SHA1(3)

💅 SHA1 — ok bestie, we're doing NIST Secure Hash Algorithm ✨

So like, this is the manual for SHA1(3), from the Library Functions Manual, and it's literally the whole tea on calculating the NIST Secure Hash Algorithm. Grab a snack, we're going through every single thing 🍓

🌸 NAME 🌸

SHA1Init, SHA1Update, SHA1Pad, SHA1Final, SHA1Transform, SHA1End, SHA1File, SHA1FileChunk, SHA1Data — calculate the NIST Secure Hash Algorithm

📚 LIBRARY 📚

Message Digest (MD4, MD5, etc.) Support Library (libmd, -lmd)

🫶 SYNOPSIS 🫶

#include <sys/types.h>
#include <sha1.h>

void
SHA1Init(SHA1_CTX *context);

void
SHA1Update(SHA1_CTX *context, const uint8_t *data, size_t len);

void
SHA1Pad(SHA1_CTX *context);

void
SHA1Final(uint8_t digest[SHA1_DIGEST_LENGTH], SHA1_CTX *context);

void
SHA1Transform(uint32_t state[5],
    const uint8_t buffer[SHA1_BLOCK_LENGTH]);

char *
SHA1End(SHA1_CTX *context, char *buf);

char *
SHA1File(const char *filename, char *buf);

char *
SHA1FileChunk(const char *filename, char *buf, off_t offset,
    off_t length);

char *
SHA1Data(const uint8_t *data, size_t len, char *buf);

✨ DESCRIPTION ✨

Okay so the SHA1 functions implement the NIST Secure Hash Algorithm (SHA-1), which is FIPS PUB 180-1. SHA-1 is used to generate a condensed representation of a message called a message digest. The algorithm takes a message less than 2^64 bits as input and produces a 160-bit digest suitable for use as a digital signature. It's giving "tiny little fingerprint of your whole entire message" 💖

🚨 Lowkey the most important part, bestie: while the SHA1 functions are considered to be more secure than the md4(3) and md5(3) functions with which they share a similar interface, they are considered insecure as of 2005, and as of 2020 chosen-prefix attacks have become practical, thus these must not be used in cryptographic contexts. Not a vibe. Do not use this for crypto. Periodt. 💀

Let me introduce you to the whole friend group, one function at a time:

• SHA1Init() — initializes a SHA1_CTX context for use with SHA1Update(), and SHA1Final(). Your little setup moment ✨
• SHA1Update() — adds data of length len to the SHA1_CTX specified by context. Feed it bytes, it eats them 🍓
• SHA1Final() — called when all data has been added via SHA1Update(), and it stores a message digest in the digest parameter. The grand finale, the mic drop 🎤
• SHA1Pad() — can be used to apply padding to the message digest as in SHA1Final(), but the current context can still be used with SHA1Update(). Like, you pad it but you're not done-done yet 🙈
• SHA1Transform() — used by SHA1Update() to hash 512-bit blocks and forms the core of the algorithm. Most programs should use the interface provided by SHA1Init(), SHA1Update() and SHA1Final() instead of calling SHA1Transform() directly. So like... it's the backstage crew, not the main character, okay? 🎀
• SHA1End() — a front end for SHA1Final() which converts the digest into an ASCII representation of the 160 bit digest in hexadecimal. Hex girl summer 🔥
• SHA1File() — calculates the digest for a file and returns the result via SHA1End(). If SHA1File() is unable to open the file a NULL pointer is returned. 😭
• SHA1FileChunk() — behaves like SHA1File() but calculates the digest only for that portion of the file starting at offset and continuing for length bytes or until end of file is reached, whichever comes first. A zero length can be specified to read until end of file. A negative length or offset will be ignored. Slice it, dice it 👑
• SHA1Data() — calculates the digest of an arbitrary string and returns the result via SHA1End(). One-shot, no thoughts, just hash 🦋

Now the buf situation, because it's lowkey a whole thing: for each of the SHA1End(), SHA1File(), and SHA1Data() functions the buf parameter should either be a string of at least 41 characters in size or a NULL pointer. In the latter case, space will be dynamically allocated via malloc(3) and should be freed using free(3) when it is no longer needed. Don't be the person who forgets the free(3), I'm begging 🫶

🌈 EXAMPLES 🌈

Bestie, the follow code fragment will calculate the digest for the string "abc" which is 0xa9993e364706816aba3e25717850c26c9cd0d89d. Watch and learn 💜

SHA1_CTX sha;
uint8_t results[SHA1_DIGEST_LENGTH];
char *buf;
int n;

buf = "abc";
n = strlen(buf);
SHA1Init(&sha);
SHA1Update(&sha, (uint8_t *)buf, n);
SHA1Final(results, &sha);

/* Print the digest as one long hex value */
printf("0x");
for (n = 0; n < SHA1_DIGEST_LENGTH; n++)
        printf("%02x", results[n]);
putchar('\n');

Alternately, the helper functions could be used in the following way. Look how much shorter this is, we love efficiency ✨

uint8_t output[SHA1_DIGEST_STRING_LENGTH];
char *buf = "abc";

printf("0x%s\n", SHA1Data(buf, strlen(buf), output));

👀 SEE ALSO 👀

Go read your other faves, bestie:

• cksum(1)
• sha1(1)
• md4(3)
• md5(3)
• rmd160(3)
• sha2(3)

And the actual source material, the academic texts:

• J. Burrows, The Secure Hash Standard, FIPS PUB 180-1.
• D. Eastlake and P. Jones, US Secure Hash Algorithm 1, RFC 3174.

🧚 HISTORY 🧚

The SHA-1 functions appeared in OpenBSD 2.0. An icon, a legend, and honestly a little bit of a cautionary tale 💅

🎀 AUTHORS 🎀

This implementation of SHA-1 was written by Steve Reid. Absolute legend behavior 👑

The SHA1End(), SHA1File(), SHA1FileChunk(), and SHA1Data() helper functions are derived from code written by Poul-Henning Kamp. Thank you for your service 🫶

💀 CAVEATS 💀

Okay, warnings time, and I need you to actually listen:

• This implementation of SHA-1 has not been validated by NIST and as such is not in official compliance with the standard. So like, it's SHA-1 flavored, not officially blessed 🙈
• If a message digest is to be copied to a multi-byte type (ie: an array of five 32-bit integers) it will be necessary to perform byte swapping on little endian machines such as the i386, alpha, and vax. Byte order jumpscare, no cap 😭

🚨 And again for the people in the back: insecure as of 2005, chosen-prefix attacks practical as of 2020, do not use in cryptographic contexts. This is the rule. The vibe check has been failed. Use sha2(3) instead, bestie 💖

That's the whole page!! This was Debian's copy, dated February 13, 2008, page SHA1(3). Mwah, go hash responsibly 💅✨🎀