SHA1 algorithm -- fully portable

[Tomás Ó hÉilidhe]

I've come across a few implementations of the algorithm but none of
them have been fully-portable.

Particularly, I'm looking for an algorithm that doesn't make
assumptions on:
1) Byte order (e.g. big-endian Vs little-endian).
2) The size of any of the integer types.
3) The presence or lack of padding in any of the integer types.

If someone has a fully-portable implementation of the algorithm I'd be
very appreciative if you could post it here, or give me a link, or
perhaps mail it to me:
tKoe AT lavKabit DOT cKom (remove the K's)

Failing that, if someone could point me to a very-well written
implementation of the algorithm that I can edit myself then I'd be
appreciative.

 

[ymuntyan]

I've come across a few implementations of the algorithm but none of
them have been fully-portable.

Particularly, I'm looking for an algorithm that doesn't make
assumptions on:
1) Byte order (e.g. big-endian Vs little-endian).
2) The size of any of the integer types.
3) The presence or lack of padding in any of the integer types.

If someone has a fully-portable implementation of the algorithm I'd be
very appreciative if you could post it here, or give me a link, or
perhaps mail it to me:
tKoe AT lavKabit DOT cKom (remove the K's)

Failing that, if someone could point me to a very-well written
implementation of the algorithm that I can edit myself then I'd be
appreciative.

http://rafb.net/p/vOSU0M89.html
Should be portable, but I don't have a nine-bit-byte
machine to test (nor did I test it much on what I have).
It is not very-well written, but that wasn't a goal

Regards,
Yevgen

 

[user923005]

I've come across a few implementations of the algorithm but none of
them have been fully-portable.

Particularly, I'm looking for an algorithm that doesn't make
assumptions on:
1) Byte order (e.g. big-endian Vs little-endian).
2) The size of any of the integer types.
3) The presence or lack of padding in any of the integer types.

If someone has a fully-portable implementation of the algorithm I'd be
very appreciative if you could post it here, or give me a link, or
perhaps mail it to me:
    tKoe AT lavKabit DOT cKom  (remove the K's)

Failing that, if someone could point me to a very-well written
implementation of the algorithm that I can edit myself then I'd be
appreciative.

Did you try ?

Follow-up added.

 

[Tomás Ó hÉilidhe]

http://rafb.net/p/vOSU0M89.html
Should be portable, but I don't have a nine-bit-byte
machine to test (nor did I test it much on what I have).
It is not very-well written, but that wasn't a goal

Regards,
Yevgen

Thanks Yevgen, but the link appears to be broken.

 

[Bill Reid]

I've come across a few implementations of the algorithm but none of
them have been fully-portable.

Particularly, I'm looking for an algorithm that doesn't make
assumptions on:
1) Byte order (e.g. big-endian Vs little-endian).
2) The size of any of the integer types.
3) The presence or lack of padding in any of the integer types.

No guarantees about the above, but have you seen the following
(I noticed this when I was answering another question here about
large-scale open source "C" projects:

http://lxr.linux.no/linux/lib/sha1.c

If someone has a fully-portable implementation of the algorithm I'd be
very appreciative if you could post it here, or give me a link, or
perhaps mail it to me:
tKoe AT lavKabit DOT cKom (remove the K's)

Failing that, if someone could point me to a very-well written
implementation of the algorithm that I can edit myself then I'd be
appreciative.

Again, no guarantees about this...who know if it even works on
Torvalds' Timex Sinclair...