CSV Extension

The csv extension provides convenience primitives for reading data from delimited strings to NetLogo lists, and for converting NetLogo lists to string values.

There are four primitives that operate the same as their equivalents in the CSV extension for desktop NetLogo.

• csv:to-string - convert a "two dimensional" list of lists into a string value.
• csv:from-string - convert a string into a list of lists with values.
• csv:to-row - convert a "one dimensional" list into a string value.
• csv:from-row - convert a string into a list of values.

Differences From Desktop CSV

Because file operations operate differently in a web browser than in a desktop application, teh csv:from-file and csv:to-file primitives are not supported and will throw errors if they are used in NetLogo Web.

You can get a CSV file using the Fetch extension, like this: fetch:user-file-async [ file-contents -> show csv:from-string file-contents ]

You can save a CSV file using the SendTo extension, like this: send-to:file "output.csv" csv:to-string [[0 1 true "hello"][1 10 false "goodbye"][2 5 true "..."]]

Simple Usage

If you're working with whole files with data on separate rows, or you want to generate such a file, you want to-string and from-string:

observer> csv:from-string ""
observer: []
observer> csv:to-string []
observer: ""
observer> csv:from-string "apples,oranges,1,2\ngreen,red,3,4"
observer: [["apples" "oranges" 1 2] ["green" "red" 3 4]]
observer> csv:to-string [[ "apples" "oranges" 1 2 ] ["green" "red" 3 4]]
observer: "apples,oranges,1,2\ngreen,red,3,4"

If you're working one row at a time with single lists, you want to-row or from-row:

observer> csv:from-row ""
observer: [""]
observer> csv:to-row []
observer: ""
observer> csv:from-row "14,10,-2"
observer: [14 10 -2]
observer> csv:to-row [14 10 "delim,here"]
observer: "14,10,\"delim,here\""

Values

Only values that can be easily converted to strings will be correctly handled by to-string and to-row. Simple values like numbers, strings, and true/false should all be correctly converted. Complex values like anonymous procedures or turtles will probably not be well-represented. I recommend "pre-converting" any complex values you want to retain:

observer> create-turtles 3 [ fd 20 set label "hi, how are ya?" ]
observer> csv:to-string (list (sort turtles) )
observer: "(turtle 0),(turtle 1),(turtle 2)" ; not very useful
observer> csv:to-string [ (list who pxcor pycor color label) ] of turtles ; convert each turtle into a list of values we want
observer: "1,-3,-2,85,\"hi, how are ya?\"\n0,-4,2,85,\"hi, how are ya?\"\n2,-5,3,65,\"hi, how are ya?\""

Delimiters

All primitives also accept an optional delimiter argument, which is a string of a single character used to separate individual values. Because the delimiter is optional, when you add one you must surround the whole primitive usage with parenthesis. Examples:

observer> (csv:from-string "1|2|3\n4|5|6" "|")
observer: [[1 2 3] [4 5 6]]
observer> (csv:to-row [14 10 "red,blue" "red"] "|")
observer: "14|10|red,blue|red"