[
https://issues.apache.org/jira/browse/ORC-226?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16146203#comment-16146203
]
ASF GitHub Bot commented on ORC-226:
------------------------------------
Github user omalley commented on a diff in the pull request:
https://github.com/apache/orc/pull/151#discussion_r135922340
--- Diff: c++/include/orc/Reader.hh ---
@@ -288,6 +288,17 @@ namespace orc {
virtual uint64_t getCompressionSize() const = 0;
/**
+ * Get ID of writer that generated the file.
+ * Current availiable Orc writers:
+ * 0 = ORC Java
+ * 1 = ORC C++
+ * 2 = Presto
+ * @param id out parameter for writer id
+ * @return true if writer id is availiable, false if otherwise
+ */
+ virtual bool getWriterId(uint32_t & id) const = 0;
--- End diff --
Actually, we do have the error stream, but we don't need to use it. I'd
prefer not to use side effects in the parameters, because it is easy to misread
in the code.
For the code and most users, the important part is which writer it is if
known. Only diagnostic tools will care about the integer value of unknown ones.
How about:
enum WriterId { ORC_JAVA_WRITER, ORC_CPP_WRITER, PRESTO_WRITER,
UNKNOWN_WRITER}
class Reader {
WriterId getWriterId();
/**
* For unknown writer ids, get the value.
*/
int getUnknownWriterIdValue();
> Support getWriterId in c++ reader interface
> -------------------------------------------
>
> Key: ORC-226
> URL: https://issues.apache.org/jira/browse/ORC-226
> Project: ORC
> Issue Type: Sub-task
> Components: C++
> Reporter: Xiening Dai
>
> We just added writer ID to identify files generated by different writers (we
> have three currently). Need an interface for reader to get this ID back.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
