From a3dadc21cc13b9d4da37dafb7b05586ca4bef6e3 Mon Sep 17 00:00:00 2001
From: tm689 <tm689@cam.ac.uk>
Date: Fri, 1 Mar 2024 11:29:02 +0000
Subject: [PATCH] doc: updating documentation for the advanced grouping
---
coordinator/ProcessorSurveys.py | 28 ++++++++++++++++------------
1 file changed, 16 insertions(+), 12 deletions(-)
diff --git a/coordinator/ProcessorSurveys.py b/coordinator/ProcessorSurveys.py
index e489732..f609856 100644
--- a/coordinator/ProcessorSurveys.py
+++ b/coordinator/ProcessorSurveys.py
@@ -3,19 +3,23 @@
Group surveys and run source gen on each groups:
Grouping is only run if it is defined in the config (otherwise it use all available surveys).
-
-Example:
- "Groups" : {
- "PROD" : ["ODK-server", "kobo-server", "CSV-CAM"],
- "WRT" : ["ODK", "CSV"]
- }
-
-It is based on the 'Origin' column of the surveys.
-All unique 'Origin' will create its own group alongside the ones defined in the config.
-Groups can be ignored by defining 'GroupsToIgnore' in the config (this has no effect on 'Origin' types).
-
+Groups need to have a name, a key to at least one column and a list of unique elements in that column to group by.
Example:
- "GroupsToIgnore" : ["ODK-server", "kobo-server", "CSV-CAM", "ODK", "CSV", "PV"]
+ "Groups" : {
+ "PROD" : {
+ "Origin" : ["CSV_01", "CSV_02", "ODK_01", "ODK_02"],
+ "PublishedLevel" : ["Unpubl", "Publ", "Unknown"]
+ },
+ "WRT-Publ" : {
+ "Origin" : ["CSV_01", "CSV_02", "ODK_01", "ODK_02"],
+ "PublishedLevel" : ["Publ"]
+ }
+ },
+ "GroupBy" : ["Origin", "PublishedLevel"],
+ "GroupsToIgnore" : ["CSV_01", "CSV_02", "ODK_01", "ODK_02", "Unpubl", "Publ"],
+
+Alongside the defined groups, individual groups get created based on the unique elements in the columns defined in "GroupBy".
+Groups can be ignored by defining "GroupsToIgnore" in the config.
'''
import csv
--
GitLab