view ucsc_xena_download.py @ 39:066ae9df2d27

labeling descriptions edit

#!/usr/bin/env python
import socket, urllib, sys, os
from galaxy import eggs #eggs needs to be imported so that galaxy.util can find docutils egg...
from galaxy.util.json import from_json_string, to_json_string
from galaxy.util import get_charset_from_http_headers
import galaxy.model # need to import model before sniff to resolve a circular import dependency
from galaxy.datatypes import sniff
import tarfile
import re

filemap = [
        ('genomic', r'genomic(Segment|Matrix)$'),
        ('clinical', r'clinicalMatrix$'),
        ]

files = {
        'genomic': sys.argv[1],
        'clinical': sys.argv[2]
        }

max_file_size = sys.argv[3]

def file_type(file):
    with open(file) as f:
        return from_json_string(f.read())['type']

def stop_err( msg ):
    sys.stderr.write( msg )
    sys.exit()

def load_input_parameters( filename, erase_file = True ):
    datasource_params = {}
    try:
        json_params = from_json_string( open( filename, 'r' ).read() )
        datasource_params = json_params.get( 'param_dict' )
    except:
        json_params = None
        for line in open( filename, 'r' ):
            try:
                line = line.strip()
                fields = line.split( '\t' )
                datasource_params[ fields[0] ] = fields[1]
            except:
                continue
    if erase_file:
        open( filename, 'w' ).close() #open file for writing, then close, removes params from file
    return json_params, datasource_params

def load_file(files):
    filename = files['genomic']
    job_params, params = load_input_parameters( filename, False )
    URL = params.get( 'URL', None ) #using exactly URL indicates that only one dataset is being downloaded
    URL_method = params.get( 'URL_method', None )
    socket.setdefaulttimeout( 600 )
    try:
        if not URL_method or URL_method == 'get':
            page = urllib.urlopen( URL )
        elif URL_method == 'post':
            page = urllib.urlopen( URL, urllib.urlencode( params ) )
    except Exception, e:
        stop_err( 'The remote data source application may be off line, please try again later. Error: %s' % str( e ) )
    if max_file_size:
        file_size = int( page.info().get( 'Content-Length', 0 ) )
        if file_size > max_file_size:
            stop_err( 'The size of the data (%d bytes) you have requested exceeds the maximum allowed (%d bytes) on this server.' % ( file_size, max_file_size ) )
    try:
        cur_filename, is_multi_byte = sniff.stream_to_open_named_file( page, os.open( filename, os.O_WRONLY | os.O_CREAT ), filename, source_encoding=get_charset_from_http_headers( page.headers ) )
    except Exception, e:
        stop_err( 'Unable to fetch %s:\n%s' % ( URL, e ) )

load_file(files)

tar = tarfile.open(files['genomic'])
names = tar.getnames()
metafiles = [n for n in names if n.endswith('.json')]
tar.extractall()
withtype = [(file_type(file), file[0:-len(".json")]) for file in metafiles]
try:
    renames = [((n for (t, n) in withtype if re.search(pat, t)).next(), name) for (name, pat) in filemap]
except StopIteration:
     stop_err( 'Missing required file type in tarball' )
for (frm, to) in renames:
    os.rename(frm, files[to])