[ https://issues.apache.org/jira/browse/ARROW-6398?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
ASF GitHub Bot updated ARROW-6398: ---------------------------------- Labels: dataset pull-request-available (was: dataset) > [C++] consolidate ScanOptions and ScanContext > --------------------------------------------- > > Key: ARROW-6398 > URL: https://issues.apache.org/jira/browse/ARROW-6398 > Project: Apache Arrow > Issue Type: Improvement > Components: C++ > Reporter: Benjamin Kietzman > Assignee: Benjamin Kietzman > Priority: Minor > Labels: dataset, pull-request-available > > Currently ScanOptions has two distinct responsibilities: it contains the data > selector (and eventually projection schema) for the current scan and it > serves as the base class for format specific scan options. > In addition, we have ScanContext which holds the memory pool for the current > scan. > I think these classes should be rearranged as follows: ScanOptions will be > removed and FileScanOptions will be the abstract base class for format > specific scan options. ScanContext will be a concrete struct and contain the > data selector, projection schema, a vector of FileScanOptions, and any other > shared scan state. -- This message was sent by Atlassian Jira (v8.3.2#803003)