Group cycle report preprocessing artifacts

This commit is contained in:
Cesa-V
2026-06-15 18:12:13 -03:00
parent 33c452b264
commit d2bf2883d7
8 changed files with 15 additions and 14 deletions

View File

@@ -1 +0,0 @@

View File

@@ -6,12 +6,13 @@ from pathlib import Path
import pdfplumber
ROOT = Path(__file__).resolve().parents[1]
ROOT = Path(__file__).resolve().parents[2]
BASE_NAME = "relatorio_ciclo_inspecoes_c105_2805_2026-06-15"
OUTPUT_DIR = Path(__file__).resolve().parent
PDF_PATH = ROOT / "raw" / f"{BASE_NAME}.pdf"
TXT_PATH = ROOT / "pre_process" / f"{BASE_NAME}_texto.txt"
JSON_PATH = ROOT / "pre_process" / f"{BASE_NAME}_inspecoes.json"
CSV_PATH = ROOT / "pre_process" / f"{BASE_NAME}_inspecoes.csv"
TXT_PATH = OUTPUT_DIR / f"{BASE_NAME}_texto.txt"
JSON_PATH = OUTPUT_DIR / f"{BASE_NAME}_inspecoes.json"
CSV_PATH = OUTPUT_DIR / f"{BASE_NAME}_inspecoes.csv"
ROW_START_RE = re.compile(r"^(?P<seq>\d+)\s+")
INTERVAL_RE = re.compile(