Reputation: 15
I currently have a list with two columns. The first column is student name, and the second column is the number of points they have.
I imported this list from multiple spreadsheets so there were many duplicates on the names of the students. I am able to remove the duplicates, but I want to keep a tally on the total points they have. For example:
Amy 10 Bob 9 Carol 15 Amy 12
would turn into:
Amy 22 Bob 9 Carol 15
This is what I have so far:
var target = SpreadsheetApp.getActiveSpreadsheet();
var sheet = target.getSheetByName("Sheet2");
var data = sheet.getRange("A2:B1000").getValues();
var newData = new Array();
var k = 0
var finallist = []
for(i in data){
k++;
var row = data[i];
var duplicate = false;
for(j in newData){
if(row[0] == newData[j][0]){
duplicate = true;
var storedHour = sheet.getRange("B"+k).getValue();
var position = finallist.indexOf(row[0]);
var originalCell = sheet.getRange("B"+(position+1));
var originalHour = originalCell.getValue();
originalCell.setValue(originalHour + storedHour);
sheet.getRange(k,2).setValue("")
sheet.getRange(k,1).setValue("")
}
}
if(!duplicate){
newData.push(row);
finallist.push(row[0])
}
}
}
The problem I'm having is that we have a really large data sample and I'm afraid it may run over Google's 5 minute maximum execution time. Is there another more efficient way to achieve my goal?
Upvotes: 1
Views: 1323
Reputation: 3728
Sorting before comparing allows looking at the next item only instead of all items for each iteration. A spillover benefit is finallist result is alphabatized. Execution time reduction significant.
function sumDups() {
var target = SpreadsheetApp.getActiveSpreadsheet();
var sheet = target.getSheetByName("Sheet2");
var data = sheet.getRange("A2:B" + sheet.getLastRow()).getValues().sort();
var finallist = [];
for(var i = 0; i<= data.length - 1; i++){
var hours = data[i][1];
while((i < data.length - 1) && (data[i][0] == data[i+1][0])) {
hours += data[i+1][1];
i++;
};
finallist.push([data[i][0], hours]);
};
Logger.log(finallist);
}
Edit: the simple data structure with the name being in the first column allows this to work. For anything more complex understanding and applying the methods shown in @Kos's answer is preferable
Upvotes: 0
Reputation:
Your code is running slow because Spreadsheets API methods (like getRange
) are time consuming and much slower then other JavaScript code.
Here is optimized function with reduced number of such Spreadsheets API calls:
function calcNumbers()
{
var target = SpreadsheetApp.getActiveSpreadsheet();
var sheet = target.getSheetByName("Sheet2");
var lastRow = sheet.getLastRow();
var dataRange = sheet.getRange(2, 1, lastRow-1, 2);
var data = dataRange.getValues();
var pointsByName = {};
for (var i = 0; i < data.length; i++)
{
var row = data[i];
var curName = row[0];
var curNumber = row[1];
// empty name
if (!curName.trim())
{
continue;
}
// if name found first time, save it to object
if (!pointsByName[curName])
{
pointsByName[curName] = Number(curNumber);
}
// if duplicate, sum numbers
else
{
pointsByName[curName] += curNumber;
}
}
// prepare data for output
var outputData = Object.keys(pointsByName).map(function(name){
return [name, pointsByName[name]];
});
// clear old data
dataRange.clearContent();
// write calculated data
var newDataRange = sheet.getRange(2, 1, outputData.length, 2);
newDataRange.setValues(outputData);
}
Upvotes: 1