forked from jondufault/statistical_consulting
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathETL.R
232 lines (230 loc) · 12.2 KB
/
ETL.R
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
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
#
#
# ETL.Extract = new.env()
# ETL.Transform = new.env()
# ETL.Load = new.env()
# calibration.env = new.env()
#
#
#
#
#
#
# with(ETL.Extract,{
# client_data_prepend = "Data/"
#
#
# grain_size_function = function(client_data_prepend){
#
# chemistry_bight_path = paste(client_data_prepend,"CHEMISTRY-BIGHT18.xlsx",sep="")
#
#
#
# chemical_results <- as.data.frame(read_excel(chemistry_bight_path,sheet='tbl_chemresults',col_types = c("text",
# "text", "text", "text", "text",
# "text", "text", "text", "text", "text",
# "text", "text", "text", "text",
# "text", "text", "text", "text",
# "text", "text", "text", "text",
# "text", "text", "text", "text",
# "text", "text", "text", "text")))
# chemical_results = chemical_results[chemical_results$sampletype=="Result",]
# chemical_results = subset(chemical_results, select=-c(sampletype))
#
# chemical_results.grainsize = chemical_results[chemical_results$analyteclass=="GrainSize",]
# chemical_results.grainsize = subset(chemical_results.grainsize,select=-c(analyteclass))
#
# chemical_results.grainsize$saID = paste(chemical_results.grainsize$stationid,chemical_results.grainsize$analytename)
#
# chemical_results.grainsize = subset(chemical_results.grainsize,select=-c(bioaccumulationsampleid,
# preparationbatchid,
# analysisbatchid,
# matrix,
# analysismethod,
# labsampleid,
# truevalue,
# percentrecovery,sampledate,created_date,qualifier,mdl,rl,lab,analysisdate,globalid,created_user,last_edited_user,last_edited_date,qacode,comments))
#
# chemical_results.grainsize = chemical_results.grainsize[order(chemical_results.grainsize$fieldduplicate, decreasing=TRUE),]
# chemical_results.grainsize = chemical_results.grainsize[!duplicated(chemical_results.grainsize$saID),]
#
# chemical_results.grainsize = chemical_results.grainsize[order(chemical_results.grainsize$labreplicate, decreasing=TRUE),]
# chemical_results.grainsize = chemical_results.grainsize[!duplicated(chemical_results.grainsize$saID),]
#
# chemical_results.grainsize = subset(chemical_results.grainsize,select=-c(fieldduplicate,labreplicate,saID,objectid,units))
#
# chemical_results.grainsize$result = as.numeric(chemical_results.grainsize$result)
# chemical_results.grainsize$result[chemical_results.grainsize$result==-88.0] =0
#
# chemical_results.grainsize$analytename = as.numeric(gsub(pattern='Phi ',replacement='',x=chemical_results.grainsize$analytename))
# chemical_results.grainsize = chemical_results.grainsize[chemical_results.grainsize$analytename >= 4.5,]
#
# grainsize_totals <- chemical_results.grainsize %>%
# spread(key = analytename, value = result)
# grainsize_totals[is.na(grainsize_totals)] <- 0
#
# grainsize_totals = data.frame(stationid = grainsize_totals$stationid,GrainSize = rowSums(grainsize_totals[,-1]))
#
# return(grainsize_totals)
# }
#
# })
#
#
#
#
#
#
#
#
#
# # Convert excel sheets to their own CSV files. Could do this automatically, except the data is formatted so it crashes readxl, and readxl is too stupid to understand how to handle the data.
# with(ETL.Extract,{
#
#
# chemistry_bight_path = paste(client_data_prepend,"CHEMISTRY-BIGHT18.xlsx",sep="")
# ref_phi_conversion_path = paste(client_data_prepend,"Ref - Phi Conversion.xlsx",sep="")
# station_completion_path = paste(client_data_prepend,"StationCompletionV8.xlsx",sep="")
# stations_list_path = paste(client_data_prepend,"B18 stations list.csv",sep="")
#
#
# grain_size = grain_size_function(client_data_prepend)
#
# chemistry_bight.tbl_chembatch = read_excel(chemistry_bight_path,
# sheet = "tbl_chembatch", col_types = c("text",
# "text", "text", "text", "date", "text",
# "text", "text", "text", "text"))
#
#
#
# chemistry_bight.tbl_chemresults = read_excel(chemistry_bight_path,
# sheet = "tbl_chemresults", col_types = c("text",
# "text", "text", "text", "text",
# "text", "text", "text", "text", "text",
# "text", "text", "text", "text",
# "text", "text", "text", "text",
# "text", "text", "text", "text",
# "text", "text", "text", "text",
# "text", "text", "text", "text")) # hammer against forehad
#
# chemistry_bight.tbl_sample_assignment_table = read_excel(chemistry_bight_path,
# sheet = "sample_assignment_table")
#
#
# ref_phi_conversion.ref_phi_conversion = read_excel(ref_phi_conversion_path,sheet="Ref___Phi_Conversion")
#
# ref_phi_conversion.for_r = read_excel(ref_phi_conversion_path,sheet="for R")
#
# ref_phi_conversion.sheet_2 = read_excel(ref_phi_conversion_path,sheet="Sheet2")
#
# station_completion.grabs = read_excel(station_completion_path,sheet="Grabs")
# station_completion.sheet_1 = read_excel(station_completion_path,sheet="Sheet1")
# station_completion.trawls = read_excel(station_completion_path,sheet="Trawls")
# station_completion.areas= read_excel(station_completion_path,sheet="areas")
#
# b18_stations_list = read_csv(stations_list_path)
#
# })
#
#
# with(ETL.Transform,{
#
# # table containing results of different tests performed on samples
# chemical_tests = ETL.Extract$chemistry_bight.tbl_chemresults
#
# # table that helps us convert phi values into a summary statistic for particle size in a
# # sample
# phi_conversion_values = ETL.Extract$ref_phi_conversion.for_r
#
# # table that contains length/area information about various regions
# # might need to reformat to get more general information about bays/estuaries/etc.
# region_information = ETL.Extract$station_completion.areas
#
# # information about specific stations where samples come from
# station_information = ETL.Extract$b18_stations_list
#
# })
#
# with(ETL.Transform,{
#
# # depth column
# station_completion.trawls = ETL.Extract$station_completion.trawls[c("stationid","depth")]
#
#
# # Only consider test results
# chemical_results = chemical_tests[chemical_tests$sampletype=="Result",]
# chemical_results = subset(chemical_results, select=-c(sampletype))
#
# # only look at inorganic analyte class
# chemical_results.inorganics = chemical_results[chemical_results$analyteclass=="Inorganics",]
# chemical_results.inorganics = subset(chemical_results.inorganics, select=-c(analyteclass))
#
# # set a unique id for each station and chemical
# chemical_results.inorganics$saID = paste(chemical_results.inorganics$stationid,chemical_results.inorganics$analytename)
#
# # remove the values that are not needed
# chemical_results.sans_unnecessary_ids = subset(chemical_results.inorganics,select=-c(bioaccumulationsampleid,
# preparationbatchid,
# analysisbatchid,
# matrix,
# analysismethod,
# labsampleid,
# truevalue,
# percentrecovery))
#
# # move the following columns to another table (in case they are needed in the future)
#
# chemical_results.ancilliary_data = subset(chemical_results.sans_unnecessary_ids,select=c(objectid,sampledate,created_date,qualifier,mdl,rl,lab,analysisdate,globalid,created_user,last_edited_user,last_edited_date,qacode,comments))
#
# chemical_results.without_ancillary_data = subset(chemical_results.sans_unnecessary_ids,select=-c(sampledate,created_date,qualifier,mdl,rl,lab,analysisdate,globalid,created_user,last_edited_user,last_edited_date,qacode,comments))
#
# # per e-mail, only keep the most recent fieldduplicate and labreplicate.
# chemical_results.only_most_recent_fieldduplicate1 = chemical_results.without_ancillary_data[order(chemical_results.without_ancillary_data$fieldduplicate, decreasing=TRUE),]
# chemical_results.only_most_recent_fieldduplicate2 = chemical_results.only_most_recent_fieldduplicate1[!duplicated(chemical_results.only_most_recent_fieldduplicate1$saID),]
#
#
# chemical_results.only_most_recent_labreplicate1 = chemical_results.only_most_recent_fieldduplicate2[order(chemical_results.only_most_recent_fieldduplicate2$labreplicate, decreasing=TRUE),]
# chemical_results.only_most_recent_labreplicate2 = chemical_results.only_most_recent_labreplicate1[!duplicated(chemical_results.only_most_recent_labreplicate1$saID),]
#
#
# chemical_results.without_duplicates_ids = subset(chemical_results.only_most_recent_labreplicate2,select=-c(fieldduplicate,labreplicate,saID,objectid,units))
#
# # lastly, any results where we get a -88 cannot detect, we'll assume to be 0.
# chemical_results.without_duplicates_ids$result = as.numeric(chemical_results.without_duplicates_ids$result)
# chemical_results.without_duplicates_ids$result[chemical_results.without_duplicates_ids$result==-88.0] =0
#
#
# chemical_results.final =chemical_results.without_duplicates_ids %>% spread(analytename,result)
# })
#
#
# with(ETL.Transform,{
# station_information$is_contaminated_site = 0
#
# station_information$is_contaminated_site[!is.na(station_information$River_Mout)] = 1
# station_information$is_contaminated_site[!is.na(station_information$POTW_Statu)] = 1
# station_information$is_contaminated_site[!is.na(station_information$POTW_Name)] = 1
#
#
# station_information.without_irrelevant_columns = subset(station_information,select=-c(River_Mout,POTW_Statu,POTW_Name,OBJECTID))
#
# station_information.final = station_information.without_irrelevant_columns
#
# })
#
#
# with(ETL.Transform,{
#
# calibration_data_wo_grain = merge(x=chemical_results.final,y=station_information.final,by="stationid")
# calibration_data = merge(x=calibration_data_wo_grain,y=ETL.Extract$grain_size,by="stationid")
#
#
#
#
# calibration_data = merge(x=calibration_data,y=station_completion.trawls,by="stationid")
# })
#
#
# save.image("etl.Rdata")
#
sys.load.image("etl.Rdata",quiet=F)