mirror of
https://github.com/google/nomulus.git
synced 2025-05-24 21:20:08 +02:00
6987d4e55c
According to RFC 2046, the body of the multipart contains:
multipart-body := [preamble CRLF]
dash-boundary transport-padding CRLF
body-part *encapsulation
close-delimiter transport-padding
[CRLF epilogue]
The preemble and epilogue are optional, and ignored. However, it's not 100%
explicit whether the CRLFs after the preamble and before the epilogue are
required. The one after the preemble is often not given if there's no preemble,
so it's conceivable that you don't *have* to give the CRLF before the epilogue
if there's no epilogue (it's also enclosed in the [], making it part of the
"optional")
However, it seems that when the TMDB "migrated to the cloud" (as they
describe it) on Aug. 13 2018, they started requiring that CRLF.
TESTED=connected to a TMDB-whitelisted server, used CURL to manually create the
message as we currently send it (without the final CRLF) with junk data and got
the error from the bug. Then sent the exact same message with the additional
CRLF, and got a different error that directly relates to the content of the
junk data.
-------------
Created by MOE: https://github.com/google/moe
MOE_MIGRATED_REVID=212637246
|
||
|---|---|---|
| .. | ||
| AppEngineServiceUtilsImplTest.java | ||
| BUILD | ||
| CidrAddressBlockTest.java | ||
| CollectionUtilsTest.java | ||
| ComparingInvocationHandlerTest.java | ||
| ConcurrentTest.java | ||
| DateTimeUtilsTest.java | ||
| DiffUtilsTest.java | ||
| DomainNameUtilsTest.java | ||
| HexDumperTest.java | ||
| PosixTarHeaderSystemTest.java | ||
| PosixTarHeaderTest.java | ||
| RequestStatusCheckerImplTest.java | ||
| RetrierTest.java | ||
| SerializeUtilsTest.java | ||
| SqlTemplateTest.java | ||
| TaskQueueUtilsTest.java | ||
| TeeOutputStreamTest.java | ||
| TypeUtilsTest.java | ||
| UrlFetchUtilsTest.java | ||