🚀 go-pugleaf

Lapo wrote:
> I would like to store some images (320x240 in JPEG format) in a file; When
> the file size will be over than 20MB, new images will overwrite the oldest
> ones.
> The problem i found is that JPEG images are not all the same size, so
> sometimes a new image can overwrite more than 1 old image.
> I read something about circular storage, but I'm not able to solve it.
> Any tips?

It's compression that accounts for the same size images having different 
file sizes. To store each image in the same sized section of a file will 
require assuming each image might take the largest possible amount of 
space. It might be possible to store the images uncompressed to be 
certain they are the same size, but both of those approaches seem 
wasteful of disk space.

One way this kind of thing is done is to store a "directory" in the file 
along with the data. This directory would contain the number of images 
and their offset in the file, but could also store the image names and 
so on. To use such files, functions to update the directory whenever an 
image is added or removed will be needed, as well as functions that can 
read the directory and use it to locate the images in the file. To 
replace an image, each image except the one that is to be deleted may be 
copied to a temporary file, then the new image added. Finally the 
original file will be deleted and the temporary one renamed.

This, by the way, is pretty much how zip files work. Since there already 
exist libraries to manipulate those files, you might consider using one 
of them to save yourself some coding. Other archive formats and 
libraries are available as well.