Python Bunch.cleanup方法代码示例

# 需要导入模块: from galaxy.util.bunch import Bunch [as 别名]
# 或者: from galaxy.util.bunch.Bunch import cleanup [as 别名]

#.........这里部分代码省略.........
    # Handle --strand
    set_options += '--strand=%s ' % options.strand
    # Handle --ambiguous
    if options.ambiguous not in [ "no" ]:
        set_options += '--ambiguous=%s ' % options.ambiguous
    # Handle --shortcuts_for_yasra
    if options.shortcuts_for_yasra not in [ 'none' ]:
        set_options += '--%s ' % ( options.shortcuts_for_yasra )
    # Specify input2 and add [fullnames] modifier if output format is diffs
    if options.format == 'diffs':
        input2 = '%s[fullnames]' % options.input2
    else:
        input2 = options.input2
    if options.format == 'tabular':
        # Change output format to general if it's tabular and add field names for tabular output
        format = 'general-'
        tabular_fields = ':score,name1,strand1,size1,start1,zstart1,end1,length1,text1,name2,strand2,size2,start2,zstart2,end2,start2+,zstart2+,end2+,length2,text2,diff,cigar,identity,coverage,gaprate,diagonal,shingle'
    elif options.format == 'sam':
        # We currently need to keep headers.
        format = 'sam'
        tabular_fields = ''
    else:
        format = options.format
        tabular_fields = ''
    # Set up our queues
    threads = int( options.threads )
    lastz_job_queue = LastzJobQueue( threads, slots=SLOTS )
    combine_data_queue = CombineDataQueue( options.output )
    if str( options.ref_source ) in [ 'history', 'self' ]:
        # Reference is a fasta dataset from the history or the dataset containing the target sequence itself,
        # so split job across the number of sequences in the dataset ( this could be a HUGE number ).
        try:
            # Ensure there is at least 1 sequence in the dataset ( this may not be necessary ).
            error_msg = "The reference dataset is missing metadata.  Click the pencil icon in the history item and 'auto-detect' the metadata attributes."
            ref_sequences = int( options.ref_sequences )
            if ref_sequences < 1:
                stop_queues( lastz_job_queue, combine_data_queue )
                stop_err( error_msg )
        except:
            stop_queues( lastz_job_queue, combine_data_queue )
            stop_err( error_msg )
        seqs = 0
        fasta_reader = FastaReader( open( options.input1 ) )
        while True:
            # Read the next sequence from the reference dataset
            seq = fasta_reader.next()
            if not seq:
                break
            seqs += 1
            # Create a temporary file to contain the current sequence as input to lastz
            tmp_in_fd, tmp_in_name = tempfile.mkstemp( suffix='.in' )
            tmp_in = os.fdopen( tmp_in_fd, 'wb' )
            # Write the current sequence to the temporary input file
            tmp_in.write( '>%s\n%s\n' % ( seq.name, seq.text ) )
            tmp_in.close()
            # Create a 2nd temporary file to contain the output from lastz execution on the current sequence
            tmp_out_fd, tmp_out_name = tempfile.mkstemp( suffix='.out' )
            os.close( tmp_out_fd )
            # Generate the command line for calling lastz on the current sequence
            command = 'lastz %s%s %s %s --format=%s%s > %s' % ( tmp_in_name, ref_name, input2, set_options, format, tabular_fields, tmp_out_name )
            # Create a job object
            job = Bunch()
            job.command = command
            job.output = tmp_out_name
            job.cleanup = [ tmp_in_name, tmp_out_name ]
            job.combine_data_queue = combine_data_queue
            # Add another job to the lastz_job_queue.  Execution will wait at this point if the queue is full.
            lastz_job_queue.put( job, block=True )
        # Make sure the value of sequences in the metadata is the same as the number of
        # sequences read from the dataset.  According to Bob, this may not be necessary.
        if ref_sequences != seqs:
            stop_queues( lastz_job_queue, combine_data_queue )
            stop_err( "The value of metadata.sequences (%d) differs from the number of sequences read from the reference (%d)." % ( ref_sequences, seqs ) )
    else:
        # Reference is a locally cached 2bit file, split job across number of chroms in 2bit file
        tbf = TwoBitFile( open( options.input1, 'r' ) )
        for chrom in tbf.keys():
            # Create a temporary file to contain the output from lastz execution on the current chrom
            tmp_out_fd, tmp_out_name = tempfile.mkstemp( suffix='.out' )
            os.close( tmp_out_fd )
            command = 'lastz %s/%s%s %s %s --format=%s%s >> %s' % \
                ( options.input1, chrom, ref_name, input2, set_options, format, tabular_fields, tmp_out_name )
            # Create a job object
            job = Bunch()
            job.command = command
            job.output = tmp_out_name
            job.cleanup = [ tmp_out_name ]
            job.combine_data_queue = combine_data_queue
            # Add another job to the lastz_job_queue.  Execution will wait at this point if the queue is full.
            lastz_job_queue.put( job, block=True )
    # Stop the lastz_job_queue.
    for t in lastz_job_queue.threads:
        lastz_job_queue.put( STOP_SIGNAL, True )
    # Although all jobs are submitted to the queue, we can't shut down the combine_data_queue
    # until we know that all jobs have been submitted to its queue.  We do this by checking
    # whether all of the threads in the lastz_job_queue have terminated.
    while threading.activeCount() > 2:
        time.sleep( 1 )
    # Now it's safe to stop the combine_data_queue.
    combine_data_queue.put( STOP_SIGNAL )

# 需要导入模块: from galaxy.util.bunch import Bunch [as 别名]
# 或者: from galaxy.util.bunch.Bunch import cleanup [as 别名]

#.........这里部分代码省略.........
                      options.O, options.E, options.X, options.Y, options.K, options.L, options.entropy )
    # Specify input2 and add [fullnames] modifier if output format is diffs
    if options.format == 'diffs':
        input2 = '%s[fullnames]' % options.input2
    else:
        input2 = options.input2
    if options.format == 'tabular':
        # Change output format to general if it's tabular and add field names for tabular output
        format = 'general-'
        tabular_fields = ':score,name1,strand1,size1,start1,zstart1,end1,length1,text1,name2,strand2,size2,start2,zstart2,end2,start2+,zstart2+,end2+,length2,text2,diff,cigar,identity,coverage,gaprate,diagonal,shingle'
    elif options.format == 'sam':
        # We currently ALWAYS suppress SAM headers.
        format = 'sam-'
        tabular_fields = ''
    else:
        format = options.format
        tabular_fields = ''

    # Set up our queues
    lastz_job_queue = LastzJobQueue( WORKERS, slots=SLOTS )
    combine_data_queue = CombineDataQueue( options.output )

    if options.ref_source == 'history':
        # Reference is a fasta dataset from the history, so split job across
        # the number of sequences in the dataset ( this could be a HUGE number )
        try:
            # Ensure there is at least 1 sequence in the dataset ( this may not be necessary ).
            error_msg = "The reference dataset is missing metadata, click the pencil icon in the history item and 'auto-detect' the metadata attributes."
            ref_sequences = int( options.ref_sequences )
            if ref_sequences < 1:
                stop_queues( lastz_job_queue, combine_data_queue )
                stop_err( error_msg )
        except:
            stop_queues( lastz_job_queue, combine_data_queue )
            stop_err( error_msg )
        seqs = 0
        fasta_reader = FastaReader( open( options.input1 ) )
        while True:
            # Read the next sequence from the reference dataset
            seq = fasta_reader.next()
            if not seq:
                break
            seqs += 1
            # Create a temporary file to contain the current sequence as input to lastz
            tmp_in_fd, tmp_in_name = tempfile.mkstemp( suffix='.in' )
            tmp_in = os.fdopen( tmp_in_fd, 'wb' )
            # Write the current sequence to the temporary input file
            tmp_in.write( '>%s\n%s\n' % ( seq.name, seq.text ) )
            tmp_in.close()
            # Create a 2nd temporary file to contain the output from lastz execution on the current sequence
            tmp_out_fd, tmp_out_name = tempfile.mkstemp( suffix='.out' )
            os.close( tmp_out_fd )
            # Generate the command line for calling lastz on the current sequence
            command = 'lastz %s%s%s %s %s --ambiguousn --nolaj --identity=%s..%s --coverage=%s --format=%s%s > %s' % \
                ( tmp_in_name, unmask, ref_name, input2, set_options, options.identity_min, 
                  options.identity_max, options.coverage, format, tabular_fields, tmp_out_name )
            # Create a job object
            job = Bunch()
            job.command = command
            job.output = tmp_out_name
            job.cleanup = [ tmp_in_name, tmp_out_name ]
            job.combine_data_queue = combine_data_queue
            # Add another job to the lastz_job_queue. Execution 
            # will wait at this point if the queue is full.
            lastz_job_queue.put( job, block=True )
        # Make sure the value of sequences in the metadata is the same as the
        # number of sequences read from the dataset ( this may not be necessary ).
        if ref_sequences != seqs:
            stop_queues( lastz_job_queue, combine_data_queue )
            stop_err( "The value of metadata.sequences (%d) differs from the number of sequences read from the reference (%d)." % ( ref_sequences, seqs ) )
    else:
        # Reference is a locally cached 2bit file, split job across number of chroms in 2bit file
        tbf = TwoBitFile( open( options.input1, 'r' ) )
        for chrom in tbf.keys():
            # Create a temporary file to contain the output from lastz execution on the current chrom
            tmp_out_fd, tmp_out_name = tempfile.mkstemp( suffix='.out' )
            os.close( tmp_out_fd )
            command = 'lastz %s/%s%s%s %s %s --ambiguousn --nolaj --identity=%s..%s --coverage=%s --format=%s%s >> %s' % \
                ( options.input1, chrom, unmask, ref_name, input2, set_options, options.identity_min, 
                  options.identity_max, options.coverage, format, tabular_fields, tmp_out_name )
            # Create a job object
            job = Bunch()
            job.command = command
            job.output = tmp_out_name
            job.cleanup = [ tmp_out_name ]
            job.combine_data_queue = combine_data_queue
            # Add another job to the lastz_job_queue. Execution 
            # will wait at this point if the queue is full.
            lastz_job_queue.put( job, block=True )

    # Stop the lastz_job_queue
    for t in lastz_job_queue.threads:
        lastz_job_queue.put( STOP_SIGNAL, True )
    # Although all jobs are submitted to the queue, we can't shut down the combine_data_queue
    # until we know that all jobs have been submitted to its queue.  We do this by checking
    # whether all of the threads in the lastz_job_queue have terminated.
    while threading.activeCount() > 2:
        time.sleep( 1 )
    # Now it's safe to stop the combine_data_queue
    combine_data_queue.put( STOP_SIGNAL )