Process data from .csv files

88 views
Skip to first unread message

mh2

unread,
Jan 28, 2019, 6:11:34 PM1/28/19
to Daikon discuss
Hello everyone,

I am using Daikon (installed on Ubuntu 16.04 virtualbox) to process .csv files. The daikon was installed successfully, and it can process .c or .java files successfully. But I want to use it to process .csv files. I've installed Text:CSV and have the check arguments checkargs.pm in the same folder with my .csv file. Then I run
convertcsv.pl sample1.csv 
It reported the following error information:
at /home/mh/Documents/daikon/daikon-5.7.2/scripts/convertcsv.pl line 380, <CSVHANDLE> line 1.

The generated .dtrace and .decl files were empty.

I also tried other .csv files like airtravel.csv, it also has the following error information:
Corrupted csv file; cannot parse variable names: "Month", "1958", "1959", "1960" at /home/mh/Documents/daikon/daikon-5.7.2/scripts/convertcsv.pl line 380, <CSVHANDLE> line 1.

Could anyone tell me what could the problem for this? How can I make Daikon work for .csv files?

Thanks!




sample1.csv
airtravel.csv

Fabrice Harel-Canada

unread,
Apr 22, 2020, 10:40:32 AM4/22/20
to Daikon discuss
Hi mh2,

I was able to successfully generate invariants for the Iris dataset using the attached notes. 

The two immediate things I'd point out is 1) making sure your checkargs.pm is in the right place, and for me, that was not by including it in the same file as the CSV and 2) make sure your paths are correct, it looks like you're doing everything from the same directory. 

Good luck!

Daikon version 5.8.0, released April 14, 2020; http://plse.cs.washington.edu/daikon.
(read 1 decls file)                                                            .
Processing trace data; reading 1 dtrace file:                                  
[10:50:51 PM]: Finished reading iris.dtrace                                    
===========================================================================    
aprogram.point:::POINT
species one of { "setosa", "versicolor", "virginica" }
sepal_length > sepal_width
sepal_length > petal_length
sepal_length > petal_width
sepal_width != petal_length
sepal_width > petal_width
petal_length > petal_width
Exiting Daikon.

Regards,
Fabrice
daikon_notes.txt
iris.csv
Reply all
Reply to author
Forward
0 new messages