path: root/src/util.c

#include "util.h"

#include <errno.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/random.h>
#include <time.h>
#include <unistd.h>

#include "sha256.h"

void *xmalloc(size_t n)
{
    void *p = malloc(n ? n : 1);
    if (!p) {
        perror("malloc");
        exit(1);
    }
    return p;
}

void *xcalloc(size_t n, size_t sz)
{
    void *p = calloc(n ? n : 1, sz ? sz : 1);
    if (!p) {
        perror("calloc");
        exit(1);
    }
    return p;
}

void *xrealloc(void *p, size_t n)
{
    void *q = realloc(p, n ? n : 1);
    if (!q) {
        perror("realloc");
        exit(1);
    }
    return q;
}

char *xstrdup(const char *s)
{
    char *p = strdup(s);
    if (!p) {
        perror("strdup");
        exit(1);
    }
    return p;
}

void buf_init(struct buf *b)
{
    b->p = NULL;
    b->len = 0;
    b->cap = 0;
}

void buf_free(struct buf *b)
{
    free(b->p);
    b->p = NULL;
    b->len = b->cap = 0;
}

static void buf_reserve(struct buf *b, size_t extra)
{
    if (b->len + extra <= b->cap)
        return;
    size_t nc = b->cap ? b->cap : 64;
    while (nc < b->len + extra)
        nc *= 2;
    b->p = xrealloc(b->p, nc);
    b->cap = nc;
}

void buf_append(struct buf *b, const void *data, size_t n)
{
    if (!n)
        return;
    buf_reserve(b, n);
    memcpy(b->p + b->len, data, n);
    b->len += n;
}

void buf_append_u16be(struct buf *b, uint16_t v)
{
    unsigned char t[2] = { (unsigned char)(v >> 8), (unsigned char)v };
    buf_append(b, t, sizeof t);
}

void buf_append_u32be(struct buf *b, uint32_t v)
{
    unsigned char t[4] = { (unsigned char)(v >> 24), (unsigned char)(v >> 16),
                           (unsigned char)(v >> 8), (unsigned char)v };
    buf_append(b, t, sizeof t);
}

void buf_append_u64be(struct buf *b, uint64_t v)
{
    unsigned char t[8];
    for (int i = 0; i < 8; i++)
        t[i] = (unsigned char)(v >> (56 - 8 * i));
    buf_append(b, t, sizeof t);
}

void util_sha256(const void *data, size_t n, unsigned char out[32])
{
    SHA256_CTX ctx;
    sha256_init(&ctx);
    sha256_update(&ctx, (const BYTE *)data, n);
    sha256_final(&ctx, out);
}

void util_hex(const unsigned char *in, size_t n, char *out)
{
    static const char hex[] = "0123456789abcdef";
    for (size_t i = 0; i < n; i++) {
        out[2 * i] = hex[in[i] >> 4];
        out[2 * i + 1] = hex[in[i] & 15];
    }
    out[2 * n] = '\0';
}

static const char B64URL[] =
    "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_";

char *util_b64url(const unsigned char *in, size_t n)
{
    size_t olen = ((n + 2) / 3) * 4 + 1;
    char *out = xmalloc(olen);
    size_t i = 0, o = 0;
    while (i + 3 <= n) {
        uint32_t v = ((uint32_t)in[i] << 16) | ((uint32_t)in[i + 1] << 8) |
                     (uint32_t)in[i + 2];
        out[o++] = B64URL[(v >> 18) & 63];
        out[o++] = B64URL[(v >> 12) & 63];
        out[o++] = B64URL[(v >> 6) & 63];
        out[o++] = B64URL[v & 63];
        i += 3;
    }
    if (i + 1 == n) {
        uint32_t v = (uint32_t)in[i] << 16;
        out[o++] = B64URL[(v >> 18) & 63];
        out[o++] = B64URL[(v >> 12) & 63];
    } else if (i + 2 == n) {
        uint32_t v = ((uint32_t)in[i] << 16) | ((uint32_t)in[i + 1] << 8);
        out[o++] = B64URL[(v >> 18) & 63];
        out[o++] = B64URL[(v >> 12) & 63];
        out[o++] = B64URL[(v >> 6) & 63];
    }
    out[o] = '\0';
    return out;
}

char *util_b64(const unsigned char *in, size_t n)
{
    static const char B64[] =
        "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
    size_t olen = ((n + 2) / 3) * 4 + 1;
    char *out = xmalloc(olen);
    size_t i = 0, o = 0;
    while (i + 3 <= n) {
        uint32_t v = ((uint32_t)in[i] << 16) | ((uint32_t)in[i + 1] << 8) |
                     (uint32_t)in[i + 2];
        out[o++] = B64[(v >> 18) & 63];
        out[o++] = B64[(v >> 12) & 63];
        out[o++] = B64[(v >> 6) & 63];
        out[o++] = B64[v & 63];
        i += 3;
    }
    if (i + 1 == n) {
        uint32_t v = (uint32_t)in[i] << 16;
        out[o++] = B64[(v >> 18) & 63];
        out[o++] = B64[(v >> 12) & 63];
        out[o++] = '=';
        out[o++] = '=';
    } else if (i + 2 == n) {
        uint32_t v = ((uint32_t)in[i] << 16) | ((uint32_t)in[i + 1] << 8);
        out[o++] = B64[(v >> 18) & 63];
        out[o++] = B64[(v >> 12) & 63];
        out[o++] = B64[(v >> 6) & 63];
        out[o++] = '=';
    }
    out[o] = '\0';
    return out;
}

int util_random(void *out, size_t n)
{
    unsigned char *p = out;
    size_t got = 0;
    while (got < n) {
        ssize_t r = getrandom(p + got, n - got, 0);
        if (r < 0) {
            if (errno == EINTR)
                continue;
            break;
        }
        got += (size_t)r;
    }
    if (got == n)
        return 0;
    int fd = open("/dev/urandom", O_RDONLY);
    if (fd < 0)
        return -1;
    got = 0;
    while (got < n) {
        ssize_t r = read(fd, p + got, n - got);
        if (r <= 0) {
            if (r < 0 && errno == EINTR)
                continue;
            close(fd);
            return -1;
        }
        got += (size_t)r;
    }
    close(fd);
    return 0;
}

char *util_random_id(const char *prefix, size_t nbytes)
{
    unsigned char raw[64];
    if (nbytes > sizeof raw)
        nbytes = sizeof raw;
    if (util_random(raw, nbytes) != 0) {
        fprintf(stderr, "fatal: no entropy source\n");
        exit(1);
    }
    char *b = util_b64url(raw, nbytes);
    size_t plen = prefix ? strlen(prefix) : 0;
    char *out = xmalloc(plen + strlen(b) + 1);
    memcpy(out, prefix ? prefix : "", plen);
    strcpy(out + plen, b);
    free(b);
    return out;
}

int util_const_eq(const void *a, const void *b, size_t n)
{
    const volatile unsigned char *x = a;
    const volatile unsigned char *y = b;
    unsigned char d = 0;
    for (size_t i = 0; i < n; i++)
        d |= (unsigned char)(x[i] ^ y[i]);
    return d == 0;
}

int64_t util_now(void)
{
    return (int64_t)time(NULL);
}

void util_iso8601(int64_t t, char *buf, size_t n)
{
    time_t tt = (time_t)t;
    struct tm tm;
    gmtime_r(&tt, &tm);
    strftime(buf, n, "%Y-%m-%dT%H:%M:%SZ", &tm);
}

char *util_str_trim(char *s)
{
    while (*s == ' ' || *s == '\t' || *s == '\r' || *s == '\n')
        s++;
    size_t len = strlen(s);
    while (len > 0) {
        char c = s[len - 1];
        if (c != ' ' && c != '\t' && c != '\r' && c != '\n')
            break;
        s[--len] = '\0';
    }
    return s;
}

static int is_leap(int y)
{
    return (y % 4 == 0 && y % 100 != 0) || y % 400 == 0;
}

int util_parse_iso_date(const char *s)
{
    if (!s || strlen(s) != 10)
        return 0;
    if (s[4] != '-' || s[7] != '-')
        return 0;
    static const int mdays[] = { 31, 28, 31, 30, 31, 30,
                                 31, 31, 30, 31, 30, 31 };
    int y = 0, m = 0, d = 0;
    for (int i = 0; i < 10; i++) {
        if (i == 4 || i == 7)
            continue;
        if (s[i] < '0' || s[i] > '9')
            return 0;
    }
    y = (s[0] - '0') * 1000 + (s[1] - '0') * 100 + (s[2] - '0') * 10 +
        (s[3] - '0');
    m = (s[5] - '0') * 10 + (s[6] - '0');
    d = (s[8] - '0') * 10 + (s[9] - '0');
    if (y < 1900 || y > 2200 || m < 1 || m > 12 || d < 1)
        return 0;
    int max = mdays[m - 1] + (m == 2 && is_leap(y) ? 1 : 0);
    return d <= max;
}

/* ------------------------------------------------------------------ */
/* base64                                                              */
/* ------------------------------------------------------------------ */

static int b64_val(char c)
{
    if (c >= 'A' && c <= 'Z')
        return c - 'A';
    if (c >= 'a' && c <= 'z')
        return c - 'a' + 26;
    if (c >= '0' && c <= '9')
        return c - '0' + 52;
    if (c == '+')
        return 62;
    if (c == '/')
        return 63;
    return -1;
}

int util_b64_decode(const char *in, size_t in_len, unsigned char **out,
                    size_t *out_len)
{
    while (in_len > 0 && (in[in_len - 1] == '\n' || in[in_len - 1] == '\r' ||
                          in[in_len - 1] == ' ' || in[in_len - 1] == '\t'))
        in_len--;
    size_t i = 0;
    while (i < in_len && (in[i] == '\n' || in[i] == '\r' || in[i] == ' ' ||
                          in[i] == '\t'))
        i++;
    if (i >= in_len)
        return -1;
    size_t cap = (in_len - i) / 4 * 3 + 3;
    unsigned char *buf = xmalloc(cap);
    size_t o = 0;
    int pad = 0;
    for (; i < in_len; i += 4) {
        if (i + 1 > in_len - 1)
            break;
        int v[4];
        for (int k = 0; k < 4 && i + (size_t)k < in_len; k++) {
            char c = in[i + (size_t)k];
            if (c == '=') {
                pad++;
                v[k] = 0;
            } else {
                v[k] = b64_val(c);
                if (v[k] < 0) {
                    free(buf);
                    return -1;
                }
                if (pad) {
                    free(buf);
                    return -1;
                }
            }
        }
        if (pad > 2) {
            free(buf);
            return -1;
        }
        buf[o++] = (unsigned char)((v[0] << 2) | (v[1] >> 4));
        if (pad < 2)
            buf[o++] = (unsigned char)(((v[1] & 15) << 4) | (v[2] >> 2));
        if (pad < 1)
            buf[o++] = (unsigned char)(((v[2] & 3) << 6) | v[3]);
    }
    *out = buf;
    *out_len = o;
    return 0;
}

/* ------------------------------------------------------------------ */
/* dates                                                               */
/* ------------------------------------------------------------------ */

int util_date_parse(const char *s, int *y, int *m, int *d)
{
    if (!util_parse_iso_date(s))
        return -1;
    *y = (s[0] - '0') * 1000 + (s[1] - '0') * 100 + (s[2] - '0') * 10 +
         (s[3] - '0');
    *m = (s[5] - '0') * 10 + (s[6] - '0');
    *d = (s[8] - '0') * 10 + (s[9] - '0');
    return 0;
}

void util_date_fmt(int y, int m, int d, char *buf, size_t n)
{
    snprintf(buf, n, "%04d-%02d-%02d", y, m, d);
}

int util_days_in_month(int y, int m)
{
    static const int mdays[] = { 31, 28, 31, 30, 31, 30,
                                 31, 31, 30, 31, 30, 31 };
    if (m < 1 || m > 12)
        return 0;
    return mdays[m - 1] + (m == 2 && is_leap(y) ? 1 : 0);
}

void util_date_add_months(const char *date, int months, char *out, size_t n)
{
    int y, m, d;
    if (util_date_parse(date, &y, &m, &d) != 0) {
        snprintf(out, n, "%s", date);
        return;
    }
    int total = y * 12 + (m - 1) + months;
    int ny = total / 12;
    int nm = total % 12 + 1;
    int max = util_days_in_month(ny, nm);
    if (d > max)
        d = max;
    util_date_fmt(ny, nm, d, out, n);
}

static int64_t days_from_civil(int y, int m, int d)
{
    y -= m <= 2;
    int64_t era = (y >= 0 ? y : y - 399) / 400;
    unsigned yoe = (unsigned)(y - era * 400);
    unsigned doy = (unsigned)((153 * (m + (m > 2 ? -3 : 9)) + 2) / 5 + d - 1);
    unsigned doe = yoe * 365 + yoe / 4 - yoe / 100 + doy;
    return era * 146097 + (int64_t)doe - 719468;
}

static void civil_from_days(int64_t z, int *y, int *m, int *d)
{
    z += 719468;
    int64_t era = (z >= 0 ? z : z - 146096) / 146097;
    unsigned doe = (unsigned)(z - era * 146097);
    unsigned yoe = (doe - doe / 1460 + doe / 36524 - doe / 146096) / 365;
    int yy = (int)(yoe + era * 400);
    unsigned doy = doe - (365 * yoe + yoe / 4 - yoe / 100);
    unsigned mp = (5 * doy + 2) / 153;
    *d = (int)(doy - (153 * mp + 2) / 5 + 1);
    *m = (int)(mp + (mp < 10 ? 3 : -9));
    *y = yy + (*m <= 2);
}

void util_date_add_days(const char *date, int days, char *out, size_t n)
{
    int y, m, d;
    if (util_date_parse(date, &y, &m, &d) != 0) {
        snprintf(out, n, "%s", date);
        return;
    }
    int64_t z = days_from_civil(y, m, d) + days;
    civil_from_days(z, &y, &m, &d);
    util_date_fmt(y, m, d, out, n);
}

int util_date_valid(const char *s)
{
    return util_parse_iso_date(s);
}

/* ------------------------------------------------------------------ */
/* CP437 conversion (subset: Latin-1 letters used in SIE files)        */
/* ------------------------------------------------------------------ */

/* Unicode codepoints for CP437 bytes 0x80..0xDF. Bytes 0xE0..0xFF are
   Greek/math symbols that do not occur in Swedish bookkeeping text and
   convert to '?'. */
static const uint16_t CP437_HIGH[0x60] = {
    0x00C7, 0x00FC, 0x00E9, 0x00E2, 0x00E4, 0x00E0, 0x00E5, 0x00E7,
    0x00EA, 0x00EB, 0x00E8, 0x00EF, 0x00EE, 0x00EC, 0x00C4, 0x00C5,
    0x00C9, 0x00E6, 0x00C6, 0x00F4, 0x00F6, 0x00F2, 0x00FB, 0x00F9,
    0x00FF, 0x00D6, 0x00DC, 0x00A2, 0x00A3, 0x00A5, 0x20A7, 0x0192,
    0x00E1, 0x00ED, 0x00F3, 0x00FA, 0x00F1, 0x00D1, 0x00AA, 0x00BA,
    0x00BF, 0x2310, 0x00AC, 0x00BD, 0x00BC, 0x00A1, 0x00AB, 0x00BB,
    0x2591, 0x2592, 0x2593, 0x2502, 0x2524, 0x2561, 0x2562, 0x2556,
    0x2555, 0x2563, 0x2551, 0x2557, 0x255D, 0x255C, 0x255B, 0x2510,
    0x2514, 0x2534, 0x252C, 0x251C, 0x2500, 0x253C, 0x255E, 0x255F,
    0x255A, 0x2554, 0x2569, 0x2566, 0x2560, 0x2550, 0x256C, 0x2567,
    0x2568, 0x2564, 0x2565, 0x2559, 0x2558, 0x2552, 0x2553, 0x256B,
    0x256A, 0x2518, 0x250C, 0x2588, 0x2584, 0x258C, 0x2590, 0x2580,
};

static size_t cp437_put_utf8(unsigned char *out, size_t o, uint16_t cp)
{
    if (cp < 0x80) {
        out[o++] = (unsigned char)cp;
    } else if (cp < 0x800) {
        out[o++] = (unsigned char)(0xC0 | (cp >> 6));
        out[o++] = (unsigned char)(0x80 | (cp & 0x3F));
    } else {
        out[o++] = (unsigned char)(0xE0 | (cp >> 12));
        out[o++] = (unsigned char)(0x80 | ((cp >> 6) & 0x3F));
        out[o++] = (unsigned char)(0x80 | (cp & 0x3F));
    }
    return o;
}

char *util_cp437_to_utf8(const unsigned char *in, size_t n)
{
    char *out = xmalloc(n * 2 + 1);
    size_t o = 0;
    for (size_t i = 0; i < n; i++) {
        unsigned char b = in[i];
        if (b < 0x80) {
            out[o++] = (char)b;
        } else if (b <= 0xDF) {
            o = cp437_put_utf8((unsigned char *)out, o, CP437_HIGH[b - 0x80]);
        } else {
            out[o++] = '?';
        }
    }
    out[o] = '\0';
    return out;
}

size_t util_utf8_to_cp437(const char *in, unsigned char *out, size_t out_sz)
{
    size_t o = 0;
    const unsigned char *p = (const unsigned char *)in;
    while (*p && o < out_sz) {
        uint32_t cp;
        size_t len;
        if (*p < 0x80) {
            cp = *p;
            len = 1;
        } else if ((*p & 0xE0) == 0xC0) {
            cp = (uint32_t)(*p & 0x1F) << 6;
            if ((p[1] & 0xC0) != 0x80)
                break;
            cp |= p[1] & 0x3F;
            len = 2;
        } else if ((*p & 0xF0) == 0xE0) {
            cp = (uint32_t)(*p & 0x0F) << 12;
            if ((p[1] & 0xC0) != 0x80 || (p[2] & 0xC0) != 0x80)
                break;
            cp |= (uint32_t)(p[1] & 0x3F) << 6;
            cp |= p[2] & 0x3F;
            len = 3;
        } else {
            p++;
            out[o++] = '?';
            continue;
        }
        if (cp < 0x80) {
            out[o++] = (unsigned char)cp;
        } else {
            unsigned char best = '?';
            for (unsigned b = 0x80; b <= 0xDF; b++) {
                if (CP437_HIGH[b - 0x80] == cp) {
                    best = (unsigned char)b;
                    break;
                }
            }
            out[o++] = best;
        }
        p += len;
    }
    return o;
}