In the previous code `PachliError` could correctly chain errors and
generate error messages, `ApiError` didn't, which is why there was the
temporary `ApiError.fmt()` extension function.
Rewrite `ApiError` to implement `PachliError` so it gets these benefits
and to reduce the number of different error-handling mechanisms in the
code.
Main changes:
- `PachliError` is now an interface so it can be extended by other
error interfaces.
- All the `ApiError` subclasses implement `PachliError`, and can
specify the error string and interpolated variables at the point of
declaration.
- Update `ListsRepository` and `ServerRepository` to return
`PachliError` subclasses.
Previous code was inconsistent about using getServerErrorMessage() and
whether or not the case where getServerErrorMessage() returns null was
handled.
Switch to using getErrorString() which does handle the null case and
always returns a usable string.
Some string resources are rendered temporarily unused by this change.
They will be used again soon, so configure lint to ignore them at the
moment.
Previous code always included the transitionKind enum as an extra, and
could cause a crash if the activity launched by the intent was not a
Pachli activity.
Fixes#700.
Edited polls only include the list of options with titles; no other
metadata (poll ID, single/multiple choice, vote counts, etc). Since the
data shape didn't match Moshi wasn't decoding the data.
Provide dedicated data classes to model the response, and add a fourth
poll display option to represent viewing an edit history snapshot.
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.
Previous code used a single animation type (slide) when transitioning,
the transition was quite slow, and didn't behave appropriately if the
device was set to a RTL writing system.
In addition, handling the back affordance didn't work well with the new
"Predictive Back" feature in Android 14
(https://developer.android.com/guide/navigation/custom-back/predictive-back-gesture).
Fix this.
## Transitions
To update the transitions the `startActivityWithSlideInAnimation()`
implementation (and associated `finishWithoutSlideOutAnimation()`) have
been replaced.
There are three transitions; `default`, `slide`, and `explode`,
represented as an enum passed to the activity intent.
The `default` transition is the activity transition from Android 14,
from the Android open source project
(https://cs.android.com/android/platform/superproject/+/android-14.0.0_r18:frameworks/base/core/res/res/anim/;bpv=1).
This is used for most transitions.
The `slide` transition is the pre-existing slide transition, with a
shorter transition time so it feels more responsive, and an RTL
implementation. This is used when there is a strong spatial component to
the navigation, for example, when going from:
- a status to its thread
- a preference menu item to its subscreen
- a filter in a list to the "edit filter" screen
- viewing your profile to editing your profile
The `explode` transition is used when the state of the app changes
significantly, such as when switching accounts.
Activities are now started with `startActivityWithTransition()` which
sets the intent and prepares the transition. `BaseActivity` checks the
intent for the transition type and makes further changes to the
transition as necessary.
## Predictive back
"Predictive back" needs to know what the back button would do before the
user interacts with it with an `onBackPressedCallback` that is either
enabled or disabled. This required refactoring some code (particularly
in `ComposeActivity`) to gather data ahead of time and enable/disable
the callback appropriately.
## Fixed bugs
- Back button wasn't stepping back through the tabs in AccountActivity
- Modifying a filter and pressing back without saving wasn't prompting
the user to save the changes
- Writing a content warning and then hiding it would still count the
text of the content warning toward's the post's length
## Other cleanups
- Use `ViewCompat.setTransitionName()` instead of setting the
`transitionName` property
- Delete the unused `fade_in` and `fade_out` animations.
- Use androidx-activity 1.9.0 to get the latest predictive back support
library code
- Show validation errors when creating / editing filters
The previous code didn't set a limit for the number of posts, links, and
hashtags to fetch on the trending pages, so used the conservative
defaults.
Increase these to the API maximums to show the user more information.
Abstract common CI setup tasks (setting up Java, Gradle, etc) in to a
single action that can be used by all CI workflows.
Run the lint, test, and assemble CI tasks in parallel for each variant
rather than in series, which cuts ~ 7 minutes (approx. 50%) off the CI
runtime.
Update code in checks and core/navigation to fix new tests.
Use `unicodeWrap` when inserting placeholders in error messages so they
set the correct text direction.
Update some strings with formatting directives to (a) include `_fmt`
in the name, and (b) use `%1$s` instead of `%s`.
Previously, modifying any tabs meant opening the left-side nav, opening
Account preferences > Tabs, and then adding / removing tabs. This is
time consuming, and difficult for new users to discover.
In addition, it was possible to remove the Home tab, and there was a
hardcoded minimum of at least two tabs.
Fix this.
When viewing a timeline that is not already in a tab an "Add to tab"
menu item is enabled, which appends the timeline to the list of existing
tabs.
When viewing a timeline in a tab (that is not the Home timeline) a
"Remove tab" menu item is enabled, which removes the tab from the list
of existing tabs.
If the user removes the active tab (either with this menu item, or
through preferences) the tab to the left of the active tab becomes the
new active tab.
A new "Manage tabs" menu item is also provided, as a shortcut to the
existing Account preferences > Tabs screen.
When managing tabs the Home timeline can not be removed; the button to
remove it is removed, and swiping is disabled on that list item. The
restriction of "at least two tabs" has also been removed.
`NotificationsActivity` has been removed, as `TimelineActivity` can
display `NotificationsFragment`.
To make the three "Trending" types (hashtags, links, and posts) more
visually distinct add two new icons for links (ic_newspaper) and posts
(ic_whatshot).
Fixes#572, #584, #585, #569
Mastodon API uses an "empty" `expires_in` value for a filter to mean
"Does not expire" (i.e., indefinite).
This was modelled as a null. Which doesn't work, because Retrofit does
not send name/value pairs in encoded forms if the value is null.
Fix this by making the API type a `String?`, and explicitly using the
empty string when indefinite expiry is used. This has to be converted
back to an Int? in a few places.
See
https://github.com/mastodon/documentation/issues/1216#issuecomment-2030222940
Previous code was inconsistent about how and when the FAB was hidden if
the user had set the relevant preference.
- Sometimes the FAB has hidden by setting the visibility to false, which
removed it with no animation.
- Sometimes the value of the preference was checked once, when the
fragment or activity was created.
- Some timelines didn't show the FAB (Hashtags, Favourites, Bookmarks,
TrendingLinks, TrendingStatuses).
- Logic for figuring out which `ComposeActivity` intent to use was
scattered across different files.
Improve this by:
- Expose changes to the `FAB_HIDE` preference in the relevant viewmodels
as a flow the UI component can collect.
- Centralise the show/hide logic in a new `ActionButtonScrollListener`
class, and always using `show()`/`hide()` to animate the transition.
- Centralise the logic for creating the `ComposeActivity` intent in
`TabViewData`.
`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.
Previous code would handle some expected exceptions (IO, HTTP) when
fetching a timeline, and show them to the user. Any other exception
would crash.
Now, surface all exceptions. Treat IO and HTTP exceptions as retryable
and show the "Retry" option, all others are considered non-retryable.
Provide a specific error string for exceptions caused by bad JSON.
Previous code used custom regular expressions to extract URLs, hashtags,
and mentions from text while the user was writing a post. These were
inconsistent with the ones that Mastodon uses so the derived character
count could be wrong.
As well as being visually incorrect this could prevent the user from
posting a status that was within the length limit, or allow them to
attempt to post a status that was over the length limit (which would
then fail).
Fix this by dropping the homegrown regular expressions and using the
same text parsing library that Mastodon users; twitter-text. This has
been converted to Kotlin and the functionality related to Twitter
specific features has been removed.
The hashtag handling has been adjusted, as Mastodon is more permissive
about the positions where hashtags can appear than Twitter is, in
particular, a hashtag does not need to be preceded with whitespace if
the tag appears after some scripts, such as Hirigana.
Previous code expected callers to typically provide the drawable and the
error message string resource, resulting in duplicate code at many
callsites.
Replace with three canned messages for empty containers, generic errors,
and network errors respectively. The images for these are fixed, the
caller may choose a different string resource for the error if there is
a more specific option.
Update and simplify the call sites.
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`.
The replies policy controls whether replies from members of the list
also appear in the list.
Display the replies policy as three radio buttons when a list is created
or updated, and send the chosen replies policy via the API.
Default value if not specified is always "list", for consistency with
the Mastodon API defaults.
While I'm here:
- Ensure the list dialog layout is inflated using the dialog's themed
context
- Use a `TextInputLayout` wrapper around the list name in the list
dialog for better UX
- Simplify the dialog layout, use LinearLayout, and standard padding and
margins
Previous code used a normal ProgressBar and a coroutine to delay
hiding/showing the bar for a snappier UI perception.
This is built-in functionality in LinearProgressIndicator, so switch to
that.
While I'm here, implement the "Select list" dialog's layout as a layout
resource.
If the user has tabs containing one or more lists, and any of those
lists are renamed or deleted then the change should be reflected in the
tabs.
To do that:
`MainActivity`:
- Re-create tabs whenever lists are loaded and there's a list in a tab
- Compare lists-in-tabs by the ID of the list when restoring the user's
tab, so that a list rename doesn't lose their position.
`NetworkListsRepository`:
- Update the user's tab preferences whenever lists are loaded, removing
tabs that contain lists that have been deleted, and updating the
list's title for lists that have been renamed.
Fixes#192
Previously to view a list the user either had to add it to a tab, or tap
through "Lists" in the navigation menu to their list of lists, and then
tap the list they want.
Fix that, and show all their lists in a dedicated section in the menu,
with a new "Manage lists" entry that's functionality identical to the
old "Lists" entry (i.e., it shows their lists and allows them to create,
delete, and edit list settings).
To do that:
- Implement a proper `ListsRepository` as the single source of truth for
list implementation throughout the app. Expose the current list of lists
as a flow, with methods to perform operations on the lists.
- Collect the `ListsRepository` flow in `MainActivity` and use that to
populate the menu and update it whenever the user's lists change.
- Rewrite the activities and fragments that manipulate lists to use
`ListRepository`.
- Always show error snackbars when list operations fail. In particular,
the HTTP code and error are always shown.
- Delete the custom `Either` implementation, it's no longer used.
- Add types for modelling API responses and errors, `ApiResponse` and
`ApiError` respectively. The response includes the headers as well as
the body, so it can replace the use of `NetworkResult` and `Response`.
The actual result of the operation is expected to be carried in a
`com.github.michaelbull.result.Result` type. Implement a Retrofit call
adapter for these types.
Unit tests for these borrow heavily from
https://github.com/connyduck/networkresult-calladapter
Additional user-visible changes:
- Add an accessible "Refresh" menu item to `ListsActivity`.
- Adding a list to a tab has a dialog with a "Manage lists" option.
Previously that would close the dialog when clicked, so the user had to
re-open it on returning from list management. Now the dialog stays open.
- The soft keyboard automatically opens when creating or editing a list.
Previous code showed a small icon for account media that the user has
hidden.
Now determine the correct size / aspect ratio for the media and use that
to compute the placeholder (either a blurhash, or the link colour for
consistency with the view on a timeline).
Fixes#513
The account logout process could fail due to API exceptions; network
errors for example, or if the user had already revoked the app's token
for that account. This would prevent the rest of the logout process
(cleaning database, etc) from completing.
Fix this by ignoring network errors during the logout process, and
always cleaning up account content in the database.
Fix a related issue where a deleted account might be recreated in a
partial state if the account's visible position was saved after it was
deleted. The recreated account couldn't do anything as it had no tokens,
but is very confusing.
The previous code forgot to close the DB after TimelineDaoTest was run,
so a warning message was displayed when the test was run locally.
Close the database using the `@After` annotation.
Fixes#511
Previously the only way to access notifications was to dedicate a tab to
them. Now notifications are available from the left-side navigation menu
so they're always accessible.
Add them to the top of the list, and swap the order of bookmarks and
favourites, assuming that users are more likely to want to see their
bookmarks than their favourites.
Move "Edit profile" to the bottom with the other settings options,
assuming that editing their profile does not happen very often, so
should not be at the top of the list.
New lint rules highlighted a potential crash; the use of named match
groups (used here when extracting server versions) requires API >= 26 or
throws an exception.
Use the group numbers instead of names when extracting the value, but
keep the group names in the regular expressions for readability.
The previous code used synchronous (i.e., non-suspending) functions to
call the /api/v2/search and /api/v2/accounts/search endpoints.
This is not necessary as the search was always performed in a separate
thread.
Remove, and replace their usage with the equivalent functions that
suspend.
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.
A filter's context (previously referred to as its `kind`) controls where
the filter is applied.
This was implemented as an enum with a specific property to control how
it would serialise when @FormUrlEncoded, and with a @Json annotation for
Moshi.
In addition, the model objects kept the filter context in its string
form throughout Pachli, requiring periodic conversion to/from the enum
type, making the code more complicated.
Fix this, by:
1. Converting the incoming JSON value to the enum type immediately, so
the rest of the code uses the enum constants exclusively.
2. Implement a Retrofit converter that serialises the enum value when
@FormUrlEncoded to the same string used in JSON serialisation
Many servers that claim to be Mastodon-API compatible are not, as
evidenced by the content they include in the responses to
`/api/v1/instance` and `/api/v2/instance` requests.
Work around the worst of the breakage by providing defaults or marking
some fields as nullable (with a default null).
Bugs have been reported against the relevant projects.
Some servers don't include a `urls` or `translation` block, which was
preventing parsing of the block, and falling back to the v1 instance
data.
Fix this by providing sensible defaults.
The `highlighted` property on a role may be absent. If it is this breaks
account parsing, including accounts in an /api/v2/instance response.
This, in turn, breaks determining server capabilities, including whether
or not translation is supported.
Set the default to `true`, which matches observed Mastodon behaviour.
Previous code expected all incoming enums values to map directly to
Kotlin enum constants.
This is a problem for servers with additional features -- e.g.,
"reaction" as a notification type.
Fix this with a new Moshi adapter that will set the incoming value to a
given constant if it's not recognised.
Apply this to the enum constants in core.network to ensure they are
handled.
Clean up enum handling in Converters.kt, ComposeViewModel.kt, and
Status.kt by using the existing `.ordinal` property and some extension
functions for idiomatic code.
Fixes#461
Previous code showed a generic placeholder for audio media on the
account's "Media" tab.
Fix this so the preview image is shown (if it's available).
- Move the "is this attachment previewable?" code to `Attachment` so it
can be reused here.
- Restructure the logic in `AccountMediaGridAdapter` to use the new
`isPreviewable()` method when deciding whether to show a preview.
- Attachments have dedicated placeholder drawables, use those when the
preview is not available.
Add an additional preference entry that triggers an update when tapped.
It also displays the earliest time of the next automatic update check as
the preference summary.
Move the code that performs the update check (and the logic for whether
to perform the check) out of `MainActivity` and in to `UpdateCheck` so
it's available from `PreferencesFragment`.
Friendica can return a null `voted_on` property, in violation of the API
spec.
Introduce a `BooleanIfNull` annotation that will convert the `null` to
`false` if encountered.
While I'm here update the other adapters as classes on their relevant
annotations instead of standalone classes to keep the code consistent.
Fixes#455
Some users report that Pachli is not retrieving/displaying notifications
in a timely fashion.
To assist in diagnosing these errors, provide an additional set of tabs
on the "About" screen that contain information about how Pachli is
fetching notifications, and if not, why not.
Allow the user to save notification related logs and other details to a
file that can be attached to an e-mail or bug report.
Recording data:
- Provide a `NotificationConfig` singleton with properties to record
different aspects of the notification configuration. Update these
properties as different notification actions occur.
- Store logs in a `LogEntryEntity` table. Log events of interest with a
new `Timber` `LogEntryTree` that is planted in all cases.
- Add `PruneLogEntryEntityWorker` to trim saved logs to the last 48
hours.
Display:
- Add a `NotificationFragment` to `AboutActivity`. It hosts two other
fragments in tabs to show details from `NotificationConfig` and the
relevant logs, as well as controls for interacting with them.
Bug fixes:
- Filter out notifications with a null tag when processing active
notifications, prevents an NPE crash
Other changes:
- Log more details when errors occur so the bug reports are more helpful
The `x` and `y` properties in `Attachment.Focus` may be null (not
documented as such, but observed in the wild).
Provide a `DefaultIfNull` adapter that can be applied to these to
replace null values with a sensible default.
The previous code didn't clear the task stack or recreate `MainActivity`
when the active user was changed.
So if the user was logged in with account A and used "Compose" from a
notification sent to account B, the active account was switched to B but
the UI chrome wasn't. After exiting `ComposeActivity` they would be
looking at the timeline for account B but with a toolbar that showed
account A.
Fix this by clearing the task stack and explicitly recreating
`MainActivity` when forwarding intents to `ComposeActivity`.
- Use format strings so any overhead of building the string is only
incurred if the message is actually logged
- Pass throwables as the first parameter so they are logged with the
stacktrace
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
Previously, `AboutActivity` had buttons and links to show the privacy
policy and licenses of dependencies.
Change this to a selection of fragments in tabs, one tab each for:
- General "About" information
- Licenses
- Privacy Policy
The information shown hasn't changed, but this lays the groundwork for
including additional tabs in the future for information like server
rules, detected capabilities, or troubleshooting information.
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
Previously when the user interacted with a status the operation (reblog,
favourite, etc) travels through multiple layers of code, carrying with
it the position of the item in the list that the user operated on.
At some point the status is retrieved from the list using its position
so that the correct status ID can be used in the network operation.
If this happens while the list is also refreshing there's a possible
race condition, and the original status' position may have changed in
the list. Looking up the status by position to determine which status to
perform the action on may cause the action to happen on the wrong
status.
Fix this by passing the status' viewdata to any actions instead of its
position. This includes all the information necessary to make the API
call, so there is no chance of a race.
This is quite an involved change because there are three types of
viewdata:
- `StatusViewData`, used for regular timelines
- `NotificationViewData`, used for notifications, may wrap a status that
can be operated on
- `ConversationViewData`, used for conversations, does wrap a status
The previous code treated them all differently, which is probably why it
operated by position instead of type.
The high level fix is to:
1. Create an interface, `IStatusViewData`, that contains the data
exposed by any viewdata that contains a status.
2. Implement the interface in `StatusViewData`, `NotificationViewData`,
and `ConversationViewData`.
3. Change the code that operates on viewdata (`SFragment`,
`StatusActionListener`, etc) to be generic over anything that implements
`IStatusViewData`.
4. Change the code that handles actions to pass the viewdata instead of
the position.
Fixes#370
Previous code could return an error on Friendica version strings like
`2024.03-dev-1547`.
Fix this:
- Extend the list of explicitly supported servers to include Fedibird,
Friendica, Glitch, Hometown, Iceshrimp, Pixelfed, and Sharkey.
- Add version parsing routines for these servers.
- Test the version parsing routines fetching every server and version
seen by Fediverse Observer (~ 2,000 servers) and ensuring that the
server and version information can be parsed.
Improve the error message:
- Show the hostname with a `ServerRepository` error
Clean up the code:
- Remove the custom `resultOf` and `mapResult` functions, they have
equivalents in newer versions of the library (like `runSuspendCatching`)
Fixes#372
Commit 993b7469 inadvertently broke this by removing the @SerializedName
annotation, so the user's display name was always null and the UI fell
back to showing the username.
Fixes#371
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
The previous code was serialising the `TabKind` enum without the `_`, so
when it was converted back to the enum name (which has a `_`) it failed
and immediately crashed.
Fixes#306
If a status was part of a thread, and it was not the "detailed" status,
and it had been translated, then the view data was marked as "show the
translation". But the translation was not loaded, so the status content
appeared as empty.
Fix that by loading the translated content of all statuses in the thead
and ensure that the translated content is rendered.
Throw an `IllegalStateException` in debug builds to catch any future
occurrences of this.
Fixes#281
Mastodon counts post lengths by considering emojis to be single
characters, no matter how many unicode code points they are composed of.
So "😜" has length 1.
Pachli was using `String.length`, which considers "😜" as length 2.
Correct the calculation by using a BreakIterator to count the characters
in the string, which treats multi-character emojis as a length 1.
Poll options had a similar problem, exacerbated by the Mastodon web UI
also having the same problem, see
https://github.com/mastodon/mastodon/issues/28336.
Fix that by creating `MastodonLengthFilter`, an `InputFilter` that does
the right thing for regular text that may contain emojis.
See also https://github.com/tuskyapp/Tusky/pull/4152, which has the fix
for status length but not polls.
---------
Co-authored-by: Konrad Pozniak <opensource@connyduck.at>
Roles for the logged in user appeared in Mastodon 4.0.0 and can be
displayed on the user's profile screen.
Show them as chips, adjusting the display of the existing "Follows you"
and "Bot" indicators to make allowances for this.
Roles can have a custom colour assigned by the server admin. This is
blended with the app colour so it is not too jarring in the display.
See also https://github.com/tuskyapp/Tusky/pull/4029
Co-authored-by: Konrad Pozniak <opensource@connyduck.at>
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 package wasn't renamed when it was moved, so was still
`app.pachli.components.timeline`, instead of the new location,
`app.pachli.core.network.model`.
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.