import os
from pathlib import Path

RANDOM_STATE = 42

GRAZIE_API_JWT_TOKEN = "eyJhbGciOiJSUzUxMiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiJHcmF6aWUgQXV0aGVudGljYXRpb24iLCJ1aWQiOiJkNmFjZGM3Zi1jZWZlLTRhMDItOWRmMi01NzY5OGRlNjYyNDAiLCJ1c2VyX3N0YXRlIjoiSU5URVJOQUwiLCJyZWdpc3RyYXRpb25fZGF0ZSI6MTY4NDMzNjI3ODI2NCwibGljZW5zZSI6IjQ1TVcwNFZBVVoiLCJsaWNlbnNlX3R5cGUiOiJqZXRicmFpbnMtYWkub3JnYW5pemF0aW9uYWwucHJvIiwiZXhwIjoxNzIwNjk0OTQ2fQ.NH5KLYgkyaC1MfFHPj8jfe3yBBR8F017QV_Nn0_5AqiWqjaaVBIBCsxkZcTbwH6FBrGm-JXYM50UAhJprI3fy-HNkwfF6nAPRqkFafxT8IZ-Epk8P9u6SnC5YjD4LM4e_-aKeuXb4WdB6K_YDIRKIp64WthCS2OzLSDPiyXaHXADOBQMfWNvorXqjuKPUPE7q6L59Wes4VaDhXMPw2XA4MHUm_cTvK2a_SixaKiawxAv-Wa8vo2KcYbd4hqtxDwnoQ6c5WfmEqD-dUYvZ8G_53WNJO6gvIv0etEBx8NIez2dPXHyNqIyam4CrMXH9_stJwf998sL7NxdG2wRLGGC4A"
GRAZIE_TIMEOUT_SEC = 1.0

HF_TOKEN = os.environ.get("HF_TOKEN")

HF_RAW_DATASET_NAME = "JetBrains-Research/commit-msg-rewriting"
HF_RAW_DATASET_SPLIT = "train"

HF_FULL_COMMITS_DATASET_NAME = "JetBrains-Research/lca-commit-message-generation"
HF_FULL_COMMITS_DATASET_SUBNAME = "commitchronicle-py-long"
HF_FULL_COMMITS_DATASET_SPLIT = "test"

HF_PREDICTIONS_DATASET_NAME = "JetBrains-Research/lca-results"
HF_PREDICTIONS_MODEL = "gpt_4_0613"

HF_SYNTHETIC_DATASET_NAME = "JetBrains-Research/synthetic-commit-msg-rewriting"
HF_SYNTHETIC_DATASET_SPLIT = "train"

LLM_MODEL = "gpt-4-1106-preview"

CACHE_DIR = Path("cache")
CACHE_DIR.mkdir(exist_ok=True)

OUTPUT_DIR = Path("output")
OUTPUT_DIR.mkdir(exist_ok=True)

END_TO_START_ARTIFACT = OUTPUT_DIR / "end_to_start.csv"
START_TO_END_ARTIFACT = OUTPUT_DIR / "start_to_end.csv"
SYNTHETIC_DATASET_ARTIFACT = OUTPUT_DIR / "synthetic.csv"
METRICS_CORRELATIONS_ARTIFACT = OUTPUT_DIR / "metrics_correlations.csv"
DATA_FOR_LABELING_ARTIFACT = OUTPUT_DIR / "data_for_labeling.csv"

OUTPUT_CHARTS_DIR = OUTPUT_DIR / "charts"
OUTPUT_CHARTS_DIR.mkdir(exist_ok=True)