Saved changes to: metrics-collect.Rproj src/dataset_utils.sh

metrics-collect.Rproj

@@ -10,4 +10,4 @@ NumSpacesForTab: 2
 Encoding: UTF-8
 
 RnwWeave: Sweave
-LaTeX: pdfLaTeX
\ No newline at end of file
+LaTeX: pdfLaTeX

src/dataset_utils.sh

+#!/usr/bin/env bash
+# author: Oksana Riba Grognuz
+
+
+function list_datasets_matching_string () {
+   USAGE="\nUsage:\n${FUNCNAME[*]} <string>\n\n\
+           List dataset identifiers for all datasets \n\
+           with any fields matching <string>.\n"
+        
+   if [ $# -eq 0 ]
+   then
+      echo -e $USAGE; return
+   fi
+    
+   local query=$1
+   curl -s "https://renkulab.io/knowledge-graph/datasets?query=$query" | jq -r '.[] | .identifier'
+}
+
+
+function datasets_properties_by_id () {
+
+   USAGE="\nUsage:\n${FUNCNAME[*]} <dataset_id> <field>\n\n\
+           Return dataset <field> value for a <dataset_id>.\n\
+           If field is not specified, then retrieve name.\n"
+        
+   if [ $# -eq 0 ]
+   then
+      echo -e $USAGE; return
+   fi
+   
+   local dataset_id=$1
+   local field=${2:-name}  
+   
+   curl -s "https://renkulab.io/knowledge-graph/datasets/$dataset_id" | jq -r '.'\"$field\"''
+   }
+   
+
+function query_datasets_by_string () {
+   
+   USAGE="\nUsage:\n${FUNCNAME[*]} <string> <field>\n\n\
+           Retrieve dataset <field> values for all datasets \n\
+           with any fields matching <string>.\n\
+           Specify <field> as: name, title, description, identifier\n\
+           If <field> is not specified, then retrieve identifier.\n"
+        
+   if [ $# -eq 0 ]
+   then
+      echo -e $USAGE; return
+   fi
+    
+   local keyword=$1
+   local field=${2:-identifier}
+   
+   for dataset_id in `list_datasets_matching_string "$keyword"`
+   do  
+       datasets_properties_by_id $dataset_id $field
+   done
+   }
+
+
+function match_datasets_keywords () {
+ 
+   USAGE="\nUsage:\n${FUNCNAME[*]} <dataset_id> <keyword> <field>\n\n\
+           Return dataset <field> value if there is an exact\n\
+           case insensitive match of <dataset_id> keywords to <keyword>.\n\
+           If field is not specified, then retrieve identifier.\n"
+        
+   if [ $# -eq 0 ]
+   then
+      echo -e $USAGE; return
+   fi
+   
+   local dataset_id=$1
+   local keyword=`echo $2 | awk '{print tolower($0)}'`
+   local field=${3:-identifier}
+   
+   curl -s "https://renkulab.io/knowledge-graph/datasets/$dataset_id" \
+   | jq -r 'select(.keywords[] | ascii_downcase == ('\"$keyword\"')) | .'\"$field\"''
+   }
+   
+   
+function query_datasets_by_keyword () {
+ 
+   USAGE="\nUsage:\n${FUNCNAME[*]} <keyword> <field>\n\n\
+           Return dataset <field> value for all datasets with an exact\n\
+           case insensitive match of dataset keywords to <keyword>.\n\
+           If field is not specified, then retrieve identifier.\n"
+        
+   if [ $# -eq 0 ]
+   then
+      echo -e $USAGE; return
+   fi
+
+   local keyword=$1
+   local field=${2:-identifier}  
+   
+   for dataset_id in `list_datasets_matching_string "$keyword"`
+   do  
+       match_datasets_keywords $dataset_id "$keyword" $field
+   done
+   }
+   
+   
+function import_datasets_by_keyword () {
+
+   USAGE="\nUsage:\n${FUNCNAME[*]} <keyword>\n\n\
+           Import all datasets annotated with <keyword>.\n\
+           Update, if datasets were already imported.\n"
+        
+   if [ $# -eq 0 ]
+   then
+      echo -e $USAGE; return
+   fi
+
+   local keyword=$1
+   
+   for dataset_id in `query_datasets_by_keyword $keyword identifier`
+   do
+      
+      local data_name=`datasets_properties_by_id $dataset_id name`
+      local data_path="${WORK_DIR}/data/${data_name}"
+      local data_url=`datasets_properties_by_id $dataset_id url`
+      
+      if [ -d "${data_path}" ]
+      then
+         renku dataset update $data_name
+      else
+         renku dataset import -y $data_url
+      fi
+   done
+   }