Deserializing a set of fields as a vector

[OmnipotentEntity]

I apologize if this is a somewhat basic question, but is there an easy way to direct rust-csv or serde from rust-csv to deserialize a set of fields as a Vec?

Example:

field1,field2,list
a,b,1,2,3,4,5,6,7
c,d,8,9,10,11,12,13,14

I'd like to deserialize this to a:

struct mydata {
    field1 : String,
    field2: String,
    list: Vec<u32>,
}

[BurntSushi]

@OmnipotentEntity Could you please describe what you actually tried, and include what happened and what you expected or wanted to happen? Without that information, it is very hard to help you. Moreover, have you read the docs for automatic deserialization, and if so, could you say what part of it was unclear so that it can be fixed?

As far as I can tell, your request "just works," but I'm not quite understanding where you ran into a problem. I'd like to understand that so that it can be fixed.

Here's an example:

extern crate csv;
extern crate serde;
#[macro_use]
extern crate serde_derive;

#[derive(Clone, Debug, Deserialize, Serialize)]
struct MyRecord {
    field1 : String,
    field2: String,
    list: Vec<u32>,
}

fn main() -> Result<(), Box<::std::error::Error>> {
    let data = "\
field1,field2,list
a,b,1,2,3,4,5,6,7
c,d,8,9,10,11,12,13,14
";

    let mut rdr = csv::ReaderBuilder::new()
        .flexible(true)
        .from_reader(data.as_bytes());
    for result in rdr.deserialize() {
        let record: MyRecord = result?;
        println!("{:?}", record);
    }
    Ok(())
}

[OmnipotentEntity]

I do apologize, it does just work, but I had to tell the reader to ignore the header. I didn't see flexible documented, and that was my oversight. Thanks for following up.

[BurntSushi]

Great! I'm going to reopen this since it sounds like there is an opportunity to improve the docs here. Where would the best place be too mention the existence of flexible reading? Should we mention it in the docs on the deserialize method?

[OmnipotentEntity]

I'm unsure about where the best place is. I eventually found it by searching the issues on this Github. Documentation on the deserialize method, would be a good place for it? Or perhaps in the "using with Serde" section of the overview?

[forrestthewoods]

I just ran into this issue and spent at least an hour debugging until I found this thread. :(

Where would the best place be too mention the existence of flexible reading?

I would recommend the reader example to include additional comments. I believe, and you can correct me if I'm wrong, that you have to either specify .has_headers(false) or .flexible(true).

Alternatively, I think flexible is slightly wrong. It's overly aggressive.

parsing CSV data will return an error if a record is found with a number of fields different from the number of fields in a previous record.

However it is perfectly acceptable for non-header rows to contain >= fields as the header row IFF the last field is a nested record/Vec. (I'm not sure the correct terminology.) This would also allow non-flexible parse to ensure that a tail Vec field has the same number of values for all data rows. Which is desirable.

[BurntSushi]

I don't see a way to implement your desired flexible semantics. PRs to improve the docs are welcome.