This question was closed, reopened and is now 4 out of 5 votes of the way to being closed again.

It seems a perfectly good question to me. The key part of the question is "how to identify the specific clusters on any disk which contain pieces of any particular given file" which is straightforward enough. RRUZ's answer nails it.

I also don't think it's too localized, because I've seen people wanting to do similar things before and there are legitimate reasons why you might need to. It's a rare requirement, but not unique.

  • 2
    Clearly it was tagged wrong, the question has nothing to do with Delphi or its IDE. You get usually get help from the Delphi community about operating system question, but there's only so much they'll put up with. I'd recommend [windows] and [filesystems] instead. Oct 8, 2012 at 20:41
  • @UphillLuge: good point. I've retagged. Oct 8, 2012 at 20:46
  • The way I see it, I had an end-product idea presented to me, so to begin, I started with the core fundamentals, seeking whatever method necessary in Delphi to achieve it. I was expecting an answer of how to use Delphi to obtain list of clusters on a disk which contain any part of any particular given file. It became clear after comments that there were 3 questions: 1) What API call for obtaining file clusters, 2) What API call for obtaining total cluster count, 3) Implementing in Delphi. While I know 3 questions isn't suitable, it's all related so I wanted to cover everything. Oct 9, 2012 at 0:31
  • Therefore, I have come to the conclusion that using Windows API Calls is a subject of its own apart from Delphi, and on the other hand each API Call might require another corresponding question for clarification. Oct 9, 2012 at 2:04
  • Does anyone want to present a case that the question is indeed too localized? If not, I'm inclined to vote to reopen - again. :-) Oct 9, 2012 at 2:24
  • Is it a question about how to find physically where the file is located on the disk, or how fragmented it is? Because the former is entirely uninteresting (because the OS abstracts this level of detail out under normal circumstances, and software developers rarely have to think about it), whereas the latter is quite interesting.
    – user102937
    Oct 9, 2012 at 3:01
  • 1
    @RobertHarvey: I think you need to know the physical location of the parts of the file in order to determine how significant the fragmentation is. (For example, if there are two fragments you may want to know whether they're really close together or on opposite ends of the partition.) So the answer is the same either way. Oct 9, 2012 at 3:08
  • @RobertHarvey and also in case you wanted to write software to recover data from a badly corrupted (in places) filesystem, it would be useful to know something about the disk layout. So the original question title is certainly interesting to some people, though perhaps not to you.
    – David Z
    Oct 9, 2012 at 4:02

1 Answer 1


I edited the question's title to more accurately reflect the body of the question, and reopened.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .