bugfix

2026-06-22 08:55:57 +08:00
parent dbb87823e8
commit 6ade6e8fa9
325 changed files with 41131 additions and 855 deletions
@@ -0,0 +1,49 @@
+-- Code index (pgvector RAG). Guarded CREATE EXTENSION so deployments without
+-- superuser/owner privileges or without the pgvector image fail loudly here
+-- (and the code degrades to keyword fallback when no embedding endpoint is set).
+CREATE EXTENSION IF NOT EXISTS vector;
+
+-- code_index_runs tracks one (workspace, commit, embedding_model) build attempt.
+CREATE TABLE code_index_runs (
+    id                    uuid PRIMARY KEY,
+    workspace_id          uuid NOT NULL REFERENCES workspaces(id) ON DELETE CASCADE,
+    worktree_id           uuid NULL REFERENCES workspace_worktrees(id) ON DELETE SET NULL,
+    branch                text NOT NULL,
+    commit_sha            text NOT NULL,
+    status                text NOT NULL DEFAULT 'pending'
+                              CHECK (status IN ('pending','running','completed','failed','stale')),
+    embedding_endpoint_id uuid REFERENCES llm_endpoints(id),
+    embedding_model       text NOT NULL,
+    dims                  int NOT NULL,
+    files_indexed         int NOT NULL DEFAULT 0,
+    chunks_indexed        int NOT NULL DEFAULT 0,
+    error                 text,
+    started_at            timestamptz,
+    finished_at           timestamptz,
+    created_at            timestamptz NOT NULL DEFAULT now(),
+    UNIQUE (workspace_id, commit_sha, embedding_model)
+);
+
+CREATE INDEX idx_code_index_runs_ws_status ON code_index_runs (workspace_id, status);
+
+-- code_chunks stores embedded line-window chunks. The vector dimension is fixed
+-- at migration time; one embedding model per deployment is enforced via config.
+CREATE TABLE code_chunks (
+    id           uuid PRIMARY KEY,
+    run_id       uuid NOT NULL REFERENCES code_index_runs(id) ON DELETE CASCADE,
+    workspace_id uuid NOT NULL,
+    commit_sha   text NOT NULL,
+    file_path    text NOT NULL,
+    lang         text,
+    start_line   int NOT NULL,
+    end_line     int NOT NULL,
+    content      text NOT NULL,
+    content_sha  bytea NOT NULL,
+    token_count  int,
+    embedding    vector(1536) NOT NULL,
+    created_at   timestamptz NOT NULL DEFAULT now()
+);
+
+CREATE INDEX idx_code_chunks_embedding ON code_chunks USING hnsw (embedding vector_cosine_ops);
+CREATE INDEX idx_code_chunks_ws_commit ON code_chunks (workspace_id, commit_sha);
+CREATE INDEX idx_code_chunks_file ON code_chunks (workspace_id, file_path);