[
https://issues.apache.org/jira/browse/THRIFT-3884?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15382394#comment-15382394
]
ASF GitHub Bot commented on THRIFT-3884:
----------------------------------------
Github user jeking3 commented on a diff in the pull request:
https://github.com/apache/thrift/pull/1051#discussion_r71162579
--- Diff: lib/erl/src/thrift_compact_protocol.erl ---
@@ -355,7 +356,7 @@ read(This0, i64) ->
read(This0, double) ->
{This1, Bytes} = read_data(This0, 8),
case Bytes of
- {ok, <<Val:64/float-signed-big, _/binary>>} -> {This1, {ok, Val}};
+ {ok, <<Val:64/float-signed-little, _/binary>>} -> {This1, {ok, Val}};
--- End diff --
Why is "float" used here instead of "Double"? I don't know erlang; just
looking at line 222 above and expecting it to be similar.
> Fix Erlang compact protocol double endianess and boolean list
> -------------------------------------------------------------
>
> Key: THRIFT-3884
> URL: https://issues.apache.org/jira/browse/THRIFT-3884
> Project: Thrift
> Issue Type: Bug
> Components: Erlang - Library
> Environment: Travis-CI
> Reporter: Aki Sukegawa
> Assignee: Aki Sukegawa
>
> * Compact double field should be little endian unlike Binary protocol.
> * boolean list could not be sent
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
