[
https://issues.apache.org/jira/browse/CSV-112?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13989018#comment-13989018
]
Benedikt Ritter commented on CSV-112:
-------------------------------------
The problem is, that we provide a key based access to the values of a
{{CSVRecord}} using the {{get(String)}} method. How should that method behave
if there are duplicate column names?
> HeaderMap inconsistent when duplicate columns names
> ---------------------------------------------------
>
> Key: CSV-112
> URL: https://issues.apache.org/jira/browse/CSV-112
> Project: Commons CSV
> Issue Type: Bug
> Components: Parser
> Affects Versions: 1.0
> Reporter: Romain Gossé
> Labels: headers, parsing
>
> Given a parser format for csv files with a header line:
> {code}
> CSVFormat myFormat =
> CSVFormat.RFC4180.withDelimiter(",").withQuoteChar('"').withQuotePolicy(Quote.MINIMAL)
>
> .withIgnoreSurroundingSpaces(true).withHeader().withSkipHeaderRecord(true);
> {code}
> And given a file with duplicate header names:
>
> Col1,Col2,Col2,Col3,Col4
> 1,2,3,4,5
> 4,5,6,7,8
> The HeaderMap returned by the parser misses an entry because of the Column
> name being used as a key, leading to wrong behavior when we rely on it.
> If this is not supposed to happen in the file regarding the CSV format, at
> least this should raise an error. If not we should come up with a more clever
> way to store and access the headers.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
