From ee7634757dda09b798dd0b8084603d5957a0e92b Mon Sep 17 00:00:00 2001
From: tm689 <tm689@cam.ac.uk>
Date: Tue, 27 Feb 2024 13:36:54 +0000
Subject: [PATCH] feat: converting countries short name to full name.
---
coordinator/ProcessorSurveysWRT.py | 33 ++++++++++++++++++++++++------
1 file changed, 27 insertions(+), 6 deletions(-)
diff --git a/coordinator/ProcessorSurveysWRT.py b/coordinator/ProcessorSurveysWRT.py
index 1201838..54446e4 100644
--- a/coordinator/ProcessorSurveysWRT.py
+++ b/coordinator/ProcessorSurveysWRT.py
@@ -20,6 +20,27 @@ from ProcessorUtils import (
logger = logging.getLogger('Processor.Surveys.WRT')
add_filters_to_sublogger(logger)
+# Downloads of the WRT based on the ODK form reports country descriptions as two letter codes.
+# The reason is unclear, as viewing online shows full country names. To be
+# consistent across all form types, this function is available for translating
+# two letter codes to full string names.
+# Only some have these have been checked explicitly against downloaded surveys.
+# Assuming the code is based on ISO 3166-1 alpha-2
+COUNTRY_ABBREVIATIONS_DICT = {
+ 'AF' : 'Afghanistan',
+ 'BD' : 'Bangladesh',
+ 'BT' : 'Bhutan',
+ 'ET' : 'Ethiopia',
+ 'IN' : 'India',
+ 'KE' : 'Kenya',
+ 'NP' : 'Nepal',
+ 'PK' : 'Pakistan',
+ 'TZ' : 'Tanzania',
+ 'UG' : 'Uganda',
+ 'ZM' : 'Zambia',
+ 'ZW' : 'Zimbabwe',
+ }
+
# Cases to convert WRT categories to expected short name category or value
cases_incident = {
'':'none',
@@ -40,12 +61,12 @@ cases_severity = {
def get_from_WRT(form_credentials: dict, startDate: str, endDate: str):
'''Make the HTTP request, then checking the status code.'''
- url = f"{form_credentials['url']}WheatRustAPI/api/Survey/GetAll"
- #url = f"{form_credentials['url']}WheatRustAPI/api/Survey/GetByPublishedLevel"
+ #url = f"{form_credentials['url']}WheatRustAPI/api/Survey/GetAll"
+ url = f"{form_credentials['url']}WheatRustAPI/api/Survey/GetByPublishedLevel"
url += '?strStopDate=' + endDate
url += '&strStartDate=' + startDate
url += '&strArrayCountryID='+ form_credentials['countryID']
- #url += '&strArrayPublishedLevel=Unpubl'
+ url += '&strArrayPublishedLevel=' + form_credentials['publishedLevel']
# set up http session
session = requests.Session()
@@ -177,7 +198,7 @@ def get_WRT_form_as_csv(form_credentials: dict, jobPath: str, config: dict, stat
"ObservationID" : 'KEY',
"OriginalID" : 'None',
"Origin" : "Origin",
- "Country" : 'surveyor_infromation-country',
+ "Country" : ('parse_cases',(('name_out','surveyor_infromation-country'),('cases', COUNTRY_ABBREVIATIONS_DICT))),
"Latitude" : 'survey_infromation-location-Latitude',
"Longitude" : 'survey_infromation-location-Longitude',
"FieldArea" : 'site_information-field_area',
@@ -226,9 +247,9 @@ def get_WRT_form_as_csv(form_credentials: dict, jobPath: str, config: dict, stat
download_success = False
# define filenames
- csv_filename = f"SurveyData_raw.csv"
+ csv_filename = f"SurveyData_raw_{form_credentials['form_name']}.csv"
- csv_processed_filename = f"SurveyDataProcessed.csv"
+ csv_processed_filename = f"SurveyDataProcessed_{form_credentials['form_name']}.csv"
csv_processed_path = f"{output_path}/{csv_processed_filename}"
if download_success and not skip_download:
--
GitLab