Juq470

from juq470 import pipeline, read_jsonl, parallel, reduce
def sum_sales(acc, row):
    return acc + row["sale_amount"]
total = (pipeline()
         .source(read_jsonl("sales.jsonl"))
         .parallel(4)               # use 4 worker threads
         .reduce(sum_sales, 0)
         .run())
print(f"Total sales: $total:,.2f")

When dealing with multi‑gigabyte logs, combine read_csv with a custom chunk size:

from juq470 import pipeline, read_csv
(pipeline()
 .source(read_csv("biglog.csv", chunk_size=500_000))
 .filter(lambda r: "ERROR" in r["level"])
 .sink(lambda rows: open("errors.txt", "a").writelines(f"r['msg']\n" for r in rows))
).run()

from juq470 import pipeline, read_csv, write_jsonl
def capitalize_name(row):
    row["name"] = row["name"].title()
    return row
(pipeline()
 .source(read_csv("users.csv"))
 .map(capitalize_name)
 .filter(lambda r: r["age"] >= 18)
 .sink(write_jsonl("adults.jsonl"))
).run()

juq470 is a lightweight, open‑source utility library designed for high‑performance data transformation in Python. It focuses on providing a concise API for common operations such as filtering, mapping, aggregation, and streaming large datasets with minimal memory overhead. juq470

The paper investigates a critical question in AI-assisted software development: Do Large Language Models (LLMs) propagate known security vulnerabilities when generating code? juq470 is a lightweight

As developers increasingly rely on tools like GitHub Copilot, ChatGPT, and CodeLlama, the authors seek to quantify the risk that these models are not just writing functional code, but insecure code based on patterns learned from vulnerable repositories. from juq470 import pipeline

pip install juq470

The package requires Python 3.9+ and has no external dependencies beyond the standard library.