autonomous-coding-toolkit 1.0.0

package/scripts/tests/test-run-plan-team.sh

@@ -0,0 +1,157 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+source "$SCRIPT_DIR/../lib/run-plan-parser.sh"
+source "$SCRIPT_DIR/../lib/run-plan-routing.sh"
+source "$SCRIPT_DIR/../lib/run-plan-team.sh"
+
+FAILURES=0
+TESTS=0
+
+assert_eq() {
+    local desc="$1" expected="$2" actual="$3"
+    TESTS=$((TESTS + 1))
+    if [[ "$expected" != "$actual" ]]; then
+        echo "FAIL: $desc"
+        echo "  expected: $expected"
+        echo "  actual:   $actual"
+        FAILURES=$((FAILURES + 1))
+    else
+        echo "PASS: $desc"
+    fi
+}
+
+WORK=$(mktemp -d)
+trap 'rm -rf "$WORK"' EXIT
+
+# Plan with parallel structure: batch 1 is foundation, 2+3 are parallel, 4 depends on both
+cat > "$WORK/parallel-plan.md" << 'PLAN'
+## Batch 1: Foundation
+
+**Files:**
+- Create: `src/lib.sh`
+
+### Task 1: Create lib
+
+## Batch 2: Feature A
+
+**Files:**
+- Create: `src/feature-a.sh`
+context_refs: src/lib.sh
+
+### Task 2: Build feature A
+
+## Batch 3: Feature B
+
+**Files:**
+- Create: `src/feature-b.sh`
+context_refs: src/lib.sh
+
+### Task 3: Build feature B
+
+## Batch 4: Integration
+
+**Files:**
+- Modify: `src/feature-a.sh`
+- Modify: `src/feature-b.sh`
+context_refs: src/feature-a.sh, src/feature-b.sh
+
+### Task 4: Wire together
+PLAN
+
+# Test compute_parallel_groups with parallel plan
+dep_graph=$(build_dependency_graph "$WORK/parallel-plan.md")
+groups=$(compute_parallel_groups "$dep_graph" 1 4)
+
+# Should produce 3 groups: [1], [2,3], [4]
+group_count=$(echo "$groups" | jq 'length')
+assert_eq "parallel groups: 3 groups" "3" "$group_count"
+
+# First group has batch 1
+first_group=$(echo "$groups" | jq -c '.[0] | sort')
+assert_eq "parallel groups: group 1 = [1]" '[1]' "$first_group"
+
+# Second group has batches 2 and 3 (in some order)
+second_group=$(echo "$groups" | jq -c '.[1] | sort')
+assert_eq "parallel groups: group 2 = [2,3]" '[2,3]' "$second_group"
+
+# Third group has batch 4
+third_group=$(echo "$groups" | jq -c '.[2] | sort')
+assert_eq "parallel groups: group 3 = [4]" '[4]' "$third_group"
+
+# Sequential plan (each depends on previous)
+cat > "$WORK/sequential-plan.md" << 'PLAN'
+## Batch 1: Setup
+
+**Files:**
+- Create: `src/main.sh`
+
+### Task 1: Setup
+
+## Batch 2: Extend
+
+**Files:**
+- Modify: `src/main.sh`
+context_refs: src/main.sh
+
+### Task 2: Extend
+
+## Batch 3: Finalize
+
+**Files:**
+- Modify: `src/main.sh`
+context_refs: src/main.sh
+
+### Task 3: Finalize
+PLAN
+
+dep_graph=$(build_dependency_graph "$WORK/sequential-plan.md")
+groups=$(compute_parallel_groups "$dep_graph" 1 3)
+
+# Sequential plan: each batch is its own group
+group_count=$(echo "$groups" | jq 'length')
+assert_eq "sequential groups: 3 groups (no parallelism)" "3" "$group_count"
+
+# Each group has exactly one batch
+for ((g = 0; g < 3; g++)); do
+    size=$(echo "$groups" | jq ".[$g] | length")
+    assert_eq "sequential groups: group $((g+1)) has 1 batch" "1" "$size"
+done
+
+# Test with subset range (start_batch=2, end_batch=3)
+dep_graph=$(build_dependency_graph "$WORK/parallel-plan.md")
+groups=$(compute_parallel_groups "$dep_graph" 2 3)
+group_count=$(echo "$groups" | jq 'length')
+# Batches 2 and 3 both depend on batch 1 which is outside our range
+# Since batch 1 is not in range, its deps should be treated as satisfied
+assert_eq "subset range: batches 2,3 form 1 group (deps outside range)" "1" "$group_count"
+
+# === Bug #1: Team mode emits shared-worktree WARNING ===
+
+TEAM_FILE="$SCRIPT_DIR/../lib/run-plan-team.sh"
+
+TESTS=$((TESTS + 1))
+if grep -q 'WARNING.*shared worktree\|WARNING.*Team mode.*shared\|WARNING: Team mode' "$TEAM_FILE"; then
+    echo "PASS: run_mode_team emits shared-worktree WARNING"
+else
+    echo "FAIL: run_mode_team should emit a WARNING about shared worktree (bug #1)"
+    FAILURES=$((FAILURES + 1))
+fi
+
+# The warning should be emitted to stderr (>&2)
+TESTS=$((TESTS + 1))
+if grep -q 'WARNING.*>&2' "$TEAM_FILE"; then
+    echo "PASS: team mode WARNING goes to stderr"
+else
+    echo "FAIL: team mode WARNING should be sent to stderr (>&2) (bug #1)"
+    FAILURES=$((FAILURES + 1))
+fi
+
+echo ""
+echo "Results: $((TESTS - FAILURES))/$TESTS passed"
+if [[ $FAILURES -gt 0 ]]; then
+    echo "FAILURES: $FAILURES"
+    exit 1
+fi
+echo "ALL PASSED"

package/scripts/tests/test-scope-infer.sh

@@ -0,0 +1,150 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+source "$SCRIPT_DIR/test-helpers.sh"
+
+INFER="$SCRIPT_DIR/../scope-infer.sh"
+
+# --- Test: --help exits 0 ---
+assert_exit "--help exits 0" 0 "$INFER" --help
+
+# --- Test: --dry-run shows proposed scope without modifying files ---
+WORK=$(mktemp -d)
+trap 'rm -rf "$WORK"' EXIT
+
+# Create a lesson with no scope field, mentioning "HA" and "entity"
+cat > "$WORK/0099-test-ha-lesson.md" <<'LESSON'
+---
+id: 99
+title: "HA entity resolution fails on restart"
+severity: should-fix
+languages: [python]
+category: data-model
+pattern:
+  type: semantic
+  description: "HA entity lookup returns stale area"
+fix: "Refresh entity registry on restart"
+---
+
+## Observation
+Home Assistant entity area resolution uses a cached registry.
+LESSON
+
+dry_output=$("$INFER" --dir "$WORK" --dry-run 2>&1 || true)
+assert_contains "--dry-run mentions ha-aria" "domain:ha-aria" "$dry_output"
+
+# Verify file was NOT modified (dry run)
+TESTS=$((TESTS + 1))
+if grep -q '^scope:' "$WORK/0099-test-ha-lesson.md" 2>/dev/null; then
+    echo "FAIL: --dry-run should not modify lesson files"
+    FAILURES=$((FAILURES + 1))
+else
+    echo "PASS: --dry-run does not modify lesson files"
+fi
+
+# --- Test: --apply writes scope field to lesson ---
+"$INFER" --dir "$WORK" --apply > /dev/null 2>&1 || true
+
+TESTS=$((TESTS + 1))
+if grep -q '^scope:' "$WORK/0099-test-ha-lesson.md" 2>/dev/null; then
+    echo "PASS: --apply writes scope field to lesson"
+else
+    echo "FAIL: --apply should write scope field to lesson"
+    FAILURES=$((FAILURES + 1))
+fi
+
+# Verify inferred scope is correct
+scope_line=$(grep '^scope:' "$WORK/0099-test-ha-lesson.md" 2>/dev/null || true)
+assert_contains "--apply infers domain:ha-aria" "domain:ha-aria" "$scope_line"
+
+# --- Test: Lesson with existing scope is not modified ---
+cat > "$WORK/0098-already-scoped.md" <<'LESSON'
+---
+id: 98
+title: "Already scoped lesson"
+severity: should-fix
+scope: [language:python]
+languages: [python]
+category: silent-failures
+pattern:
+  type: semantic
+  description: "test"
+fix: "test"
+---
+LESSON
+
+apply_output=$("$INFER" --dir "$WORK" --apply 2>&1 || true)
+scope_line=$(grep '^scope:' "$WORK/0098-already-scoped.md" 2>/dev/null || true)
+assert_contains "existing scope preserved" "language:python" "$scope_line"
+
+# --- Test: Python-only lesson with no domain signals → language:python ---
+cat > "$WORK/0097-python-only.md" <<'LESSON'
+---
+id: 97
+title: "Generic Python anti-pattern"
+severity: should-fix
+languages: [python]
+category: async-traps
+pattern:
+  type: syntactic
+  regex: "some_pattern"
+  description: "test"
+fix: "test"
+---
+
+## Observation
+This is a generic Python lesson with no domain signals.
+LESSON
+
+"$INFER" --dir "$WORK" --apply > /dev/null 2>&1 || true
+scope_line=$(grep '^scope:' "$WORK/0097-python-only.md" 2>/dev/null || true)
+assert_contains "python-only → language:python" "language:python" "$scope_line"
+
+# --- Test: No signals → universal ---
+cat > "$WORK/0096-universal.md" <<'LESSON'
+---
+id: 96
+title: "Generic coding practice"
+severity: nice-to-have
+languages: [all]
+category: test-anti-patterns
+pattern:
+  type: syntactic
+  regex: "some_other_pattern"
+  description: "test"
+fix: "test"
+---
+
+## Observation
+This applies to all projects everywhere.
+LESSON
+
+"$INFER" --dir "$WORK" --apply > /dev/null 2>&1 || true
+scope_line=$(grep '^scope:' "$WORK/0096-universal.md" 2>/dev/null || true)
+assert_contains "no signals → universal" "universal" "$scope_line"
+
+# --- Test: Summary output shows counts ---
+WORK2=$(mktemp -d)
+trap 'rm -rf "$WORK" "$WORK2"' EXIT
+
+cat > "$WORK2/0001-test.md" <<'LESSON'
+---
+id: 1
+title: "Test lesson"
+severity: should-fix
+languages: [python]
+category: silent-failures
+pattern:
+  type: syntactic
+  regex: "test"
+  description: "test"
+fix: "test"
+---
+Generic content.
+LESSON
+
+summary_output=$("$INFER" --dir "$WORK2" --dry-run 2>&1 || true)
+assert_contains "summary shows count" "Inferred scope for" "$summary_output"
+
+report_results

package/scripts/tests/test-setup-ralph-loop.sh

@@ -0,0 +1,63 @@
+#!/usr/bin/env bash
+# Test setup-ralph-loop.sh
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+source "$SCRIPT_DIR/test-helpers.sh"
+
+SETUP_SCRIPT="$SCRIPT_DIR/../setup-ralph-loop.sh"
+WORK=$(mktemp -d)
+trap 'rm -rf "$WORK"' EXIT
+
+# Run setup-ralph-loop in a temp dir so it writes .claude/ there
+run_setup() {
+    (cd "$WORK" && bash "$SETUP_SCRIPT" "$@" 2>&1)
+}
+
+get_yaml_promise() {
+    sed -n 's/^completion_promise: *//p' "$WORK/.claude/ralph-loop.local.md"
+}
+
+# === Test: Basic prompt creates state file ===
+rm -rf "$WORK/.claude"
+output=$(run_setup "Hello world")
+assert_contains "basic: creates state file" "Ralph loop activated" "$output"
+assert_contains "basic: shows prompt" "Hello world" "$output"
+
+# === Test: Completion promise with double quotes is safely quoted (#22) ===
+rm -rf "$WORK/.claude"
+output=$(run_setup 'Build it' --completion-promise 'say "hello" done')
+yaml_val=$(get_yaml_promise)
+# jq produces JSON-quoted string: "say \"hello\" done"
+assert_contains "double quotes: safely quoted" '\"hello\"' "$yaml_val"
+
+# === Test: Completion promise with backslashes is safely quoted (#22) ===
+rm -rf "$WORK/.claude"
+output=$(run_setup 'Build it' --completion-promise 'path\to\thing')
+yaml_val=$(get_yaml_promise)
+# jq produces: "path\\to\\thing"
+assert_contains "backslash: safely quoted" '\\' "$yaml_val"
+
+# === Test: Completion promise with single quotes is safely quoted (#22) ===
+rm -rf "$WORK/.claude"
+output=$(run_setup 'Build it' --completion-promise "it's done")
+yaml_val=$(get_yaml_promise)
+assert_contains "single quote: safely quoted" "it's done" "$yaml_val"
+
+# === Test: Null completion promise stays null ===
+rm -rf "$WORK/.claude"
+output=$(run_setup "Build it")
+yaml_val=$(get_yaml_promise)
+assert_eq "null promise: stays null" "null" "$yaml_val"
+
+# === Test: --help exits 0 ===
+output=$(bash "$SETUP_SCRIPT" --help 2>&1 || echo "EXIT:$?")
+assert_contains "--help: shows usage" "USAGE:" "$output"
+assert_not_contains "--help: exit 0" "EXIT:" "$output"
+
+# === Test: No prompt exits 1 ===
+rm -rf "$WORK/.claude"
+output=$(run_setup 2>&1 || echo "EXIT:$?")
+assert_contains "no prompt: exits 1" "EXIT:1" "$output"
+
+report_results

package/scripts/tests/test-telegram-env.sh

@@ -0,0 +1,38 @@
+#!/usr/bin/env bash
+# Test telegram.sh — ACT_ENV_FILE support
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+REPO_ROOT="$(cd "$SCRIPT_DIR/../.." && pwd)"
+
+source "$SCRIPT_DIR/test-helpers.sh"
+
+# --- Setup ---
+WORK=$(mktemp -d)
+trap 'rm -rf "$WORK"' EXIT
+
+# Create a fake .env
+cat > "$WORK/test.env" <<'ENV'
+TELEGRAM_BOT_TOKEN=test-token-123
+TELEGRAM_CHAT_ID=test-chat-456
+ENV
+
+# --- Test 1: ACT_ENV_FILE overrides default ---
+unset TELEGRAM_BOT_TOKEN TELEGRAM_CHAT_ID 2>/dev/null || true
+ACT_ENV_FILE="$WORK/test.env" source "$REPO_ROOT/scripts/lib/telegram.sh"
+ACT_ENV_FILE="$WORK/test.env" _load_telegram_env
+assert_eq "ACT_ENV_FILE loads token" "test-token-123" "$TELEGRAM_BOT_TOKEN"
+assert_eq "ACT_ENV_FILE loads chat id" "test-chat-456" "$TELEGRAM_CHAT_ID"
+
+# --- Test 2: Explicit argument still works ---
+unset TELEGRAM_BOT_TOKEN TELEGRAM_CHAT_ID 2>/dev/null || true
+_load_telegram_env "$WORK/test.env"
+assert_eq "Explicit arg loads token" "test-token-123" "$TELEGRAM_BOT_TOKEN"
+
+# --- Test 3: Missing file returns error ---
+unset TELEGRAM_BOT_TOKEN TELEGRAM_CHAT_ID 2>/dev/null || true
+exit_code=0
+_load_telegram_env "$WORK/nonexistent.env" 2>/dev/null || exit_code=$?
+assert_eq "Missing env file returns 1" "1" "$exit_code"
+
+report_results

package/scripts/tests/test-telegram.sh

@@ -0,0 +1,121 @@
+#!/usr/bin/env bash
+# Test telegram.sh shared library
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+source "$SCRIPT_DIR/../lib/telegram.sh"
+
+FAILURES=0
+TESTS=0
+
+assert_eq() {
+    local desc="$1" expected="$2" actual="$3"
+    TESTS=$((TESTS + 1))
+    if [[ "$expected" != "$actual" ]]; then
+        echo "FAIL: $desc"
+        echo "  expected: $expected"
+        echo "  actual:   $actual"
+        FAILURES=$((FAILURES + 1))
+    else
+        echo "PASS: $desc"
+    fi
+}
+
+assert_exit() {
+    local desc="$1" expected_exit="$2"
+    shift 2
+    local actual_exit=0
+    "$@" || actual_exit=$?
+    TESTS=$((TESTS + 1))
+    if [[ "$expected_exit" != "$actual_exit" ]]; then
+        echo "FAIL: $desc"
+        echo "  expected exit: $expected_exit"
+        echo "  actual exit:   $actual_exit"
+        FAILURES=$((FAILURES + 1))
+    else
+        echo "PASS: $desc"
+    fi
+}
+
+WORK=$(mktemp -d)
+trap 'rm -rf "$WORK"' EXIT
+
+# === _load_telegram_env tests ===
+
+# Missing file
+assert_exit "_load_telegram_env: missing file returns 1" 1 \
+    _load_telegram_env "$WORK/nonexistent"
+
+# File without keys
+echo "SOME_OTHER_KEY=value" > "$WORK/empty.env"
+assert_exit "_load_telegram_env: missing keys returns 1" 1 \
+    _load_telegram_env "$WORK/empty.env"
+
+# File with both keys
+cat > "$WORK/valid.env" << 'ENVFILE'
+TELEGRAM_BOT_TOKEN=test-token-123
+TELEGRAM_CHAT_ID=test-chat-456
+ENVFILE
+# File with export prefix
+cat > "$WORK/export.env" << 'ENVFILE'
+export TELEGRAM_BOT_TOKEN=export-token-789
+export TELEGRAM_CHAT_ID=export-chat-012
+ENVFILE
+assert_exit "_load_telegram_env: export prefix returns 0" 0 \
+    _load_telegram_env "$WORK/export.env"
+assert_eq "_load_telegram_env: export token parsed" "export-token-789" "$TELEGRAM_BOT_TOKEN"
+assert_eq "_load_telegram_env: export chat_id parsed" "export-chat-012" "$TELEGRAM_CHAT_ID"
+
+# File with both keys (no export)
+assert_exit "_load_telegram_env: valid file returns 0" 0 \
+    _load_telegram_env "$WORK/valid.env"
+assert_eq "_load_telegram_env: token loaded" "test-token-123" "$TELEGRAM_BOT_TOKEN"
+assert_eq "_load_telegram_env: chat_id loaded" "test-chat-456" "$TELEGRAM_CHAT_ID"
+
+# === Quote stripping (#7): values wrapped in double quotes ===
+
+cat > "$WORK/quoted-double.env" << 'ENVFILE'
+TELEGRAM_BOT_TOKEN="quoted-token-abc"
+TELEGRAM_CHAT_ID="quoted-chat-def"
+ENVFILE
+assert_exit "_load_telegram_env: double-quoted values returns 0" 0 \
+    _load_telegram_env "$WORK/quoted-double.env"
+assert_eq "_load_telegram_env: double-quoted token stripped" "quoted-token-abc" "$TELEGRAM_BOT_TOKEN"
+assert_eq "_load_telegram_env: double-quoted chat_id stripped" "quoted-chat-def" "$TELEGRAM_CHAT_ID"
+
+# === Quote stripping (#7): values wrapped in single quotes ===
+
+cat > "$WORK/quoted-single.env" << 'ENVFILE'
+TELEGRAM_BOT_TOKEN='single-token-ghi'
+TELEGRAM_CHAT_ID='single-chat-jkl'
+ENVFILE
+assert_exit "_load_telegram_env: single-quoted values returns 0" 0 \
+    _load_telegram_env "$WORK/quoted-single.env"
+assert_eq "_load_telegram_env: single-quoted token stripped" "single-token-ghi" "$TELEGRAM_BOT_TOKEN"
+assert_eq "_load_telegram_env: single-quoted chat_id stripped" "single-chat-jkl" "$TELEGRAM_CHAT_ID"
+
+# === Quote stripping (#7): export with double quotes ===
+
+cat > "$WORK/export-quoted.env" << 'ENVFILE'
+export TELEGRAM_BOT_TOKEN="export-quoted-mno"
+export TELEGRAM_CHAT_ID="export-quoted-pqr"
+ENVFILE
+assert_exit "_load_telegram_env: export double-quoted returns 0" 0 \
+    _load_telegram_env "$WORK/export-quoted.env"
+assert_eq "_load_telegram_env: export double-quoted token stripped" "export-quoted-mno" "$TELEGRAM_BOT_TOKEN"
+assert_eq "_load_telegram_env: export double-quoted chat_id stripped" "export-quoted-pqr" "$TELEGRAM_CHAT_ID"
+
+# === _send_telegram without credentials ===
+
+unset TELEGRAM_BOT_TOKEN TELEGRAM_CHAT_ID
+assert_exit "_send_telegram: no creds returns 0 (skip)" 0 \
+    _send_telegram "test message"
+
+# === Summary ===
+echo ""
+echo "Results: $((TESTS - FAILURES))/$TESTS passed"
+if [[ $FAILURES -gt 0 ]]; then
+    echo "FAILURES: $FAILURES"
+    exit 1
+fi
+echo "ALL PASSED"

package/scripts/tests/test-telemetry.sh

@@ -0,0 +1,46 @@
+#!/usr/bin/env bash
+# Test scripts/telemetry.sh — telemetry capture, show, export, reset
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+REPO_ROOT="$(cd "$SCRIPT_DIR/../.." && pwd)"
+TELEMETRY="$REPO_ROOT/scripts/telemetry.sh"
+
+source "$SCRIPT_DIR/test-helpers.sh"
+
+# --- Setup ---
+WORK=$(mktemp -d)
+trap 'rm -rf "$WORK"' EXIT
+mkdir -p "$WORK/logs"
+
+# --- Test 1: record writes to telemetry.jsonl ---
+bash "$TELEMETRY" record --project-root "$WORK" \
+    --batch-number 1 --passed true --strategy superpowers \
+    --duration 120 --cost 0.42 --test-delta 5 2>&1 || true
+assert_eq "record creates telemetry.jsonl" "true" \
+    "$([ -f "$WORK/logs/telemetry.jsonl" ] && echo true || echo false)"
+
+# --- Test 2: record appends valid JSON ---
+line=$(head -1 "$WORK/logs/telemetry.jsonl")
+jq_exit=0
+echo "$line" | jq . >/dev/null 2>&1 || jq_exit=$?
+assert_eq "record writes valid JSON" "0" "$jq_exit"
+
+# --- Test 3: show produces dashboard output ---
+output=$(bash "$TELEMETRY" show --project-root "$WORK" 2>&1 || true)
+assert_contains "show displays header" "Telemetry Dashboard" "$output"
+
+# --- Test 4: export produces anonymized output ---
+bash "$TELEMETRY" export --project-root "$WORK" > "$WORK/export.json" 2>&1 || true
+assert_eq "export creates output" "true" "$([ -s "$WORK/export.json" ] && echo true || echo false)"
+
+# --- Test 5: reset clears telemetry ---
+bash "$TELEMETRY" reset --project-root "$WORK" --yes 2>&1 || true
+if [[ -f "$WORK/logs/telemetry.jsonl" ]]; then
+    line_count=$(wc -l < "$WORK/logs/telemetry.jsonl")
+    assert_eq "reset clears telemetry" "0" "$line_count"
+else
+    pass "reset removes telemetry file"
+fi
+
+report_results