[
https://issues.apache.org/jira/browse/IOTDB-750?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17128025#comment-17128025
]
Julian Feinauer commented on IOTDB-750:
---------------------------------------
The resultData type or encoding is conflicting....
I am puzzled as e.g. there is the test
{code:java}
@Test
public void testAddLeftNodePath() throws MetadataException {
MTree root = new MTree();
root.setStorageGroup("root.laptop");
try {
root.createTimeseries("root.laptop.d1.s1", TSDataType.INT32,
TSEncoding.RLE,
TSFileDescriptor.getInstance().getConfig().getCompressor(),
Collections.EMPTY_MAP, null);
} catch (MetadataException e) {
e.printStackTrace();
fail(e.getMessage());
}
boolean hasException = false;
try {
root.createTimeseries("root.laptop.d1.s1.b", TSDataType.INT32,
TSEncoding.RLE,
TSFileDescriptor.getInstance().getConfig().getCompressor(),
Collections.EMPTY_MAP, null);
} catch (MetadataException e) {
hasException = true;
}
Assert.assertTrue(hasException);
}
{code}
in MTreeTest in the commit which should fail as this should be allowed, right?
> Allow Sub-Devices in IoTDB Server
> ---------------------------------
>
> Key: IOTDB-750
> URL: https://issues.apache.org/jira/browse/IOTDB-750
> Project: Apache IoTDB
> Issue Type: New Feature
> Components: Core/Engine
> Reporter: Julian Feinauer
> Assignee: Julian Feinauer
> Priority: Major
> Labels: pull-request-available
> Attachments: image-2020-06-08-11-56-45-575.png,
> image-2020-06-08-11-57-00-833.png
>
>
> Currently, a Node is considered a Device if it has a measurement. Then it is
> not possible that this devices has a "subpath" which again is a device but
> only measurements.
>
> This should be changed so that a Device can also have "subdevices".
>
> Example:
> {code:java}
> CREATE TIMESERIES root.sg1.d1.s1 with datatype=INT32,encoding=RLE;
> CREATE TIMESERIES root.sg1.d1.d2.s1 with datatype=INT32,encoding=RLE;
> CREATE TIMESERIES root.sg1.d1.d2.s2 with datatype=INT32,encoding=RLE;
> {code}
> Currently the first command succeeds but the second and third fail.
> But they should also be possible.
>
> Note: Example above is from [~jixuan1989]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)