Compare commits

...

2 Commits

Author SHA1 Message Date
4249e902b2 Feat: add *.duckdb in gitignore 2024-06-18 06:30:27 +02:00
ab36931c06 Feat: create pdf ingest pipeline 2024-06-18 06:26:51 +02:00
2 changed files with 35 additions and 0 deletions

2
.gitignore vendored
View File

@ -159,3 +159,5 @@ cython_debug/
# and can be added to the global gitignore or merged into this file. For a more nuclear
# option (not recommended) you can uncomment the following to ignore the entire idea folder.
#.idea/
*.duckdb

33
dlt/pdf_pipeline.py Normal file
View File

@ -0,0 +1,33 @@
import dlt
from pathlib import Path
from pdf_oralia.extract import from_pdf
import pdfplumber
DATA_PATH = Path("datas/")
assert DATA_PATH.exists()
RAW_CRG_PDF = DATA_PATH / "pdfs"
assert RAW_CRG_PDF.exists()
@dlt.resource(name="crg")
def crg_pdf(filename):
print(filename)
pdf = pdfplumber.open(filename)
try:
df_charge, df_loc = from_pdf(pdf)
except ValueError as e:
print(f"\tExtract Error: {e}")
pass
else:
for row in df_charge.to_dict("records"):
yield row
if __name__ == "__main__":
pipeline = dlt.pipeline(
pipeline_name='raw', destination="duckdb", dataset_name="crg"
)
for pdf_file in RAW_CRG_PDF.glob("**/*.pdf"):
load_info = pipeline.run(crg_pdf(pdf_file), table_name='charge')
print(load_info)