Github user fsparv commented on a diff in the pull request:
https://github.com/apache/lucene-solr/pull/335#discussion_r173638701
--- Diff:
solr/core/src/test/org/apache/solr/cloud/AliasIntegrationTest.java ---
@@ -238,51 +239,52 @@ public void testModifyMetadataV2() throws Exception {
checkFooAndBarMeta(aliasName, zkStateReader);
}
- public void testModifyMetadataV1() throws Exception {
+ @Test
+ public void testModifyPropertiesV1() throws Exception {
// note we don't use TZ in this test, thus it's UTC
final String aliasName = getTestName();
ZkStateReader zkStateReader = createColectionsAndAlias(aliasName);
final String baseUrl =
cluster.getRandomJetty(random()).getBaseUrl().toString();
- HttpGet get = new HttpGet(baseUrl +
"/admin/collections?action=MODIFYALIAS" +
+ HttpGet get = new HttpGet(baseUrl +
"/admin/collections?action=ALIASPROP" +
"&wt=xml" +
"&name=" + aliasName +
- "&metadata.foo=baz" +
- "&metadata.bar=bam");
+ "&properties.foo=baz" +
--- End diff --
I tend to like dot notation to be nounish things. A set of things
collectively know as "properties" with member element ".foo" rather than an
adjective "property" describing ".foo"
"properties" : {
"foo": "baz",
"bar": "bam"
}
vs
"property" : {
"foo": "baz",
"bar": "bam"
}
The v1 api flattening should match what we do for v2 I think?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
