If the QML modules for QtLocation and QtPositioning are
missing the QML in mapwidget.cpp will fail to load,
which can lead to crashes.
To solve the issue check if the QML has loaded and set
a flag 'isReady' to true. If the loading has failed
load another QML which is for showing a red error text
in the lines of `MapWidget.qml failed to load!`.
If the map QML has failed, use a macro in all relevant
MapWidget members to turn them into a NOP. This approach
leaves the rest of the codebase intact - e.g. no checks
in classes which connect to the MapWidget class.
Fixes#596
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
This commit is empty when shown with '-w' - it just updates the
indentation after commit c00804eff6 ("QML UI: always start edit at top
of page").
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When starting to edit / add a dive, the Flickable needs to be positioned
at the top of the page, not the last position shown.
For clarity I'll do the re-indentation in the next commit.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The hard switch was not ideal. This isn't perfect, yet, but a step in
the right direction. The 'transitions' to change the visibility
properties are a bit odd, but that's how it's done in the examples as
well.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We need a white path drawn on the dark action button, but a black path
drawn for the main menu.
(looks like a white space change snuck in here)
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This addresses some review comment on whitespace and translated
string formatting.
In the string formatting, a tiny additional change is made.
I wanted the email address in the explanation text in a bold
font. Using the HTML <b> for this, removed the /n newline
characters in the output. Apparantly, mixing these two
formatting styles does not work. No problem, replaced the
/n to HTML style too.
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
This commit tries to implement most of issue #515. It reworks the
one credential page, which its dynamic PIN part, into two pages.
Main driver of selecting one of the two pages is the showPin
boolean. Page 1 contains the email/passwd field (and the
option to use a no cloud setup). Page 2 only contains the PIN
part (and the option to cancel the process).
The Kirigami central button does not seem very handy here. We
need, for example, a cancel, sign-in and register, only register,
etc. buttons, which are not easy to handle in specific icons.
Therefore, normal pushbuttons are chosen to deal with user
interaction, and the Kirigami button is removed from these
pages.
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
Previously, we could edit the cloud credentials in basically two
places. At startup of the app from a fresh install (and no previous
data on the device), and from the settings. Issue #515 proposes
only one way.
However, we need a way to access the new credential UI pages, so
that the pages at a fresh install of the app can be reused,
for example for account switching.
This commit replaces the settings cloud credentials block by
a simple (not editable) display of the current credentials, and
a button to access the initial pages, for all management tasks
on the credentials.
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
It appears that the onCompleted of the StartPage item is triggered
before the onCompleted of the rootItem. This is logical as the
Startpage is a child of the rootItem. And, yes, this does matter.
As the divelist also contains the logic for initial cloud
registration (and is the default page shown in a state where
the cloud credentials are valid (CS_VERIFIED state)), we need to
know the correct credential state at start of the app.
The move of this one line of code makes sure of that, in addition
to setting the credential state from the preferences. Now, the
setupActions function can reference correct credential data.
This is further preparation for a better cloud creation
process from mobile.
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
Following on beb0d5703a, the context menu seems to work fine
with a much older QtQuick import - version 2.0.
The 2.7 import is technically a development leftover
and a minimal version should have been considered earlier.
On older Qt setups (e.g. 5.5.x) this might throw a:
'module "QtQuick" version 2.6 is not installed'
Reported-by: Stefan Fuchs <sfuchs@gmx.de>
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Having two different enums around with more or less the same
definition has lead to unclear code. After removing two not needed
states on the mobile end, the remaining step to one enum for the
credential state becomes almost is simple rename operation.
Unfortunately, I do not know a way to embed a plain C enum
from pref.h into the QMLManager object. So after this, there
are still 2 enums around, but now identical.
This commit is not changing any functionality.
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
This is a no-brainer removal of the VALID_EMAIL state used in QMLManager.
All current usage of this state is "if state is VALID or VALID_EMAIL",
so there is no distinction between the two states.
It is even a little different. The comment suggests "when we can open
a local cloud storage, tied to a cloud account (so explicitly not
the no-cloud status), we have at least a valid email". While this
is formally true, this implies that there is also a cloud account
on the cloud server (ie. the cloud account is in a VERIFIED state).
In other words: currently, there can't exist a valid local storage
that is tied to a valid email adress, without valid cloud account
on the server.
Notice that this touches the discussion on GitHub for commit
e76f527fe5 (pull request #520). Can we implement the creation
of a valid cloud account without data link to the cloud server?
Currently, we need the server to confirm the email address (for
example for uniqueness reasons on server side). Obviously, we could
hack our way out of this, but we have a perfect solution already
in place. Create a no-cloud account, and transfer that later to
a true and valid cloud account.
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
The googlemaps plugin is about to support tile language
via the PluginParameter 'googlemaps.maps.language'.
To be able to pass the subsurface UI language (obtained
from the uiLanguage() helper) the Plugin has to be created
dynamically, only *after* the MapWidgetHelper is created.
MapWidgetHelper::pluginObject() now provides a QString which
contains the Plugin object and also include the uiLanguage
ISO value.
This string is used in mapwidget.qml as:
map.plugin = Qt.createQmlObject(pluginObject, rootItem)
This creates the Plugin object dynamically with the proper
UI language string, but also requires a couple of small changes:
- move the declaration of map.mapType after the
Qt.createQmlObject() call
- assign map.activeMapType after map.mapType has a value
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
The QML Map has a couple of methods - toCoordinate(someQPointF) and
fromCoordinate(someQGeoCoordinate). Ideally, if one passes a point to
toCoordinate() and then attempts to convert the resulted coordinates
back to the point, the same point in the Map view port with minimal error
should be retrieved. That's not always the case - e.g. near
47.400200 -123.142066, which means that the methods are not exactly
*reliable* and there might be Map class bugs at hand.
The new zoom-in and zoom-out improvements try to work around the above:
- for both centerOnRectangle() and centerOnCoordinate(), if no good
zoom-out level is found (newZoomOut), the zoom-out is set to the default
value of defaultZoomOut. In practice, this prevents the case where the map
does not zoom-out at all when going from one place to another
- centerOnRectangle() now uses rectangle diagonals to estimate a fit,
instead of checking if 2 points (top-left and bottom-right) are visible
in the viewport. The usage of fromCoordinate() was giving bad results
in this case and comparing distances (diagonals) is more reliable
but more expensive on the CPU.
Due to the inconsistencies of toCoordinate() and fromCoordinate()
the dynamic zoom-in and zoom-out are still not ideal, but the current
implementation is somewhat usable with decent accuracy.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
This appears to fix the mystery crashes that can occur when deleting a dive
from the dive list.
Fixes: #497
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
This gets rid of TypeError messages, but does not appear to affect behaviour
qrc:///qml/DiveDetailsEdit.qml:254: TypeError: Cannot read property 'dive' of null
qrc:///qml/DiveDetailsEdit.qml:228: TypeError: Cannot read property 'dive' of null
qrc:///qml/DiveDetailsEdit.qml:216: TypeError: Cannot read property 'dive' of null
qrc:///qml/DiveDetailsEdit.qml:204: TypeError: Cannot read property 'dive' of null
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
MapWidgetContextMenu now has a new action (SELECT_VISIBLE_LOCATIONS),
that will invoke the MapWidgetHelper method selectVisibleLocations().
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
First, this function calculates the zoom out effect until both the
current Map center and the target rectangle are visible - see the
"calculate zoom out" part.
Then it calculates a zoom level, so that the target rectangle
fits the viewport, but also so that the zoom is not too much (clamped).
see the "calculate zoom in" part.
NOTE: "centerStored" (the variable used to store the current map center)
is created using QtPositioning.coordinate(), because the code needs a new
object and not a reference of the map.center QGeoCoordinate object.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Add stopZoomAnimations(), which is really just a precaution
function to make sure we have stopped all animation of the
map before doing calculations! Since the animation *should*
be running in a separate thread, this would make sure the Map "center"
and "zoomLevel" properties do not change.
Add pointIsVisible(), which is a helper to determine
if a point created by map.fromCoordinate() method is inside
the viewport. fromCoordinate() has to be called without the
"false" (clip) flag for this to work.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
MapWidgetHelper::centerOnDiveSite() now checks if more than
one dive sites are selected and finds the most top-left and
bottom-right ones in the coordinate system to form
a rectangle.
It also supports the special cases where a selected dive site
does not have coordinates or the case where only a single dive site
with GPS coordinates are selected.
TODO: implement mapwidget.qml::centerOnRectangle()
This QML function will receive a QGeoCoordinate based rectangle which
has to be centered in the viewport with animation.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
For instance, if the 'googlemaps' geoservices plugin is missing
a new message is now shown:
"qml: MapWidget.qml: cannot find a plugin with the name 'googlemaps'"
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Do so until the following issue is resolved:
https://github.com/vladest/googlemaps/issues/9
With the satellite map ([1]), the black tiles at relatively high
zoom levels over the Red Sea cannot be seen. Instead it shows white
tiles at the ultra-maximum zoom level at places.
As a side note, the "satellite" map does not have the country labels
which the "hybrid" map has.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
No idea why this was restricted to be so narrow. Also, making
the font of the label smaller and lighter seems visually more
pleasing.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This makes sure that the autocompleting comboboxes never displays the scroll list where the user can pick an entry, and as such forces the view of a autocompleting textfield from the users point of view. The reasoning is that when scrolling in the drop down list it it possible to scroll to far and thus abort the ongoing edit by flicking away the edit page.
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
Mobile devices hare rather limited when it comes to screen size,
there is no point in reducing the availabel space more than neccecary.
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
This handles a cornercase where a user starts editing and then cances the divedit followed by a return to edit mode. The previously acive textfield would then be active again even though all changes are lost.
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
For some reason the textfields on dive edit page did not loose focus when editing ended by press on the return key.
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
When the user makes a selection using the autocomplete function while editing a dive the keyboard should close and the selection list should disappear.
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
Add star icons from the material repository.
https://material.io/icons/
The icons are under Apache License Version 2.0
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
This patch makes use of the geolocation plugin "googlemaps"
for Qt Location with source code from here:
https://github.com/vladest/googlemaps
The change from the ESRI plugin is that it requires new indexes
for the Hybrid ([3]) and Street ([0]) map types.
There are more zoom levels in this plugins, but our default zoom-in
settings seem to transition well to it.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
