def install_tool(self, executable):
"""Install the given tool, using the script:
bioconvert/install_script/install_executable.sh
if the executable is not already present
:param executable to install
:return: nothing
"""
import bioconvert
from bioconvert import bioconvert_data
if shutil.which(executable) is None:
logger.info("Installing tool : " + executable)
bioconvert_path = bioconvert.__path__[0]
script = bioconvert_data('install_' + executable + '.sh',
where="../misc")
subprocess.call(['sh', script])
def install_tool(self, executable):
"""Install the given tool, using the script:
bioconvert/install_script/install_executable.sh
if the executable is not already present
:param executable: executable to install
:return: nothing
"""
# imported but not unused (when we don't have bioconvert_path)
# import bioconvert
from bioconvert import bioconvert_data
if shutil.which(executable) is None:
logger.info("Installing tool : " + executable)
# Assigned but never used, says flake8
# bioconvert_path = bioconvert.__path__[0]
script = bioconvert_data(
'install_' + executable + '.sh', where="../misc")
subprocess.call(['sh', script])
class shell:
_process_args = {}
_process_prefix = ""
_process_suffix = ""
@classmethod
def executable(cls, cmd):
if os.path.split(cmd)[-1] == "bash":
cls._process_prefix = "set -euo pipefail; "
cls._process_args["executable"] = cmd
@classmethod
def prefix(cls, prefix):
cls._process_prefix = prefix
@classmethod
def suffix(cls, suffix):
cls._process_suffix = suffix
def __new__(cls,
cmd,
*args,
async=False,
iterable=False,
read=False,
**kwargs):
if "stepout" in kwargs:
raise KeyError("Argument stepout is not allowed in shell command.")
stdout = sp.PIPE if iterable or async or read else STDOUT
close_fds = sys.platform != 'win32'
logger.info(cmd)
proc = sp.Popen("{} {} {}".format(cls._process_prefix, cmd.rstrip(),
cls._process_suffix),
bufsize=-1,
shell=True,
stdout=stdout,
close_fds=close_fds,
**cls._process_args)
ret = None
if iterable:
return cls.iter_stdout(proc, cmd)
if read:
ret = proc.stdout.read()
elif async:
return proc
retcode = proc.wait()
if retcode:
raise sp.CalledProcessError(retcode, cmd)
return ret
def wrapped(inst, *args, **kwargs):
output_compressed = None
if inst.outfile.endswith(".gz"):
(inst.outfile, output_compressed) = splitext(inst.outfile)
elif inst.outfile.endswith(".bz2"):
(inst.outfile, output_compressed) = splitext(inst.outfile)
elif inst.outfile.endswith(".dsrc"): # !!! only for fastq files
(inst.outfile, output_compressed) = splitext(inst.outfile)
# Now inst has the uncompressed output file name
# computation
results = func(inst, *args, **kwargs)
# Compress output and restore inst output file name
if output_compressed == ".gz":
# TODO: this uses -f ; should be a
logger.info("Compressing output into .gz")
inst.shell("pigz -f -p {} {}".format(inst.threads, inst.outfile))
inst.outfile = inst.outfile + ".gz"
elif output_compressed == ".bz2":
logger.info("Compressing output into .bz2")
inst.shell("pbzip2 -f -p{} {}".format(inst.threads, inst.outfile))
inst.outfile = inst.outfile + ".bz2"
elif output_compressed == ".dsrc": # !!! only for FastQ files
logger.info("Compressing output into .dsrc")
inst.shell("dsrc c -t{} {} {}.dsrc".format(inst.threads,
inst.outfile,
inst.outfile))
inst.outfile = inst.outfile + ".dsrc"
return results
def wrapped(inst, *args, **kwargs):
infile_name = inst.infile
output_compressed = None
if inst.outfile.endswith(".gz"):
(inst.outfile, output_compressed) = splitext(inst.outfile)
elif inst.outfile.endswith(".bz2"):
(inst.outfile, output_compressed) = splitext(inst.outfile)
elif inst.outfile.endswith(".dsrc"): # !!! only for fastq files
(inst.outfile, output_compressed) = splitext(inst.outfile)
# Now inst has the uncompressed output file name
if infile_name.endswith(".gz"):
# decompress input
# TODO: https://stackoverflow.com/a/29371584/1878788
logger.info("Generating uncompressed version of %s " % infile_name)
(ungz_name, _) = splitext(infile_name)
(_, base_suffix) = splitext(ungz_name)
with TempFile(suffix=base_suffix) as ungz_infile:
inst.infile = ungz_infile.name
inst.shell("unpigz -c -p {} {} > {}".format(
inst.threads, infile_name, inst.infile))
# computation
results = func(inst, *args, **kwargs)
inst.infile = infile_name
else:
results = func(inst, *args, **kwargs)
# Compress output and restore inst output file name
if output_compressed == ".gz":
# TODO: this uses -f ; should be a
logger.info("Compressing output into .gz")
inst.shell("pigz -f -p {} {}".format(inst.threads, inst.outfile))
inst.outfile = inst.outfile + ".gz"
elif output_compressed == ".bz2":
logger.info("Compressing output into .bz2")
inst.shell("pbzip2 -f -p{} {}".format(inst.threads, inst.outfile))
inst.outfile = inst.outfile + ".bz2"
elif output_compressed == ".dsrc": # !!! only for FastQ files
logger.info("Compressing output into .dsrc")
inst.shell("dsrc c -t{} {} {}.dsrc".format(inst.threads,
inst.outfile,
inst.outfile))
inst.outfile = inst.outfile + ".dsrc"
return results
def _method_biopython(self, *args, **kwargs):
logger.info("Executing biopython")
records = SeqIO.parse(self.infile, 'fastq')
SeqIO.write(records, self.outfile, 'fasta')
def __init__(self, infile, outfile, force=False):
""".. rubric:: constructor
:param str infile: The path of the input file.
:param str outfile: The path of The output file
:param bool force: overwrite output file if it exists already
otherwise raises an error
"""
if os.path.exists(infile) is False:
msg = "Incorrect input file: %s" % infile
_log.error(msg)
raise ValueError(msg)
# check existence of output file. If it exists,
# fails except if force argument is set to True
if os.path.exists(outfile) is True:
msg = "output file {} exists already".format(outfile)
_log.warning("output file exists already")
if force is False:
_log.critical(
"output file exists. If you are using bioconvert, use --force "
)
raise ValueError(msg)
else:
_log.warning("output file will be overwritten")
# Only fastq files can be compressed with dsrc
if outfile.endswith(".dsrc"):
# only valid for FastQ files extension
# dsrc accepts only .fastq file extension
if outfile.endswith(".fastq.dsrc") is False:
msg = "When compressing with .dsrc extension, " +\
"only files ending with .fastq extension are " +\
"accepted. This is due to the way dsrc executable +"\
"is implemented."
_log.critical(msg)
raise IOError
# case1: fastq.gz to fasta.bz2
# Here, we want to decompress, convert, compress.
# so we need the extension without .gz or .bz2
# We should have inext set to fastq and outext
# set to fasta.bz2
self.inext = getext(infile, remove_compression=True)
self.outext = getext(outfile, remove_compression=True)
# Case 2, fastq.gz to fastq.bz2
# data is not changed, just the type of compression, so we want
# to keep the original extensions, here inext and outext will contain
# .gz and .bz2
if self.inext == self.outext:
_log.info("decompression/compression mode")
self.inext = getext(infile)
self.outext = getext(outfile)
self.mapper = Registry()
# From the input parameters 1 and 2, we get the module name
try:
_log.info("Input: {}".format(self.inext))
_log.info("Output: {}".format(self.outext))
class_converter = self.mapper[(self.inext, self.outext)]
self.name = class_converter.__name__
except KeyError:
# This module name was not found
msg = "Requested input format ({}) to output format ({}) is not available in bioconvert"
_log.critical(msg.format(self.inext, self.outext))
_log.critical(
"Use --formats to know the available formats and --help for examples"
)
sys.exit(1)
self.converter = class_converter(infile, outfile)
_log.info("Using {} class".format(self.converter.name))
