The user has to specify the language they're posting in, and sometimes
they might get it wrong (e.g., replying to a post that also had the
language set incorrectly, forgetfulness, etc).
This has accessiblity issues (only following statuses in a given
language fails, translation can fail, etc).
Prevent this by trying to detect the language the status is written in
when the user tries to post it. If the detected language and the set
language do not match, and the detection is 60+% confident, warn the
user the status language might be incorrect, and offer to correct it
before posting.
How this works differs by device and API level.
- API 23 - 28, fdroid and github build flavours
- Not supported. A no-op language detector is used.
- API 29 and above, fdroid and github build flavours
- Uses Android TextClassifier to detect the likely language
- AP 23 and above, google build flavour
- Uses ML Kit language identification
To do this:
- Add `LanguageIdentifier`, with methods to do the identification, and
`LanguageIdentifier.Factory` to create the identifiers.
- Inject the factory in `ComposeActivity`
- Detect the language when the user posts, showing a dialog if there's a
sufficiently large discrepancy.
The ML Kit dependencies (language models) will be installed by the Play
libraries, so there's some machinery to check that they're installed,
and kick off the installation if not. If they can't be installed then
the language check is bypassed.
Update the privacy policy, as the ML Kit libraries may send some data to
Google.
`ImageDownsizer.downsizeImage()`:
- Remove the return value, it was ignored
- Throw `FileNotFoundException` when `openInputStream` returns null
`ImageDownsizer.getImageOrientation()`:
- Throw `FileNotFoundException` when `openInputStream` returns null
`MediaUploader.prepareMedia()`:
- Copy URI contents using Okio buffers / source / sink
`UriExtensions`:
- Rename from `IOUtils`
- Implement `Uri.copyToFile()` using Okio buffers / source / sink
- Replace `ProgressRequestBody()` with `Uri.asRequestBody()` using Okio
buffers / source / sink
`DraftHelper.copyToFolder()`
- Use Okio buffers / source / sink
`CompositeWithOpaqueBackground`
- Use constants `SIZE_BYTES` and `CHARSET` instead of magic values
- Use `Objects.hash` when hashing multiple objects
Based on work by Christophe Beyls in
- https://github.com/tuskyapp/Tusky/pull/4366
- https://github.com/tuskyapp/Tusky/pull/4372
Implement suggestions as a new `feature:suggestions` module, with
associated activity, fragment, etc.
Suggested accounts are shown with their normal information, as well as
information about the number of follows / followers, and a guide to
posting frequency, so the user can make a more informed decision about
whether to follow or not.
Crash was occuring because the instance info hadn't been fetched, trying
to take the last item of an empty list.
To fix:
- Expose the instance info as a state flow, with a default. New instance
info is fetched whenever the active account changes.
- Do the same for the emojis supported by the server.
- Update call sites as appropriate.
- Mark `InstanceInfoRepository` as `@Singleton` so it isn't repeatedly
created causing fresh content fetches.
The tests needed updating to get this to work.
- Extract the network fake modules in to a network-test module so
multiple other modules can use them.
- Rewrite `InstanceInfoRepositoryTest` to use Hilt and use Turbine to
test the new flows.
Checking this showed cosmetic bugs in the About layout when instance
info is missing, clean those up.
`TabData` recorded the type of the timeline the user had added to a tab.
`TimelineKind` is another type that records general information about
configured timelines, with identical properties.
There's no need for both, so remove `TabData` and use `TimelineKind` in
its place.
`TimelineKind` is itself mis-named; it's not just the timeline's kind
but also holds data necessary to display that timeline (e.g., the list
ID if it's a `.UserList`, or the hashtags if it's a `.Hashtags`) so
rename to `Timeline` to better reflect its usage. Move it to a new
`core.model` module.
Move `ListsActivity`, along with fragments and viewmodels, to a new
`feature:lists` module.
Previous code used the `item_follow_request` layout, which was not
ideal, so update it to use a dedicated layout, `item_account_in_list`.
The UI uses strings and views originally defined in the main app, so
move them elsewhere so they can be re-used.
- `BackgroundMessageView` moves to `core.ui`.
- `Lazy` moves to `core.common`.
- `ThrowableExtensions` split; the extensions specific to throwables
from network activity move to `core.network`, others move to `core.ui`.
- `BindingHolder` moves to `core.ui`
- Shared drawables and strings move to `core.ui`.
Once desugaring is enabled it needs to be enabled for up/down the
dependency chain, so enable it in the shared configuration defined by
the build convention code.
Highlighted a failing test that wasn't being run, so fix that too.
Moshi is faster to decode JSON at runtime, is actively maintained, has a
smaller memory and method footprint, and a slightly smaller APK size.
Moshi also correctly creates default constructor arguments instead of
leaving them null, which was a source of `NullPointerExceptions` when
using Gson.
The conversion broadly consisted of:
- Adding `@JsonClass(generateAdapter = true)` to data classes that
marshall to/from JSON.
- Replacing `@SerializedName(value = ...)` with `@Json(name = ...)`.
- Replacing Gson instances with Moshi in Retrofit, Hilt, and tests.
- Using Moshi adapters to marshall to/from JSON instead of Gson `toJson`
/ `fromJson`.
- Deleting `Rfc3339DateJsonAdapter` and related code, and using the
equivalent adapter bundled with Moshi.
- Rewriting `GuardedBooleanAdapter` as a more generic `GuardedAdapter`.
- Deleting unused ProGuard rules; Moshi generates adapters using code
generation, not runtime reflection.
The conversion surfaced some bugs which have been fixed.
- Not all audio attachments have attachment size metadata. Don't show
the attachment preview if the metadata is missing.
- Some `throwable` were not being logged correctly.
- The wrong type was being used when parsing the response when sending a
scheduled status.
- Exceptions other than `HttpException` or `IoException` would also
cause a status to be resent. If there's a JSON error parsing a response
the status would be repeatedly sent.
- In tests strings containing error responses were not valid JSON.
- Workaround Mastodon a bug and ensure `filter.keywords` is populated,
https://github.com/mastodon/mastodon/issues/29142
Release builds normally strip out all logging to reduce the number of
disk writes and reduce UI jank.
These logs would still be useful in user error reports from orange
builds. To preseve them:
- Implement a simple `RingBuffer`.
- Create `TreeRing`, a `Timber` `Tree` logger that logs to a
`RingBuffer` instance in orange release builds.
- Create `TreeRingCollector`, called when ACRA reports are generated,
which includes the contents of the ring buffer in the report.
- Enable desugaring to allow the use of java.time libraries on older
Android versions.
- Disable ProGuard obfuscation of class names as the obfuscation adds
additional de-obfuscation steps when handling error reports from users.
Getting error reports with logs of strange behaviour is useful even if
the app doesn't crash.
Move crash reporting in to `core.activity`, and provide a menu option
(in orange builds) to trigger a non-fatal crash report that is handled
the same way (i.e., sent by e-mail) as a regular crash report.
`BaseActivity` has to be able to create and handle menus, so adjust
subclasses to call the superclass when necessary.
Update `tools/mvstring` to be able to move strings between different
flavour directories, not just `main`.
Continue modularisation by moving activities in the "About" feature to a
new `feature.about` module.
Implement `feature.about:
- Move `AboutActivity`, `LicenseActivity`, and `PrivacyPolicyActivity`
here.
- Update `markdown2resource` plugin to work with libraries
Implement `core.data`:
- Types and repositories used through the app
- Move `InstanceInfo` and `InstanceInfoRepository` here so they are
available to `feature.about`.
Implement `core.ui`:
- App-specific views, spans, and other UI content
- Move `ClickableSpanTextView` and `NoUnderlineURLSpan` here so they are
available to `feature.about`.
Continue modularisation by moving core activity classes that almost all
activities depende on to a `core.activity` module. This includes
core "helper" classes as well.
Implement core.activity:
- Contains BaseActivity, BottomSheetActivity
- Contains LinkHelper and other utility classes used by activities
Implement core.common.extensions:
- Move ViewBindingExtensions and ViewExtensions here
Implement core.common.util:
- Move BlurHashDecoder and VersionName here
Implement core.designsystem:
- Holds common resources (animations, colours, drawables, etc) used
through the app
- Import "core.designsystem.R as DR" through the app to distinguish
from the module's own resources
Implement feature.login:
- Move the LoginActivity and related code/resources to its own module
Implement tools/mvstring
- Moves string resources (and all translations) from one module to
another
The previous code unilaterally enabled filter functionality. Some
Mastodon-like servers -- like GoToSocial -- do not support filters, and
this resulted in user visible error messages when connecting to those
servers.
To fix this:
- Extend the set of supported server capabilities to include client and
server side filtering.
- Disable the filter preferences if the server does not support filters
and show a message explaining why it's disabled.
Extend the capabilities model to support this:
- Fetch server software name and version from the nodeinfo endpoints
(implementing the nodeinfo API and schema)
- Extend the use of kotlin-result to provide hierarchies of Error
classes and demonstrate how to chain errors and display more informative
messages without using exceptions.
Fixes#343
Add a dependency on ACRA (in orange builds only), and catch crashes.
The user is given the option to e-mail the crash report data to the
support address, and can view and edit/redact the data before doing so.
The previous code generally started an activity by having the activity
provide a method in a companion object that returns the relevant intent,
possibly taking additional parameters that will be included in the
intent as extras.
E.g., if A wants to start B, B provides the method that returns the
intent that starts B.
This introduces a dependency between A and B.
This is worse if B also wants to start A.
For example, if A is `StatusListActivity` and B is`ViewThreadActivity`.
The user might click a status in `StatusListActivity` to view the
thread, starting `ViewThreadActivity`. But from the thread they might
click a hashtag to view the list of statuses with that hashtag. Now
`StatusListActivity` and `ViewThreadActivity` have a circular
dependency.
Even if that doesn't happen the dependency means that any changes to B
will trigger a rebuild of A, even if the changes to B are not relevant.
Break this dependency by adding a `:core:navigation` module with an
`app.pachli.core.navigation` package that contains `Intent` subclasses
that should be used instead. The `quadrant` plugin is used to generate
constants that can be used to launch activities by name instead of by
class, breaking the dependency chain.
The plugin uses the `Activity` names from the manifest, so when an
activity is moved in the future the constant will automatically update
to reflect the new package name.
If the activity's intent requires specific extras those are passed via
the constructor, with companion object methods to extract them from the
intent.
Using the intent classes from this package is enforced by a lint
`IntentDetector` which will warn if any intents are created using a
class literal.
See #291
The existing code base is a single monolithic module. This is relatively
simple to configure, but many of the tasks to compile the module and
produce the final app have to run in series.
This is unnecessarily slow.
This change starts to split the code in to multiple modules, which are:
- :core:account - AccountManager, to break a dependency cycle
- :core:common - low level types or utilities used in many other modules
- :core:database - database types, DAOs, and DI infrastructure
- :core:network - network types, API definitions, and DI infrastructure
- :core:preferences - shared preferences definitions and DI
infrastructure
- :core:testing - fakes and rules used across different modules
Benchmarking with gradle-profiler shows a ~ 17% reduction in incremental
build times after an ABI change. That will improve further as more code
is moved to modules.
The rough mechanics of the changes are:
- Create the modules, and move existing files in to them. This causes a
lot of churn in import arguments.
- Convert build.gradle files to build.gradle.kts
- Separate out the data required to display a tab (`TabViewData`) from
the data required to configure a tab (`TabData`) to avoid circular
dependencies.
- Abstract the repeated build logic shared between the modules in to
a set of plugins under `build-logic/`, to simplify configuration of
the application and library builds.
- Be explicit that some nullable types are non-null at time of use.
Nullable properties in types imported from modules generally can't be
smart cast to non-null. There's a detailed discussion of why this
restriction exists at
https://discuss.kotlinlang.org/t/what-is-the-reason-behind-smart-cast-being-impossible-to-perform-when-referenced-class-is-in-another-module/2201.
The changes highlight design problems with the current code, including:
- The main application code is too tightly coupled to the network types
- Too many values are declared unnecessarily nullable
- Dependency cycles between code that make modularisation difficult
Future changes will add more modules.
See #291.
Nik Clayton