Age | Commit message (Collapse) | Author |
|
Bug: 345656107
Change-Id: Ibe9e712b1f102e56ea611d88559c4866e7ac2a39
|
|
FolderOrigin-RevId: /tmp/copybara-origin/src
Change-Id: Ib37cbc097e05ce205b4c198b65877e69b7141b02
|
|
Bug: 345656106
Change-Id: Iad86e1b204d621f5892e073a10d600d060999bc6
|
|
FolderOrigin-RevId: /tmp/copybara-origin/src
Change-Id: I29e70962b36e9c964484170d28af449df1f9a18f
|
|
Bug: 345656003
Change-Id: I13a0b5d74315448b361cdc8d1dca17457a7b1c7a
|
|
FolderOrigin-RevId: /tmp/copybara-origin/src
Change-Id: I727d2277512236d7d0db42e102d291b6204b38e5
|
|
aosp/3109819 did not fix the merge conflict correctly
Test: build
Bug: 342969117
Change-Id: I39fa106de86850b078dc2406223f08f4ba9fab3b
|
|
Change-Id: Ifc570035bb3fcddba358bfd1c9dc93646249391f
|
|
FolderOrigin-RevId: /tmp/copybara-origin/src
Change-Id: Ie1ecf4a8ee6c6263302f0e88e86baa49a8052442
|
|
aosp/3104740 missed to fix the merge conflicts
Test: m NetHttpCoverageTests
Bug: 342972034
Change-Id: I083b7b35b3f755928a98924c77d685581ea7510e
|
|
Change-Id: I9b8f799b3a8afc5e4cb0150d7cc74017be5feccc
|
|
FolderOrigin-RevId: /tmp/copybara-origin/src
Change-Id: Ifb7b548cde690e10cc102366bc538e744efa902b
|
|
Removed the zlib divergence where we had to replace line like
`#include "third_party/zlib/zlib.h"` with `#include <zlib.h>`. This was
done by creating an actual header file whose path is `third_party/zlib
zlib.h` that will include AOSP's header.
Test: m cronet_aml_net_net && atest net_unittests_tester
Bug: 321680067
Bug: 304217500
Change-Id: Ie794ff71c0d5f9c9083024bb5a3e05f1fadf6171
|
|
'aosp/upstream-staging' into main
Revert submission 3031237-revert_v3
Reason for revert: Fixed in aosp/3036676
Reverted changes: /q/submissionid:3031237-revert_v3
Change-Id: I256d356c90ba71a1cba4fa2b492be060021a4c15
|
|
This reverts commit 181bb755f589c53bd3e47be5f2a1883b9c33a562.
Change-Id: I08f9366b868455e9e4294e0563b86b423ec395a8
|
|
Revert submission 3005178-revert-3001326-revert-2998737-revert-2980591-cronet_121_0_6167_71-JBOUFGJXPX-ICXECJPUPA-OOMALUYYUV
Reason for revert: aosp/3022283 fixed coverage issue
Reverted changes: /q/submissionid:3005178-revert-3001326-revert-2998737-revert-2980591-cronet_121_0_6167_71-JBOUFGJXPX-ICXECJPUPA-OOMALUYYUV
Change-Id: I131a916abcaaeff1dbf0ebf346d252680d78d5cc
|
|
Revert submission 3001326-revert-2998737-revert-2980591-cronet_121_0_6167_71-JBOUFGJXPX-ICXECJPUPA
Reason for revert: Build breakage as b/330450460
Reverted changes: /q/submissionid:3001326-revert-2998737-revert-2980591-cronet_121_0_6167_71-JBOUFGJXPX-ICXECJPUPA
Change-Id: I36a3e37647ceb3561bbd139b61c99d1de220667c
|
|
Revert submission 2998737-revert-2980591-cronet_121_0_6167_71-JBOUFGJXPX
Reason for revert: aosp/3003361 fixed failing test. See b/329605950.
Reverted changes: /q/submissionid:2998737-revert-2980591-cronet_121_0_6167_71-JBOUFGJXPX
Test: atest NetHttpCoverageTests
Bug: b/322154153
Change-Id: I1df1f22bbb0cf72ac1cd5aed8d5f3c8084ad1942
|
|
Revert submission 2980591-cronet_121_0_6167_71
Reason for revert: Seems to be the cause for b/328855816
Reverted changes: /q/submissionid:2980591-cronet_121_0_6167_71
Change-Id: Ia90577797bfd9baf2e12e4b6bfaf83635573183b
|
|
Also regenerate desc_*.json to reflect dep file changes
Bug: b/322154153
Change-Id: I8c2aa24f3e42cc40645f5e2320e8eb0e76091bea
|
|
Bug: b/322154153, b/307234565
Change-Id: I2e128d94163dc2182c9e36e166954b7cde32abdd
|
|
Bug: b/322154153
Change-Id: I61dbf0bb966e0479d8e7a58294a914a46e5f59e6
|
|
Test: atest NetHttpTests
Change-Id: I860e2735258bd7015a82608f40ebaf53c616165f
|
|
From aosp/2927838
Change-Id: Ieca19063f0968b8f2acd95f7067768d841863b9d
|
|
Change-Id: I959864c5b5531aca216b9965bf471e2c7b6e1025
|
|
* This is not a clean cherry-pick, I simply copied the entire file as solving
merge-conflicts is not really necessary, once the CL itself make its way through
the import then the divergence will be removed.
Test: atest NetHttpTests
Change-Id: I0f13c2d71311e82f10c3efcaf2d12385df0932b2
|
|
FolderOrigin-RevId: /tmp/copybara-origin/src
Change-Id: Ieee3d31a0e9c404daa9a1630919e62b52ce61d1b
|
|
* Alot of the code living in ExperimentalCronetEngine has been moved to live in ExperimentalOptionsTranslatingCronetEngineBuilder
upstream but this was not reflected downstream.
Test: atest NetHttpTests && atest CtsNetHttpTestCases
Bug: 304217500
Change-Id: I12a6690cbad74cfd717986bcce69fc4af9f5e2a7
|
|
Bug: 304217500
Change-Id: Ib74b816862f4f853b3aeffdb9f1647c11ed3f00f
|
|
* Reverted all of the test utils + test files back to their original
form as found in aosp/upstream-import branch at version 121.0.6103.2
* Added @SkipPresubmit tags back again on some tests, this is the only divergence
left. It will be removed once aosp/2910846 gets merged then we can change
`exclude-annotation` to `exclude-filter` in TEST_MAPPING.
Bug: 304217500
Test: atest NetHttpTests
Change-Id: Idf8b76078aaf99613fc23ad65f9d0c6410f0d4a1
|
|
See https://android-review.git.corp.google.com/c/platform/external/cronet/+/2687908/comment/46c6fff9_6778590e/ and https://cs.android.com/android/platform/superproject/main/+/main:external/cronet/Android.extras.bp;l=271
This was supposed to be removed when a Cronet version >= 117.0.5911.0 was imported into AOSP which
is the case now.
Change-Id: Ibea7c9ff370c043b03b09f4523f726232c7bb8a7
|
|
Since there is no longer a need to locally apply patches to Chromium before
generating desc_*.json file, revert the BUILD.gn files to the original versions
with no local changes.
Change-Id: I8d5d274e01d1c60d6b797b8cabdb36e5cb7138fc
|
|
FakeCronetEngine tests were previously deleted, I don't have context on why they were
deleted, but it is probably better to keep those files around, they are not part of
Android.bp but they will reduce the import process effort and reduce divergence further.
Change-Id: I7273781ea3397676bc71f83a1a727a0578fe11e6
|
|
Apihelpers were deleted from upstream-staging, I'm not quite sure why they were deleted
in the first place, maybe because we don't ship apihelpers? This change introduces apihelpers
but they are not included in API surface that we ship and they can not be included by **mistake**
as adding any new API method requires updating a `text` file and get API council approval.
From the above, I can conclude that it is better to checkout those files to reduce the
import process merge conflicts and reduce the divergence further.
Change-Id: I656bebb7dc11e579898ccfe07817f82f6b791f82
|
|
See https://chromium-review.googlesource.com/c/chromium/src/+/5152885
Bug: 311373434
Bug: 314275954
Change-Id: I0a3e8485852cfcdee3e15193625486bf05a2d19c
|
|
JavaUrlRequest is now being compiled because it is needed by JavaCronetEngine, all of this code
will not be used at all. It is only there for the sake of compilation and removing downstream patches
from `CronetTestRule`.
Bug: 314275954
Change-Id: Icf1b83e393c64311fd46b09c9128e2cbaf13135d
|
|
This is used for testing and we should probably change this test upstream to use getDefault
to avoid future merge conflicts
Bug: 314275954
Change-Id: Ic03cf57881df49ef58ea5589d88555a147075f31
|
|
Accidentally removed FileUtils import which is needed to ensure we use
the same implementation as chromium.
Bug: 314275954
Change-Id: Ib744540871d8ebb1d0010968d6f3e98e65271c56
|
|
* Forgot to remove @RunNativeOnly annotations as they no longer existed.
* Removed @Ignore annotations and switched to disabling those tests using AndroidTest.xml as seen here (aosp/2893182), this reduces the merge conflicts and divergence while still achieving the same end-results.
Bug: 314275954
Change-Id: Id2aef699e11463b7d1be9ee9432146eaffa43405
|
|
The `TAG` field should be public for testing.
Bug: 314275954
Change-Id: I9c51188219b71af28a4e586d13e91d9b76d57498
|
|
This is needed to use `CronetTestRule` that is imported from Chromium without any downstream changes.
This will not be used at all, it is only there for the sake of compiling.
Bug: 311373434
Bug: 314275954
Change-Id: I1caacecad52a1d344fd85356059a2992e5f100e1
|
|
This is needed in order to use `CronetTestRule` that is imported from Chromium without any downstream changes.
In practice, this is not going to be used at all, it is just there for the sake of compilation.
Bug: 311373434
Bug: 314275954
Change-Id: Ic5752284b106636da51d9588e9ac9232b828e5b4
|
|
CronetTestRule had some downstream changes that caused merge-conflict, I figured out that
the downstream changes are not necessarily as we can use the upstream version of CronetTestRule
and import JavaCronetEngine whilst disabling Fallback implementation tests as seen in upcoming CLs
Bug: 311373434
Change-Id: I341a4af6ee652b634903bbdbb9899101542c0c6c
|
|
Bug: 314275954
Change-Id: Ib62ad4b0e6a261d9f4b276a69b68e51c69f9e210
|
|
Those getters/setters were implemented as per the API Council's requirement to ship Cronet to Android U as
seen here (aosp/2646733) and here (aosp/2647252). When we introduced Cronet Wrapper API, it was implemented
upstream (Chromium) so it matched the Cronet's API that exist upstream. However, HttpEngine API is not identical
to Cronet's API, which has caused an error.
Bug: 314275954
Change-Id: I1ceb61a98cfe30329afb1234bf9f0ff54dba9afb
|
|
Change-Id: Ib1d37aa7813d49232adddf67fa3d8284cf37eb4a
|
|
FolderOrigin-RevId: /tmp/copybara-origin/src
Change-Id: I690becfaba7ad4293eba08b4f9d1aa7f953fce20
|
|
Change-Id: I63507f9caac4edeac95e47283706de6d53d6dc7c
|
|
When calling System.loadLibrary(), ART includes a check to ensure the
native library has not already been loaded from a different class loader
context: "The library will be associated with class_loader. The JNI spec
says we can't load the same library into more than one class loader."
This error is encountered regardless of the conflicting libraries being
in different linker namespaces (which would later correctly load /
resolve symbols).
This is an issue specifically for HttpEngine which calls
System.loadLibrary("cronet.114.0.5735.84") just as upstream Cronet does.
If an app happens to statically link upstream Cronet with version
114.0.5735.84, it will crash when trying to use HttpEngine.
The immediate solution is to rename HttpEngine's libcronet.so to
libhttpengine.so.
Bug: 298674099
Test: TH
Change-Id: I64f48f5337caa36e53d33d5a300ee0ed25aa29ec
|
|
* Those tests are failing downstream because AOSP runs all the tests in a single batch
where Cronet is already initialized and loaded so the code-path to loadLibrary is never triggered.
Change-Id: I690ef6e52eba3883fb5ba96694fda4e607d9832c
|