anoopj commented on code in PR #16174:
URL: https://github.com/apache/iceberg/pull/16174#discussion_r3289210325
##########
core/src/main/java/org/apache/iceberg/util/LocationUtil.java:
##########
@@ -57,4 +57,54 @@ public static String tableLocation(TableIdentifier
tableIdentifier, boolean useU
return tableIdentifier.name();
}
}
+
+ /**
+ * Returns true if the location contains a URI scheme (e.g. {@code s3:},
{@code hdfs:}, {@code
+ * file:}), per <a
href="https://datatracker.ietf.org/doc/html/rfc3986#section-3.1";>RFC 3986
+ * section 3.1</a>.
+ */
+ private static boolean hasScheme(String location) {
+ for (int i = 0; i < location.length(); i += 1) {
+ char ch = location.charAt(i);
+ if (ch == ':') {
+ return i > 0;
+ }
+
+ if (!Character.isLetterOrDigit(ch) && ch != '+' && ch != '-' && ch !=
'.') {
+ return false;
+ }
+ }
+
+ return false;
+ }
+
+ /**
+ * Resolves a location against a table location. If the location has a URI
scheme, it is returned
+ * as-is. Otherwise, the table location and the relative location are joined
by the URI separator
+ * character {@code /}.
+ */
+ public static String resolveLocation(String tableLocation, String location) {
Review Comment:
Sorry, forgot to respond to that part. The first version of the PR had that
null check. But @rdblue pointed out that `resolveLocation()` will be called in
a tight loop, so we shouldn't do a precondition check. The code should just do
a NPE, and we are fine with that. We could add a null check somewhere upstream
in reader, where it won't be in a tight loop.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
