wrapper:
    "v0.85.1/bio/fastqc"

shell:
    """
    multiqc --force -o data/processed/qc -n multiqc.html {params.fastqc_dir} {params.trinity_dir}> {log} 2>&1
    """

shell:
    """
    Trinity --seqType fq --SS_lib_type {params.SS_lib_type}  --max_memory {resources.mem_gb} --CPU {threads} --samples_file {input} --output data/processed/trinity/trinity_{wildcards.name}-{wildcards.condition} \
    --min_kmer_cov {params.min_kmer_cov} --bflyCPU {threads} --bflyHeapSpaceMax {resources.mem_gb} --normalize_max_read_cov {params.normalize_max_read_cov} --trimmomatic > {log} 2>> {log}
    """

import pandas as pd
from pathlib import Path
import sys

def get_trinity_samples(run_name, condition, outfile, samples_path="config/samples.tsv", units_path="config/units.tsv"):
    """
    """
    samples = (
        pd.read_csv(samples_path, sep="\t", dtype={"ID": str})
        .set_index("ID", drop=False)
        .sort_index()
        )

    units = (
        pd.read_csv(units_path, sep="\t", dtype={"ID": str})
        .set_index("ID", drop=False)
        .sort_index()
        )

    trinity = []

    subset = samples[samples.loc[:, "Condition"] == condition]
    for s in subset.index:
        s_dict = {}
        s_dict["Condition"] = samples.loc[s, "Condition"]
        s_dict["Replicate"] = units.loc[s, "unit_name"]
        s_dict["F"] = units.loc[s, "fq1"]
        s_dict["R"] = units.loc[s, "fq2"]
        trinity.append(s_dict)

    df = pd.DataFrame(trinity)
    df.to_csv(f"{outfile}", sep="\t", index=False, header=False)
    return None

if __name__ == "__main__":
    get_trinity_samples(sys.argv[1], sys.argv[2], sys.argv[3])

__author__ = "Tessa Pierce"
__copyright__ = "Copyright 2018, Tessa Pierce"
__email__ = "[email protected]"
__license__ = "MIT"

from snakemake.shell import shell
from os import path
import tempfile

log = snakemake.log_fmt_shell(stdout=True, stderr=True)
extra = snakemake.params.get("extra", "")
mode = snakemake.params.get("mode")
assert mode is not None, "please input a run mode: genome, transcriptome or proteins"
lineage = snakemake.params.get("lineage")
assert lineage is not None, "please input the path to a lineage for busco assessment"

stripped_output = snakemake.output[0].rstrip("/")
out = path.basename(stripped_output)
out_dirname = path.dirname(stripped_output)
out_path = " --out_path {} ".format(out_dirname) if out_dirname else ""

download_path_dir = snakemake.params.get("download_path", "")
download_path = (
    " --download_path {} ".format(download_path_dir) if download_path_dir else ""
)

# note: --force allows snakemake to handle rewriting files as necessary
# without needing to specify *all* busco outputs as snakemake outputs
shell(
    "busco --in {snakemake.input} --out {out} --force "
    "{out_path} "
    "--cpu {snakemake.threads} --mode {mode} --lineage {lineage} "
    "{download_path} "
    "{extra} {log}"
)

__author__ = "Johannes Köster, Jorge Langa"
__copyright__ = "Copyright 2016, Johannes Köster"
__email__ = "[email protected]"
__license__ = "MIT"


from snakemake.shell import shell
from snakemake_wrapper_utils.java import get_java_opts

# Distribute available threads between trimmomatic itself and any potential pigz instances
def distribute_threads(input_files, output_files, available_threads):
    gzipped_input_files = sum(1 for file in input_files if file.endswith(".gz"))
    gzipped_output_files = sum(1 for file in output_files if file.endswith(".gz"))
    potential_threads_per_process = available_threads // (
        1 + gzipped_input_files + gzipped_output_files
    )
    if potential_threads_per_process > 0:
        # decompressing pigz creates at most 4 threads
        pigz_input_threads = (
            min(4, potential_threads_per_process) if gzipped_input_files != 0 else 0
        )
        pigz_output_threads = (
            (available_threads - pigz_input_threads * gzipped_input_files)
            // (1 + gzipped_output_files)
            if gzipped_output_files != 0
            else 0
        )
        trimmomatic_threads = (
            available_threads
            - pigz_input_threads * gzipped_input_files
            - pigz_output_threads * gzipped_output_files
        )
    else:
        # not enough threads for pigz
        pigz_input_threads = 0
        pigz_output_threads = 0
        trimmomatic_threads = available_threads
    return trimmomatic_threads, pigz_input_threads, pigz_output_threads


def compose_input_gz(filename, threads):
    if filename.endswith(".gz") and threads > 0:
        return "<(pigz -p {threads} --decompress --stdout {filename})".format(
            threads=threads, filename=filename
        )
    return filename


def compose_output_gz(filename, threads, compression_level):
    if filename.endswith(".gz") and threads > 0:
        return ">(pigz -p {threads} {compression_level} > {filename})".format(
            threads=threads, compression_level=compression_level, filename=filename
        )
    return filename


extra = snakemake.params.get("extra", "")
java_opts = get_java_opts(snakemake)
log = snakemake.log_fmt_shell(stdout=True, stderr=True)
compression_level = snakemake.params.get("compression_level", "-5")
trimmer = " ".join(snakemake.params.trimmer)

# Distribute threads
input_files = [snakemake.input.r1, snakemake.input.r2]
output_files = [
    snakemake.output.r1,
    snakemake.output.r1_unpaired,
    snakemake.output.r2,
    snakemake.output.r2_unpaired,
]

trimmomatic_threads, input_threads, output_threads = distribute_threads(
    input_files, output_files, snakemake.threads
)

input_r1, input_r2 = [
    compose_input_gz(filename, input_threads) for filename in input_files
]

output_r1, output_r1_unp, output_r2, output_r2_unp = [
    compose_output_gz(filename, output_threads, compression_level)
    for filename in output_files
]

shell(
    "trimmomatic PE -threads {trimmomatic_threads} {java_opts} {extra} "
    "{input_r1} {input_r2} "
    "{output_r1} {output_r1_unp} "
    "{output_r2} {output_r2_unp} "
    "{trimmer} "
    "{log}"
)