strom_demo tweaks

run.py

@@ -2,10 +2,6 @@
 #RUN ALL  - STROMATOLITES
 #==============================================================================
 
-#path: /Users/jhusson/local/bin/deepdive-0.7.1/deepdive-apps/stromatolites
-
-#==============================================================================
-
 import os, time, subprocess, yaml
 
 #tic
@@ -18,10 +14,6 @@
 #load credentials file
 with open('./credentials', 'r') as credential_yaml:
     credentials = yaml.load(credential_yaml)
-
-
-#ensure working directory is proper
-#os.chdir("/Users/jhusson/local/bin/deepdive-0.7.1/deepdive-apps/stromatolites")
 
 #INITALIZE THE POSTGRES TABLES
 print 'Step 1: Initialize the PSQL tables ...'
@@ -56,7 +48,7 @@
 print 'Step 8: Define the relationships between stromatolite phrases and distant stratigraphic entities/mentions ...'
 os.system('python ./udf/ext_strat_target_distant.py')
 
-#DEFINE RELATIONSHIPS BETWEEN TARGET AND DISTANT STRATIGRAPHIC NAMES
+#FIND BEGINNING OF REFERENCE LIST
 print 'Step 9: Delineate reference section from main body extractions ...'
 os.system('python ./udf/ext_references.py')
 
@@ -68,10 +60,6 @@
 print 'Step 11: Find adjectives describing strom target words ...'
 os.system('python ./udf/ext_target_adjective.py')
 
-#POSTGRES DUMP
-print 'Step 12: Dump select results from PSQL ...'
-output = 'pg_dump -U '+ credentials['postgres']['user'] + ' -t results -t strat_target -t strat_target_distant -t age_check -t refs_location -t bib -t target_adjectives -d ' + credentials['postgres']['database'] + ' > ./output/output.sql'
-subprocess.call(output, shell=True)
 
 #summary of performance time
 elapsed_time = time.time() - start_time

udf/ext_results.py

@@ -2,7 +2,7 @@
 #GENERATE RESULTS TABLE
 #==============================================================================
 
-import time, random, re, yaml, psycopg2, copy
+import time, random, re, yaml, psycopg2, copy, csv
 from psycopg2.extensions import AsIs
 
 start_time = time.time()
@@ -164,6 +164,24 @@
 #push update
 connection.commit()
 
+#write culled results to CSV
+cursor.execute("""
+         SELECT result_id,docid,sentid,target_word,strat_phrase_root,strat_flag,strat_name_id,in_ref,source,phrase
+        	FROM results 
+        	WHERE (is_strat_name='yes' AND source='in_sent')
+           OR (is_strat_name='yes' AND source='out_sent' AND in_ref='no')
+     """)
+
+results=cursor.fetchall()
+
+with open('./output/results.csv', 'a') as outcsv:   
+    #configure writer to write standard csv file
+    writer = csv.writer(outcsv, delimiter=',', quoting=csv.QUOTE_ALL, lineterminator='\n')
+    writer.writerow(['result_id','docid','sentid','target_word','strat_phrase_root','strat_flag','strat_name_id','in_ref','source','phrase'])
+    for item in results:
+        #Write item to outcsv
+        writer.writerow([item[0], item[1], item[2],item[3], item[4], item[5],item[6], item[7], item[8], item[9]])
+
 #close the postgres connection
 connection.close()