@daniellins/power-claude 0.16.3

package/template/pods/dev-ia/tools/db-ops/data-loader.sh

@@ -0,0 +1,616 @@
+#!/usr/bin/env bash
+# =============================================================================
+# data-loader.sh - CSV and Seed Loader (Worker Script)
+# =============================================================================
+# execution_type: Worker
+# pattern: EXEC-W-001
+# tasks: db-load-csv, db-seed
+# token_savings: 88%
+# =============================================================================
+
+set -euo pipefail
+
+# Colors
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+BLUE='\033[0;34m'
+NC='\033[0m'
+
+# Defaults
+DB_URL=""
+DRY_RUN=false
+VALIDATE=false
+VERBOSE=false
+DELIMITER=","
+IMPORT_MODE="safe"
+ENVIRONMENT=""
+MODE=""
+TABLE_REF=""
+DATA_FILE=""
+SEED_FILE=""
+TABLE_SCHEMA="public"
+TABLE_NAME=""
+QUALIFIED_TABLE=""
+QUALIFIED_STAGING=""
+
+declare -a TEMP_FILES=()
+declare -a HEADER_COLUMNS=()
+declare -a PRIMARY_KEY_COLUMNS=()
+declare -a NON_PRIMARY_COLUMNS=()
+
+usage() {
+    cat << EOF
+Usage:
+  $(basename "$0") [OPTIONS] <table> <csv_file>
+  $(basename "$0") [OPTIONS] --seed <sql_file>
+  $(basename "$0") [OPTIONS] <sql_file>
+
+Load CSV data into PostgreSQL/Supabase or apply seed SQL safely.
+
+CSV OPTIONS:
+  --mode <safe|fast|upsert>  Import strategy (default: safe)
+  --delimiter <char>         CSV delimiter (default: ,)
+
+SEED OPTIONS:
+  --seed <path>              Treat input as seed SQL file
+  --env <name>               Label target environment for logging
+
+COMMON OPTIONS:
+  --dry-run                  Execute inside rollback
+  --validate                 Run extra safety validation before execution
+  -v, --verbose              Verbose output
+  -h, --help                 Show this help
+
+EXAMPLES:
+  $(basename "$0") users data/users.csv
+  $(basename "$0") users data/users.csv --delimiter '|'
+  $(basename "$0") users data/users.csv --dry-run
+  $(basename "$0") --seed supabase/seeds/dev.sql
+  $(basename "$0") supabase/seeds/dev.sql --validate --dry-run
+
+ENVIRONMENT:
+  SUPABASE_DB_URL or DATABASE_URL  Required. PostgreSQL connection string.
+EOF
+}
+
+log_info() { echo -e "${BLUE}ℹ${NC} $1"; }
+log_success() { echo -e "${GREEN}✓${NC} $1"; }
+log_warning() { echo -e "${YELLOW}⚠${NC} $1"; }
+log_error() { echo -e "${RED}✗${NC} $1"; }
+
+cleanup() {
+    local file
+    for file in "${TEMP_FILES[@]}"; do
+        [[ -e "$file" ]] && rm -f "$file"
+    done
+}
+
+trap cleanup EXIT
+
+register_temp_file() {
+    TEMP_FILES+=("$1")
+}
+
+escape_sql_literal() {
+    printf '%s' "$1" | sed "s/'/''/g"
+}
+
+quote_ident() {
+    printf '"%s"' "$1"
+}
+
+join_sql_list() {
+    local joined=""
+    local item
+    for item in "$@"; do
+        if [[ -n "$joined" ]]; then
+            joined+=", "
+        fi
+        joined+="$item"
+    done
+    printf '%s' "$joined"
+}
+
+quoted_identifier_list() {
+    local quoted_columns=()
+    local column
+    for column in "$@"; do
+        quoted_columns+=("$(quote_ident "$column")")
+    done
+    join_sql_list "${quoted_columns[@]}"
+}
+
+array_contains() {
+    local needle="$1"
+    shift
+    local item
+    for item in "$@"; do
+        [[ "$item" == "$needle" ]] && return 0
+    done
+    return 1
+}
+
+check_env() {
+    DB_URL="${SUPABASE_DB_URL:-${DATABASE_URL:-}}"
+    if [[ -z "$DB_URL" ]]; then
+        log_error "SUPABASE_DB_URL or DATABASE_URL not set"
+        exit 1
+    fi
+}
+
+validate_identifier() {
+    local value="$1"
+    if [[ ! "$value" =~ ^[A-Za-z_][A-Za-z0-9_]*$ ]]; then
+        log_error "Invalid SQL identifier: $value"
+        exit 1
+    fi
+}
+
+resolve_table_reference() {
+    TABLE_SCHEMA="public"
+    TABLE_NAME="$1"
+
+    if [[ "$1" == *.* ]]; then
+        TABLE_SCHEMA="${1%%.*}"
+        TABLE_NAME="${1##*.}"
+    fi
+
+    validate_identifier "$TABLE_SCHEMA"
+    validate_identifier "$TABLE_NAME"
+    QUALIFIED_TABLE="$(quote_ident "$TABLE_SCHEMA").$(quote_ident "$TABLE_NAME")"
+}
+
+resolve_absolute_path() {
+    local input_path="$1"
+    if [[ "$input_path" = /* ]]; then
+        printf '%s' "$input_path"
+    else
+        printf '%s/%s' "$(pwd)" "$input_path"
+    fi
+}
+
+collect_header_columns() {
+    local header_line raw_column column
+
+    header_line=$(head -n 1 "$DATA_FILE" | tr -d '\r')
+    if [[ -z "$header_line" ]]; then
+        log_error "CSV header is empty: $DATA_FILE"
+        exit 1
+    fi
+
+    HEADER_COLUMNS=()
+    IFS="$DELIMITER" read -r -a raw_columns <<< "$header_line"
+
+    for raw_column in "${raw_columns[@]}"; do
+        column=$(printf '%s' "$raw_column" | sed 's/^[[:space:]]*//;s/[[:space:]]*$//')
+        column="${column#\"}"
+        column="${column%\"}"
+        validate_identifier "$column"
+        HEADER_COLUMNS+=("$column")
+    done
+
+    if [[ "${#HEADER_COLUMNS[@]}" -eq 0 ]]; then
+        log_error "No columns detected in CSV header"
+        exit 1
+    fi
+}
+
+load_primary_key_columns() {
+    PRIMARY_KEY_COLUMNS=()
+
+    while IFS= read -r column; do
+        [[ -n "$column" ]] && PRIMARY_KEY_COLUMNS+=("$column")
+    done < <(
+        psql "$DB_URL" -t -A <<SQL
+SELECT a.attname
+FROM pg_index i
+JOIN pg_class c ON c.oid = i.indrelid
+JOIN pg_namespace n ON n.oid = c.relnamespace
+JOIN pg_attribute a ON a.attrelid = c.oid AND a.attnum = ANY(i.indkey)
+WHERE i.indisprimary
+  AND n.nspname = '$TABLE_SCHEMA'
+  AND c.relname = '$TABLE_NAME'
+ORDER BY array_position(i.indkey, a.attnum);
+SQL
+    )
+}
+
+validate_table_exists() {
+    local exists
+    exists=$(psql "$DB_URL" -t -A <<SQL
+SELECT EXISTS (
+  SELECT 1
+  FROM information_schema.tables
+  WHERE table_schema = '$TABLE_SCHEMA'
+    AND table_name = '$TABLE_NAME'
+);
+SQL
+    )
+
+    if [[ "$exists" != "t" ]]; then
+        log_error "Table not found: $TABLE_SCHEMA.$TABLE_NAME"
+        exit 1
+    fi
+}
+
+validate_csv_inputs() {
+    local available_columns missing_columns="" column
+
+    if [[ ! -f "$DATA_FILE" ]]; then
+        log_error "CSV file not found: $DATA_FILE"
+        exit 1
+    fi
+
+    validate_table_exists
+    collect_header_columns
+
+    available_columns=$(psql "$DB_URL" -t -A <<SQL
+SELECT column_name
+FROM information_schema.columns
+WHERE table_schema = '$TABLE_SCHEMA'
+  AND table_name = '$TABLE_NAME'
+ORDER BY ordinal_position;
+SQL
+    )
+
+    for column in "${HEADER_COLUMNS[@]}"; do
+        if ! printf '%s\n' "$available_columns" | grep -Fxq "$column"; then
+            if [[ -n "$missing_columns" ]]; then
+                missing_columns+=", "
+            fi
+            missing_columns+="$column"
+        fi
+    done
+
+    if [[ -n "$missing_columns" ]]; then
+        log_error "CSV columns not found in $TABLE_SCHEMA.$TABLE_NAME: $missing_columns"
+        exit 1
+    fi
+}
+
+build_column_lists() {
+    local column quoted_columns=()
+
+    for column in "${HEADER_COLUMNS[@]}"; do
+        quoted_columns+=("$(quote_ident "$column")")
+    done
+    COLUMN_LIST_SQL=$(join_sql_list "${quoted_columns[@]}")
+
+    NON_PRIMARY_COLUMNS=()
+    for column in "${HEADER_COLUMNS[@]}"; do
+        if ! array_contains "$column" "${PRIMARY_KEY_COLUMNS[@]}"; then
+            NON_PRIMARY_COLUMNS+=("$column")
+        fi
+    done
+}
+
+build_merge_sql() {
+    local conflict_columns=()
+    local update_assignments=()
+    local column primary_key_present=true
+
+    if [[ "${#PRIMARY_KEY_COLUMNS[@]}" -eq 0 ]]; then
+        log_warning "No primary key detected; falling back to INSERT"
+        printf 'INSERT INTO %s (%s) SELECT %s FROM %s' \
+            "$QUALIFIED_TABLE" "$COLUMN_LIST_SQL" "$COLUMN_LIST_SQL" "$QUALIFIED_STAGING"
+        return
+    fi
+
+    for column in "${PRIMARY_KEY_COLUMNS[@]}"; do
+        if ! array_contains "$column" "${HEADER_COLUMNS[@]}"; then
+            primary_key_present=false
+            break
+        fi
+        conflict_columns+=("$(quote_ident "$column")")
+    done
+
+    if [[ "$primary_key_present" != true ]]; then
+        log_warning "CSV payload does not include the full primary key; falling back to INSERT"
+        printf 'INSERT INTO %s (%s) SELECT %s FROM %s' \
+            "$QUALIFIED_TABLE" "$COLUMN_LIST_SQL" "$COLUMN_LIST_SQL" "$QUALIFIED_STAGING"
+        return
+    fi
+
+    if [[ "${#NON_PRIMARY_COLUMNS[@]}" -eq 0 ]]; then
+        printf 'INSERT INTO %s (%s) SELECT %s FROM %s ON CONFLICT (%s) DO NOTHING' \
+            "$QUALIFIED_TABLE" \
+            "$COLUMN_LIST_SQL" \
+            "$COLUMN_LIST_SQL" \
+            "$QUALIFIED_STAGING" \
+            "$(join_sql_list "${conflict_columns[@]}")"
+        return
+    fi
+
+    for column in "${NON_PRIMARY_COLUMNS[@]}"; do
+        update_assignments+=("$(quote_ident "$column") = EXCLUDED.$(quote_ident "$column")")
+    done
+
+    printf 'INSERT INTO %s (%s) SELECT %s FROM %s ON CONFLICT (%s) DO UPDATE SET %s' \
+        "$QUALIFIED_TABLE" \
+        "$COLUMN_LIST_SQL" \
+        "$COLUMN_LIST_SQL" \
+        "$QUALIFIED_STAGING" \
+        "$(join_sql_list "${conflict_columns[@]}")" \
+        "$(join_sql_list "${update_assignments[@]}")"
+}
+
+write_csv_sql_file() {
+    local sql_file="$1"
+    local absolute_file escaped_file escaped_delimiter merge_sql
+
+    absolute_file=$(resolve_absolute_path "$DATA_FILE")
+    escaped_file=$(escape_sql_literal "$absolute_file")
+    escaped_delimiter=$(escape_sql_literal "$DELIMITER")
+
+    if [[ "$IMPORT_MODE" == "fast" ]]; then
+        {
+            echo "BEGIN;"
+            printf "\\copy %s (%s) FROM '%s' WITH (FORMAT csv, HEADER true, DELIMITER '%s')\n" \
+                "$QUALIFIED_TABLE" "$COLUMN_LIST_SQL" "$escaped_file" "$escaped_delimiter"
+            printf "SELECT COUNT(*) AS target_rows_after FROM %s;\n" "$QUALIFIED_TABLE"
+            if $DRY_RUN; then
+                echo "ROLLBACK;"
+            else
+                echo "COMMIT;"
+            fi
+        } > "$sql_file"
+        return
+    fi
+
+    STAGING_TABLE="_db_ops_${TABLE_NAME}_$$"
+    QUALIFIED_STAGING="$(quote_ident "$STAGING_TABLE")"
+    merge_sql=$(build_merge_sql)
+
+        {
+            echo "BEGIN;"
+            printf "CREATE TEMP TABLE %s (LIKE %s INCLUDING ALL) ON COMMIT DROP;\n" \
+                "$QUALIFIED_STAGING" "$QUALIFIED_TABLE"
+            printf "\\copy %s (%s) FROM '%s' WITH (FORMAT csv, HEADER true, DELIMITER '%s')\n" \
+                "$QUALIFIED_STAGING" "$COLUMN_LIST_SQL" "$escaped_file" "$escaped_delimiter"
+            printf "SELECT COUNT(*) AS staged_rows FROM %s;\n" "$QUALIFIED_STAGING"
+
+        if [[ "${#PRIMARY_KEY_COLUMNS[@]}" -gt 0 ]]; then
+            local pk_list_sql
+            pk_list_sql=$(quoted_identifier_list "${PRIMARY_KEY_COLUMNS[@]}")
+            printf "DO \$\$\nBEGIN\n"
+            printf "  IF EXISTS (\n"
+            printf "    SELECT 1\n"
+            printf "    FROM (\n"
+            printf "      SELECT %s\n" "$pk_list_sql"
+            printf "      FROM %s\n" "$QUALIFIED_STAGING"
+            printf "      GROUP BY %s\n" "$pk_list_sql"
+            printf "      HAVING COUNT(*) > 1\n"
+            printf "    ) duplicated_keys\n"
+            printf "  ) THEN\n"
+            printf "    RAISE EXCEPTION 'Duplicate primary key rows detected in CSV payload';\n"
+            printf "  END IF;\n"
+            printf "END\n\$\$;\n"
+        fi
+
+        printf "%s;\n" "$merge_sql"
+        printf "SELECT COUNT(*) AS target_rows_after FROM %s;\n" "$QUALIFIED_TABLE"
+        if $DRY_RUN; then
+            echo "ROLLBACK;"
+        else
+            echo "COMMIT;"
+        fi
+    } > "$sql_file"
+}
+
+run_csv_import() {
+    local sql_file
+    local line_count
+
+    if [[ "${#DELIMITER}" -ne 1 ]]; then
+        log_error "Delimiter must be a single character"
+        exit 1
+    fi
+
+    if $VALIDATE; then
+        IMPORT_MODE="safe"
+    fi
+
+    case "$IMPORT_MODE" in
+        safe|fast|upsert) ;;
+        *)
+            log_error "Invalid import mode: $IMPORT_MODE"
+            exit 1
+            ;;
+    esac
+
+    resolve_table_reference "$TABLE_REF"
+    validate_csv_inputs
+    load_primary_key_columns
+    build_column_lists
+
+    line_count=$(wc -l < "$DATA_FILE" | tr -d ' ')
+
+    log_info "CSV file: $DATA_FILE"
+    log_info "Target table: $TABLE_SCHEMA.$TABLE_NAME"
+    log_info "Mode: $IMPORT_MODE"
+    log_info "Header columns: $(join_sql_list "${HEADER_COLUMNS[@]}")"
+    log_info "CSV rows (including header): $line_count"
+
+    sql_file=$(mktemp "/tmp/db-ops-data-loader-XXXXXX.sql")
+    register_temp_file "$sql_file"
+    write_csv_sql_file "$sql_file"
+
+    psql "$DB_URL" -v ON_ERROR_STOP=1 -f "$sql_file"
+
+    if $DRY_RUN; then
+        log_warning "Dry-run complete. Changes were rolled back."
+    else
+        log_success "CSV load completed."
+    fi
+}
+
+seed_has_explicit_transaction() {
+    grep -Eiq '^[[:space:]]*(BEGIN|START[[:space:]]+TRANSACTION|COMMIT|ROLLBACK)\b' "$SEED_FILE"
+}
+
+validate_seed_inputs() {
+    if [[ ! -f "$SEED_FILE" ]]; then
+        log_error "Seed file not found: $SEED_FILE"
+        exit 1
+    fi
+
+    if $VALIDATE; then
+        if grep -Eiq '(^|[^A-Z_])(TRUNCATE|DELETE[[:space:]]+FROM)\b' "$SEED_FILE"; then
+            log_error "Seed validation failed: destructive statements detected"
+            exit 1
+        fi
+
+        if ! grep -Eiq 'ON[[:space:]]+CONFLICT|WHERE[[:space:]]+NOT[[:space:]]+EXISTS|MERGE[[:space:]]+INTO' "$SEED_FILE"; then
+            log_warning "Seed file does not advertise an idempotent pattern (ON CONFLICT / WHERE NOT EXISTS / MERGE)"
+        else
+            log_success "Seed file exposes an idempotent pattern"
+        fi
+    fi
+}
+
+run_seed() {
+    local sql_file
+
+    validate_seed_inputs
+
+    log_info "Seed file: $SEED_FILE"
+    if [[ -n "$ENVIRONMENT" ]]; then
+        log_info "Target environment: $ENVIRONMENT"
+    fi
+    if [[ "$ENVIRONMENT" == "production" || "$ENVIRONMENT" == "prod" ]]; then
+        log_warning "Production seed requested. Ensure this was explicit."
+    fi
+
+    if seed_has_explicit_transaction; then
+        if $DRY_RUN; then
+            log_error "Dry-run is not supported when the seed file manages its own transaction"
+            exit 1
+        fi
+        psql "$DB_URL" -v ON_ERROR_STOP=1 -f "$SEED_FILE"
+        log_success "Seed applied."
+        return
+    fi
+
+    sql_file=$(mktemp "/tmp/db-ops-seed-loader-XXXXXX.sql")
+    register_temp_file "$sql_file"
+
+    {
+        echo "BEGIN;"
+        cat "$SEED_FILE"
+        if $DRY_RUN; then
+            echo "ROLLBACK;"
+        else
+            echo "COMMIT;"
+        fi
+    } > "$sql_file"
+
+    psql "$DB_URL" -v ON_ERROR_STOP=1 -f "$sql_file"
+
+    if $DRY_RUN; then
+        log_warning "Dry-run complete. Seed changes were rolled back."
+    else
+        log_success "Seed applied."
+    fi
+}
+
+parse_args() {
+    local positional=()
+
+    while [[ $# -gt 0 ]]; do
+        case "$1" in
+            --seed)
+                SEED_FILE="${2:-}"
+                MODE="seed"
+                shift 2
+                ;;
+            --delimiter)
+                DELIMITER="${2:-}"
+                shift 2
+                ;;
+            --mode)
+                IMPORT_MODE="${2:-}"
+                shift 2
+                ;;
+            --env)
+                ENVIRONMENT="${2:-}"
+                shift 2
+                ;;
+            --dry-run)
+                DRY_RUN=true
+                shift
+                ;;
+            --validate)
+                VALIDATE=true
+                shift
+                ;;
+            -v|--verbose)
+                VERBOSE=true
+                shift
+                ;;
+            -h|--help)
+                usage
+                exit 0
+                ;;
+            --)
+                shift
+                while [[ $# -gt 0 ]]; do
+                    positional+=("$1")
+                    shift
+                done
+                ;;
+            *)
+                positional+=("$1")
+                shift
+                ;;
+        esac
+    done
+
+    if [[ -n "$SEED_FILE" ]]; then
+        if [[ "${#positional[@]}" -gt 0 ]]; then
+            log_error "Unexpected positional arguments after --seed"
+            exit 1
+        fi
+        return
+    fi
+
+    if [[ "${#positional[@]}" -eq 1 ]]; then
+        if [[ -f "${positional[0]}" ]]; then
+            MODE="seed"
+            SEED_FILE="${positional[0]}"
+            return
+        fi
+        log_error "Single positional argument must be a seed file path"
+        exit 1
+    fi
+
+    if [[ "${#positional[@]}" -eq 2 ]]; then
+        MODE="csv"
+        TABLE_REF="${positional[0]}"
+        DATA_FILE="${positional[1]}"
+        return
+    fi
+
+    usage
+    exit 1
+}
+
+main() {
+    parse_args "$@"
+    check_env
+
+    case "$MODE" in
+        csv)
+            run_csv_import
+            ;;
+        seed)
+            run_seed
+            ;;
+        *)
+            usage
+            exit 1
+            ;;
+    esac
+}
+
+main "$@"