aboutsummaryrefslogtreecommitdiff
path: root/tests/test-device.cc
Commit message (Collapse)AuthorAgeFilesLines
* Make device with power supply has higher sorting priority. (LP: #1100546)Charles Kerr2017-05-151-47/+53
|
* Fix bug that chose the wrong header icon if a connected device has a charge ↵Charles Kerr2017-05-151-108/+268
| | | | but its charging/discharging state is unknown. (LP: #1470080).
* Fix the fallback power icon so the battery icon doesn't show red above 20% ↵Charles Kerr2017-05-151-27/+39
| | | | even when the Suru icon theme isn't present. Fixes: LP:#1559731.
* Fork from Ubuntu's indicator-power code project.Mike Gabriel2015-11-121-2/+2
|
* round power level to the nearest 10% to take advantage of higher-resolution ↵Charles Kerr2015-01-121-2/+20
| | | | power level icons
* fix typo introduced while trying to fix clang -Weverything warningsCharles Kerr2014-09-081-1/+1
|
* silence warnings generated by clang -WeverythingCharles Kerr2014-09-071-29/+29
|
* sync unit tests.Charles Kerr2014-07-291-8/+8
|
* replace 'std::vector<unsigned>' with 'std::vector<unsigned int>' for readabilityCharles Kerr2014-07-171-1/+1
|
* fix some compiler warnings generated by clang static analyzerCharles Kerr2014-07-131-2/+2
|
* prefer the 'battery-020-charging' icon to the 'battery-low-charging' one ↵Charles Kerr2014-05-051-5/+10
| | | | because the former is more precise so its icon is more likely to be close to the actual actual battery level.
* sync to lp:indicator-powerCharles Kerr2014-03-251-21/+36
|\
| * when expecting a NULL string, use EXPECT_EQ(NULL, str)Charles Kerr2014-03-131-5/+17
| |
| * in the new indicator_power_device_get_*() functions, use heap-allocated ↵Charles Kerr2014-03-131-21/+24
| | | | | | | | strings rather than relying on g_snprintf().
* | when choosing a primary device and more than one battery is found, ↵Charles Kerr2014-03-191-65/+95
|/ | | | accumulate their percentages and time-remaining properties as per the spec revisions in bug #880881.
* update the header / menuitem text / accessible text to reflect the changes ↵Charles Kerr2014-03-041-40/+112
| | | | in https://wiki.ubuntu.com/Power?action=diff&rev2=44&rev1=43#Title
* use the correct battery icon when charging/dischargingCharles Kerr2013-10-011-12/+12
|
* update the tests to look at the 24 hour thresholdCharles Kerr2013-08-231-3/+3
|
* copyediting: remove trailing spaces in source codeCharles Kerr2013-08-221-11/+11
|
* add show-percentage feature, basing off Haw Loeung's patch.Charles Kerr2013-08-221-33/+79
|
* get test-devices building and passing again.Charles Kerr2013-06-171-10/+10
|
* first steps at fixing the tests that broke during GMenuificationCharles Kerr2013-06-161-3/+4
|
* remove g_type_init() calls and bump glib minimum to 2.35.4, where ↵Charles Kerr2013-01-171-26/+0
| | | | g_type_init() is deprecated
* explicitly cast integral values to the correct type when passing them to a ↵Charles Kerr2012-11-161-4/+4
| | | | varargs function.
* modify DeviceTest.Labels s.t. it fails (currently failing) if the 'short' ↵Charles Kerr2012-11-071-1/+1
| | | | string for an AC Adapter is nonempty.
* in TestDevice.ChoosePrimary, add another test (currently failing) to confirm ↵Charles Kerr2012-11-061-0/+1
| | | | that when indicator-datetime is choosing from devices that are neither charging nor discharging, batteries are more interesting than other devices, and other devices are more interesting than UP_DEVICE_KIND_LINE_POWER
* add unit test (currently failing) to test for the bug reported in LP bug ↵Charles Kerr2012-11-061-11/+15
| | | | #1075192
* in TestDevice, make the log variable names a little more consistentCharles Kerr2012-10-281-14/+14
|
* set DeviceTest's gwarning/gcritical log func to one that prints no messages ↵Charles Kerr2012-10-281-0/+31
| | | | but accumulates the log count. This way we can both (a) silence in-console warnings that look like bugs in the test but are actually desirable warnings generated by fuzz testing, and (b) assert that we get exactly as many warnings/criticals as we expect to get.
* expand DeviceTest.ChoosePrimary to test choosing from a wider variety of ↵Charles Kerr2012-10-281-67/+44
| | | | device states.
* add test (currently failing) to detect the logic error reported in bug #1071757Charles Kerr2012-10-261-0/+16
|
* fix comment typoCharles Kerr2012-10-261-1/+1
|
* Select the primary device based on the spec's criteria. Add tests to confirm.Charles Kerr2012-10-261-0/+75
|
* the device kind string wasn't being used in some icon names.Charles Kerr2012-10-251-145/+196
|
* when discharging the battery, prefer the percent-charged icons over the ↵Charles Kerr2012-10-031-7/+15
| | | | full,good,low icons because the former show a more accurate view of the charge.
* copyediting: text formatting to make the icon list more readableCharles Kerr2012-09-051-3/+6
|
* reimplement indicator_power_device_get_icon_names() since our CA is ↵Charles Kerr2012-06-061-17/+17
| | | | incompatible with reusing code from GSD.
* use C++-style comments in test-device.ccCharles Kerr2012-06-011-32/+32
|
* In indicator_power_device_get_time_details(), add a test for non-devices ↵Charles Kerr2012-06-011-1/+13
| | | | being passed in.
* Fix edge case in indicator_power_device_get_icon_names() that returned a bad ↵Charles Kerr2012-06-011-0/+3
| | | | value if the caller passed in a NULL pointer as a device. Added regression test.
* Add coverage test for AC Adapters in indicator_power_device_get_time_details()Charles Kerr2012-06-011-0/+8
|
* Add unit tests for indicator_power_device_get_time_details()Charles Kerr2012-06-011-2/+87
|
* remove IndicatorPowerDevice's now-unused device icon string.Charles Kerr2012-05-311-12/+0
|
* Add indicator_power_device_get_icon_names().Charles Kerr2012-05-311-5/+196
| | | | | | | The main goal of this change is to make it possible to test the device's icon. A secondary goal is to clarify in the code how indicator-power's icons differ from the ones recommended by GSD.
* call g_object_run_dispose() in a standalone test to get coverage on the NULL ↵Charles Kerr2012-05-261-0/+1
| | | | / non-NULL branches of dispose()'s g_clear_pointer() calls
* ah, c++ is too smart for the last commit's implicit conversion.Charles Kerr2012-05-241-1/+1
|
* test passing bad non-NULL pointers to Device's accessorsCharles Kerr2012-05-241-6/+16
|
* add unit tests confirming that the device accessor functions won't crash ↵Charles Kerr2012-05-241-0/+14
| | | | when you pass NULL for the device
* remove the tests that attempted to exercise the code removed in r151Charles Kerr2012-05-241-33/+2
|
* add unit tests for getting/setting device propertiesCharles Kerr2012-05-241-0/+90
|