package com.example.aiapp import androidx.compose.runtime.Immutable import kotlinx.coroutines.Dispatchers import kotlinx.coroutines.withContext /** * What the transcript renders: the event stream folded into displayable rows (see [foldEvent]). The * stream is the only data source -- opening a session screen replays from seq 0, and a reconnect * resumes from the last seq seen, so there is no separate history fetch to drift from it. */ @Immutable sealed class TranscriptItem { /** * The transcript sequence number this row started at, and its identity on screen. * * The list is drawn newest-first, so every new message is an insertion at index 0 and every * page of history is an insertion at the far end. Without an identity that survives both, the * list is addressed by position: whatever somebody had scrolled to keeps its index while the * content underneath it slides, which reads as the view scrolling on its own. * * A seq is the right identity because it is what the transcript itself is ordered by, it never * changes, and it is already carried by every event. A row built from several events -- a * streaming message, a tool call and its result -- keeps the seq of the first, so it holds * still while the rest of it arrives. */ abstract val seq: Long data class UserMsg( override val seq: Long, val text: String, /** Refs of what was attached, drawn inside the bubble. */ val images: List = emptyList(), ) : TranscriptItem() data class AssistantMsg(override val seq: Long, val text: String) : TranscriptItem() data class ToolRun( override val seq: Long, val id: String, /** * The run of adjacent calls this one belongs to, named once when the call is folded in and * never recomputed. * * Carried rather than derived because a run can gain members at *either* end -- a new call * arriving beside it, or a page of history arriving in front of it -- so no function of its * current members is stable. It is the first call's id at the moment the run started, which * is a name rather than a description: [joinPages] hands it to older calls that turn out to * belong to the same run, instead of renaming the run they joined. */ val runId: String, val tool: String, val input: String, val output: String, val done: Boolean, /** * The questions this call is waiting on, in the order they were asked. * * On the call's own row rather than beside it: an ask used to arrive as a second card * repeating the input verbatim, so the reader saw the same command twice and had to work * out that it was one event. The backend says which call a question is about, so this is a * fact rather than a match on the input. * * A list because AskUserQuestion asks up to four at once, and they are one decision to make * -- a permission is the case of exactly one, not a different shape. */ val asks: List = emptyList(), /** * Images this call's result carried, drawn under it. * * Beside it they had to be paired by position, and position is the thing a page boundary * breaks -- a screenshot loaded on one page and its call on the next read as unrelated. */ val images: List = emptyList(), ) : TranscriptItem() data class QuestionCard( override val seq: Long, val id: String, val prompt: String, /** A few words naming what this is about, when the asker offered one. */ val header: String?, val options: List, /** Whether several options may be chosen at once. */ val multiSelect: Boolean, /** What was chosen, once something was; empty until then. */ val answers: List, ) : TranscriptItem() data class ErrorMsg(override val seq: Long, val message: String) : TranscriptItem() /** An image by server-side ref, fetched from the session's files route. */ data class ImageItem(override val seq: Long, val ref: String) : TranscriptItem() /** * A message another agent sent this session. * * Its own row rather than a [UserMsg]: see [PeerMessageRow] for why the voice matters. */ data class PeerNote(override val seq: Long, val from: String, val text: String) : TranscriptItem() /** * A command the session ran on itself -- `/compact`, `/rename`. * * Kept in the transcript rather than only shown while it waits, because it explains what * follows: a conversation that suddenly has half the context, or a session with a new name. */ data class CommandRow(override val seq: Long, val text: String) : TranscriptItem() /** Placeholder row for events this build can't render (newer kinds). */ data class Note(override val seq: Long, val text: String) : TranscriptItem() /** * A clear that happened: everything above it left the session's context and stayed on screen. * * Carries only its position, because that is all it means. */ data class ClearedNote(override val seq: Long) : TranscriptItem() /** * A compaction that happened, and what it recovered. * * In the transcript rather than only in the status line, because the status is gone the moment * it finishes and this is the part worth keeping: it is the explanation for a gap in the * conversation, and for a minute or two in which the session was busy with nothing to show. * * The wire also says what triggered it, and this deliberately does not carry that: the row says * the two sizes and nothing else (see [compactionSummary]), so keeping the trigger here would * be a field nothing can read. */ data class CompactedNote( override val seq: Long, val preTokens: Long?, val postTokens: Long?, ) : TranscriptItem() } /** * The run a call joins: the one it lands next to, or a new one named after itself. * * Only ever consulted when the call is first folded in. That is what makes the name stable -- a run * keeps whatever it was called when it started, however many calls arrive at either end of it * afterwards. * * A question to the reader is in a run of its own, which is what puts it on the transcript as a row * rather than inside a collapsed "Called 6 tools" card. Two things follow from being alone: it is * always visible, since a run of one is drawn as itself rather than as a group; and the calls * around it fall into a group before it and a group after it, so where the reader was asked * something is legible in the shape of the transcript without opening anything. It ends the run * before it as well as starting a fresh one after -- the moment somebody was asked is a boundary in * the work, not a gap in the middle of one run. */ private fun runIdFor(items: List, id: String, tool: String): String { val previous = items.lastOrNull() as? TranscriptItem.ToolRun ?: return id if (tool == ASK_USER_QUESTION || previous.tool == ASK_USER_QUESTION) return id return previous.runId } /** * Puts a page of older items in front of the ones already loaded, healing whatever the page * boundary cut in two. * * Two things straddle a boundary: a tool call separated from its result, and a message separated * from the rest of itself. Both were one thing before the transcript was cut into pages, and both * have to be one thing again -- a reply drawn as two messages is the same defect as a call drawn * twice, arriving from the same cause. * * A boundary lands wherever it lands, and roughly half the time that is between a call and its * result. The newer page then holds a `ToolEnd` whose start it never saw, which [foldEvent] draws * as a row of its own -- correctly, because a call that renders as nothing is indistinguishable * from one that never happened. When the older page arrives it brings the real `ToolStart`, and * concatenating the two lists left *both*: the same call twice, once as a proper card and once as a * nameless placeholder. Visible as a run of four calls reporting "Called 5 tools", and worse than * the miscount -- the extra row is at the join, so it also moves everything the reader was looking * at. * * Merged by the call's own id rather than by position, because position is exactly what a page * boundary destroys. The older row wins on what a start knows (the tool's name, its input) and the * newer on what an end knows (the output, and whether it finished), which is the only way round * that loses nothing. * * The third thing is the *run*, and it is the one that used to be missed. Every page ends up here, * but [adoptRun] only ran on the path where a split call had been found -- so the boundary that * falls cleanly between two finished calls, which is most of them, went straight to concatenation * and left the older page's calls under the run name they were folded with. On screen: one run of * tool calls drawn as two groups, with the seam wherever the reader happened to have paged. The two * early returns were an optimisation on a list the size of one page, and they were skipping work * rather than saving it. */ fun joinPages(earlier: List, later: List): List { val (older, newer) = healSplitMessage(earlier, later) val startedEarlier = older.filterIsInstance().mapTo(mutableSetOf()) { it.id } val endedLater = newer .filterIsInstance() .associateBy { it.id } .filterKeys { it in startedEarlier } val healed = older.map { row -> val half = (row as? TranscriptItem.ToolRun)?.let { endedLater[it.id] } if (row is TranscriptItem.ToolRun && half != null) { row.copy( output = half.output, done = half.done, // Kept from both halves: a question or an image can be attached to either, // depending on which side of the boundary its event fell. asks = row.asks + half.asks, images = row.images + half.images, ) } else { row } } val kept = newer.filterNot { it is TranscriptItem.ToolRun && it.id in endedLater } return adoptRun(healed, kept) + kept } /** * Rejoins a message the page boundary cut, and hands back the two pages to concatenate. * * [foldEvent] never leaves two assistant messages next to each other inside one page -- deltas * accumulate into the message before them -- so two meeting at a join are always the two halves of * one reply, and leaving them apart drew a single answer as two, with a paragraph break through the * middle of a sentence. * * The newer half keeps its identity, for the reason [adoptRun] gives: it is the row already on * screen, and renaming that is how the list loses its anchor. It grows by what the older half * brings, which is safe here and nowhere else -- the join is at the oldest end of what is loaded, * so the growth extends off the top of the screen, away from the row the list anchors to. */ private fun healSplitMessage( earlier: List, later: List, ): Pair, List> { val head = earlier.lastOrNull() val tail = later.firstOrNull() if (head !is TranscriptItem.AssistantMsg || tail !is TranscriptItem.AssistantMsg) { return earlier to later } return earlier.dropLast(1) to (listOf(tail.copy(text = head.text + tail.text)) + later.drop(1)) } /** * Hands the older calls at the join the name of the run they are joining. * * The two pages were folded separately, so a run split by the boundary came back as two runs with * two names. Naming the joined run after the *older* half would be the obvious way round and is the * wrong one: the newer half is the part already on screen, and renaming it is renaming the row the * reader is looking at, which is how a list loses its anchor and steps under them. So the arriving * calls take the name of the ones already there, and nothing visible changes identity. */ private fun adoptRun( earlier: List, later: List, ): List { val first = later.firstOrNull() as? TranscriptItem.ToolRun ?: return earlier // A question is in a run of its own on both sides of the join, the same as it would be had // the two pages been folded as one -- see `runIdFor`. Without this the heal would merge a // group straight through the row the reader was asked something on. if (first.tool == ASK_USER_QUESTION) return earlier val joining = first.runId val tail = earlier.takeLastWhile { it is TranscriptItem.ToolRun && it.tool != ASK_USER_QUESTION } if (tail.isEmpty()) return earlier return earlier.dropLast(tail.size) + tail.map { (it as TranscriptItem.ToolRun).copy(runId = joining) } } /** * A peer message goes above the turn it started, not where it happened to arrive. * * The live Claude Code path cannot record it in place: the CLI says nothing about a peer message * until the turn's `result`, so the event lands below the whole reply it caused -- the answer * printed above the question. The server stamps it with where that turn began * ([SessionEvent.PeerMessage.turnStart]) and the note takes that seq, so it sorts into the list * where it belongs rather than being drawn out of order at the end. * * Taking the turn's opening seq as its own is also what keeps the list sorted, which anchors and * paging both depend on. That seq belongs to a status change, and a status draws no row, so there * is nothing for it to collide with. * * Without a stamp -- a message replayed out of a session file, which is already in the right place * -- it stays where it arrived. */ private fun placePeerNote( items: List, seq: Long, event: SessionEvent.PeerMessage, ): List { val at = event.turnStart ?: return items + TranscriptItem.PeerNote(seq, event.from, event.text) val note = TranscriptItem.PeerNote(at, event.from, event.text) val index = items.indexOfFirst { it.seq > at } if (index < 0) return items + note val behind = (items.getOrNull(index - 1) as? TranscriptItem.ToolRun)?.runId return items.subList(0, index) + note + splitRun(items.subList(index, items.size), behind) } /** * The calls the note now sits in front of, renamed if they were sharing a run with the calls behind * it. * * A run is named from what a call landed next to (see [runIdFor]), and nothing there knows about * turns -- so a turn opening with a tool call, straight after one that ended with one, folds them * into a single run. Left alone, [groupToolRuns] would flush at the note and hand both halves the * same name: two rows with one key, which a keyed list cannot draw at all. * * The later half is the one renamed, which is the opposite of a page join ([adoptRun]) and right * for the opposite reason. There the two halves were always one run and the newer was already on * screen; here they were never one turn's work, and both halves change appearance at the same * moment the note appears between them. */ private fun splitRun(tail: List, behind: String?): List { val first = tail.firstOrNull() as? TranscriptItem.ToolRun ?: return tail if (behind == null || first.runId != behind) return tail val run = tail.takeWhile { it is TranscriptItem.ToolRun && it.runId == behind } return run.map { (it as TranscriptItem.ToolRun).copy(runId = first.id) } + tail.drop(run.size) } fun foldEvent(items: List, entry: SeqEvent): List = when (val event = entry.event) { is SessionEvent.UserMessage -> items + TranscriptItem.UserMsg(entry.seq, event.text, event.images) is SessionEvent.AssistantText -> { // Deltas accumulate into the message they're streaming, which keeps the seq of the // first of them: a row whose identity changed with every delta would be a new row on // every frame, and the list would jump for the whole of a streamed answer. val last = items.lastOrNull() if (last is TranscriptItem.AssistantMsg) { items.dropLast(1) + last.copy(text = last.text + event.delta) } else { items + TranscriptItem.AssistantMsg(entry.seq, event.delta) } } is SessionEvent.ToolStart -> items + TranscriptItem.ToolRun( entry.seq, event.id, runIdFor(items, event.id, event.tool), event.tool, event.input, "", done = false, ) is SessionEvent.ToolUpdate -> updateTool(items, event.id) { it.copy(output = event.output) } is SessionEvent.ToolEnd -> // Created when its start is not here, rather than dropped. A // fold that only ever *updates* loses the whole call when the // start fell outside the loaded window, and a tool call that // renders as nothing is indistinguishable from one that never // happened. The name is unknown from an end alone; loading the // page before this one replaces the row with the real thing. if (items.any { it is TranscriptItem.ToolRun && it.id == event.id }) { updateTool(items, event.id) { it.copy(output = event.output, done = true) } } else { items + TranscriptItem.ToolRun( entry.seq, event.id, // The name is not known from an end alone, so a call that was an ask // cannot be recognised as one here; loading the page before this // replaces the row with the real thing, which is when it splits out. runIdFor(items, event.id, "tool"), "tool", "", event.output, done = true, ) } is SessionEvent.Question -> { val card = TranscriptItem.QuestionCard( entry.seq, event.id, event.prompt, event.header, event.options, event.multiSelect, emptyList(), ) // A question with no tool behind it -- AskUserQuestion, or an ask // whose call fell outside the loaded window -- is a card of its // own, which is what every question was before this. if ( event.about != null && items.any { it is TranscriptItem.ToolRun && it.id == event.about } ) { updateTool(items, event.about) { it.copy(asks = it.asks + card) } } else { items + card } } is SessionEvent.Answered -> // Resolved wherever it is drawn: a card of its own, or a tool // row's ask. Missing the second left an Allow/Deny pair live on // a question already answered from another device. items.map { when { it is TranscriptItem.QuestionCard && it.id == event.id -> it.copy(answers = event.answers) it is TranscriptItem.ToolRun && it.asks.any { ask -> ask.id == event.id } -> it.copy( asks = it.asks.map { ask -> if (ask.id == event.id) ask.copy(answers = event.answers) else ask } ) else -> it } } is SessionEvent.PeerMessage -> placePeerNote(items, entry.seq, event) is SessionEvent.CommandSent -> items + TranscriptItem.CommandRow(entry.seq, event.text) // Screen-level state, not transcript rows -- see SessionScreen. is SessionEvent.CommandQueued -> items // No row of its own: a message that is still waiting is drawn as a pending bubble below // the transcript, and becomes an ordinary one where the session read it. is SessionEvent.MessageQueued -> items // The bubble goes away and nothing takes its place: the message was never read, so there // is nothing it belongs above. is SessionEvent.MessageDropped -> items is SessionEvent.Settings -> items is SessionEvent.Status -> items is SessionEvent.Error -> items + TranscriptItem.ErrorMsg(entry.seq, event.message) is SessionEvent.Image -> // Under the call that produced it when there is one, and a row of // its own when there is not -- a person's own attachment belongs // to no call, and neither does one whose call fell outside the // loaded window. if ( event.about != null && items.any { it is TranscriptItem.ToolRun && it.id == event.about } ) { updateTool(items, event.about) { it.copy(images = it.images + event.ref) } } else { items + TranscriptItem.ImageItem(entry.seq, event.ref) } is SessionEvent.Cleared -> items + TranscriptItem.ClearedNote(entry.seq) is SessionEvent.Compacted -> items + TranscriptItem.CompactedNote(entry.seq, event.preTokens, event.postTokens) is SessionEvent.Unknown -> items + TranscriptItem.Note(entry.seq, "[${event.type}]") // Screen-level state, not transcript rows -- see SessionScreen. is SessionEvent.UsageDelta -> items } private fun updateTool( items: List, id: String, change: (TranscriptItem.ToolRun) -> TranscriptItem.ToolRun, ): List = items.map { if (it is TranscriptItem.ToolRun && it.id == id) change(it) else it } /** * Where markdown is parsed ahead of being drawn: two threads, never all of them. * * The default dispatcher sizes itself to the machine, which is right for work somebody is waiting * on and wrong for work nobody is. A page of history is hundreds of parses arriving at once, and * taking every core for them leaves the thread that draws the frame queueing behind one -- measured * on a Pixel 9 Pro XL as 21ms of `waited` at the 90th percentile, which is the frame failing to * *start* rather than taking too long once it had. */ @OptIn(kotlinx.coroutines.ExperimentalCoroutinesApi::class) private val parsingThreads = Dispatchers.Default.limitedParallelism(2) /** * Parses the markdown among [rows], off whatever thread is drawing. * * Called where a page of transcript is folded rather than where a row is composed, which is the * whole point: the work happens seconds before the reader reaches the rows it was done for. See * [ParsedReplies]. * * What is warmed mirrors what the rows draw, unit by unit -- prose split into its blocks, a memory * note whole, a peer message split the same way prose is -- because a string warmed under a key no * row ever looks up is a miss that nothing reports; see [transcriptUnits], which is the flatten * this has to agree with. It reads the same [ParsedReplies.partsOf] and [ParsedReplies.blocksOf] * caches the flatten does, so a message is scanned once however many pages hand it back through * here, while the whole loaded transcript crosses this on every page. * * Every kind of row that draws markdown belongs in the `when` below. That is the rule the peer * message was missing: this used to filter for assistant replies alone, so the one row type nobody * had thought about paid its whole parse in the frame it appeared in, with no counter saying which * row it was. */ suspend fun warm(replies: ParsedReplies, rows: List) { withContext(parsingThreads) { val texts = rows.flatMap { row -> when (row) { is TranscriptItem.AssistantMsg -> replies.partsOf(row.text).flatMap { part -> when (part) { is MessagePart.Prose -> replies.blocksOf(part.text) // Drawn as one MarkdownText, so its whole text is the key // looked up. is MessagePart.Remembered -> listOf(part.text) } } // A message from another agent is markdown too, and it is the longest thing // in a transcript often enough that leaving it out was the whole of why one // cost a fifth of a second to open: it was the only markdown in the app // parsed on the thread that draws. Its blocks, not its text, because // [PeerMessageRow] draws it a block at a time. is TranscriptItem.PeerNote -> replies.blocksOf(row.text) else -> emptyList() } } if (texts.isNotEmpty()) replies.warm(texts) } }