[jira] [Commented] (CSV-160) CSVParser reading the whole reader even when it is supposed to read record by record

["Gary Gregory (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CSV-160?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14973868#comment-14973868 ] 

Gary Gregory commented on CSV-160:
----------------------------------

You are assuming that the implementation of the parser should behave a certain way when it now behaves in some other way. Once you give the parser a reader, it can do as it sees fit with it. Relying on the state of the reader beyond that is not part of the parser's contract. I would argue that it would handcuff the parser's implementation beyond anything reasonable to expect the reader to point to here or there at any time after the parser takes it over.

> CSVParser reading the whole reader even when it is supposed to read record by record
> ------------------------------------------------------------------------------------
>
>                 Key: CSV-160
>                 URL: https://issues.apache.org/jira/browse/CSV-160
>             Project: Commons CSV
>          Issue Type: Bug
>    Affects Versions: 1.2
>            Reporter: Nitiraj Singh Rathore
>         Attachments: TestCSVParser.java
>
>
> When we create an iterator from CSV parser and then use iterator.next() method for the first time. At this point the reader should be pointing to the second line and if we read the Reader from outside CSVParser then it should give all lines except the first one. But after first next() through iterator the Reader is already pointing to the end of the stream.
> Please find attached the Java main file to test this.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)