feat(server,kernel): repo-map indexer + droppable L3 context injection

2026-06-04 01:48:05 +04:00
parent 0ca7d4c86b
commit 8b6eedcf87
9 changed files with 242 additions and 4 deletions
@@ -298,6 +298,10 @@ fun main() {
            l3MemoryStore = l3MemoryStore,
            journalRepository = decisionJournalRepository,
            eventStore = eventStore,
+            indexer = com.correx.apps.server.memory.RepoMapIndexer(
+                maxDepth = correxConfig.project.maxDepth,
+                ignoreGlobs = correxConfig.project.ignoreGlobs,
+            ),
        )
    } else {
        null
@@ -129,8 +129,14 @@ class ServerModule(
    ) {
        activeSessionJobs.computeIfAbsent(sessionId) {
            moduleScope.launch {
-                // Seed prior-session project memory before the run so stages see it in context.
-                projectMemory?.let { pm -> runCatching { pm.retrieveAndSeed(sessionId, pm.repoRoot()) } }
+                // Record the repo map + seed prior-session memory before the run so stages
+                // see both in context.
+                projectMemory?.let { pm ->
+                    runCatching {
+                        pm.indexAndRecord(sessionId, pm.repoRoot())
+                        pm.retrieveAndSeed(sessionId, pm.repoRoot())
+                    }
+                }
                runCatching {
                    orchestrator.run(sessionId, graph, sessionConfig)
                    // Distil this run's decisions into durable project memory on completion.
@@ -3,6 +3,7 @@ package com.correx.apps.server.memory
 import com.correx.core.config.ProjectConfig
 import com.correx.core.events.events.EventMetadata
 import com.correx.core.events.events.NewEvent
+import com.correx.core.events.events.RepoMapComputedEvent
 import com.correx.core.events.events.SteeringNoteAddedEvent
 import com.correx.core.events.stores.EventStore
 import com.correx.core.events.types.EventId
@@ -39,12 +40,47 @@ class ProjectMemoryService(
    private val journalRepository: DefaultDecisionJournalRepository,
    private val eventStore: EventStore,
    private val distiller: ProjectMemoryDistiller = ProjectMemoryDistiller(),
+    private val indexer: RepoMapIndexer = RepoMapIndexer(),
 ) {
    private fun tag(repoRoot: String) = "project:$repoRoot"

    /** Resolved repo-root key: configured [ProjectConfig.root], else the working dir. */
    fun repoRoot(): String = config.root.ifBlank { System.getProperty("user.dir") ?: "." }

+    /**
+     * Walk [repoRoot] and record the ranked file/symbol map as a [RepoMapComputedEvent] once
+     * per session. The full map is recorded (the log); only a top-K slice is injected into
+     * context downstream. Replay reads the recorded event, never re-scanning (invariant #9).
+     */
+    suspend fun indexAndRecord(sessionId: SessionId, repoRoot: String) {
+        if (!config.enabled) return
+        runCatching {
+            val entries = indexer.index(java.nio.file.Path.of(repoRoot))
+            if (entries.isEmpty()) return
+            eventStore.append(
+                NewEvent(
+                    metadata = EventMetadata(
+                        eventId = EventId(UUID.randomUUID().toString()),
+                        sessionId = sessionId,
+                        timestamp = Clock.System.now(),
+                        schemaVersion = 1,
+                        causationId = null,
+                        correlationId = null,
+                    ),
+                    payload = RepoMapComputedEvent(
+                        sessionId = sessionId,
+                        repoRoot = repoRoot,
+                        entries = entries,
+                        computedAt = Clock.System.now(),
+                    ),
+                ),
+            )
+        }.exceptionOrNull()?.let { e ->
+            if (e is CancellationException) throw e
+            log.warn("repo-map indexing failed for {}: {}", repoRoot, e.message)
+        }
+    }
+
    /** Distil the session's decision journal into durable, repo-tagged L3 entries. */
    suspend fun persist(sessionId: SessionId, repoRoot: String) {
        if (!config.enabled) return
@@ -0,0 +1,79 @@
+package com.correx.apps.server.memory
+
+import com.correx.core.events.events.RepoMapEntry
+import java.nio.file.Files
+import java.nio.file.Path
+import kotlin.io.path.extension
+import kotlin.io.path.getLastModifiedTime
+import kotlin.io.path.isDirectory
+import kotlin.io.path.isRegularFile
+import kotlin.io.path.name
+import kotlin.io.path.readText
+import kotlin.io.path.relativeTo
+
+/**
+ * Walks a repo and produces a ranked file/symbol index. Reads the filesystem (a
+ * nondeterministic environment observation) — the caller records the result as a
+ * [com.correx.core.events.events.RepoMapComputedEvent] so replay reads recorded facts and
+ * never re-scans (invariant #9). Paths + top-level symbol names only, never file bodies.
+ *
+ * Scoring is recency-based: the most-recently-modified source file scores 1.0, the oldest
+ * ~0.0, linearly between. Recency is a cheap centrality proxy — files under active work rank
+ * highest, which is what a fresh session most wants to see.
+ */
+class RepoMapIndexer(
+    private val maxDepth: Int = 4,
+    private val ignoreGlobs: List<String> = emptyList(),
+) {
+    fun index(repoRoot: Path): List<RepoMapEntry> {
+        if (!repoRoot.isDirectory()) return emptyList()
+        val ignore = ignoreGlobs.toSet()
+
+        val files = Files.walk(repoRoot, maxDepth).use { stream ->
+            stream.filter { it.isRegularFile() }
+                .filter { it.extension in SYMBOL_PATTERNS.keys }
+                .filter { path -> path.relativeTo(repoRoot).none { it.name in ignore } }
+                .toList()
+        }
+        if (files.isEmpty()) return emptyList()
+
+        val mtimes = files.associateWith { runCatching { it.getLastModifiedTime().toMillis() }.getOrDefault(0L) }
+        val min = mtimes.values.min()
+        val max = mtimes.values.max()
+        val span = (max - min).coerceAtLeast(1L)
+
+        return files
+            .map { path ->
+                RepoMapEntry(
+                    path = path.relativeTo(repoRoot).toString(),
+                    score = (mtimes.getValue(path) - min).toDouble() / span,
+                    symbols = extractSymbols(path),
+                )
+            }
+            .sortedByDescending { it.score }
+    }
+
+    private fun extractSymbols(path: Path): List<String> {
+        val regex = SYMBOL_PATTERNS[path.extension] ?: return emptyList()
+        val text = runCatching { path.readText() }.getOrElse { return emptyList() }
+        return regex.findAll(text)
+            .mapNotNull { it.groups["name"]?.value }
+            .distinct()
+            .take(MAX_SYMBOLS_PER_FILE)
+            .toList()
+    }
+
+    private companion object {
+        const val MAX_SYMBOLS_PER_FILE = 40
+
+        // Top-level declarations only — best-effort per language. Bodies/locals are never matched.
+        val SYMBOL_PATTERNS: Map<String, Regex> = mapOf(
+            "kt" to Regex("""(?m)^\s*(?:public |internal |open |abstract |sealed |data )*(?:class|interface|object|fun)\s+(?<name>[A-Za-z_][A-Za-z0-9_]*)"""),
+            "java" to Regex("""(?m)^\s*(?:public |private |protected |final |abstract |static )*(?:class|interface|enum|record)\s+(?<name>[A-Za-z_][A-Za-z0-9_]*)"""),
+            "py" to Regex("""(?m)^(?:def|class)\s+(?<name>[A-Za-z_][A-Za-z0-9_]*)"""),
+            "go" to Regex("""(?m)^\s*(?:func\s+(?:\([^)]*\)\s*)?|type\s+)(?<name>[A-Za-z_][A-Za-z0-9_]*)"""),
+            "js" to Regex("""(?m)^\s*(?:export\s+)?(?:default\s+)?(?:async\s+)?(?:function|class)\s+(?<name>[A-Za-z_$][A-Za-z0-9_$]*)"""),
+            "ts" to Regex("""(?m)^\s*(?:export\s+)?(?:default\s+)?(?:async\s+)?(?:function|class|interface|type|enum)\s+(?<name>[A-Za-z_$][A-Za-z0-9_$]*)"""),
+        )
+    }
+}
@@ -0,0 +1,60 @@
+package com.correx.apps.server.memory
+
+import org.junit.jupiter.api.Assertions.assertEquals
+import org.junit.jupiter.api.Assertions.assertFalse
+import org.junit.jupiter.api.Assertions.assertTrue
+import org.junit.jupiter.api.io.TempDir
+import java.nio.file.Files
+import java.nio.file.Path
+import kotlin.io.path.createDirectories
+import kotlin.io.path.writeText
+import org.junit.jupiter.api.Test
+
+class RepoMapIndexerTest {
+
+    @Test
+    fun `indexes source files with top-level symbols, ranks recent highest, prunes ignores`(@TempDir root: Path) {
+        root.resolve("Old.kt").writeText("class Old\nfun helper() {}\n")
+        Thread.sleep(10)
+        root.resolve("New.kt").writeText("interface New\nobject Thing\n")
+
+        // Pruned directory must not appear.
+        root.resolve("build").createDirectories()
+        root.resolve("build/Generated.kt").writeText("class Generated\n")
+
+        // Non-source file must be ignored.
+        root.resolve("README.md").writeText("# hi\n")
+
+        val entries = RepoMapIndexer(maxDepth = 4, ignoreGlobs = listOf("build")).index(root)
+
+        val paths = entries.map { it.path }
+        assertTrue(paths.contains("New.kt"))
+        assertTrue(paths.contains("Old.kt"))
+        assertFalse(paths.any { it.contains("build") }, "ignored dir leaked: $paths")
+        assertFalse(paths.contains("README.md"), "non-source file leaked")
+
+        // Most-recently-modified ranks first.
+        assertEquals("New.kt", entries.first().path)
+
+        val newEntry = entries.first { it.path == "New.kt" }
+        assertTrue(newEntry.symbols.containsAll(listOf("New", "Thing")))
+        assertTrue(entries.first { it.path == "Old.kt" }.symbols.contains("Old"))
+    }
+
+    @Test
+    fun `empty or missing root yields empty map`(@TempDir root: Path) {
+        assertTrue(RepoMapIndexer().index(root).isEmpty())
+        assertTrue(RepoMapIndexer().index(root.resolve("does-not-exist")).isEmpty())
+    }
+
+    @Test
+    fun `maxDepth bounds recursion`(@TempDir root: Path) {
+        root.resolve("a/b/c").createDirectories()
+        root.resolve("Top.kt").writeText("class Top\n")
+        root.resolve("a/b/c/Deep.kt").writeText("class Deep\n")
+
+        val shallow = RepoMapIndexer(maxDepth = 1).index(root).map { it.path }
+        assertTrue(shallow.contains("Top.kt"))
+        assertFalse(shallow.any { it.contains("Deep") }, "depth cap breached: $shallow")
+    }
+}
@@ -418,6 +418,9 @@ object ConfigLoader {
            enabled = asBoolean(projectSection["enabled"], false),
            root = asString(projectSection["root"], ""),
            memoryK = asInt(projectSection["memory_k"], 5),
+            maxDepth = asInt(projectSection["max_depth"], 4),
+            ignoreGlobs = asStringList(projectSection["ignore_globs"]).ifEmpty { ProjectConfig.DEFAULT_IGNORES },
+            injectTopK = asInt(projectSection["inject_top_k"], 30),
        )

        val modelsSettings = ModelsSettings(
@@ -26,7 +26,16 @@ data class ProjectConfig(
    val enabled: Boolean = false,
    val root: String = "",
    val memoryK: Int = 5,
-)
+    val maxDepth: Int = 4,
+    val ignoreGlobs: List<String> = DEFAULT_IGNORES,
+    val injectTopK: Int = 30,
+) {
+    companion object {
+        val DEFAULT_IGNORES: List<String> = listOf(
+            ".git", "node_modules", "build", "target", ".gradle", "dist", ".idea", ".opencode",
+        )
+    }
+}

 /**
 * A user-declared artifact kind. [schemaPath] points to a JSON file holding the
@@ -30,6 +30,7 @@ import com.correx.core.events.events.ArtifactValidatingEvent
 import com.correx.core.events.events.EventMetadata
 import com.correx.core.events.events.EventPayload
 import com.correx.core.events.events.InferenceCompletedEvent
+import com.correx.core.events.events.RepoMapComputedEvent
 import com.correx.core.events.events.InferenceFailedEvent
 import com.correx.core.events.events.InferenceStartedEvent
 import com.correx.core.events.events.InferenceTimeoutEvent
@@ -117,6 +118,7 @@ import kotlin.coroutines.cancellation.CancellationException

 private const val MAX_TOOL_ROUNDS = 10
 private const val STAGE_COMPLETE_TOOL = "stage_complete"
+private const val REPO_MAP_INJECT_TOP_K = 30

@SuppressWarnings(
    "ForbiddenComment",
@@ -303,7 +305,9 @@ abstract class SessionOrchestrator(
                role = EntryRole.SYSTEM,
            ),
        )
-        var accumulatedEntries = systemPrompt + journalEntries + schemaEntries + promptEntries + steeringEntries
+        val repoMapEntries = buildRepoMapEntries(sessionId)
+        var accumulatedEntries =
+            systemPrompt + journalEntries + repoMapEntries + schemaEntries + promptEntries + steeringEntries
        val contextPack = contextPackBuilder.build(
            id = ContextPackId(UUID.randomUUID().toString()),
            sessionId = sessionId,
@@ -807,6 +811,38 @@ abstract class SessionOrchestrator(
        }
    }

+    /**
+     * Renders a top-K slice of the recorded repo map as a droppable L3 context entry. Reads
+     * the latest [RepoMapComputedEvent] from the log (replay-safe — never re-scans the FS).
+     * Not pinned: under budget pressure this yields before the working set (paths + symbols
+     * only, so the model can `file_read` for detail).
+     */
+    private suspend fun buildRepoMapEntries(sessionId: SessionId): List<ContextEntry> {
+        val map = eventStore.read(sessionId)
+            .mapNotNull { it.payload as? RepoMapComputedEvent }
+            .lastOrNull() ?: return emptyList()
+        val top = map.entries.sortedByDescending { it.score }.take(REPO_MAP_INJECT_TOP_K)
+        if (top.isEmpty()) return emptyList()
+        val content = buildString {
+            appendLine("## Repo map (top files by recency — read files for detail)")
+            top.forEach { entry ->
+                val symbols = if (entry.symbols.isEmpty()) "" else ": ${entry.symbols.joinToString(", ")}"
+                appendLine("- ${entry.path}$symbols")
+            }
+        }.trimEnd()
+        return listOf(
+            ContextEntry(
+                id = ContextEntryId(UUID.randomUUID().toString()),
+                layer = ContextLayer.L3,
+                content = content,
+                sourceType = "repoMap",
+                sourceId = "repo-map",
+                tokenEstimate = estimateTokens(content),
+                role = EntryRole.SYSTEM,
+            ),
+        )
+    }
+
    private suspend fun emitToolArtifacts(
        sessionId: SessionId,
        stageId: StageId,
@@ -80,6 +80,11 @@ capabilities = { General = 1.0, Coding = 0.7, Reasoning = 0.6, Summarization = 0
 enabled = false
 root = ""          # repo root key; empty = current working dir
 memory_k = 5       # distilled decisions retrieved per session start
+# Repo-map indexer (emits a ranked file/symbol index as a recorded observation at session
+# start; a top-K slice rides in context as droppable L3 memory — paths + symbol names only).
+max_depth = 4                                   # directory recursion cap
+inject_top_k = 30                               # entries injected into context (ranked by score)
+# ignore_globs = [".git", "node_modules", "build", "target", ".gradle", "dist", ".idea"]

 # Router configuration (optional, defaults shown below)
 [router]