I don't think I really understood the syntax before. Just to check I've got it right now, the release candidate is looking at all folders, and shows all their sizes. But, it is set to filter folders that start "program*," and within those, it considers JPG and ZIP files. SpaceSniffer's main statistics at the top of the screen now show those files in those folders only, and those are the ones where text is in bright white, and specially coloured. Other folders and files are shown with their sizes, but not considered in the main statistics.
Is that correct?
-rjbull
Well... not. I think that this way it should be difficult to understand. Maybe i lacked of explanation in my previous post. In the posted example, the filter is composed of two main parts: a folder mask and two file masks.
If you consider the disk structure as a tree, the purpose of the folder mask is to cut all branches whose paths don't contain "program*" in at least one nesting level.
After the tree has been cutted down, the file mask will search for the file extensions. It's like a two dimensional filter: the first axis is the folder dimension, the second axis is the files dimension. Both are perpendicular and not directly related.
Consider this example:
root
|
+Program Files
| |
| +Data
|
+Users
| |
| +Programs
| |
| +Data
|
+Windows
| |
| +System
The previous folder mask will consider all files under "Program Files\..." and all files under "Users\Programs\..." but not other branches because they don't contain "program*".
The statistic of a folder is always the sum of the size of their direct children, hence in the screenshot you will see that the C:\ drive contains 1.7Gb of JPG or ZIP files that are stored under a program* folder (at any nesting level). Actually, the C:\ drive is 300 Gb in size (without filters).
Suppose that your disk structure contains many folders with similar names, but you don't know how many and where they are (and at which nesting level they are). Without a folder mask you must examine each branch of the disk structure to find them. With folder mask you can find every "temp" folder with a click.
I think that the semanthics is simple (perhaps only to me because i programmed the algorithm, eheh), but if you have any other interpretation please let me know. I need to understand if the behavior of SpaceSniffer is what people expects.
Thank you! Umberto