Info on histogram diff

[raygard]

Not an issue with your code, but the only way I know to give you this FYI.

In your blog post (nice post BTW) you mention histogram diff and say you "still have to implement it and understand it better before writing about it." Some months ago I looked for some good info on how it really works and could not find it, so I studied the original jgit implementation until I understood it (not easy), then wrote about it and implemented it in C (C99).

The code is licensed 0BSD so it can be freely used with no attribution required (but would be nice). I haven't learned Go yet but intend to. Feel free to model anything you want on my code as closely or loosely as you wish.

P.S. when I said "I’ve been interested in file comparison (diff) algorithms for quite a while", I mean since 1972 or so, when a co-worker had a program called "correction card generator" on a Univac 1108, that produced a patch deck from two source decks. (Actually card image files on a hard drive.)

[znkr]

Thank you for the note, I really appreciate it. I have actually read your first article before writing mine and learned a lot. I didn't see your second article or your code though. Time constraints are the only things that stopped me from implementing it so far, but I'll eventually get to it sooner or later.