Update README.md

Add info about the removal of rows and the prefix suffix.

README.md

@@ -1,6 +1,6 @@
 # archivist-utilities
 
-This is a tool that implements the functionality described at https://wiki.ucl.ac.uk/display/CTTEAM/Mappings
+This is a tool that implements the functionality described at https://wiki.ucl.ac.uk/display/CTTEAM/Mappings, where there is also a link to the Excel template to use. 
 
 The tool is accessible at https://closer-cohorts.github.io/archivist-utilities/
 
@@ -10,6 +10,8 @@ The worksheet containing data from which the qv, tv and tq mappings are created
 
 The worksheet containing data from which the dv mappings are created MUST be the second worksheet in the file. This worksheet can have any name.
 
+Any rows which do not contain all the coloumns listed below will not be present in the output.
+
 ## QV mappings file
 
 The following column headers must be present in the first worksheet in the input Excel file, in order to create the qv text files:
@@ -21,7 +23,7 @@ The following column headers must be present in the first worksheet in the input
 
 The headers are not case sensitive, i.e. a column called 'question name' instead of 'Question Name' is acceptable.
 
-A separate qv text file will be created for each unique questionnaire prefix value in the first worksheet in the input Excel file.
+A separate qv text file will be created for each unique questionnaire prefix value in the first worksheet in the input Excel file. The questionnaire prefix will automatically have _ccs01 suffixed. 
 
 For example, if some rows in the input worksheet had the value 'heaf_17_fup4' for the 'Questionnaire prefix' column, and some other rows had the value 'heaf_17_fup5' for that column, this would result in 2 text files being generated: 
 
@@ -53,7 +55,9 @@ The following column headers must be present in the first worksheet in the input
  - Question name
  - Topic id
 
-The headers are not case sensitive, i.e. a column called 'question name' instead of 'Question Name' is acceptable.
+The headers are not case sensitive, i.e. a column called 'question name' instead of 'Question Name' is acceptable. For rows which have grid cells (e.g. qc_1$1;1), all but one will be removed and this will have the suffix renoved (e.g. qc_1) to allow for loading into Archivist which only allows one topic for question. 
+
+emove all but one of the rows which have $ suffix (which have the same topic ID), keeping one row and deleting the $x;x.
 
 A separate tq text file will be created for each unique questionnaire prefix value in the input Excel file.