import click
import clodius.cli.aggregate as cca
import ntpath
import os.path as op
import sys

from higlass_manage.common import fill_filetype_and_datatype
from higlass_manage.common import import_file
from higlass_manage.common import get_temp_dir
from higlass_manage.start import _start


@click.command()
@click.argument("filename")
@click.option(
    "--hg-name",
    default="default",
    help="The name of the higlass container to import this file to",
)
@click.option(
    "--filetype", default=None, help="The type of file to ingest (e.g. cooler)"
)
@click.option(
    "--datatype", default=None, help="The data type of in the input file (e.g. matrix)"
)
@click.option(
    "--assembly", default=None, help="The assembly that this data is mapped to"
)
@click.option("--name", default=None, help="The name to use for this file")
@click.option("--uid", default=None, help="The uuid to use for this file")
@click.option(
    "--no-upload",
    default=None,
    is_flag=True,
    help="Do not copy the file to the media directory. File must already be in the media directory.",
)
@click.option(
    "--chromsizes-filename",
    default=None,
    help="A set of chromosome sizes to use for bed and bedpe files",
)
@click.option(
    "--has-header",
    default=False,
    is_flag=True,
    help="Does the input file have column header information (only relevant for bed or bedpe files)",
)
@click.option(
    "--project-name",
    default=None,
    help="Group this tileset with others by specifying a project name",
)
def ingest(
    filename,
    hg_name,
    filetype=None,
    datatype=None,
    assembly=None,
    name=None,
    chromsizes_filename=None,
    has_header=False,
    uid=None,
    no_upload=None,
    project_name=None,
):
    """
    Ingest a dataset
    """
    _ingest(
        filename,
        hg_name,
        filetype,
        datatype,
        assembly,
        name,
        chromsizes_filename,
        has_header,
        uid,
        no_upload,
        project_name,
    )


def _ingest(
    filename,
    hg_name,
    filetype=None,
    datatype=None,
    assembly=None,
    name=None,
    chromsizes_filename=None,
    has_header=False,
    uid=None,
    no_upload=None,
    project_name=None,
):

    try:
        get_temp_dir(hg_name)
    except Exception:
        print("HiGlass not running. Starting...")
        _start(hg_name=hg_name)

    if not no_upload and (not op.exists(filename) and not op.islink(filename)):
        print("File not found:", filename, file=sys.stderr)
        return None

    # guess filetype and datatype if they're None
    (filetype, datatype) = fill_filetype_and_datatype(filename, filetype, datatype)

    temp_dir = get_temp_dir(hg_name)
    (to_import, filetype) = aggregate_file(
        filename,
        filetype,
        assembly,
        chromsizes_filename,
        has_header,
        no_upload,
        temp_dir,
    )

    return import_file(
        hg_name,
        to_import,
        filetype,
        datatype,
        assembly,
        name,
        uid,
        no_upload,
        project_name,
    )


def aggregate_file(
    filename, filetype, assembly, chromsizes_filename, has_header, no_upload, tmp_dir
):
    if filetype == "bedfile":
        if no_upload:
            raise Exception(
                "Bedfile files need to be aggregated and cannot be linked. Consider not using the --link-file option",
                file=sys.stderr,
            )

        if assembly is None and chromsizes_filename is None:
            print(
                "An assembly or set of chromosome sizes is required when importing bed files. Please specify one or the other using the --assembly or --chromsizes-filename parameters",
                file=sys.stderr,
            )
            return

        output_file = op.join(tmp_dir, ntpath.basename(filename) + ".beddb")

        print("Aggregating bedfile")
        cca._bedfile(
            filename,
            output_file,
            assembly,
            importance_column="random",
            has_header=has_header,
            chromosome=None,
            max_per_tile=50,
            delimiter=None,
            chromsizes_filename=chromsizes_filename,
            offset=0,
            tile_size=1024,
        )

        to_import = output_file

        # because we aggregated the file, the new filetype is beddb
        filetype = "beddb"
        return (to_import, filetype)
    elif filetype == "bedpe":
        if no_upload:
            raise Exception(
                "Bedpe files need to be aggregated and cannot be linked. Consider not using the --link-file option",
                file=sys.stderr,
            )
        if assembly is None and chromsizes_filename is None:
            print(
                "An assembly or set of chromosome sizes is required when importing bed files. Please specify one or the other using the --assembly or --chromsizes-filename parameters",
                file=sys.stderr,
            )
            return

        output_file = op.join(tmp_dir, filename + ".bed2ddb")

        print("Aggregating bedpe (output_file: {}".format(output_file))
        cca._bedpe(
            filename,
            output_file,
            assembly,
            importance_column="random",
            has_header=has_header,
            chromosome=None,
            max_per_tile=50,
            chromsizes_filename=chromsizes_filename,
            tile_size=1024,
        )

        to_import = output_file

        # because we aggregated the file, the new filetype is beddb
        filetype = "bed2ddb"
        return (to_import, filetype)
    else:
        return (filename, filetype)