* Centralize library management logic and introduce support for plain text and HTML formats

* Centralize library management logic and introduce support for plain text and HTML formats

* Expand unit test coverage for library state management, UI models, and MainViewModel features.

* Add comprehensive unit tests for PDF reader core logic, preferences, and data persistence

* Add unit tests for EPUB parsing, content loading, search functionality, and reader JavaScript bridges.

* Add unit tests for OPDS parsing and Smart Collection engine, and integrate Kover plugin

* Add comprehensive unit tests

* Centralize library snapshot serialization in the `shared` module and improve filtering and sorting logic.

* Implement text selection, highlighting, and reading state persistence for PDF and EPUB engines in desktop version

* Folder import support for desktop app

* Introduce Smart Shelves with rule-based filtering in desktop version

* Implement shared EPUB annotation serialization and highlight rendering

* Centralize file type capabilities and platform-specific support logic

* Refactor reader state management to use a central reducer

* Implement customizable reader toolbar and advanced formatting settings in shared

* Implement locator-based navigation and customizable highlight palette for desktop app

* Enhance reader customization and expand search functionality in desktop app

* Redesign reader settings and tools into a tabbed control panel in desktop app

* Enhance reader navigation and highlight precision in desktop app

* Implement bidirectional position synchronization and dynamic highlights in the desktop reader

* Implement shared state management and enhanced search for the PDF reader in desktop app

* Add vertical scroll support to the desktop PDF reader

* Implement ink, text, and eraser annotation support in desktop PDF viewer

* Implement PDF bookmarks, Table of Contents, and annotation editing in desktop app

* Implement link handling and navigation for PDF and EPUB readers in desktop app

* Implement PDF jump history for navigation in desktop app

* Enhance PDF ink rendering and annotation capabilities in desktop app

* Implement advanced PDF text annotations with inline editing and rich styling in desktop app

* Add move handle and movement logic for PDF text annotations in desktop app

* Implement local folder synchronization and metadata sidecar support in desktop app

* Implement book metadata extraction and drag-and-drop import for Desktop

* Implement dynamic and custom app theme management for desktop

* Introduce canonical PDF annotation codec and support for multi-segment highlights

* Implement rich text editing and pagination support for the PDF reader in desktop app

* Improve PDF rich text pagination, synchronization, and observability in desktop

* Hide trailing structural page breaks in rich text editor

* Implement a unified JVM book loader and expand supported formats on Desktop

* Add comic archive support for Desktop and enhance MOBI parsing

* Implement shared OPDS catalog support and UI for Android and Desktop

* Improve native WebView lifecycle and surface transition management on Desktop

* Enable Compose Swing interop blending and simplify Desktop WebView management

* Integrate BYOK AI features and Cloud TTS for desktop

* Enhance Desktop TTS with streaming audio and improved secure storage for AI key

* Implement scoped Cloud TTS with synchronized highlighting for EPUB and PDF in desktop app

* Implement custom font management and utility screens in desktop app

* Implement PDFium-based PDF annotation export

* Remove PdfBox dependency and standardize PDF export via Pdfium

* Implement local audio caching and playback controls for Gemini Cloud TTS in desktop app

* Implement reader themes and custom texture support in desktop app

* Redesign non-reader UI with responsive navigation and enhanced library management in desktop app

* Introduce ReaderWorkspaceShell to unify EPUB and PDF reader layouts in desktop app

* Exclude manual-only files from automated sync and import

* Implement customizable Text-to-Speech (TTS) word replacements

* Optimize reader performance with persistent layout caching and decoupled theme rendering

* Improve position restoration during reader reconfiguration in epub pagination

* Use independent thickness for eraser tool and stylus override
This commit is contained in:
Aryan 2026-05-10 10:07:37 +05:30 committed by GitHub
parent 88c7fa7b5c
commit 8366d76dcd
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
214 changed files with 53372 additions and 4702 deletions

View file

@ -186,11 +186,13 @@ class TtsController(context: Context) : Player.Listener {
)
val textList = ArrayList(chunks.map { it.text })
val spokenTextList = ArrayList(chunks.map { it.spokenText.ifBlank { it.text } })
val cfiList = ArrayList(chunks.map { it.sourceCfi })
val offsetList = ArrayList(chunks.map { it.startOffsetInSource })
val args = Bundle().apply {
putStringArrayList(KEY_TEXT_CHUNKS, textList)
putStringArrayList(KEY_SPOKEN_TEXT_CHUNKS, spokenTextList)
putStringArrayList(KEY_SOURCE_CFIS, cfiList)
putIntegerArrayList(KEY_START_OFFSETS, offsetList)
putString(KEY_SPEAKER_ID, _ttsState.value.speakerId)

View file

@ -61,6 +61,7 @@ val SET_PLAYBACK_PARAMS_COMMAND = SessionCommand("com.aryan.reader.tts.SET_PLAYB
const val TTS_NOTIFICATION_DIAG_TAG = "TTS_NOTIFICATION_DIAG"
const val KEY_TEXT_CHUNKS = "KEY_TEXT_CHUNKS"
const val KEY_SPOKEN_TEXT_CHUNKS = "KEY_SPOKEN_TEXT_CHUNKS"
const val KEY_SOURCE_CFIS = "KEY_SOURCE_CFIS"
const val KEY_START_OFFSETS = "KEY_START_OFFSETS"
const val KEY_SPEAKER_ID = "KEY_SPEAKER_ID"
@ -205,6 +206,7 @@ class TtsPlaybackManager(
)
val cfis = args.getStringArrayList(KEY_SOURCE_CFIS)
val offsets = args.getIntegerArrayList(KEY_START_OFFSETS)
val spokenTexts = args.getStringArrayList(KEY_SPOKEN_TEXT_CHUNKS)
val speakerId = args.getString(KEY_SPEAKER_ID, DEFAULT_SPEAKER_ID)
val bookTitle = args.getString(KEY_BOOK_TITLE)
val chapterTitle = args.getString(KEY_CHAPTER_TITLE)
@ -218,10 +220,23 @@ class TtsPlaybackManager(
val richChunks = if (cfis != null && offsets != null && chunks.size == cfis.size && chunks.size == offsets.size) {
chunks.mapIndexed { index, text ->
val safeOffset = offsets.getOrNull(index) ?: -1
TtsChunk(text, cfis[index], safeOffset)
val spokenText = spokenTexts?.getOrNull(index)?.ifBlank { text } ?: text
TtsChunk(
text = text,
sourceCfi = cfis[index],
startOffsetInSource = safeOffset,
spokenText = spokenText,
)
}
} else {
chunks.map { TtsChunk(it, "", -1) }
chunks.mapIndexed { index, text ->
TtsChunk(
text = text,
sourceCfi = "",
startOffsetInSource = -1,
spokenText = spokenTexts?.getOrNull(index)?.ifBlank { text } ?: text,
)
}
}
val authToken = args.getString(KEY_AUTH_TOKEN)
@ -337,7 +352,11 @@ class TtsPlaybackManager(
}
val slicedText = currentChunk.text.substring(relativeOffset)
val newChunk = currentChunk.copy(text = slicedText, startOffsetInSource = offset)
val newChunk = currentChunk.copy(
text = slicedText,
startOffsetInSource = offset,
spokenText = slicedText,
)
val mutableChunks = textChunks.toMutableList()
mutableChunks[currentIdx] = newChunk
@ -540,7 +559,8 @@ class TtsPlaybackManager(
"Preparing first chunk. startAtIndex=$startAtIndex, playWhenReady=$playWhenReady"
)
val ttsAudioData = generateAudioChunk(bookTitle ?: "Unknown Book", chapterTitle, startAtIndex, textChunks.size, firstChunk.text, currentSpeakerId, currentTtsMode, currentAuthToken)
val spokenText = firstChunk.spokenText.ifBlank { firstChunk.text }
val ttsAudioData = generateAudioChunk(bookTitle ?: "Unknown Book", chapterTitle, startAtIndex, textChunks.size, spokenText, currentSpeakerId, currentTtsMode, currentAuthToken)
Timber.tag("TTS_CLOUD_DIAG").i("generateAudioChunk returned in ${System.currentTimeMillis() - chunkStartTime}ms")
if (ttsAudioData.error == "INSUFFICIENT_CREDITS") {
@ -572,7 +592,7 @@ class TtsPlaybackManager(
if (id != null) chunkStreamIds[startAtIndex] = id
}
val pathToUse = streamUri ?: audioFile!!.absolutePath
val mediaItem = createMediaItem(serverText, pathToUse, startAtIndex, updatedChunk)
val mediaItem = createMediaItem(updatedChunk.text, pathToUse, startAtIndex, updatedChunk)
withContext(Dispatchers.Main) {
val prepStartTime = System.currentTimeMillis()
@ -589,7 +609,7 @@ class TtsPlaybackManager(
_ttsState.value = _ttsState.value.copy(
isLoading = false,
isPlaying = playWhenReady,
currentText = serverText,
currentText = updatedChunk.text,
chapterTitle = chapterTitle,
chapterIndex = chapterIndex,
totalChapters = totalChapters,
@ -618,6 +638,9 @@ class TtsPlaybackManager(
if (wordTimings.isNullOrEmpty()) {
return originalChunk
}
if (originalChunk.spokenText != originalChunk.text) {
return originalChunk.copy(timedWords = emptyList())
}
val timedWords = mutableListOf<TimedWord>()
var currentSearchIndex = 0
@ -823,7 +846,8 @@ class TtsPlaybackManager(
val prefetchStartTime = System.currentTimeMillis()
Timber.tag("TTS_CLOUD_DIAG").i("Starting prefetch generation for chunk $targetIndex")
val ttsAudioData = generateAudioChunk(bookTitle ?: "Unknown Book", chapterTitle, targetIndex, textChunks.size, nextChunk.text, currentSpeakerId, currentTtsMode, currentAuthToken)
val spokenText = nextChunk.spokenText.ifBlank { nextChunk.text }
val ttsAudioData = generateAudioChunk(bookTitle ?: "Unknown Book", chapterTitle, targetIndex, textChunks.size, spokenText, currentSpeakerId, currentTtsMode, currentAuthToken)
Timber.tag("TTS_CLOUD_DIAG").i("Prefetch audio setup for chunk $targetIndex took ${System.currentTimeMillis() - prefetchStartTime}ms")
@ -842,7 +866,7 @@ class TtsPlaybackManager(
if ((audioFile != null || streamUri != null) && serverText != null) {
val updatedChunk = processWordTimings(nextChunk, serverText, ttsAudioData.wordTimings)
val pathToUse = streamUri ?: audioFile!!.absolutePath
val nextMediaItem = createMediaItem(serverText, pathToUse, targetIndex, updatedChunk)
val nextMediaItem = createMediaItem(updatedChunk.text, pathToUse, targetIndex, updatedChunk)
withContext(Dispatchers.Main) {
if (audioFile != null) {