Commit graph

15531 commits

Author SHA1 Message Date
Dirk Hohndel
26de89364c Revert "Windows: build against user space FTDI library"
This reverts commit 53341c037d.

The commit isn't wrong, but it breaks Travis and I can't seem to get the
newer MXE build to work on Travis. So while I figure out how to work
around THAT, let's just revert this and come back to it once Travis is
ready.

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-10-02 13:23:29 -07:00
Dirk Hohndel
ac8936004e update-manager: actually remember that we asked the user
Otherwise we get that annoying question every time the git version changes.

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-10-02 13:20:44 -07:00
retiform
251d2589ad Fix typo in INSTALL file.
Changes git submoule to git submodule. In the INSTALL file.
Just a minor spelling fix.

Signed-off-by: Cole Rogers <colerogers@protonmail.com>
2018-10-02 19:28:18 +02:00
Jocke
b29f6ef2c0 Mobile: remove all related data when forgetting DCs
We need to delete all related data when forgetting dive computers or we will
have an issue if we connect a DC from the same vendor but of a different model.

Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
2018-10-02 18:40:47 +02:00
Dirk Hohndel
c31d517b8f Update to latest libdivecomputer
Retry packet send for BLE connections for Aqualung i770R.

(oops, Linus pushed right when I pushed the previous update)

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-10-01 16:32:47 -07:00
Dirk Hohndel
6a1d17534a Update to latest libdivecomputer
Add initial support (BLE only) for Aqualung i770R.

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-10-01 15:59:08 -07:00
Dirk Hohndel
79d0185c62 Bluetooth: use_random_address helper isn't needed on Windows
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-10-01 15:55:19 -07:00
Jocke
a8953da02d Mobile download from DC: show the correct DC rather than the latest
Since we now keep track of up to 4 DCs we don't want to display the last used one
but rather the one that is connected.

Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
2018-10-01 06:32:41 -07:00
Jocke
c504c08097 Mobile download from DC: improve selection logic
Improve the logic when auto selecting a DC for download.
Some USB cables only supply vendor information  but we can select the correct
model if we have downloaded from it before.
For BT/BLE our discovery process adds the device name to the address, so we need to keep that in mind when we try to match against what we seen before.
When we have a positive match for a DC we have seen before we deactivate the
corresponding button of our saved DCs.

Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
2018-10-01 06:32:41 -07:00
Jan Mulder
9b199602c6 Mobile QML UI: bump Kirigami imports to 2.4
Primarily for reasens of keeping up with upstream. And hopefully
bugfixes and added functionality.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-10-01 05:29:46 -07:00
Jan Mulder
5681f686ba Revert "Revert "QML UI: newer Kirigami and fallout""
This reverts commit a5db03c2b3.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-10-01 05:29:46 -07:00
Jan Mulder
e321d1563e QML UI: clip divedetails Listview
A technically trival commit, but one with long story. This commit
basically reverts dd1d90b529 (1.5 year ago). While upgrading
Kirigami after Kirigami commit 26b8bdea24c39, we suddenly have
overlapping divelist and details pages in case they are both
on the pageStack (this occurrs when navigating from divedetails
to the divelist using breadcrumb navigation). At this point, its
not clear (to me) if this the by design of Kirigami, or an unintended
effect of the mentioned Kirigami commit.

This all said. Simply clipping resolves our issue of overlapping
pages, and it does not harm.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-10-01 05:29:46 -07:00
Jan Mulder
bec029c766 QML UI: trivial resize of datebox
Something that I simply overlooked earlier with respect to scaling the
divelist. The trip databox did scale a bit, but it was not nicely
related to the hight of the trip header. So there was a tiny
overflow on the small scale on a small device. Fixed here.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-10-01 05:28:18 -07:00
Dirk Hohndel
53341c037d Windows: build against user space FTDI library
This needs more testing, but should provide an alternative to relying on
the Windows FTDI driver.

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-30 08:28:16 -07:00
Dirk Hohndel
f0ecd1fbb7 Windows: enable BLE support
This currently requires the wip/win branch of qtconnectivity. Eventually
this should be merged into upstream Qt.

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-30 08:28:16 -07:00
Jan Mulder
c6a4250a58 QML UI: correct trip separator line
Partially cosmetic and partially a bug fix. 1) the seperator line
between trips and dives that are not in a trip was drawn in the
background color => the line was invisible. 2) When looking very
closely, there was a 1-2 pixel wide error between the seperator line
between trips and dives that are not in a trip. 3) there was a comment
that the trip separator needed to be extra thick. IMHO, this
looks ugly, and is superfluous as there is a nice sidebar along
the dives that belong to the trip. Finally, the line shall not
be displayed when not in a trip.

So, basically, the line (the QML rectangle) is completely rewritten,
to take care of all issues. There is 1 hack: the line color is
taken from the dive separator line. But its fully unclear to me
where that color is defined in Qt/QML or Kirigami, so I hardcoded
the proper color. That just works.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-09-30 08:16:52 -07:00
Jan Mulder
40a65a2e6b QML UI: center delete from divelist button
A small cosmetic change. The delete from divelist button was "glued"
to the top of the line. Not nice, so just center it vertically, and
make the button a tiny bit smaller, so that it fits nicely on the line.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-09-30 08:16:52 -07:00
Jan Mulder
751286e4e7 QML UI, cleanup: no reason for 2 overlapping mouseareas
Fully usure why the code was as it was. The trip header had 2
overlapping mouse areas, to expand the trip and vise versa. Simply
remove the smallest one.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-09-30 08:16:52 -07:00
Jan Mulder
bce952615c QML UI, cleanup: do not use iconSizes for unrelated sizing
A very unimportant change, but found when looking through the code
for places where size of icons where used. The one changed here
has nothing to do with icon related placement of a string, so
its replaced by a way more logical placement of the affected string.

Simply center the "no dives in the dive list" for an empty logbook
on the screen, instead of at some random place in the upper left
corner. Like I said: very unimportant, but it just looks nicer
in the UI.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-09-30 08:16:52 -07:00
Dirk Hohndel
0422cd3662 Bluetooth: don't use random addresses on Windows
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-30 08:09:37 -07:00
Dirk Hohndel
1f59629a76 Bluetooth: use standard Qt code on Windows
We shouldn't need our hand crafted code anymore.

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-30 08:09:37 -07:00
Berthold Stoeger
d3d06bc580 Cleanup: split out free_dive() function from delete_single_dive()
Currently, we can only delete dives that are indexed in the main
dive table. In the future, we will have to delete dives outside
of this table (e.g. for undo). Therefore, split out the free_dive()
function from delete_single_dive(), which takes an index into
the main dive table.

In the process, adopt the dive freeing-code from clear_dive(),
which frees more data than the code in delete_single_dive().
This potentially fixes a memory-leak.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-29 15:24:08 -07:00
Berthold Stoeger
574065b314 Cleanup: reinstate override modifiers
This reverts commit 1c4a859c8d,
where the override modifiers were removed owing to the noisy
"inconsistent override modifiers" which is default-on in clang.

This warning was disabled in 77577f717f,
so we can reinstate the overrides.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-29 15:23:25 -07:00
Berthold Stoeger
d7e3d68f36 Cleanup: Devirtualize WebServices::downloadTimedOut()
This member function was not overriden in a derived class. No point
in it being virtual.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-29 15:16:11 -07:00
Berthold Stoeger
7c22e4d805 Cleanup: don't derive UserSurveyServices from WebServices
UserSurveyServices derives from WebServices and therefore has
to define three pure virtual functions [startDownload(),
startUpload(), buttonClicked()] as no-ops. Interestingly,
a comment in the header says
	"need to declare them as no ops or Qt4 is unhappy"
which is of course not true as these functions are not
declared by Qt.

There seems to be no point in deriving from WebServices,
therefore don't do it. These function definitions can then
be removed.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-29 15:16:11 -07:00
Dirk Hohndel
7c320854b3 Bluetooth: turn on logging earlier
This way we should get logging output for the discovery phase as well.

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-29 14:16:37 -07:00
Berthold Stoeger
fbc4b2ac72 Mobile: don't call clear_dive() on deleteDive
clear_dive() will be called anyway in the subsequent call to
copy_dive().

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-29 14:02:07 -07:00
Berthold Stoeger
41089b74a9 Cleanup: don't use QByteArray::data() to create copy
QByteArray::data() provides access to the underlying data
for direct manipulation. Thus, the construct
	csv = fileNamePtr.data();
found in MainWindow::importTxtFiles() suggests that modifications
to csv also affect fileNamePtr. This is *not* the case, because
csv itself is a QByteArray. It is therefore constructed from
the data.

Replace this treacherous construct by a simple assignment.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-29 14:01:22 -07:00
Berthold Stoeger
3923f54e10 Cleanup: don't needlessly recalculate dive list on CSV import
On CSV import, the dive list was recalculated after the import
dialog was shown. This is pointless, as no dives are yet imported.
Remove.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-29 14:01:22 -07:00
Berthold Stoeger
6e24762a6c Cleanup: Don't leak DiveLogImportDialog objects
The non-modal DiveLogImportDialog was only implicitly deleted when
the MainWindow was destroyed. Instead hook into the accept() and
reject() functions and schedule for deletion with deleteLater().

Quite the horrible proposition, but in line with Qt's object model.
Consider making the dialog modal instead. There seems to be no
upside for this being modal.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-29 14:01:22 -07:00
Dirk Hohndel
b80fd26e81 build-system: fix appdata creation
The pipe in the  execution apparently didn't fail, so the date wasn't set
when building from a tar file.

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-29 12:09:11 -07:00
Dirk Hohndel
3a79eb11e8 Update latest translations from Transifex
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-29 12:09:11 -07:00
Dirk Hohndel
e0e20edd87 Update list of supported dive computers
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-28 15:40:03 -07:00
Dirk Hohndel
d13f2ad8d5 Update README and ReleaseNotes for 4.8.3 / 2.1.4
Slight updates / rephrasing of the CHANGELOG entries as they get moved
to ReleaseNotes.

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-28 15:36:59 -07:00
Dirk Hohndel
dc4b0c910e Pull latest translations from Transifex
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-28 15:23:56 -07:00
Dirk Hohndel
7708492232 user manual: update bundled HTML of Spanish translation
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-28 15:23:56 -07:00
Berthold Stoeger
32a4ce6169 Parser: parse text-based files into arbitrary table
In d815e0c947 a dive_table pointer
was added to the parsing functions to allow parsing into tables
other than the global dive table. This will be necessary for undo of
import and implementation a cleaner interface. A few cases, notably
CSV and proprietary formats were forgotten.

Implement parsing into arbitrary tables also for these cases.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-28 15:06:00 -07:00
Dirk Hohndel
a5db03c2b3 Revert "QML UI: newer Kirigami and fallout"
Turns out that Jan found an issue with the latest Kirigami, so let's go back to
the known good one.

This reverts commit 17ec95e70c.

Suggested-by: Jan Mulder <jlmulder@xs4all.nl>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-28 14:53:22 -07:00
Dirk Hohndel
e99df4af2c Revert "build-system: fix kirigami.pri for iOS build"
Turns out that Jan found an issue with the latest Kirigami, so let's go back to
the known good one.

This reverts commit 40766db459.

Suggested-by: Jan Mulder <jlmulder@xs4all.nl>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-28 14:51:54 -07:00
Jan Mulder
56b24707a5 CHANGELOG: update for mobile
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-09-28 11:09:17 -07:00
Dirk Hohndel
ce41324e68 Update translation source strings
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-28 09:02:20 -07:00
Dirk Hohndel
b69672429f Update libdivecomputer
Mares bluetooth support tweaks

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-28 09:01:47 -07:00
Berthold Stoeger
3137fbe973 Cleanup: remove redundant prototypes from dive.h
A few of these prototypes were already in import-csv.h.
Put them in an 'extern "C" { ... }' block.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-28 07:47:39 -07:00
Berthold Stoeger
2e6df623bb Cleanup: Make add_sample_data() local to translation unit
This function in parse_csv.c was not used anywhere. Make
it of static linkage.

Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
2018-09-28 07:47:39 -07:00
Robert C. Helling
43c9178bf7 Mac: Turn on automatic GPU switching
We are not OpenGL heavy so this saves battery life.

Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-28 07:34:50 -07:00
Dirk Hohndel
40766db459 build-system: fix kirigami.pri for iOS build
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-27 10:44:37 -07:00
Jan Mulder
17ec95e70c QML UI: newer Kirigami and fallout
Update to the master of today, and no issues detected on mobile-on-desktop
and Android.

Only, the ugly border is back as the magic hack of 0b16b547ae failed
due to the patch file that errored. So that is fixed too.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-09-27 08:42:47 -07:00
Dirk Hohndel
2d1ef04e2f Update to latest libdivecomputer
Fix dc_iostream_{read,write} debugging implementation
Mares Icon HD family: send the command as one single write buffer

Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
2018-09-27 06:52:31 -07:00
Jan Mulder
0ad7df9859 mobile, QML UI: Settings page scalability
A relatively big change for such a simple page. Most relevant
changes are:

1) Do not use Kirigami.Header anymore. It appears that this header
has word wrapping on (and we cannot override that). This is
annoying on this page, as headings seems randomly be split over
2 lines, even in cases where there is more than enhough room to
display it on one line. And as the Kirigami.Header is just a
trivial wrapper of a Text field, we can simple replace it.

2) A lot of the toplevel GridLayouts had width properties set. These
are not needed (and confused my debugging code), so they are removed
withput any visual change. As a general rule, do not try to set
properties that are not needed. In general, it can only lead to
binding loops or undefined behavior.

3) Add a font size to our Theme. The step from regular to title size
was a little too big.

4) And, obviously, numerous font.pointSize lines are added to actually
resize the font.

Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
2018-09-27 06:36:22 -07:00
Linus Torvalds
890d4c3d64 qt-ble: allow reading of partial packet data
The existing BLE dive computers treat BLE as the packetized protocol it
is, and read whole packets at a time.

However, the Mares BlueLink backend treats it as just a basic "serial
over BLE" transport, and for historical reasons reads the reply packets
in smaller chunks.

This allows that kind of IO behavior, where if the divecomputer backend
reads just a part of a packet, we'll split the packet, return the part
the user asked for, and push back the leftover packet onto the received
packet queue.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-09-26 12:13:20 -07:00