cortesi / scurve Goto Github PK

@cortesi
I was working on something similar a few years back, but the image magic library I used sucked, so it got held back and forgotten. I have some ideas for improvements though.

• It is extremely helpful to be able to resize the width of images (on-the-fly), because doing this allow certain sections to align perfectly. In fact for certain types of bit mapped images, you can actually see them!
• The data are often aligned to certain popular values (128, 256, 512,1024 etc), setting the width to multiples of these also help the visualization.
• When visualizing large blobs, it is useful to average the byte values. So that 1 "pixel" can represent the average of multiples of (same set as above). That way, for example, a section of mostly ASCII, will still be represented as ASCII and so on.
• ARM code (and surely many other code) has typical sequences of bytes, that could be visualized as code etc. (alternatively, we could combine this with objdump and/or readelf output, to know what sections we are in.

There are more, but since IDK if anyone will ever see this...

Hello,
I am trying to use the Z-order and Hilbert filling curve as a way to index co-ordinates.
The problem that I am facing is that when using a negative co-ordinate as input for the algorithm I get back unusual numbers as index.

Does your implementation support negative inputs or should I for example normalize the input dataset in some way to make it non-negative?

Where is the 'file' class in line166 d = file(args[0]).read() of binvis.py

h = hilbert.Hilbert(3, 3)

for i in range(15, 17):
x,y,z = h[i]
print(x, y, z)

This prints
2 0 1
3 0 2

Is that an invalid increment since both the X and Z values change in one step?