I have a very large 2 dimensional array (for this case lets just make it 20 rows x 5 columns). If I am on the nth row, I have to make check to see if the nth+1 row (lets say column 3) is less than the nth row. The problem is that the nth+1 row (data[nth+1] = []) has not been created so the script bombs.
(1) is there a way of telling if a row in an array has been created? or
(2) is there a way of telling the nth+1 array location has not been created? I tried isError_() but that just bombed script.
Example:
function tester() {
var data = [ [] , [] ];
data[1] = [];
data[1][1] = 10;
for (i=1 ; i < 5 ; i++) {
if (data[1][1] < data[i+1][1]) { Browser.msgBox("You have a problem", Browser.Buttons.OK);}
}
}
Also: does anyone have a better suggestion for a script editor than the google sheets script editor. It seems to be light years behind EXCELS VBA editor? EXCEL lets you pick the 3 or 4 variables you want to see and not the total list of all variable, which you then have to search through. Also, as you move through an function using breaks, the list of variable keeps changing and moving variable names around, quite confusing. Thank you for any suggestions.
Array length:
You can use the length property of the Array objects to limit the amount of iterations in your loop based on the length of your outer array, data.
You would have to change this:
for (i=1 ; i < 5 ; i++) {
To this:
for (var i = 1 ; i < data.length - 1; i++) {
Note:
If you initialize the loop with var i = 1, the first element in data gets ignored (the array indexes start at 0). Change that to var i = 0 if you want to avoid this.
Reference:
Array.prototype.length
Related
CONTEXT
I have a source table with multiple columns (Source B table in example spreadsheet).
On column E we have max one name and on column F we can have from 0 to multiple names separated by a comma.
AIM
When there are values in F, add one row per each name (E,F - the last one can have more than one separated by a comma) and duplicate the common values.
It should keep the rows where there are no values in F.
The final result will have one less column than Source table.
WHAT I'VE TRIED
=ARRAYFORMULA(QUERY(IFERROR(SPLIT(FLATTEN(IF(ISBLANK(E3:F6);;
A3:A6&"♦"&B3:B6&"♦"&C3:C6&"♦"&D3:E6&"♦"&E3:E6&"♦"&G3:G6&"♦"&H3:H6&"♦"&I3:I6&"♦"&J3:J6&"♦"&K3:K6&"♦"&L3:L6)); "♦"));
"where Col3 <> 0"; 0))
Problem
This formula was applied to Source A table (in the example spreadsheet and that's = Source B table but without comma separated values in F), which didn't have more than one value in F at the time, and:
duplicates the common values as expected but it's not showing the values from F. Just duplicates the ones from E.
ads blank row if the source row doesn't have values in F
because in column K only one row has a value, it messes up the final data
doesn't do anything different with or without comma separated values in F
Example spreadsheet
--- EDIT ---
I've found this other post with a script that I also tested on the example sheet. I've reduced the number of columns to ease the test and because I think it's aiming for the last column being the one that is supposed to be splitted, if applicable.
It does split but it repeats the "Main" client on the left column and the splited one on the right, where I would like the outcome to have all clients on the same column.
To give you an idea, you can try the following script, I used the "Source for Script" Sheet and created another sheet called "Result". As you can see in the script, I used arrays to collect the data, then added it to the "Result" Sheet.
function splitNewRow() {
var source = SpreadsheetApp.getActiveSpreadsheet().getSheetByName("Source for Script");
var data = source.getRange(2, 1, source.getLastRow(), source.getLastColumn()).getValues();
var outcomeSheet = SpreadsheetApp.getActiveSpreadsheet().getSheetByName("Result");
newArray = [];
newRow = [];
for (i=0; i < data.length; i++){
var claim = source.getRange(i+2,1).getValue();
var mainClient = source.getRange(i+2,2).getValue();
var others = source.getRange(i+2, 3).getValue();
hasComma = others.indexOf(",") != -1;
if (others != "" && hasComma == true){
newArray = [];
newArray = others.split(",");
newArray.push(mainClient);
for (j=0; j < newArray.length; j++){
newRow = [claim, newArray[j]];
outcomeSheet.appendRow(newRow);
}
}
else if(others != "" && hasComma == false){
newRow = [claim, mainClient];
outcomeSheet.appendRow(newRow);
newRow = [claim, others];
outcomeSheet.appendRow(newRow);
}
else{
newRow = [claim, mainClient];
outcomeSheet.appendRow(newRow);
}
}
}
After running the script, you get the following:
If you have any questions, let me know.
I have a spreadsheet with criteria, a start and end date, and a value. The goal is to find the lowest value for each unique criteria and start date without overlapping dates (exclusive of end date). I made a pivot table to make it easier for myself but I know there is probably a way to highlight all valid rows that meet the above requirements with some formula or conditional formatting.
I have attached a google drive link where the spreadsheet can be found here and I have some images of the sheet as well. I know that it might be possible with conditional formatting but I just don't know how to combine everything I want it to do in a single formula.
Example below:
Row 2 is a valid entry because it has the lowest value for Item 1 starting on 03-15-2021, same with row 9.
Row 5 is valid because the start date does not fall within the date range of row 2 (exclusive of end date)
Row 7 is not valid because the start date is between the start and end date of row 6
You may add a bounded script to your project. Then you can call it either with a picture/drawing that has the function assigned (button-like), or adding a menu to Google Sheets.
From what you said in the question and the comments, this seems to do what you are trying. Notice that this requires the V8 runtime (which should be the default).
function validate() {
// Get the correct sheet
const spreadsheet = SpreadsheetApp.getActiveSpreadsheet()
const sheet = spreadsheet.getSheetByName('Sheet1')
// Get the data
const length = sheet.getLastRow() - 1
const range = sheet.getRange(2, 1, length, 4)
const rows = range.getValues()
const data = Array.from(rows.entries(), ([index, [item, start, end, value]]) => {
/*
* Row Index
* 1 Criteria 1
* 2 Item 1 0
* 3 Item 1 1
* 4 Item 1 2
*
* row = index + 2
*/
return {
row: index + 2,
criteria: item,
start: start.getTime(),
end: end.getTime(),
value: value
}
})
// Sort the data by criteria (asc), start date (asc), value (asc) and end date (asc)
data.sort((a, b) => {
let order = a.criteria.localeCompare(b.criteria)
if (order !== 0) return order
order = a.start - b.start
if (order !== 0) return order
order = a.value - b.value
if (order !== 0) return order
order = a.end - b.end
return order
})
// Iterate elements and extract the valid ones
// Notice that because we sorted them, the first one of each criteria will always be valid
const valid = []
let currentCriteria
let currentValid = []
for (let row of data) {
if (row.criteria !== currentCriteria) {
// First of the criteria
valid.push(...currentValid) // Move the valids from the old criteria to the valid list
currentValid = [row] // The new list of valid rows is only the current one (for now)
currentCriteria = row.criteria // Set the criteria
} else {
const startDateCollision = currentValid.some(valid => {
row.start >= valid.start && row.start < valid.end
})
if (!startDateCollision) {
currentValid.push(row)
}
}
}
valid.push(...currentValid)
// Remove any old marks
sheet.getRange(2, 5, length).setValue('')
// Mark the valid rows
for (let row of valid) {
sheet.getRange(row.row, 5).setValue('Valid')
}
}
Algorithm rundown
We get the sheet that we have the data in. In this case we do it by name (remember to change it if it's not the default Sheet1)
We read the data and transform it in a more an array of objects, which for this case makes it easier to manage
We sort the data. This is similar to the transpose you made but in the code. It also forces a priority order and groups it by criteria
Iterate the rows, keeping only the valid:
We keep a list of all the valid ones (valid) and one for the current criteria only (currentValid) because we only have to check data collisions with the ones in the same criteria.
The first iteration will always enter the if block (because currentCriteria is undefined).
When changing criteria, we dump all the rows in currentValid into valid. We do the same after the loop with the last criteria
When changing criteria, the CurrentValid is an array with the current row as an element because the first row will always be valid (because of sorting)
For the other rows, we check if the starting date is between the starting and ending date of any of the valid rows for that criteria. If it's not, add it to this criteria's valid rows
We remove all the current "Valid" in the validity row and fill it out with the valids
The cornerstone of the algorithm is actually sorting the data. It allows us to not have to search for the best row, as it's always the next one. It also ensures things like that the first row of a criteria is always valid.
Learning resources
Javascript tutorial (W3Schools)
Google App Scripts
Overview of Google Apps Script
Extending Google Sheets
Custom Menus in Google Workspace
Code references
Class SpreadsheetApp
Class Sheet
Sheet.getRange (notice the 3 overloads)
let ... of (MDN)
Spread syntax (...) (MDN)
Arrow function expressions (MDN)
Array.from() (MDN)
Array.prototype.push() (MDN)
Array.prototype.sort() (MDN)
Date.prototype.getTime() (MDN)
String.prototype.localeCompare() (MDN)
I am trying to do a SUM of last 12 rows in the column (I'll be adding more rows into this column so I wanted to automate the calculation).
First of all, I am able to get the value of last cell with some value in this column by =SUMPRODUCT(MAX((B1:B200<>"")*ROW(B1:B200))) - result is stored in C1. However, I am not sure how to use this value inside the SUM formula, I was thinking something like =SUM(B(get value of C1)-12:B(get value of C1).
I tried multiple things but none of them have worked - I also don't mind using a different approach if it gets the job done.
You can create your own custom function to do that using Google Apps Script (GAS).
Try the following:
function onEdit(e){
var row = e.range.getRow();
var col = e.range.getColumn();
if ( col==2 && e.source.getActiveSheet().getName() == "Sheet1" ){
e.source.getActiveSheet().getRange("C1").setValue(sumLast12());
}
}
function sumLast12() {
var sheet = SpreadsheetApp.getActive().getSheetByName("Sheet1");
var sheet_size = sheet.getLastRow();
var elmt = sheet.getRange("B1:B"+sheet_size).getValues().flat([1]);
var elmt12 = elmt.slice(-12);
var sum = 0;
for( var i = 0; i < elmt12.length; i++ ){
sum += parseInt( elmt12[i], 10 );
}
return sum;
}
Explanation:
In order to activate this functionality go the menu bar on top of the
spreadsheet file and click on Tools => Script editor and copy the
aforementioned code into a blank script document (see attached
screenshot for more information) and save the document (cntrl+s).
After the script has been saved, everytime you edit a cell in column
B (either by adding a new value on the bottom or modify an existing value, the script will automatically update the value in
cell C1 with the sum of the last 12 values in column B.
Note that if you don't want to change my code, name the sheet you are working with as Sheet1.
Does this work?
=SUM(FILTER(B:B,ROW(B:B)>=MAX(ROW(B:B))-12)
I am a hobbyest making mods in TableTop Simulator using LUA and have a question that I can not seam to work out.
I have a number of "objects" which is a table in TTS that contains various data for those objects. For example.. obj.position = {x,y,z}... and can be accessed at the axis level as well.
obj.position = {5,10,15} -- x,y,z
obj.position.x == 5
This is an example. The makers of TTS have made it so you can access all the parts like that. So I can acess the object.. and then its various parts. There is a heap, like name, mesh, difuse and a ton more. roations{x,y,z} etc etc
Anyway. I have a table of objects... and would like to order those objects based on the positional data of the x axis.. so highest to lowest. So if I have a table and obj1 in that table is x=3 and obj2 is x=1 and obj3 = x=2 it would be sorted as obj2,obj3,obj1
Pseudo code:
tableOfObjects = {obj1,obj2,obj3}
--[[
tableOfObjectsp[1] == obj1
tableOfObjectsp[2] == obj2
tableOfObjectsp[3] == obj3
tableOfObjectsp[1].position.x == 3
tableOfObjectsp[2].position.x == 1
tableOfObjectsp[4].position.x == 2
--]]
---After Sort it would look this list
tableOfObjects = {obj1,obj3,obj2}
--[[
tableOfObjectsp[1] == obj1
tableOfObjectsp[2] == obj3
tableOfObjectsp[3] == obj2
tableOfObjectsp[1].position.x == 3
tableOfObjectsp[2].position.x == 2
tableOfObjectsp[3].position.x == 1
--]]
I hope I am making sense. I am self taught in the last few months!
So basically I have a table of objects and want to sort the objects in that table based on a single value attached to each individual object in the table. In this case the obj.position.x
Thanks!
You need table.sort. The first argument is the table to sort, the second is a function to compare items.
Example:
t = {
{str = 42, dex = 10, wis = 100},
{str = 18, dex = 30, wis = 5}
}
table.sort(t, function (k1, k2)
return k1.str < k2.str
end)
This article has more information
table.sort(tableOfObjects, function(a, b) return a.position.x > b.position.x end)
This line will sort your table tableOfObjects in descending order by the x-coordinate.
To reverse order, replace > by <.
From the Lua reference manual:
table.sort (list [, comp])
Sorts list elements in a given order, in-place, from list[1] to
list[#list]. If comp is given, then it must be a function that
receives two list elements and returns true when the first element
must come before the second in the final order (so that, after the
sort, i < j implies not comp(list[j],list[i])). If comp is not given,
then the standard Lua operator < is used instead.
Note that the comp function must define a strict partial order over
the elements in the list; that is, it must be asymmetric and
transitive. Otherwise, no valid sort may be possible.
The sort algorithm is not stable: elements considered equal by the
given order may have their relative positions changed by the sort.
So in other words table.sort will sort a table in ascending order by its values.
If you want to order descending or by something other than the table value (like the x-coordinate of your table value's position in your case) you have to provide a function that tells Lua which element will come first.
you can create a function that handles this exact thing:
local function fix_table(t)
local x_data = {};
local inds = {};
local rt = {};
for i = 1, #t do
x_data[#x_data + 1] = t[i].position.x;
inds[t[i].position.x] = t[i];
end
local min_index = math.min(table.unpack(x_data));
local max_index = math.max(table.unpack(x_data));
for i = min_index, max_index do
if inds[i] ~= nil then
rt[#rt + 1] = inds[i];
end
end
return rt;
end
local mytable = {obj1, obj2, obj3};
mytable = fix_table(mytable);
fix_table first takes in every x value inside of the given table, and also places a new index inside the table inds according to each x value (so that they will be ordered from least to greatest), then it gets the smallest value in the x_data array table, which is used to traverse the inds table in order. fix_table checks to make sure that inds[i] is not equal to nil before it increases the size of the return table rt so that every value in rt is ordered from greatest to least, starting at index 1, and ending at index #rt, finally rt is returned.
I hope this helped.
I have a table made of several rows and a variable number of columns in each rows.
If I want to add data in a new row, I just do
table[#table+1] = {['d1'] = data, ['d2'] = data, ... }
Now what I'd like to do is if I want to add to line 1 (for example):
table[1] = {['d' .. #columns+1] = data}
except that does not work and I can't find the solution.
My problem right now is that when my code adds data to an already existing row, it overwrites the existing data, which is not what I want.
For example this is currently the result for one line of the table:
-- Table: {4} { ["d3"]=154.04, },
instead of having a 'd1', 'd2' and finally 'd3' I just have 'd3'.
The code
table[1] = {['d' .. #columns] = data}
replaces the value at table[1] with the table on the right.
Try this instead:
table[1]['d' .. #columns] = data
For those who might have the same question, here is the closest answer I came to:
the # operator only counts integer keys. So # won't count any table containing string keys such as table[p1], which was my problem.
I solved it using the following function:
function tlength(T)
local count = 0
for _ in pairs(T) do count = count + 1 end
return count
end
Now I can count any number of items in a column/subcolumn and add to it.
The problem I had with using numerical indices is that it makes the code unreadable, I didn't know what whas what. With this I can still use string indices and count them. It's a trade-off. One more function or the # operator.
Cheers