summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Import revised translations.Eric Fischer2010-08-1720-165/+85
| | | | Change-Id: I5897c2ba31070f3c6b8225720db12eef5364777f
* am fa5175e5: am adb6887d: Clean up error codes returned by download manager.Steve Howard2010-08-174-99/+112
|\ | | | | | | | | | | | | Merge commit 'fa5175e572258fba0cd55e1c8a01b654178e3f1e' * commit 'fa5175e572258fba0cd55e1c8a01b654178e3f1e': Clean up error codes returned by download manager.
| * am adb6887d: Clean up error codes returned by download manager.Steve Howard2010-08-174-99/+112
| |\ | | | | | | | | | | | | | | | | | | Merge commit 'adb6887d3270d180c94eaf90878d5b67d74a8f28' into gingerbread-plus-aosp * commit 'adb6887d3270d180c94eaf90878d5b67d74a8f28': Clean up error codes returned by download manager.
| | * Clean up error codes returned by download manager.Steve Howard2010-08-164-99/+112
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This set of changes cleans up the error codes returned by the download manager in various failure cases, aiming for improved consistency. Error codes are part of the public API so it's important to get this right now. The main changes here are: * Refactoring the flow of error status information throughout DownloadThread to make it more explicit, by having StopRequest accept a status code in its constructor and eliminating State.mFinaStatus. * Eliminating the use of valid HTTP 4xx statuses when those statuses weren't actually returned by the server. Now, if the returned error code is a valid HTTP status code, that means it was returned by the server. These cases have been replaced with more sensible artificial error codes, including a new ERROR_CANNOT_RESUME when an interrupted download can't be resumed. * Improvements to some of the error handling code paths -- ensuring we don't clear the cache for external downloads, ensuring we don't fail with CANNOT_RESUME when the download hasn't actually started yet, removing the restriction on acceptable mime types for public API downloads. Change-Id: I0d825845fe0fe7ed5df74bad26e8d34ac0d1cc4e
* | | am e80c3555: am e61798da: Extend PublicApiAccessTest to exercise ↵Steve Howard2010-08-162-0/+28
|\| | | | | | | | | | | | | | | | | | | | | | | | | | DownloadManager. Merge commit 'e80c35557bd72489323303af9d93c2c33c88cce9' * commit 'e80c35557bd72489323303af9d93c2c33c88cce9': Extend PublicApiAccessTest to exercise DownloadManager.
| * | am e61798da: Extend PublicApiAccessTest to exercise DownloadManager.Steve Howard2010-08-162-0/+28
| |\| | | | | | | | | | | | | | | | | | | Merge commit 'e61798da80558450f580ed948d0d469bd6423d8e' into gingerbread-plus-aosp * commit 'e61798da80558450f580ed948d0d469bd6423d8e': Extend PublicApiAccessTest to exercise DownloadManager.
| | * Extend PublicApiAccessTest to exercise DownloadManager.Steve Howard2010-08-162-0/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | This change adds a new test case to PublicApiAccessTest to enqueue a request through DownloadManager, ensuring that the values constructed by DownloadManager fit within the allowed bounds. It also fixes a bug with allowing http header values exposed by the new test. Change-Id: I94fec57d7a41298ac42ddaab338516e6a60c4e75
* | | Merge "Import revised translations."Eric Fischer2010-08-1020-40/+120
|\ \ \
| * | | Import revised translations.Eric Fischer2010-08-1020-40/+120
| | | | | | | | | | | | | | | | Change-Id: I5ff563b44875bf1668a906c45420e303be614881
* | | | am b986eca2: (-s ours) am 99de0992: Import revised translations. DO NOT MERGEEric Fischer2010-08-100-0/+0
|\ \ \ \ | |/ / / |/| / / | |/ / | | | | | | | | | Merge commit 'b986eca2ca6764f1f3ed60ebfb82a8d8306acbac' * commit 'b986eca2ca6764f1f3ed60ebfb82a8d8306acbac': Import revised translations. DO NOT MERGE
| * | am 99de0992: Import revised translations. DO NOT MERGEEric Fischer2010-08-1020-40/+120
| |\| | | | | | | | | | | | | | | | | | | Merge commit '99de0992f03c5b4a1611d25db1df0773433dea7c' into gingerbread-plus-aosp * commit '99de0992f03c5b4a1611d25db1df0773433dea7c': Import revised translations. DO NOT MERGE
| | * Import revised translations. DO NOT MERGEEric Fischer2010-08-1020-40/+120
| | | | | | | | | | | | Change-Id: Ib2ce8f6ee48925ab4371c39266ef4696cb812f91
* | | am 1d332ceb: am da5a4183: Merge "Make all public API downloads visible." ↵Steve Howard2010-08-065-24/+58
|\| | | | | | | | | | | | | | | | | | | | | | | | | | into gingerbread Merge commit '1d332ceb5756ef2e09ae97f1735aa9b529063874' * commit '1d332ceb5756ef2e09ae97f1735aa9b529063874': Make all public API downloads visible.
| * | am da5a4183: Merge "Make all public API downloads visible." into gingerbreadSteve Howard2010-08-065-24/+58
| |\| | | | | | | | | | | | | | | | | | | Merge commit 'da5a4183312f520cf35b154d33cb7205e6ba5557' into gingerbread-plus-aosp * commit 'da5a4183312f520cf35b154d33cb7205e6ba5557': Make all public API downloads visible.
| | * Merge "Make all public API downloads visible." into gingerbreadSteve Howard2010-08-065-24/+58
| | |\
| | | * Make all public API downloads visible.Steve Howard2010-07-285-24/+58
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change makes all downloads through the public API visible by default. It removes the API that had allowed applications to control notifications while the download runs. This has been replaced with a hidden API, since such behavior is needed by SystemUpdater and Market (for self-updates). Additionally, the behavior is now protected by a new permission. I'm making this permission signatureOrSystem, and changing the non-purgeable permission to the same (it should've been that, I just didn't know). I'm also adding string descriptions to appease the translation folks. Change-Id: I192e8b19ff9b0e425257cef0db081c3d75996ea5
* | | | Import revised translations.Eric Fischer2010-08-0520-0/+80
| | | | | | | | | | | | | | | | Change-Id: Ic4df1a4487552342e398a4c0ae94c33b7a448976
* | | | am 5f587436: (-s ours) am a3c9f66a: Import revised translations. DO NOT MERGEEric Fischer2010-08-050-0/+0
|\| | | | | | | | | | | | | | | | | | | | | | | | | | | Merge commit '5f587436aad18acc3bbf6577872cc80b0af980d2' * commit '5f587436aad18acc3bbf6577872cc80b0af980d2': Import revised translations. DO NOT MERGE
| * | | am a3c9f66a: Import revised translations. DO NOT MERGEEric Fischer2010-08-0420-0/+80
| |\| | | | | | | | | | | | | | | | | | | | | | | | | | Merge commit 'a3c9f66a80c4d801b893f7867041f34d902156fe' into gingerbread-plus-aosp * commit 'a3c9f66a80c4d801b893f7867041f34d902156fe': Import revised translations. DO NOT MERGE
| | * | Import revised translations. DO NOT MERGEEric Fischer2010-08-0420-0/+80
| | |/ | | | | | | | | | Change-Id: Ic6f84636b3bbb635ecc790711ba5828dd094f878
* | | am 3253c05a: am 0d4f9f10: Get wifi limit from secure settings.Steve Howard2010-08-035-7/+14
|\| | | | | | | | | | | | | | | | | | | | Merge commit '3253c05afccc0cb62c29958d023d863222a5845c' * commit '3253c05afccc0cb62c29958d023d863222a5845c': Get wifi limit from secure settings.
| * | am 0d4f9f10: Get wifi limit from secure settings.Steve Howard2010-08-025-7/+14
| |\| | | | | | | | | | | | | | | | | | | Merge commit '0d4f9f105c3a31475a36e9a1edfa868c66ff455f' into gingerbread-plus-aosp * commit '0d4f9f105c3a31475a36e9a1edfa868c66ff455f': Get wifi limit from secure settings.
| | * Get wifi limit from secure settings.Steve Howard2010-07-285-7/+14
| | | | | | | | | | | | Change-Id: I750654c28cb3d9f9aa67bd56e4d8d770dbfde4b4
* | | am 92ae639f: am dea0a5f8: Add test for many interruptions to a single download.Steve Howard2010-07-303-103/+94
|\| | | | | | | | | | | | | | | | | | | | Merge commit '92ae639f324084a23a54812ab92304b009a93340' * commit '92ae639f324084a23a54812ab92304b009a93340': Add test for many interruptions to a single download.
| * | am dea0a5f8: Add test for many interruptions to a single download.Steve Howard2010-07-303-103/+94
| |\| | | | | | | | | | | | | | | | | | | Merge commit 'dea0a5f8e5804d3aba40eaa7de763dd88058384e' into gingerbread-plus-aosp * commit 'dea0a5f8e5804d3aba40eaa7de763dd88058384e': Add test for many interruptions to a single download.
| | * Add test for many interruptions to a single download.Steve Howard2010-07-273-103/+94
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Adding a new test case for downloads that undergo many interruptions (as may happen with a very large download that takes many hours). Includes some refactoring in the test suite. Early on, this test exposed a race condition in which the download manager got some I/O exception while reading from the MockWebServer. I went in and improved/refactored much of the error logging code in DownloadThread to try and track this down. Unfortunately, once I finished, the race condition no longer seems to be reproducible, even with hundreds of runs of the test case. So I've given up on it for now. In any event, error logging is better and much duplicate code has been eliminated.
* | | am 915912de: resolved conflicts for merge of f85aa9ef to gingerbread-plus-aospSteve Howard2010-07-302-73/+98
|\| | | | | | | | | | | | | | | | | | | | Merge commit '915912de0373b78de4cd7cc1f45dad760ef94243' * commit '915912de0373b78de4cd7cc1f45dad760ef94243': Improved support for 302/307 redirects.
| * | resolved conflicts for merge of f85aa9ef to gingerbread-plus-aospSteve Howard2010-07-302-73/+98
| |\| | | | | | | | | | Change-Id: I0adb03f16a6f0cc648886e02c79ffde63e18f99f
| | * Improved support for 302/307 redirects.Steve Howard2010-07-272-69/+94
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change the download manager's handling of 302/307 temporary redirects so that after an interruption of any kind, the delayed retry/resume will return to the original URI. This complies better with the HTTP spec. This change also makes the download manager handle other redirects immediately rather than using the delay that's otherwise applied to download errors. I made one more method extraction in DownloadThread to simplify this change, pulling the top-level logic for a single request into executeDownload(). It was then just a matter of introducing a new RetryDownload exeception, similar to StopRequest, and making the redirection code use it. Change-Id: Ic719c5725a9fd2e5eebe4dc03453ee71d9f27cd4
* | | am 7a8bf08f: am f6b4c24b: Fix bug with closing output stream for external ↵Steve Howard2010-07-273-5/+19
|\| | | | | | | | | | | | | | | | | | | | | | | | | | downloads. Merge commit '7a8bf08fe936a1fdb0408dea1ec2f39b83acc5d7' * commit '7a8bf08fe936a1fdb0408dea1ec2f39b83acc5d7': Fix bug with closing output stream for external downloads.
| * | am f6b4c24b: Fix bug with closing output stream for external downloads.Steve Howard2010-07-273-5/+19
| |\| | | | | | | | | | | | | | | | | | | Merge commit 'f6b4c24b4a876daa3f4e91d6da418983222b9dfd' into gingerbread-plus-aosp * commit 'f6b4c24b4a876daa3f4e91d6da418983222b9dfd': Fix bug with closing output stream for external downloads.
| | * Fix bug with closing output stream for external downloads.Steve Howard2010-07-273-5/+19
| | | | | | | | | | | | | | | | | | | | | | | | I added a unit test to cover this, and it caught another issue with disallowing external destinations outside of the default downloads directory (which are now allowed with the new API). Change-Id: I4df6442bebb06458ad28c85f6bc8cbcbf3ce67a1
* | | am b6ba0582: am e6a05a1a: Serialize threading for download manager testing.Steve Howard2010-07-2710-264/+269
|\| | | | | | | | | | | | | | | | | | | | Merge commit 'b6ba05828ace9546e5e27d3745e7cbcba323b66d' * commit 'b6ba05828ace9546e5e27d3745e7cbcba323b66d': Serialize threading for download manager testing.
| * | am e6a05a1a: Serialize threading for download manager testing.Steve Howard2010-07-2710-264/+269
| |\| | | | | | | | | | | | | | | | | | | Merge commit 'e6a05a1aa4697440e9630d12b741b3bae321fe49' into gingerbread-plus-aosp * commit 'e6a05a1aa4697440e9630d12b741b3bae321fe49': Serialize threading for download manager testing.
| | * Serialize threading for download manager testing.Steve Howard2010-07-2710-264/+269
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The download manager uses threading in a simple way -- it launches two threads, UpdateThread and DownloadThread, and both are "fire and forget". This is fortunate for testing, since it means we can eliminate multithreading and simply run each thread in order, and everything still works. This change does just that, abstracting Thread.start() behind SystemFacade and making FakeSystemFacade put new threads into a queue and then run through them serially. This simplifies much of the test code and makes it all much more predictable. I've simplified the test code as much as possible here and moved a few more tests over to PublicApiFunctionalTest, leaving only a minimum in DownloadManagerFunctionalTest, which will eventually be deleted altogether. I've also improved testing in some areas -- for example, we can now test that running notifications get cancelled after the download completes in a robust way. There is one test case that checks for race conditions and requires multithreading. I've moved this into a new ThreadingTest class, which uses a custom FakeSystemFacade that allows multithreading. I've extracted AbstractPublicApiTest for the newly shared code. Change-Id: Ic1d5c76bfa9913fe053174c3d8b516790ca8b25f
* | | am f5373e6b: am 93155e1d: Stub out and test system notifications.Steve Howard2010-07-268-23/+103
|\| | | | | | | | | | | | | | | | | | | | Merge commit 'f5373e6ba917cc60db4ee796aed8bc0e37a3f5e9' * commit 'f5373e6ba917cc60db4ee796aed8bc0e37a3f5e9': Stub out and test system notifications.
| * | am 93155e1d: Stub out and test system notifications.Steve Howard2010-07-268-23/+103
| |\| | | | | | | | | | | | | | | | | | | Merge commit '93155e1da7e89d4925e244f5afa94afb8ada7381' into gingerbread-plus-aosp * commit '93155e1da7e89d4925e244f5afa94afb8ada7381': Stub out and test system notifications.
| | * Stub out and test system notifications.Steve Howard2010-07-238-23/+103
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change abstracts NotificationManager interactions behind SystemFacade and takes advantage of that to test notifications, to a limited degree. It also fixes a silly typo in AbstractDownloadManagerFunctionalTest, and it introduces an extra sleep between tests to avoid some flakiness. I'll look for a better solution to that problem after this change goes in. Change-Id: I3a0307f828955cd45b0e3581ad499da28cc0556e
* | | am f572f8d5: am 691fc58e: Test + utilize change notification in DownloadProviderSteve Howard2010-07-262-6/+54
|\| | | | | | | | | | | | | | | | | | | | Merge commit 'f572f8d54340da87fcf19b4a526e6215ac1a8686' * commit 'f572f8d54340da87fcf19b4a526e6215ac1a8686': Test + utilize change notification in DownloadProvider
| * | am 691fc58e: Test + utilize change notification in DownloadProviderSteve Howard2010-07-262-6/+54
| |\| | | | | | | | | | | | | | | | | | | Merge commit '691fc58e9a61d99ab5018c5ddc8349ed413e4470' into gingerbread-plus-aosp * commit '691fc58e9a61d99ab5018c5ddc8349ed413e4470': Test + utilize change notification in DownloadProvider
| | * Test + utilize change notification in DownloadProviderSteve Howard2010-07-232-6/+54
| | | | | | | | | | | | | | | | | | | | | | | | | | | Now that I've learned how ContentResolver.notifyChange() works, add a test to explicitly check it's being called in DownloadProvider, and take advantage of it in waitForDownloadToStop to get rid of the ugly old polling mechanism. Change-Id: Iba71bdf1b0d31454cc4e186ceb4fdd9fdb5faad5
* | | am 0a450da1: am 71aab521: Support for non-purgeable downloads through the ↵Steve Howard2010-07-263-5/+31
|\| | | | | | | | | | | | | | | | | | | | | | | | | | public API. Merge commit '0a450da1d9e95f27f479aa45b442960baac49bb1' * commit '0a450da1d9e95f27f479aa45b442960baac49bb1': Support for non-purgeable downloads through the public API.
| * | am 71aab521: Support for non-purgeable downloads through the public API.Steve Howard2010-07-263-5/+31
| |\| | | | | | | | | | | | | | | | | | | Merge commit '71aab521efba9b28779541440c797220ec98ac97' into gingerbread-plus-aosp * commit '71aab521efba9b28779541440c797220ec98ac97': Support for non-purgeable downloads through the public API.
| | * Support for non-purgeable downloads through the public API.Steve Howard2010-07-233-5/+31
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change adds a permission, android.permission.DOWNLOAD_CACHE_NON_PURGEABLE. When an app has this permission, any downloads it requests through the public API to the download cache will automatically become non-purgeable, i.e. they'll never be automatically deleted by the download manager to free up space. This is intended for use only by the system updater. Change-Id: I35cdd44f7e5d46bc70443d1a9743f61a51395ddb
* | | am 8667ebd1: am b06b739b: Make DownloadProvider accessible for public API usage.Steve Howard2010-07-267-67/+311
|\| | | | | | | | | | | | | | | | | | | | Merge commit '8667ebd109ea212a7645ad6c214d4d765bb70b75' * commit '8667ebd109ea212a7645ad6c214d4d765bb70b75': Make DownloadProvider accessible for public API usage.
| * | am b06b739b: Make DownloadProvider accessible for public API usage.Steve Howard2010-07-267-67/+311
| |\| | | | | | | | | | | | | | | | | | | Merge commit 'b06b739b078ce4b00600487cfec31659647bf31f' into gingerbread-plus-aosp * commit 'b06b739b078ce4b00600487cfec31659647bf31f': Make DownloadProvider accessible for public API usage.
| | * Make DownloadProvider accessible for public API usage.Steve Howard2010-07-227-67/+311
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change removes the requirement that apps have the ACCESS_DOWNLOAD_MANAGER permission in order to access DownloadProvider. This enables the public API to work. Instead, DownloadProvider enforces the new permissions model for the public API: * insert() requires INTERNET permission * insert() checks that input fits within the restricted input allowed for the public API * insert() also strictly checks the file URI provided with DESTINATION_FILE_URI (and still requires WRITE_EXTERNAL_STORAGE permission if that is supplied) Note that if an app has the ACCESS_DOWNLOAD_MANAGER permission, legacy behavior is retained. I've added a test to cover this new access, and updated the existing permissions tests. I also fixed a bug in WHERE clause construction in update() and delete(), and refactored the code to eliminate duplication. Change-Id: I53a08df137b35c2788c36350276c9dff24858af1
* | | am c2448d48: am 0d8d8910: Make COLUMN_URI readable and tighten UID restrictions.Steve Howard2010-07-222-39/+15
|\| | | | | | | | | | | | | | | | | | | | Merge commit 'c2448d488ff4e476762d0a07bb0757ca99c05445' * commit 'c2448d488ff4e476762d0a07bb0757ca99c05445': Make COLUMN_URI readable and tighten UID restrictions.
| * | am 0d8d8910: Make COLUMN_URI readable and tighten UID restrictions.Steve Howard2010-07-222-39/+15
| |\| | | | | | | | | | | | | | | | | | | Merge commit '0d8d89105c00edbad95a268aaae65f2ff94ed5a1' into gingerbread-plus-aosp * commit '0d8d89105c00edbad95a268aaae65f2ff94ed5a1': Make COLUMN_URI readable and tighten UID restrictions.
| | * Make COLUMN_URI readable and tighten UID restrictions.Steve Howard2010-07-212-39/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I need to make COLUMN_URI readable by apps, since the public API exposes that field. In order to avoid any possible security issues, I got rid of the feature that potentially allowed apps to view downloads from other UIDs. No one was using that feature and the public API exposes no such feature (yet). While at it, I cleaned up some related code in update() and delete(). Change-Id: I5384115d2a865255d009fbe37449488fd2269389
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-07 11:50:25 +0000