Had a look at the array version which throws an index error,

I think the error is coming from the BARINDEX – array[n] and  that the BARINDEX equation doesn’t recognise array[n] as an integer.

Even though the element value may be an integer,  array[n] is not an int data type, and I think that may be something to do with it.
So the error is coming from the barindex side and not the array side, pre-formatting array or isset etc, not going to affect it.

In your other post you discussed

drawtext(“text #array[n]#”,x,y         and         drawtext(array[n],x,y)

where the former doesn’t display the array value, but the latter does.

I think the same thing is happening with

Barindex – array[n]    and          arr = array[n]     ,      barindex – arr

Changing the array’s in the barindex equations with normal variables versions seems to avoid the index error.

Don’t no if displayed results are right , i’ll leave that up to you to determine.

Once RSILen = 14
Once prd = 1
Once pBcTi = -1
Once pBmTi = -1
 
src = RSI[RSILen](close)
 

If src[prd] = lowest[2*prd+1](src) then
pBcTi=pBcTi+1
$pBcT[pBcTi]= Barindex-1
 
if pBcTi > 2 and isset($pBcT[pBcTi]) then

 var = $pBcT[pBcTi-1]
var2 = $pBcT[pBcTi-2]
var3 = $pBcT[pBcTi]

if src[barindex-var] <= src[barindex-var2] and src[barindex-var] <= src[barindex-var3] then
pBmTi=pBmTi+1
$pBmT[pBmTi]= $pBcT[pBcTi-1]
drawpoint($pBcT[pBcTi-1],src[barindex-var],3) coloured("red",100)
endif
endif
endif
 
Return scr

Your’re right !  … under certain scenarios, it appears not to work, but then others it does.

I’m looking a bit deeper!

Placing the original index terms in the equations, to causes the index error, but enclosing the whole code in,

IF barindex < n then   …  endif

Avoids the index error when n is smaller than barindex, increasing n beyond barindex, error doesn’t return.

changing to

IF barindex = barindex then    …. endif

appears to negate the error as well. Not sure why this avoids index error, unless it helps aligning to a bar somehow!

This loosely give me the idea,   that the error is after the OHLC data of the chart build,   and into the current candle tick area.

Comparing the output of your latest code and a version modified in above way, show odd missing points when both compared.

However, if you re-add the modified error version, the missing points show up in the right place on a chart re-build.

This again points to some deviation when ticks are being used.

Not got any further with it yet, need to think about a while and figure out how to test further.