use SHA1 with COBOL - sha1

Question

I was wondering if there's any way to apply SHA1 hash with COBOL.
If there's at least some info of how the SHA1 algorithm works it will be usefull.
Thanks

Answer 1

You didn't say which Cobol platform. If you are on z/OS, there are a variety of cryptography services that are easily called from Cobol. And SHA1 is available among those services.

Answer 2

I did a small sample embedding Python in COBOL, and picked an MD5 checksum as an example.
I wouldn't necessarily go with Python, but if you are lucky enough to be able to use OpenCOBOL, then all the features of libcrypto are a simple CALL away.
For completeness, having mentioned the Python angle, but again that's pretty hefy baggage if the goal is simply cryptography. In which case OpenSSL would be a much tighter fit. This listing is very likely NOT suitable for your needs, but it shows off the power of CALL and the C application binary interface. Please excuse me if this is just noise.
From SourceForge:
Very high level Python embedding is pretty straight forward, been there, done that.
>>SOURCE FORMAT IS FIXED
*> *******************************************************
*> Author: Brian Tiffin
*> Date: 20130126
*> Purpose: Embed Python
*> Tectonics: cobc -x cobpy.cob -lpython2.6
*> *******************************************************
identification division.
program-id. cobpy.
procedure division.
call "Py_Initialize"
on exception
display "link cobpy with -lpython2.6" end-display
end-call
call "PyRun_SimpleString" using
by reference
"from time import time,ctime" & x"0a" &
"print('Today is', ctime(time()))" & x"0a" & x"00"
on exception continue
end-call
call "Py_Finalize" end-call
goback.
end program cobpy.
Giving
$ cobc -x cobpy.cob -lpython2.6
$ ./cobpy
('Today is', 'Sat Jan 26 20:01:41 2013')
Python dutifully displayed the tuple.
But what fun is Python if it is just for high level script side effects? Lots, but still.
Pure embedding.
>>SOURCE FORMAT IS FIXED
*> *******************************************************
*> Author: Brian Tiffin
*> Date: 20130126
*> Purpose: Embed Python
*> Tectonics: cobc -x cobkat.cob -lpython2.6
*> NOTES: leaks, no Py_DECREF macros called.
*> *******************************************************
identification division.
program-id. cobkat.
data division.
working-storage section.
77 python-name usage pointer.
77 python-module usage pointer.
77 python-dict usage pointer.
77 python-func usage pointer.
77 python-stringer usage pointer.
77 python-args usage pointer.
77 python-value usage pointer.
01 cobol-buffer-pointer usage pointer.
01 cobol-buffer pic x(80) based.
01 cobol-string pic x(80).
01 cobol-integer usage binary-long.
01 command-line-args pic x(80).
*> *******************************************************
procedure division.
call "Py_Initialize"
on exception
display "link cobpy with -lpython" end-display
end-call
*> Python likes module names in Unicode
call "PyUnicodeUCS4_FromString" using
by reference "pythonfile" & x"00"
returning python-name
on exception
display "unicode problem" end-display
end-call
*> import the module, using PYTHONPATH
call "PyImport_Import" using
by value python-name
returning python-module
on exception
display "this would be borked" end-display
end-call
if python-module equal null
display "no pythonfile.py in PYTHONPATH" end-display
end-if
*> within the module, an attribute is "pythonfunction"
call "PyObject_GetAttrString" using
by value python-module
by reference "pythonfunction" & x"00"
returning python-func
on exception continue
end-call
*>
*> error handling now skimped out on
*>
*> pythonfunction takes a single argument
call "PyTuple_New" using
by value 1
returning python-args
end-call
*> of type long, hard coded to the ultimate answer
call "PyLong_FromLong" using
by value 42
returning python-value
end-call
*> set first (only) element of the argument tuple
call "PyTuple_SetItem" using
by value python-args
by value 0
by value python-value
end-call
*> call the function, arguments marshalled for Python
call "PyObject_CallObject" using
by value python-func
by value python-args
returning python-value
end-call
*> we know we get a long back, hopefully 1764
call "PyLong_AsLong" using
by value python-value
returning cobol-integer
end-call
display "Python returned: " cobol-integer end-display
*> **************************************************** *<
*> a function taking string and returning string
call "PyObject_GetAttrString" using
by value python-module
by reference "pythonstringer" & x"00"
returning python-stringer
end-call
call "PyTuple_New" using
by value 1
returning python-args
end-call
*> Use the OpenCOBOL command argument
accept command-line-args from command-line end-accept
call "PyString_FromString" using
by reference
function concatenate(
function trim(command-line-args)
x"00")
returning python-value
end-call
*> Set the function argument tuple to the cli args
call "PyTuple_SetItem" using
by value python-args
by value 0
by value python-value
end-call
*> call the "pythonstringer" function
call "PyObject_CallObject" using
by value python-stringer
by value python-args
returning python-value
end-call
*> return as String (with the MD5 hex digest tacked on)
call "PyString_AsString" using
by value python-value
returning cobol-buffer-pointer
end-call
*> one way of removing null while pulling data out of C
set address of cobol-buffer to cobol-buffer-pointer
string
cobol-buffer delimited by x"00"
into cobol-string
end-string
display "Python returned: " cobol-string end-display
*> and clear out <*
call "Py_Finalize" end-call
goback.
end program cobkat.
with pythonfile.py
#
# Simple Python sample for OpenCOBOL embedding trial
#
def pythonfunction(i):
return i * i
import hashlib
def pythonstringer(s):
sum = hashlib.md5()
sum.update(s)
return s + ": " + sum.hexdigest()
Giving
$ ./cobkat Python will use this for MD5 hash
no pythonfile.py in PYTHONPATH
Attempt to reference unallocated memory (Signal SIGSEGV)
Abnormal termination - File contents may be incorrect
Oops
$ export PYTHONPATH=.
$ ./cobkat Python will use this for MD5 hash
Python returned: +0000001764
Python returned: Python will use this for MD5 hash: c5577e3ab8dea11adede20a1949b5fb3
Haven't done one of these in a while, fun.
Cheers,
Brian
Oh, in case you're reading along, 1764 is the ultimate answer, squared.

Answer 3

Here's a more on point source listing. Requires OpenCOBOL 1.1CE for the OC_CBL_DUMP
First up, is paying homage to the authors of OpenSSL. They deserve credits.
OpenSSL License
---------------
/* ====================================================================
* Copyright (c) 1998-2011 The OpenSSL Project. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
*
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
*
* 3. All advertising materials mentioning features or use of this
* software must display the following acknowledgment:
* "This product includes software developed by the OpenSSL Project
* for use in the OpenSSL Toolkit. (http://www.openssl.org/)"
*
* 4. The names "OpenSSL Toolkit" and "OpenSSL Project" must not be used to
* endorse or promote products derived from this software without
* prior written permission. For written permission, please contact
* openssl-core#openssl.org.
*
* 5. Products derived from this software may not be called "OpenSSL"
* nor may "OpenSSL" appear in their names without prior written
* permission of the OpenSSL Project.
*
* 6. Redistributions of any form whatsoever must retain the following
* acknowledgment:
* "This product includes software developed by the OpenSSL Project
* for use in the OpenSSL Toolkit (http://www.openssl.org/)"
*
* THIS SOFTWARE IS PROVIDED BY THE OpenSSL PROJECT ``AS IS'' AND ANY
* EXPRESSED OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE OpenSSL PROJECT OR
* ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
* OF THE POSSIBILITY OF SUCH DAMAGE.
* ====================================================================
*
* This product includes cryptographic software written by Eric Young
* (eay#cryptsoft.com). This product includes software written by Tim
* Hudson (tjh#cryptsoft.com).
*
*/
Original SSLeay License
-----------------------
/* Copyright (C) 1995-1998 Eric Young (eay#cryptsoft.com)
* All rights reserved.
*
* This package is an SSL implementation written
* by Eric Young (eay#cryptsoft.com).
* The implementation was written so as to conform with Netscapes SSL.
*
* This library is free for commercial and non-commercial use as long as
* the following conditions are aheared to. The following conditions
* apply to all code found in this distribution, be it the RC4, RSA,
* lhash, DES, etc., code; not just the SSL code. The SSL documentation
* included with this distribution is covered by the same copyright terms
* except that the holder is Tim Hudson (tjh#cryptsoft.com).
*
* Copyright remains Eric Young's, and as such any Copyright notices in
* the code are not to be removed.
* If this package is used in a product, Eric Young should be given attribution
* as the author of the parts of the library used.
* This can be in the form of a textual message at program startup or
* in documentation (online or textual) provided with the package.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* "This product includes cryptographic software written by
* Eric Young (eay#cryptsoft.com)"
* The word 'cryptographic' can be left out if the rouines from the library
* being used are not cryptographic related :-).
* 4. If you include any Windows specific code (or a derivative thereof) from
* the apps directory (application code) you must include an acknowledgement:
* "This product includes software written by Tim Hudson (tjh#cryptsoft.com)"
*
* THIS SOFTWARE IS PROVIDED BY ERIC YOUNG ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* The licence and distribution terms for any publically available version or
* derivative of this code cannot be changed. i.e. this code cannot simply be
* copied and put under another distribution licence
* [including the GNU Public Licence.]
*/
And some COBOL to exercise the cryptography and the two forms of SHA1 hashing
>>source format is free
*> ***************************************************************
*> Author: Brian Tiffin
*> Date: 20130321
*> Purpose: Compute an SHA1 digest, whole
*> Tectonics: cobc -x sha1a.cob -lcrypto
*> ***************************************************************
IDENTIFICATION DIVISION.
program-id. sha1a.
data division.
working-storage section.
01 sha1-digest pic x(20).
01 digestable pic x(80) value "this message needs to be verified".
*> ***************************************************************
procedure division.
*> Compute disgest from block of memory
call "SHA1" using
by reference digestable
by value function length(function trim(digestable))
by reference sha1-digest
on exception
display "link sha1.cob with OpenSSL's libcrypto" end-display
end-call
*> Dump the hash, as it'll unlikely be printable
call "CBL_OC_DUMP" using
by reference sha1-digest
on exception continue
end-call
goback.
end program sha1a.
With a sample run of
$ cobc -x sha1a.cob
$ ./sha1a
link sha1.cob with OpenSSL's libcrypto
Offset HEX-- -- -- -5 -- -- -- -- 10 -- -- -- -- 15 -- CHARS----1----5-
000000 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20
000016 20 20 20 20
$ cobc -x sha1a.cob -lcrypto
$ ./sha1a
Offset HEX-- -- -- -5 -- -- -- -- 10 -- -- -- -- 15 -- CHARS----1----5-
000000 c7 3b 52 0c 61 39 9b f9 a5 2f fe 3f 11 90 5e 10 .;R.a9.../.?..^.
000016 3b 0d 15 c5 ;...
And a more complete example, building the digest from multiple updates.
Assumptions here: This will crack on files with trailing spaces.
>>source format is free
*> ***************************************************************
*> Author: Brian Tiffin
*> Date: 20130321
*> Purpose: Compute an SHA1 digest, by piece
*> Tectonics: cobc -x sha1.cob -lcrypto
*> ***************************************************************
IDENTIFICATION DIVISION.
program-id. sha1.
environment division.
configuration section.
input-output section.
file-control.
select samplefile
assign to "sha1.cob"
organization is line sequential
file status is sample-status
.
DATA DIVISION.
file section.
fd samplefile.
01 input-line pic x(2048).
working-storage section.
01 sha1-context usage pointer.
01 sha1-libresult usage binary-long.
88 sha1-success value 1 when set to false is 0.
01 sha1-digest pic x(20).
01 sample-status pic 9999.
01 sample-file-state pic 9.
88 no-more-sample value 9 when set to false is 0.
01 sha-ctx-structure pic x(1024).
*> ***************************************************************
PROCEDURE DIVISION.
*> Compute disgest from a sequential file
open input samplefile
if sample-status not equal to zero
display "Status of " sample-status " returned from open" end-display
display "rest of sample run will be garbage" end-display
end-if
*> Init the SHA1 internals
set sha1-context to address of sha-ctx-structure
call "SHA1_Init" using
by value sha1-context
returning sha1-libresult
on exception
display "Can't find SHA1_Init. hint: cobc -x sha1 -lcrypto" end-display
end-call
if not sha1-success
display "Could not initialize SHA1 structures" end-display
display "normally you'd want to stop run and call the emergency hotline to wake up the support techs, but this is an example and blindly continues." end-display
end-if
*> loop across some data, ignoring issue of trailing spaces on input lines
read samplefile at end set no-more-sample to true end-read
if input-line equal spaces then
move x"0a" to input-line(1:1)
else
move function concatenate(function trim(input-line trailing), x"0a") to input-line
end-if
perform until no-more-sample
call "SHA1_Update" using
by value sha1-context
by content function trim(input-line trailing)
by value function length(function trim(input-line trailing))
on exception display "internal update failure of SHA1_Update" end-display
end-call
if not sha1-success
display "Could not update SHA1 structures" end-display
display "normally you'd want to stop run." end-display
end-if
read samplefile at end set no-more-sample to true end-read
if input-line equal spaces then
move x"0a" to input-line(1:1)
else
move function concatenate(function trim(input-line trailing), x"0a") to input-line
end-if
end-perform
*> finalize the disgest
call "SHA1_Final" using
by reference sha1-digest
by value sha1-context
on exception display "you're kidding right? internal failure of SHA1_Final" end-display
end-call
close samplefile
*> Dump the hash, as it'll unlikely be printable
call "CBL_OC_DUMP" using
by reference sha1-digest
on exception continue
end-call
goback.
END PROGRAM sha1.
and another sample run of
$ cobc -x sha1.cob -lcrypto
$ ./sha1
Offset HEX-- -- -- -5 -- -- -- -- 10 -- -- -- -- 15 -- CHARS----1----5-
000000 d4 04 4b ed 02 e8 ef 54 e0 c4 73 0b 6b 51 85 bc ..K....T..s.kQ..
000016 85 73 d3 16 .s..
$ openssl sha1 sha1.cob
SHA1(sha1.cob)= d4044bed02e8ef54e0c4730b6b5185bc8573d316

Answer 4

SHA1 rquires a lot of bitwise operations (XOR,AND,OR) which are not generally supported in COBOL (they are supported by a few compilers).
Your best bet would be to simply adjust one of the many C implementations so it can be easily called as a COBOL subroutine.
Information on your platform and compiler would be useful.

Answer 5

Yes, there's a way to apply SHA1 hash with COBOL.
I have written the SHA256 hashing algorithm in COBOL, with heaps of telemetry everywhere to let you know exactly what is going on at all points along the way.
If you can do SHA256 in COBOL, you can do SHA-1 in COBOL.
Don't squib it by hashing small input character strings.
Make sure your program works for reams of input characters, so that it can be used for document authentication.
If you've understood the entire specs, then your result for that will be correct as well.
And then, see if you can hash the hash - like bitcoin does.
That is a little bit more tricky than it looks like on face value.
I have assumed that you have to do the coding in COBOL itself,
so that YOU do the real work - not just call a subprogram which someone else has written.
Anyone could do that, so you wouldn't be asking if it was just that.
Look on Github to see how the SHA256 algorithm works.
It also shows example animated intermediate calculations, not just the final result.
IMO finding the full specification is BY FAR the hardest thing.
There's a number of youtube videos, but they are describing only a few tiny parts of the full story.
But if you have ANY weakness in your COBOL skills, especially table processing, but also in structuring your code, EVEN IF YOU FINALLY UNDERSTAND THE SPECS, you'll have a lot of trouble converting SPECS into working code, depending on how you want to get your code written. Transform centred design helps.
This link is a good help as well.
https://hackernoon.com/how-sha-2-works-a-step-by-step-tutorial-sha-256-46103t6k
Use it AND the Github stuff together, but they don't both use the same example input string.
In summary, this about your analysis skills, design skills, coding skills and testing skills - if you want it to be.
Good Luck.
Let us all know how you go.
Added 18th July 2021
FYI
https://github.com/DoHITB/CryptoCobol/blob/main/SHA1HEX.CBL