory with 3 test files. The workflow currently looks like this. I have attached a directory with 3 test files. Originally the files are in a directory called “csakufd”.
something_new.knwf (534.7 KB)
TaurusStar001.txt (1004 Bytes)
TaurusStar002.txt (1004 Bytes)
TaurusStar003.txt (1004 Bytes)
Hi @Zaharovka
If I understood the requirements, please see if this is what you intend to do.
something_new2.knwf (566.1 KB)
Br
Instead of the Value column header of the table you sent, you should have the file name, which also includes a path, so that the table can be identified.
Yes, this is a proper form. Thank you for your help and patience.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.
