Details
Description
If we build an URI using java.lang.URI, the encoding of the query parameters are different compared to build the URI using URIBuilder. The URIBuilder uses e.g. "+" to encode spaces (by default) while java.lang.URI uses (the expected) "%20" hex encoding for spaces.
Test that show the problem:
@Test public void testAddParameterEncodingEquivalence() throws Exception { final URI uri = new URI("http", null, "localhost", 80, "/", "param=stuff with spaces", null); final URIBuilder uribuilder = new URIBuilder().setScheme("http").setHost("localhost").setPort(80).setPath("/").addParameter( "param", "stuff with spaces"); final URI result = uribuilder.build(); Assert.assertEquals(uri, result); }
Think, there is a platform mismatch the URIBuilder is thought to be used: the current implementation focus on server site to build URI to be used in http post/put; not for a simple http get as it is also been used on client site (by browsers, or vaadin apps). Would be fine if we could control the use target from outsite; e.g. as a build() call parameter, or separate buildUrlEncoded() using "%20" and build() use the current implementation?
Attachments
Issue Links
- is related to
-
HTTPCORE-739 org.apache.hc.core5.net.URIBuilder does not decode plus characters (`+`) in the query part
- Resolved