function translatefits, image, imagedims, vectran

;determines if it is necessary to add black space for translation
;this is done only if the translation is positive (up, right or both directions)
if imagedims[0]+vectran[0] lt imagedims[0] then tranmodx=0 else tranmodx=vectran[0]
if imagedims[1]+vectran[1] lt imagedims[1] then tranmody=0 else tranmody=vectran[1]

dummy=image
image=lonarr(imagedims[0]+tranmodx, imagedims[1]+tranmody)
  for j=0, (imagedims[0]-1) do begin
    for k=0, (imagedims[1]-1) do begin
      image[j,k]=dummy[j,k]
    endfor
  endfor

imagedims=size(image, /dimensions)

dummy=image
if vectran[0] lt 0 then begin ;here the subscript modifications are calculated
  j1=0                        ;basically these are needed so that the image
  j2=abs(vectran[0])          ;can be translated in all directions
  j3=0                        ;the algorhytm is the most complex one in the code
  j4=imagedims[0]             ;but it can be understood through the example given at
  endif else begin            ;the end of the function
  j1=vectran[0]
  j2=0
  j3=imagedims[0]
  j4=0
  endelse
if vectran[1] lt 0 then begin
  k1=0
  k2=abs(vectran[1])
  k3=0
  k4=imagedims[1]
  endif else begin
  k1=vectran[1]
  k2=0
  k3=imagedims[1]
  k4=0
  endelse
dummy=image
  for j=0, (imagedims[0]-1-(j1+j2)) do begin
      for k=0, (imagedims[1]-1-(k1+k2)) do begin
        image[j+j1,k+k1]=dummy[j+j2,k+k2]
      endfor
  endfor
  for j=0, (imagedims[0]-1-(j1+j2)) do begin
      for k=(imagedims[1]-(k1+k2)), (imagedims[1]-1) do begin
        image[j+j1,k+k1-k3]=dummy[j+j2,k+k2-k4]
      endfor
  endfor
  for j=(imagedims[0]-(j1+j2)), (imagedims[0]-1) do begin
      for k=0, (imagedims[1]-1-(k1+k2)) do begin
        image[j+j1-j3,k+k1]=dummy[j+j2-j4,k+k2]
      endfor
  endfor
  for j=(imagedims[0]-(j1+j2)), (imagedims[0]-1) do begin
      for k=(imagedims[1]-(k1+k2)), (imagedims[1]-1) do begin
        image[j+j1-j3,k+k1-k3]=dummy[j+j2-j4,k+k2-k4]
      endfor
  endfor
return, image
end

;so let's suppose we have a 5x5 image and -4,+10 translation vector. the new image will
;have dimensions 5x15, adding black space at the end of the image. now let's look at
;the coeficient clauses. vectran[0] is negative, so:
;  j1=0                        
;  j2=abs(vectran[0])
;  j3=0
;  j4=imagedims[0]
;now let's just look at the part of translation
;
;  for j=0, (imagedims[0]-1-(j1+j2)) do begin
;      for k=0, (imagedims[1]-1-(k1+k2)) do begin
;        image[j+j1,k+k1]=dummy[j+j2,k+k2]
;      endfor
;  endfor
;
;so the counter j goes from 0 to imagedims[0]-1-(j1+j2)
;imagedims[0]-1 is the maximum index
;(j1+j2) is always a positive value and eather j1 or j2 have value of 0
;if the x translation is positive then j2 will be zero
;if the x translation is negative (as in our case) then j1 will be zero
;accordingly the counters on will start from j1 of the image array if the translation
;is positive and from j2 of the dummy array if the translation is negative, effectively
;simulating translation to positive or negative direction
;
;        image[j+j1,k+k1]=dummy[j+j2,k+k2]
;
;same stands for k coeficients and y axis
;
;the first part of the code will only transfer one part of the picture
;so the j3 and j4 koeficients are used to substract the image dimensions from the
;current counter
;
;        image[j+j1-j3,k+k1-k3]=dummy[j+j2-j4,k+k2-k4]
;
;note that in those parts of the code counters go from (imagedims[0]-(j1+j2)) to
;(imagedims[0]-1)
;
;to sum up - translation is a pixel (or rather [i,j] array element) to pixel 
;transfer of data with accordingly changed coordinates for each pixel