aboutsummaryrefslogtreecommitdiff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* The notifying thread does not need to hold the lock on the same mutex as the ↵Charles Kerr2016-09-221-1/+1
| | | | one held by the waiting thread(s); in fact doing so is a pessimization.
* fix r25 copypaste error, lock on the correct mutexCharles Kerr2016-09-221-1/+1
|
* GAdbdClient::Impl::on_public_key_response() doesn't need a condition ↵Charles Kerr2016-09-222-6/+8
| | | | variable wait, so use std::lock_guard rather than std::unique_lock
* add a plethora of log statements to help figure out what's causing the silo ↵Charles Kerr2016-09-222-0/+135
| | | | test failures
* adbd-client's std::condition_variable::wait() call seems to be the ↵Charles Kerr2016-09-212-7/+12
| | | | system_error culprit, so wrapping in try-catch to be cautious
* in adbd-client's dtor, check the worker thread's joinable() state before ↵Charles Kerr2016-09-211-2/+7
| | | | calling join() on it
* in PKIdleData, take the std::string argument as a const& to make cppcheck ↵Charles Kerr2016-09-211-1/+1
| | | | happy on yakkety
* add tracer g_debug() calls for the benefit of the integration testsCharles Kerr2016-03-241-1/+4
|
* fix UsbManager dtor issue found by valgrindCharles Kerr2016-03-231-1/+11
|
* fix missing field initialization compiler warningCharles Kerr2016-03-231-1/+1
|
* keep the adbd socket open even when the lockscreen is closed. hold the ↵Charles Kerr2016-03-231-17/+27
| | | | pkrequest state in USBManager until the screen's unlocked.
* replace text 'Deny' with 'Don't Allow' for consistency with other permission ↵Charles Kerr2016-03-231-1/+1
| | | | prompts
* add tests for not showing snap decisions in greeter modeCharles Kerr2016-03-221-32/+16
|
* don't show the snap decision until we're out of the greeterCharles Kerr2016-03-223-23/+59
|
* fix typoCharles Kerr2016-03-221-2/+2
|
* get greeter's IsActive property workingCharles Kerr2016-03-213-21/+73
|
* add Greeter skeletonCharles Kerr2016-03-214-0/+178
|
* in UsbManager, reset AdbdClient on usb disconnectCharles Kerr2016-03-212-19/+26
|
* if our USB device is disconnected while prompting the user for ADBD, cancel ↵Charles Kerr2016-03-218-22/+192
| | | | the prompt.
* turn off verbose debuggingCharles Kerr2016-03-182-6/+3
|
* add some extra debug statements to usb-manager.cpp to track user response ↵Charles Kerr2016-03-181-1/+4
| | | | and the act of writing the pk out to disk
* fix warning messageCharles Kerr2016-03-171-1/+1
|
* de-dupe use of dbus namesCharles Kerr2016-03-172-19/+57
|
* tweak class description commentsCharles Kerr2016-03-162-0/+6
|
* in usb-manager.cpp, remove the remember_choice workaround since is already ↵Charles Kerr2016-03-161-4/+3
| | | | working around it
* use cmake-extras' EnableCoverageReport instead of home-rolled rulesCharles Kerr2016-03-161-2/+0
|
* use cmake's find_package(Threads) output everywhere instead of just in src/Charles Kerr2016-03-151-7/+1
|
* listen on /dev/socket/adbd, not /dev/socket/adbCharles Kerr2016-03-151-1/+1
|
* oops, last commit's diagnosis was incorrect. The timing test issue came from ↵Charles Kerr2016-03-151-11/+11
| | | | async dbus handling interfering with fast setup/teardown of automated tests. Revert the last change and fix by setting up the dbus signal subscription immediately upon getting the dbus connection.
* add some debug stubsCharles Kerr2016-03-152-2/+7
|
* cmake's Thread::Thread import doesn't exist in vivid's version of cmake ↵Charles Kerr2016-03-151-1/+7
| | | | (3.0.2), so fall back gracefully to -pthread there
* tweak a couple of 'auto' declarations that g++ 5.3.1 understands but 4.9.2 ↵Charles Kerr2016-03-151-2/+2
| | | | on vivid doesn't
* clean up compile_options and warnings. Add DBUS debug log messages to ctest ↵Charles Kerr2016-03-111-1/+4
| | | | environment variables.
* add UsbManagerCharles Kerr2016-03-104-10/+148
|
* cmake and test directory cleanupCharles Kerr2016-03-101-30/+31
|
* add human-readable fingerprint extraction from the adb public keysCharles Kerr2016-03-094-3/+43
|
* add fingerprint snap decisions; test with notification dbusmockCharles Kerr2016-03-083-3/+299
|
* add out-of-line virtual method definitions to Indicator to silence clang++ ↵Charles Kerr2016-03-083-8/+46
| | | | warnings
* add ADB server/client + testsCharles Kerr2016-03-064-1/+344
|
* fix RotationLock shutdown issue that messed with getting clean test runsCharles Kerr2016-03-041-0/+1
|
* fix explicit ctor warning from cppcheck to get tests passingCharles Kerr2015-07-151-1/+1
|
* use the same action-state-to-variant mapping as in the rtm-14.09 branchCharles Kerr2015-02-281-1/+1
|
* Use com.ubuntu.touch.system's "rotation-lock" boolean for the rotation lock ↵Charles Kerr2014-10-094-22/+18
|\ | | | | | | | | | | indicator. Fixes: 1378078 Approved by: Ted Gould, PS Jenkins bot
| * don't use 'auto tmp = ...' when tmp is going to be passed to g_free()Charles Kerr2014-10-061-1/+1
| |
| * make a debugging message go through g_debug() instead of g_message()Charles Kerr2014-10-061-1/+1
| |
| * fix the settings-to-action-state mappingsCharles Kerr2014-10-061-18/+11
| |
| * fix warnings reported by flint++ -- explicit ctors, classes with virtual ↵Charles Kerr2014-10-062-1/+6
| | | | | | | | methods but no virtual dtor, #ifdef include guards
| * replace 'orientation-lock' enum with 'rotation-lock' boolCharles Kerr2014-10-061-13/+11
| |
* | change header title from 'Rotation lock' to 'Rotation' as per I:RTMUFCharles Kerr2014-10-031-1/+1
|/
* in src/exporter.cpp, use G_DBUS_NAME_OWNER_FLAGS_NONE when owning the busnameCharles Kerr2014-08-211-1/+1
|