aboutsummaryrefslogtreecommitdiff
path: root/wqflask/dataSharing/SharingInfo.py
blob: 10abcefa373ba0c3dd45a37795ed3908f0e87988 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
# Copyright (C) University of Tennessee Health Science Center, Memphis, TN.
#
# This program is free software: you can redistribute it and/or modify it
# under the terms of the GNU Affero General Public License
# as published by the Free Software Foundation, either version 3 of the
# License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# See the GNU Affero General Public License for more details.
#
# This program is available from Source Forge: at GeneNetwork Project
# (sourceforge.net/projects/genenetwork/).
#
# Contact Drs. Robert W. Williams and Xiaodong Zhou (2010)
# at rwilliams@uthsc.edu and xzhou15@uthsc.edu
#
#
#
# This module is used by GeneNetwork project (www.genenetwork.org)
#
# Created by GeneNetwork Core Team 2010/08/10
#
# Last updated by GeneNetwork Core Team 2010/10/20

import httplib

from dbFunction import webqtlDatabaseFunction
import SharingBody


#########################################
#      Sharing Info
#########################################
class SharingInfo:

		def __init__(self, GN_AccessionId, InfoPageName):
				self.GN_AccessionId = GN_AccessionId
				self.InfoPageName = InfoPageName

		def getInfo(self):
				cursor = webqtlDatabaseFunction.getCursor()
				if (not cursor):
						return
				sql = "select Id, GEO_Series, Status, Title, Organism, Experiment_Type, Summary, Overall_Design, Contributor, Citation, Submission_Date, Contact_Name, Emails, Phone, URL, Organization_Name, Department, Laboratory, Street, City, State, ZIP, Country, Platforms, Samples, Species, Normalization, InbredSet, InfoPageName, DB_Name, Organism_Id, InfoPageTitle, GN_AccesionId, Tissue, AuthorizedUsers, About_Cases, About_Tissue, About_Download, About_Array_Platform, About_Data_Values_Processing, Data_Source_Acknowledge, Progreso from InfoFiles where "
				if(self.GN_AccessionId):
						sql += "GN_AccesionId = %s"
						cursor.execute(sql, self.GN_AccessionId)
				elif (self.InfoPageName):
						sql += "InfoPageName = %s"
						cursor.execute(sql, self.InfoPageName)
				else:
						raise 'No correct parameter found'
				info = cursor.fetchone()
				# fetch datasets file list
				try:
					conn = httplib.HTTPConnection("atlas.uthsc.edu")
					conn.request("GET", "/scandatasets.php?GN_AccesionId=%s" % (info[32]))
					response = conn.getresponse()
					data = response.read()
					filelist = data.split()
					conn.close()
				except Exception:
					filelist = []
				return info, filelist
				
		def getBody(self, infoupdate=""):
				info, filelist = self.getInfo()
				if filelist:
					htmlfilelist = '<ul style="line-height:160%;">\n'
					for i in range(len(filelist)):
							if i%2==0:
									filename = filelist[i]
									filesize = filelist[i+1]
									htmlfilelist += "<li>"
									htmlfilelist += '<a href="ftp://atlas.uthsc.edu/users/shared/Genenetwork/GN%s/%s">%s</a>' % (self.GN_AccessionId, filename, filename)
									htmlfilelist += '&nbsp;&nbsp;&nbsp;'
									#r=re.compile(r'(?<=\d)(?=(\d\d\d)+(?!\d))')
									#htmlfilelist += '[%s&nbsp;B]' % r.sub(r',',filesize)
									if 12<len(filesize):
										filesize=filesize[0:-12]
										filesize += ' T'
									elif 9<len(filesize):
										filesize=filesize[0:-9]
										filesize += ' G'
									elif 6<len(filesize):
										filesize=filesize[0:-6]
										filesize += ' M'
									elif 3<len(filesize):
										filesize=filesize[0:-3]
										filesize += ' K'
									htmlfilelist += '[%sB]' % filesize
									htmlfilelist += "</li>\n"
					htmlfilelist += "</ul>"
				else:
					htmlfilelist = "Data sets are not available or are not public yet."
				return SharingBody.sharinginfo_body_string % (info[31], info[32], infoupdate, info[32], info[1], info[3], info[30], info[4], info[27], info[33], info[2], info[23], info[26], info[11], info[15], info[16], info[18], info[19], info[20], info[21], info[22], info[13], info[12], info[14], info[14], htmlfilelist, info[6], info[35], info[36], info[37], info[38], info[39], info[40], info[5], info[7], info[8], info[9], info[10], info[17], info[24])