comparison runner.py @ 21:25fa179d9d0a draft

planemo upload for repository https://github.com/richard-burhans/galaxytools/tree/main/tools/segalign commit e4e05d23d9da18ea87bc352122ca9e6cfa73d1c7

20:96ff17622b17

            break

        run_args = ["python", f"{args.tool_directory}/diagonal_partition.py", str(chunk_size)]
        for word in line.split():
            run_args.append(word)
        process = subprocess.run(run_args, stdin=subprocess.DEVNULL, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)

        for line in process.stdout.splitlines():
            output_q.put(line)

        for line in process.stderr.splitlines():

        if process.returncode != 0:
            sys.exit(f"Error: diagonal partitioner {instance} exited with returncode {process.returncode}")


def estimate_chunk_size(args: argparse.Namespace) -> int:
    chunk_size = -1
    line_size = -1

    if args.debug:
        r_beg = resource.getrusage(resource.RUSAGE_SELF)

                except FileNotFoundError:
                    continue

            fdict[entry.name.split(".split", 1)[0]] += file_size

    # if noot enough segment files for estimation, continue
    if len(fdict) > 2:
        chunk_size = int(statistics.quantiles(fdict.values())[-1] // line_size)

    if args.debug:
        ns: int = time.monotonic_ns() - beg
        r_end = resource.getrusage(resource.RUSAGE_SELF)
        print(f"estimate chunk size clock time: {ns} ns", file=sys.stderr, flush=True)

21:25fa179d9d0a

            break

        run_args = ["python", f"{args.tool_directory}/diagonal_partition.py", str(chunk_size)]
        for word in line.split():
            run_args.append(word)
        process = subprocess.run(run_args, stdin=subprocess.DEVNULL, stdout=subprocess.PIPE, stderr=subprocess.PIPE, bufsize=1, text=True)

        for line in process.stdout.splitlines():
            output_q.put(line)

        for line in process.stderr.splitlines():

        if process.returncode != 0:
            sys.exit(f"Error: diagonal partitioner {instance} exited with returncode {process.returncode}")


def estimate_chunk_size(args: argparse.Namespace) -> int:
    # only used when segment size is being estimated
    MAX_CHUNK_SIZE = 50000
    chunk_size = -1
    line_size = -1

    if args.debug:
        r_beg = resource.getrusage(resource.RUSAGE_SELF)

                except FileNotFoundError:
                    continue

            fdict[entry.name.split(".split", 1)[0]] += file_size

    if len(fdict) < 7:
        # outliers can heavily skew prediction if <7 data points
        # to be safe, use 50% quantile
        chunk_size = int(statistics.quantiles(fdict.values())[1] // line_size)
    else:
        # otherwise use 75% quantile
        chunk_size = int(statistics.quantiles(fdict.values())[-1] // line_size)

    # if not enough data points, there is a chance of getting unlucky
    # minimize worst case by using MAX_CHUNK_SIZE

    chunk_size = min(chunk_size, MAX_CHUNK_SIZE)

    if args.debug:
        ns: int = time.monotonic_ns() - beg
        r_end = resource.getrusage(resource.RUSAGE_SELF)
        print(f"estimate chunk size clock time: {ns} ns", file=sys.stderr, flush=True)