pi-vcc 0.4.0

package/tests/extract-goals.test.ts

@@ -0,0 +1,86 @@
+import { describe, it, expect } from "vitest";
+import { extractGoals } from "../src/extract/goals";
+import type { NormalizedBlock } from "../src/types";
+
+describe("extractGoals", () => {
+  it("returns empty for no blocks", () => {
+    expect(extractGoals([])).toEqual([]);
+  });
+
+  it("returns empty when no user blocks", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "assistant", text: "hello" },
+    ];
+    expect(extractGoals(blocks)).toEqual([]);
+  });
+
+  it("extracts first user message lines as goals", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Fix login bug\nCheck auth flow" },
+    ];
+    const goals = extractGoals(blocks);
+    expect(goals).toEqual(["Fix login bug", "Check auth flow"]);
+  });
+
+  it("takes up to 6 lines from first user block", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "fix the login bug\ncheck auth flow\nupdate the tests\nrefactor utils\nclean up" },
+    ];
+    expect(extractGoals(blocks)).toHaveLength(5);
+  });
+
+  it("ignores subsequent user blocks", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "first goal" },
+      { kind: "assistant", text: "ok" },
+      { kind: "user", text: "second request" },
+    ];
+    expect(extractGoals(blocks)).toEqual(["first goal"]);
+  });
+
+  it("detects scope change with explicit pivot keywords", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Fix login bug" },
+      { kind: "assistant", text: "ok" },
+      { kind: "user", text: "Actually, instead let's refactor the auth module" },
+    ];
+    const goals = extractGoals(blocks);
+    expect(goals).toContain("Fix login bug");
+    expect(goals).toContain("[Scope change]");
+    expect(goals.some((g) => g.includes("refactor"))).toBe(true);
+  });
+
+  it("detects scope change from new task statements", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Fix login bug" },
+      { kind: "assistant", text: "done" },
+      { kind: "user", text: "Now implement the user registration flow" },
+    ];
+    const goals = extractGoals(blocks);
+    expect(goals).toContain("[Scope change]");
+  });
+
+  it("keeps latest scope change only", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Fix login bug" },
+      { kind: "assistant", text: "done" },
+      { kind: "user", text: "Actually, fix the signup page instead" },
+      { kind: "assistant", text: "ok" },
+      { kind: "user", text: "Change of plan, implement password reset" },
+    ];
+    const goals = extractGoals(blocks);
+    const scopeIdx = goals.indexOf("[Scope change]");
+    expect(goals[scopeIdx + 1]).toContain("password reset");
+  });
+
+  it("skips noise short user messages as goals", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "ok" },
+      { kind: "assistant", text: "hello" },
+      { kind: "user", text: "Fix the authentication module" },
+    ];
+    const goals = extractGoals(blocks);
+    expect(goals[0]).toContain("Fix the authentication");
+    expect(goals.some((g) => g === "ok")).toBe(false);
+  });
+});

package/tests/extract-preferences.test.ts

@@ -0,0 +1,30 @@
+import { describe, it, expect } from "vitest";
+import { extractPreferences } from "../src/extract/preferences";
+import type { NormalizedBlock } from "../src/types";
+
+describe("extractPreferences", () => {
+  it("returns empty for no blocks", () => {
+    expect(extractPreferences([])).toEqual([]);
+  });
+
+  it("captures preference patterns from user", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "I prefer TypeScript over JavaScript" },
+    ];
+    expect(extractPreferences(blocks).length).toBe(1);
+  });
+
+  it("ignores assistant blocks", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "assistant", text: "I always use best practices" },
+    ];
+    expect(extractPreferences(blocks)).toEqual([]);
+  });
+
+  it("captures please use pattern", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "please use bun instead of node" },
+    ];
+    expect(extractPreferences(blocks).length).toBe(1);
+  });
+});

package/tests/extract-references.test.ts

@@ -0,0 +1,475 @@
+import { describe, it, expect } from "vitest";
+import { extractReferences, formatReferences } from "../src/extract/references";
+import type { NormalizedBlock } from "../src/types";
+
+// ── extractReferences ──
+
+describe("extractReferences", () => {
+  // ── URLs ──
+
+  it("returns empty for no blocks", () => {
+    expect(extractReferences([]).urls).toEqual([]);
+    expect(extractReferences([]).githubRefs).toEqual([]);
+  });
+
+  it("extracts http URL from user message", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Check out http://example.com for the docs" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toContain("http://example.com");
+  });
+
+  it("extracts https URL from user message", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "See https://docs.site.com/api/v2#section" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toContain("https://docs.site.com/api/v2#section");
+  });
+
+  it("extracts IP:port URL", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "The server is at http://100.114.135.99:4747" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toContain("http://100.114.135.99:4747");
+  });
+
+  it("strips trailing punctuation from URLs", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "See https://example.com/docs. And also https://other.com/path," },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toContain("https://example.com/docs");
+    expect(refs.urls).toContain("https://other.com/path");
+  });
+
+  it("strips trailing closing paren from URLs", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Check the API (https://api.example.com/v2)" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toContain("https://api.example.com/v2");
+  });
+
+  it("extracts multiple URLs from one message", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "See https://a.com and http://b.com and https://c.com" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toHaveLength(3);
+  });
+
+  it("deduplicates same URL across blocks", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Use https://example.com" },
+      { kind: "assistant", text: "I checked https://example.com" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toEqual(["https://example.com"]);
+  });
+
+  it("extracts URLs from assistant blocks too", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "assistant", text: "The docs are at https://docs.example.com/api" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toContain("https://docs.example.com/api");
+  });
+
+  it("extracts URL from markdown link syntax", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Check [the docs](https://docs.example.com/guide) for details" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toContain("https://docs.example.com/guide");
+  });
+
+  it("does NOT extract file:// URLs", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Open file:///path/to/file" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).not.toContain("file:///path/to/file");
+    expect(refs.urls).toHaveLength(0);
+  });
+
+  it("skips tool_result blocks for URL extraction", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "tool_result", name: "bash", text: "curl https://example.com/api\nResponse: 200 OK", isError: false },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toHaveLength(0);
+  });
+
+  it("skips tool_call blocks for URL extraction", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "tool_call", name: "bash", args: { command: "curl https://example.com/api" } },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toHaveLength(0);
+  });
+
+  it("caps URLs at 10 entries", () => {
+    const urls = Array.from({ length: 15 }, (_, i) => `https://site${i}.com`);
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: urls.join(" ") },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toHaveLength(10);
+  });
+
+  // ── GitHub refs ──
+
+  it("extracts bare issue number #42", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Fix issue #42" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.githubRefs).toContain("#42");
+  });
+
+  it("extracts PR reference PR #7", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Merge PR #7" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.githubRefs).toContain("PR #7");
+  });
+
+  it("extracts pr#12 variant (normalized to 'pr #12')", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Review pr#12 when ready" },
+    ];
+    const refs = extractReferences(blocks);
+    // The PR regex captures 'pr' and '12' separately, joining as 'pr #12'
+    expect(refs.githubRefs).toContain("pr #12");
+  });
+
+  it("extracts owner/repo path", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "The repo is at buihongduc132/pi-plugins" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.githubRefs).toContain("buihongduc132/pi-plugins");
+  });
+
+  it("does NOT match file paths as owner/repo", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Edit src/components/button.tsx" },
+    ];
+    const refs = extractReferences(blocks);
+    // src/components doesn't look like owner/repo (owner must be alnum with hyphens)
+    expect(refs.githubRefs.every(r => !r.startsWith("src/"))).toBe(true);
+  });
+
+  it("extracts full GitHub URL as both URL and github ref", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "See https://github.com/owner/repo/issues/42" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toContain("https://github.com/owner/repo/issues/42");
+    expect(refs.githubRefs).toContain("owner/repo#42");
+  });
+
+  it("deduplicates github refs", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Fix #42" },
+      { kind: "assistant", text: "Also related to #42" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.githubRefs.filter(r => r === "#42")).toHaveLength(1);
+  });
+
+  it("caps github refs at 8 entries", () => {
+    const issues = Array.from({ length: 12 }, (_, i) => `#${i + 1}`).join(" ");
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: `Fix ${issues}` },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.githubRefs.length).toBeLessThanOrEqual(8);
+  });
+
+  // ── Versions ──
+
+  it("extracts v1.2.3 version", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Upgrade to v1.2.3" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.versions).toContain("v1.2.3");
+  });
+
+  it("extracts v0.3.12 version", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "We're at v0.3.12" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.versions).toContain("v0.3.12");
+  });
+
+  it("extracts bare semver 2.0.1", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "The version is 2.0.1 now" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.versions).toContain("2.0.1");
+  });
+
+  it("does NOT match IP address octets as versions", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "The server is at 192.168.1.100" },
+    ];
+    const refs = extractReferences(blocks);
+    // 192.168.1 is not a version (no v prefix and in IP context), 1.100 not standalone
+    expect(refs.versions).toHaveLength(0);
+  });
+
+  it("deduplicates versions", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Using v1.0.0" },
+      { kind: "assistant", text: "Confirmed v1.0.0" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.versions).toEqual(["v1.0.0"]);
+  });
+
+  it("caps versions at 5 entries", () => {
+    const versions = Array.from({ length: 8 }, (_, i) => `v${i}.0.0`).join(", ");
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: `Versions: ${versions}` },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.versions.length).toBeLessThanOrEqual(5);
+  });
+
+  // ── Branches ──
+
+  it("extracts feat/ branch", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Work on feat/new-auth module" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.branches).toContain("feat/new-auth");
+  });
+
+  it("extracts fix/ branch", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Check the fix/login-bug branch" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.branches).toContain("fix/login-bug");
+  });
+
+  it("extracts hotfix/, release/, chore/, refactor/, docs/ branches", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Branches: hotfix/urgent-fix, release/v2, chore/cleanup, refactor/api, docs/readme" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.branches).toContain("hotfix/urgent-fix");
+    expect(refs.branches).toContain("release/v2");
+    expect(refs.branches).toContain("chore/cleanup");
+    expect(refs.branches).toContain("refactor/api");
+    expect(refs.branches).toContain("docs/readme");
+  });
+
+  it("extracts test/ branch", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Work on test/unit-tests" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.branches).toContain("test/unit-tests");
+  });
+
+  it("does NOT extract random word/something as branch", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Just some/random text" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.branches).toHaveLength(0);
+  });
+
+  it("deduplicates branches", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "On feat/new-auth" },
+      { kind: "assistant", text: "Checking feat/new-auth" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.branches).toEqual(["feat/new-auth"]);
+  });
+
+  it("caps branches at 5 entries", () => {
+    const branches = Array.from({ length: 8 }, (_, i) => `feat/feature-${i}`).join(" ");
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: branches },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.branches.length).toBeLessThanOrEqual(5);
+  });
+
+  // ── Commit refs ──
+
+  it("extracts 7-char hex commit ref", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "This was fixed in abc1234" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.commitRefs).toContain("abc1234");
+  });
+
+  it("extracts 12-char hex commit ref", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "See commit abc123456789" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.commitRefs).toContain("abc123456789");
+  });
+
+  it("does NOT extract hex from tool_result blocks", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "tool_result", name: "bash", text: "[main abc1234] fix: login bug", isError: false },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.commitRefs).toHaveLength(0);
+  });
+
+  it("deduplicates commit refs", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: "Fixed in abc1234" },
+      { kind: "assistant", text: "abc1234 was the fix commit" },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.commitRefs).toEqual(["abc1234"]);
+  });
+
+  it("caps commit refs at 5 entries", () => {
+    const commits = Array.from({ length: 8 }, (_, i) =>
+      `abc${i.toString().padStart(4, "0")}`
+    ).join(" ");
+    const blocks: NormalizedBlock[] = [
+      { kind: "user", text: commits },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.commitRefs.length).toBeLessThanOrEqual(5);
+  });
+
+  // ── General skipping ──
+
+  it("skips thinking blocks", () => {
+    const blocks: NormalizedBlock[] = [
+      { kind: "thinking", text: "Check https://example.com and #42", redacted: false },
+    ];
+    const refs = extractReferences(blocks);
+    expect(refs.urls).toHaveLength(0);
+    expect(refs.githubRefs).toHaveLength(0);
+  });
+
+  it("returns all-zeros for empty blocks array", () => {
+    const refs = extractReferences([]);
+    expect(refs).toEqual({
+      urls: [],
+      githubRefs: [],
+      versions: [],
+      branches: [],
+      commitRefs: [],
+    });
+  });
+});
+
+// ── formatReferences ──
+
+describe("formatReferences", () => {
+  it("returns empty array when no references", () => {
+    expect(formatReferences({
+      urls: [],
+      githubRefs: [],
+      versions: [],
+      branches: [],
+      commitRefs: [],
+    })).toEqual([]);
+  });
+
+  it("formats URLs with prefix", () => {
+    const lines = formatReferences({
+      urls: ["https://example.com"],
+      githubRefs: [],
+      versions: [],
+      branches: [],
+      commitRefs: [],
+    });
+    expect(lines).toEqual(["URL: https://example.com"]);
+  });
+
+  it("formats GitHub refs with prefix", () => {
+    const lines = formatReferences({
+      urls: [],
+      githubRefs: ["#42", "buihongduc132/pi-plugins"],
+      versions: [],
+      branches: [],
+      commitRefs: [],
+    });
+    expect(lines).toContain("GitHub: #42, buihongduc132/pi-plugins");
+  });
+
+  it("formats versions with prefix", () => {
+    const lines = formatReferences({
+      urls: [],
+      githubRefs: [],
+      versions: ["v1.2.3"],
+      branches: [],
+      commitRefs: [],
+    });
+    expect(lines).toContain("Version: v1.2.3");
+  });
+
+  it("formats branches with prefix", () => {
+    const lines = formatReferences({
+      urls: [],
+      githubRefs: [],
+      versions: [],
+      branches: ["feat/new-auth"],
+      commitRefs: [],
+    });
+    expect(lines).toContain("Branch: feat/new-auth");
+  });
+
+  it("formats commit refs with prefix", () => {
+    const lines = formatReferences({
+      urls: [],
+      githubRefs: [],
+      versions: [],
+      branches: [],
+      commitRefs: ["abc1234"],
+    });
+    expect(lines).toContain("CommitRef: abc1234");
+  });
+
+  it("combines multiple categories", () => {
+    const lines = formatReferences({
+      urls: ["https://example.com", "https://other.com"],
+      githubRefs: ["#42"],
+      versions: ["v1.0.0"],
+      branches: ["feat/new"],
+      commitRefs: ["abc1234"],
+    });
+    expect(lines).toHaveLength(5);
+    expect(lines[0]).toBe("URL: https://example.com, https://other.com");
+    expect(lines[1]).toBe("GitHub: #42");
+    expect(lines[2]).toBe("Version: v1.0.0");
+    expect(lines[3]).toBe("Branch: feat/new");
+    expect(lines[4]).toBe("CommitRef: abc1234");
+  });
+
+  it("skips empty categories entirely", () => {
+    const lines = formatReferences({
+      urls: ["https://example.com"],
+      githubRefs: [],
+      versions: [],
+      branches: [],
+      commitRefs: [],
+    });
+    expect(lines).toHaveLength(1);
+    expect(lines[0]).toBe("URL: https://example.com");
+  });
+});