Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

011018_843AM_addingChangesFromCn1Repository #18

Merged
merged 217 commits into from
Oct 1, 2018

Conversation

DurankGts
Copy link
Owner

No description provided.

shai-almog and others added 30 commits June 26, 2018 07:16
…g. Regression was introduced in 5d1824d .  In order to fix this, we had to move some some display properties regarding notification channels into build hints.  These properties haven't been documented yet.

This also includeds fix for Display.execute(filePath) in SDK 27.  #2461
Both of these fixes require forthcoming patches to the build server also.
…es support in android and iOS for attaching images to push notifications, but it also includes all of the groundwork necessary to expand this to other features of rich push notifications, like actions.

This also includes some improvements to the push simulator to support all of the different push notification types.
This commit is only part of a series of commits to various projects for adding support for rich push notifications.  The build server, certificate wizard, Javascript port, and push server all require updates - commits forthcoming for those projects.

Support for push notifications required a new push type: 99
The message body for push type 99 is an XML snippet of the form:
<push type='X' body='thebody' category='category'><img src='https://example.com/image.jpg'/></push>

The type and body attributes correspond with the push type of the main push content.  In this way, all of the existing push types can be expressed using this structure, and they can all have rich content added to them.  E.g.

<push type='1' body='Hello World'/>

Would be the same as just sending a message with body 'Hello World' and type=1.
…upport on iOS and JavaSE. This also adds improvements to the push simulator to show a notification if the app is paused with flow and content similar to how push notifications are received on device when the app is closed. Push notifications that include a supported category will be shown in a JOptionPane (if the app is paused) with options for each action in the category.

A forthcoming commit to the IPhoneBuilder is necessary for this new functionality to work.
… allows files to be opened in the gallery on the simulator in place rather than making a temp copy of the file. shannah/cn1-filechooser#1

This is required for this commit to work in the CN1FileChooser project. shannah/cn1-filechooser@8e94fa3
…hcoming update to build server and push server in order to work.
…t tests don't fail in time between now and when build server is updated.
…i-colon-delimited list of frameworks that should be treated as optional ParparVM. Frameworks included in this list also need to be in add libs list - it acts as a filter on that list ot make the frameworks optional. UserNotifications.framework is set as optional by default. This should fix #2469 / https://stackoverflow.com/questions/51197720/my-cn1-app-crash-on-startup
Fixed image screenshots to include the background of the form
…CM classes that are injected by the build server. #2389
…Ref #2467  This doesn't seem to work yet, but the ultimate solution will include these modifications and they are benign so, I'm committing them now.

This commit also includes a change to JavaSE port for password fields so that they use JPasswordField in the simulator rather than JTextField.
Made StringBuilder.append(char[]) public.
Fixed javadocs for StringBuilder to remove references to StringBuffer -- apparently they were largely copied over from StringBuffer.
Removed some methods from StringBuilder that were clearly mistakes.  In particular StringBuilder StringBuffer(CharSequence) which appears to have been intended to be a constructor, but in the original copy/paste from StringBuffer it was turned into a member method - and the constructor that it was supposed to be was already implemented separately.  #2465
Implement this RFE: #2458
shai-almog and others added 28 commits September 13, 2018 23:14
…tandard browser navigation callback runs into trouble with there are too many callbacks happening too quickly - some tend to get dropped. On iOS, this improved mechanism requires the JavascriptCore library to be linked so it is disabled by default. It can be enabled using the CN1_USE_JAVASCRIPTCORE define which can be uncommented in CodenameOne_GLViewController.h. This will be enabled in the build server using the ios.useJavascriptCore=true build hint which is false by default. Eventually this should probably be on by default, but I left it off since JavascriptCore requires iOS 7 or higher.
…red in front of the CN1 canvas. This is because z-indexing is flakey on UWP. This flag can be enabled with the build hint windows.renderWebviewsInFront=true.

Fixed onLoad event in UWP web view so that it fires consistently when the DOM is finished loading.
Fixed Browser.executeAndReturnString on UWP.  Previously, it didn't work at all.
…prepare to play. This is necessary to download the poster frame of a video.
just for the case of debugging. When e.g. a request fails and the url of the request should be logged
… the max texture size. This should give the developer a clue when images are rendered completely black as to what the problem is.

Added define flag that will optionally auto-scale textures that are too large  -- this is set to off and there is no corresponding build hint as this is probably a bad idea... but it is there to be turned on if we change our minds later.
Added fallback in DrawImage.m so that it an image is too large, it will use the size that has been specified as the rendered size.  When drawing images, it tries to rely on OpenGL for scaling if possible.  This means that if you load a very large image, but call scaledWidth(100) and try to render that, it will still create a giant texture.  Not sure if it is better to always just create a 100px texture in this case, but, for now, I have made it fall back to 100px only if the actual image size exceeds the max texture size.  This is probably an area thatdeserves experimentation to see if there is a better heuristic to maximize performance and minimize memory usage.
Updated javadocs for Image.createImage(String) to indicate that this will work for FileSystemStorage.
Applied small fix in iOS implementation of createImage(String) to handle file paths properly.  It may have already worked in some cases - but did not work in others.
…photo to fail. #2550

Fixed deadlock when using capture photo on ios device that doesn't have a camera (e.g. the iOS simulator).
Fixed some issues with ios gallery multiselect.  Previously, if the user selected an image that is stored on icloud, it would fail.  Now it will correctly download the image automatically.
…he same time that the 'next' button is pressed on Android. #2544
… This ends up playing nicer with other audio sources in the app so that they don't interrupt each other. This is configurable with the display property ios.useApplicationAudioSession='false'.
…olors don't require image border generation.
…r all of the corners have the same radius, or top corners have the same radius (and bottom has no radius), or bottom corners have same radius (and top has no radius) - otherwise it falls back to image border. Ultimately we should move to supporting topRightMode, topLeftMode, bottomRightMode, and bottomLeftMode for independent corner radii, but the resource file currently doesn't support these settings... and didn't want to delve into that right now.
…when a component in the view heirarchy sets a transform and doesn't clean up afterward. This fixes issue #2557

It is possible that this papers over other issues with Android related to clipping and transforms - but it fixes this test case.
…ted() since the isSubscribed() method will work as long as a receipt store is implemented.
…ent's parent's background would be painted over top of the component's siblings when painting its background. This caused some elements in the form layer pane to disappear during animations. Fixes #2564
@DurankGts DurankGts merged commit 9376eb1 into DurankGts:master Oct 1, 2018
DurankGts pushed a commit that referenced this pull request May 31, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants