split the defragment options

[yapayapayapa]

-a N     Select an action to perform. The number N is a value from 1 to 11, default is 3:
1 = Analyze only, do not fixup and do not optimize.
2 = Analyze and fixup, do not optimize.
3 = Analyze, fixup, and fast optimization (default).
5 = Analyze and force together.
6 = Analyze and move to end of disk.
7 = Analyze and sort files by name (folder + filename).
8 = Analyze and sort files by size (smallest first).
9 = Analyze and sort files by last access (newest first).
10 = Analyze and sort files by last change (oldest first).
11 = Analyze and sort files by creation time (oldest first).
 

i think that this is a too large list for a single option, and isn't the more efficient, what about other forms of listing? and if i dont want reclaim MFT? and if i only wish sort the files without defrag? i suggest split that options list on:

- anlizing:
0=use the windows api (less efficient but safe, default)
?=use another (more efficient) form (like read the MFT and the FAT as you plan in the help files)

- defragment
0=no defragment
1=defragment (default of course xD)
?=another future defrag forms that dont use the windows api (like coping the files on other hardisk an copy back...) or use others windows-free software api (defrag the pagefiles and registry...)

- "reclaim MFT" option

-sorting
0= no sorting (default)
?= forms of sorting (name a-z, name z-a, size small-big, size big-small...), if we have sorting dont do optimize

- optimization
0= no optimizing
1= fast optimizing
2= slow optimizing (fill all gaps on the disk if all files are readable, if it cant, part another readable file and fill the spaces if is posible)
3= calculated optimizing (before all, calculate the posible filling of the disk, important files first, small files (1-3 clusters) for fill gaps, large files last (the program dont do that by defrag delays); fill all gaps on the disk if all files are readable, if it cant, fill the spaces with non compacted small files, if that files are over, move the defragmented files and fillup with a fragmented cluster-sized file, the last option is part another (too big) fragmented readable file and fill spaces if is posible)
?=another form of optimization (like force together, move to the end of the disk...) if we have optimizing dont do sorting

its only a idea, maybe i think something that someone thinked before and maybe dont be considered (too many code rewriting) then i aprecciate a non flaming answer

[jeroen]

Thank you very much for sharing your thoughts, I appreciate it. I am also not happy with the commandline. It is simply not flexible enough to handle all the possibilities. For example, what if somebody wants to use a different sorting method for the spacehogs than for the regular files? Add to that at least 20 other things I want to implement, and the commandline becomes horribly complicated and totally unusable. That's why version 4 of JkDefrag will have a totally different way of specifying options, it has a small scripting language.

[yapayapayapa]

thanks by your reply, i appreciate it too

create a scripting language isn't the most easy solution, is more reliable a config file (like a *.ini) that do this... or a registry entry, but lose portability... or a gui, but lose the automatic and unattended execution

[jeroen]

create a scripting language isn't the most easy solution, is more reliable a config file (like a *.ini) that do this... or a registry entry, but lose portability... or a gui, but lose the automatic and unattended execution

The scripting language is intended for power users only. Version 4 will still accept the v3 commandline arguments, and will contain a default script so it can still be used fully automatically and unattended.