| Commit message (Collapse) | Author | Files | Lines |
|
Change-Id: Id7337defebfd196e839f37ce199c94c7df0ed949
|
|
Change-Id: Id2fa594ab5ed35119a1c924e8c97894fff4245bc
|
|
Change-Id: I56b4a99f670ad38708ad83d76b93a631295e3333
|
|
This is much faster. It doesn't create messages, it doesn't post
stuff in a queue, it doesn't delay execution of other messages.
Also, it decouples the timer from the functionality of the Handler
so this can be pulled up in another change to consolidate with
mLastKeyTime or so.
Change-Id: I9044fc031373e66b9b40eee8025633613d0945dc
|
|
This will avoid some two-way IPC calls.
Change-Id: If9eb3e99b3d08346aadd4bbed2a1761f8c2527b9
|
|
Stop using the numerical value, use the actual method
Change-Id: Iffa7f2d21a0471c7862afd3234cc9c5dd86feb2e
|
|
Bug: 6202480
Change-Id: I19e04f17614e70867f8383d08667e350d753cde9
|
|
to avoid NPEs.
Bug: 6188932
Change-Id: I377ceaac867df879dc7984dae44e96b369858cca
|
|
Bug: 6635482
Bug: 6644787
Bug: 6645780
Bug: 6646060
Bug: 6645917
Change-Id: I61e62a7b9840bb4767ce40f521004fae7dbe905a
|
|
For some reason, there were several ways, not exactly identical,
to get this setting. The ones that used mAutoCorrectionEnabled
would kick in when the input field was specifying no correction,
so it would be a little strange (although harmless in the practice
because the settings set in this way would not get used later,
because the correct test would be done at that time).
Also perform a very small refactoring
Change-Id: Ica9f32b238d98009ae1852d3c1e940398f5d341c
|
|
Change-Id: I512b04e23490413a44b1ca0517102fe2d9138df3
|
|
Change-Id: Ie143ad008ca8deb1c5034cb80ec1cb28a6b0682d
|
|
Change-Id: I5da48442c0b39aaff817c3b6d067a9b13fda0643
|
|
Change-Id: If8d03bea1b6369cb3043da6321048d052995d664
|
|
Bug: 6637614
Change-Id: I34d26563e59d3b09bf35b8173dac5645ccb6a39f
|
|
Change-Id: I2ab623c74a6c9b45558af0241dbd985e851428cd
|
|
Change-Id: I865d01b8e1f936a0715bb28ce6ab2dd8ecd5a315
|
|
Change-Id: I3a0a7fc4901be1171d0877bb3a02633d6f6c1c2d
|
|
...for readability. After all, it makes sense that "current settings"
is a "SettingsValues", while it's harder to understand why we'd
manipulate an object named "SettingsValues".
Change-Id: Ide9a17b970a6cc00b80b9a217f986268c739c662
|
|
Change-Id: If34112c69acd5122d87554043d87d8f4dbdf5d15
|
|
Change-Id: I11a8e1aca3541c35d2b39b27716687fcd8ff3dc6
|
|
At last this guy is final, at last
Change-Id: I76911f8d47493f8a1719f75fc7e949be5d7da1a2
|
|
Change-Id: I43c0505e9a1652bfc568f23b787fc68a5c5e9220
|
|
Change-Id: I0339f39d49bc6390d098017f32d92c776974e03e
|
|
Change-Id: I8527776a4bd8236a85bb77f9791f9b76977ad181
|
|
Change-Id: I73acfee9098a87b9cde8aca720586ec748b2a82f
|
|
- make all functions in unigram_dictionary const
Change-Id: I94a64865aa793435e6c81890e68d999f6374074e
|
|
Change-Id: I9982ff325bef56694402caef28a77683c52ccd71
|
|
The goal is to simplify the code in LatinIME.java as well as having
a handy place to put debug calls to see interaction with TextView.
Change-Id: I255227e7e7343e0c2f3dcd1f185e5020d6186732
|
|
- proximity_info is now stateless
Change-Id: I62725bfe05b161fa8fc050ea6b50867e10a354e2
|
|
Should we maybe rename back *BinaryDict* to *Dict* in another
change for simplicity?
Change-Id: I8f7dcb78b9bdf1a13ce403a997fbb8619f2b453b
|
|
Change-Id: Icd540742073d49d12e70b2d8bd99aaf7ccb5802d
|
|
Change-Id: I03236386aea13fbd4fb8eaeee18e0008aa136502
|
|
Change-Id: I6cad8f4ae58eeb08d8a29137987cb3da60019844
|
|
Change-Id: I7f676d036d2773bd3caa5087626b72fc599b1f08
|
|
Change-Id: I264f829c1584a37995dbc5ee4ba6b6783d5d4834
|
|
Bug: 6584882
Change-Id: I1f76388fa8cc16171bc3fc60c935ddcf0a7c4e92
|
|
Bug: 6626700
Change-Id: I3190d7404e45e704be79ee2031e79b18475feb8c
|
|
Bug: 6625029
Change-Id: I6ea02e8e46d2bae29080cf9b4ad852a1b6fdd6f5
|
|
Bug: 6616436
Change-Id: I8d66a37f295134c5b9875b2a305a9be7442bd75d
|
|
Bug: 6548943
Change-Id: I7b4fbe20615a28151a74875be43b9f4a20e0dce9
|
|
Bug: 6584677
Change-Id: Ifd9b0f995f4e0cf4d35a7ecde767826ab976f0b8
|
|
- lets users flag a particular time in the research log
- lets users delete the log for this session
also makes the UsabilityLog setting control whether the ResearchLog logs or not.
multi-project commit with I89067e7d3b8daca7179333f1dbe82224c26920fe
Bug: 6188932
Change-Id: I89864ef3ab53b0efe1ea8d75247be08712f0c399
|
|
Change-Id: I7718b03a6f995edfb3cce07f761a4f56d573aa1b
|
|
Change-Id: I6c285a7484375ce4aa4569af6249027f84711d60
|
|
This change also re-introduces colored action key, that was disbaled
by I1c7335c5.
Bug: 6568834
Change-Id: I0c04b86049325c84492d90457cee0c16c57937f8
|
|
Bug: 6568834
Change-Id: If48635ead60dc119656f1b10c9b05a1918ebd668
|
|
Bug: 6620122
Change-Id: I2852f77d78b242c1a96c53cdd1e4c0e299f8ea4f
|
|
Bug: 6568834
Change-Id: I138464b5445e13eaf062ce84a1ba332dfd1afafa
|
|
Bug: 6603257
Change-Id: I73b65e921cff8337a00594525f67f6aec2178881
|