Don't try to get a bootloader filename from an empty list. This happens
if the device doesn't have a bootloader file on disk.
Change-Id: Idd4c265e7ae298913c5feefb3963120867a0e858
Replace QLabel with QSvgWidget so the widget draws the svg directly
instead of first creating a pixmap from it.
This also avoids an issue when building with mxe which causes the image
to not show due to a missing svg plugin (which doesn't happen for svg
icons.)
Change-Id: Ic4e265f0567984d1ffe7a3e1cc641df27755e90b
The old deployment script doesn't work anymore due to the change to
cmake, and since we build the distribution packages (zip / dmg /
AppImage) with cmake directly there's not much need left for the old
deployment.
Change-Id: Ide20887c5bc2e22aabbfb47374d70529609fbc3c
Instead of checking for certain hardcoded strings, fetch
the actual capability bitmask by instantiating an installer
and querying it.
Change-Id: I7883d9c1e90da37ee7c0189732ac626685adbfa1
- Fix binary name in Info.plist.
- Run dmgbuild as Python module, in some cases it isn't available directly.
- Add missing dmgbuild configuration for themeeditor.
Change-Id: I2db50959b88283f43e2203cf71b130cdb33103dd
Make tests implemented as data driven tests show a sensible test data
tag value, so test can get identified more easily.
Change-Id: Ifeb42ed11ce9a8efc6491b2c3b9fe31802a67d33
We're not dealing with binary plist files, to biplist is not the one we
want. Also, plistlib is part of Python.
Change-Id: I529516da0d6b04809a7c0d38b20c6f40900560f8
Since we're using signals to connect to the Logger we don't need an
interface class, and the only thing still used was the log level enum.
Put those values in a dedicated header instead.
Change-Id: Ic3aee14ba13b81e283b477d11b1c15a5c26ffeab
We don't need this file, and the implementations are stubs, causing a
compile warning. Remove it from the build.
Change-Id: I4ee2594c8fbc3ecf4661cc4144f8d9072bd6d983
Getting the values from a QMultiMap() returns a list, but not the same
on consecutive calls. Thus calling it twice to get the iterators will
yield different lists, causing the iterator to fail.
Fixes a segfault when destructing the logger.
Change-Id: Ie8ea8d3a9acf6fbd3969fbb49cbc50a1c1fa2854
Add a "deploy" target that will create a distributable file.
- Linux: AppImage.
- Windows: zip file.
- MacOS: dmg.
Change-Id: Id8ae9c021bc5bbb1abf066205b57d943c3f3b327
The test timeout is used to abort the tests if something goes wrong and
should not occur during normal test run. On some systems the current
value is too small, causing a timeout and thus test failure before
HttpGet responds. Increasing the timeout fixes this.
Change-Id: Ib61dd97d9f866c03cca9d2f0d7cb151f7076410d
Simplify, and replace use of QRegExp with QRegularExpression for Qt6
compatibility.
Also fix the test running on Windows. RockboxInfo constructs the
filename from path and filename, so we cannot pass an empty path, since
that results in an invalid path. On Linux / MacOS this works only
because we use an absolute path.
Change-Id: Ieaf30a2df005291d3e997aabf42d64ee832381c2
Rewrite in cmake, and include all necessary libs. This removes calling
separate Makefiles which is rather error-prone on non-Linux systems.
Change-Id: Ic2af4a3810e4b000e3ad33304229e2349a930e16
The former is not part of Qt6 anymore, while the latter has been
introduced with Qt5. We don't support Qt4 anymore, so move to
QRegularExpression.
Change-Id: Icc15ef422790f3740660c5581294aa0536f46b1f
rbutil uses several components from the utils folder, and can be
considered part of utils too. Having it in a separate folder is an
arbitrary split that doesn't help anymore these days, so merge them.
This also allows other utils to easily use libtools.make without the
need to navigate to a different folder.
Change-Id: I3fc2f4de19e3e776553efb5dea5f779dfec0dc21
It turns out the prescaler fields in OST_CTRL are 2 bits wide,
not 3. The programming manual (as usual) is ambiguous and its
diagram shows 2-bit wide fields, but the bit positions in the
text give a 3-bit wide field. Ingenic's Linux code and my own
testing shows that they are, in fact, 2 bits wide.
This caused the OST2 divisor to be 16 instead of 4; the OST1
divisor was correct. This means that all udelay/mdelay calls
took 4x longer than they should've. After this change the OST2
prescaler will be 4, as intended, and udelay/mdelay calls will
wait for the intended duration.
Change-Id: I2ac0a9190f49b59a840c649bf586131f5f9fde82
