import datetime
import itertools
import logging
import os
import sys
from collections import defaultdict
from pathlib import Path
from typing import DefaultDict, Sequence

from codemodder import __version__, providers, registry
from codemodder.cli import parse_args
from codemodder.codemods.api import BaseCodemod
from codemodder.codemods.semgrep import SemgrepRuleDetector
from codemodder.codetf import CodeTF
from codemodder.context import CodemodExecutionContext
from codemodder.dependency import Dependency
from codemodder.llm import TokenUsage, log_token_usage
from codemodder.logging import configure_logger, log_list, log_section, logger
from codemodder.project_analysis.file_parsers.package_store import PackageStore
from codemodder.project_analysis.python_repo_manager import PythonRepoManager
from codemodder.result import ResultSet
from codemodder.sarifs import detect_sarif_tools
from codemodder.semgrep import run as run_semgrep


def find_semgrep_results(
    context: CodemodExecutionContext,
    codemods: Sequence[BaseCodemod],
    files_to_analyze: list[Path] | None = None,
) -> ResultSet:
    """Run semgrep once with all configuration files from all codemods and return a set of applicable rule IDs"""
    if not (
        yaml_files := list(
            itertools.chain.from_iterable(
                [
                    codemod.detector.get_yaml_files(codemod._internal_name)
                    for codemod in codemods
                    if codemod.detector
                    and isinstance(codemod.detector, SemgrepRuleDetector)
                ]
            )
        )
    ):
        return ResultSet()

    return run_semgrep(context, yaml_files, files_to_analyze)


def log_report(context, output, elapsed_ms, files_to_analyze, token_usage):
    log_section("report")
    logger.info("scanned: %s files", len(files_to_analyze))
    all_failures = context.get_failed_files()
    logger.info(
        "failed: %s files (%s unique)",
        len(all_failures),
        len(set(all_failures)),
    )
    all_changes = context.get_changed_files()
    logger.info(
        "changed: %s files (%s unique)",
        len(all_changes),
        len(set(all_changes)),
    )
    logger.info("report file: %s", output)
    log_token_usage("All", token_usage)
    logger.info("total elapsed: %s ms", elapsed_ms)
    logger.info("  semgrep:     %s ms", context.timer.get_time_ms("semgrep"))
    logger.info("  parse:       %s ms", context.timer.get_time_ms("parse"))
    logger.info("  transform:   %s ms", context.timer.get_time_ms("transform"))
    logger.info("  write:       %s ms", context.timer.get_time_ms("write"))


def apply_codemods(
    context: CodemodExecutionContext,
    codemods_to_run: Sequence[BaseCodemod],
    remediation: bool,
) -> TokenUsage:
    log_section("scanning")
    token_usage = TokenUsage()

    if not context.files_to_analyze:
        logger.info("no files to scan")