/* See LICENSE file for copyright and license details. */ #include <stdio.h>#include <stdlib.h>#include <string.h>#include <stdarg.h>#include <ctype.h>#include <time.h> #include "util.h" static const char base64_table[] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"; voiddie(const char *fmt, ...){ va_list ap; va_start(ap, fmt); vfprintf(stderr, fmt, ap); va_end(ap); if (fmt[0] && fmt[strlen(fmt)-1] == ':') { fputc(' ', stderr); perror(NULL); } else { fputc('\n', stderr); } exit(1);} voidwarn(const char *fmt, ...){ va_list ap; va_start(ap, fmt); fprintf(stderr, "warning: "); vfprintf(stderr, fmt, ap); va_end(ap); fputc('\n', stderr);} void *xmalloc(size_t size){ void *p = malloc(size); if (!p) die("malloc:"); return p;} void *xrealloc(void *ptr, size_t size){ void *p = realloc(ptr, size); if (!p) die("realloc:"); return p;} char *xstrdup(const char *s){ char *p = strdup(s); if (!p) die("strdup:"); return p;} char *base64_encode(const unsigned char *data, size_t input_len, size_t *output_len){ size_t olen = 4 * ((input_len + 2) / 3); char *encoded = xmalloc(olen + 1); size_t i, j; for (i = 0, j = 0; i < input_len;) { unsigned int a = i < input_len ? data[i++] : 0; unsigned int b = i < input_len ? data[i++] : 0; unsigned int c = i < input_len ? data[i++] : 0; unsigned int triple = (a << 16) | (b << 8) | c; encoded[j++] = base64_table[(triple >> 18) & 0x3F]; encoded[j++] = base64_table[(triple >> 12) & 0x3F]; encoded[j++] = base64_table[(triple >> 6) & 0x3F]; encoded[j++] = base64_table[triple & 0x3F]; } /* Add padding */ size_t mod = input_len % 3; if (mod) { encoded[olen - 1] = '='; if (mod == 1) encoded[olen - 2] = '='; } encoded[olen] = '\0'; if (output_len) *output_len = olen; return encoded;} intstr_starts_with(const char *str, const char *prefix){ return strncmp(str, prefix, strlen(prefix)) == 0;} intstr_ends_with(const char *str, const char *suffix){ size_t slen = strlen(str); size_t suflen = strlen(suffix); if (suflen > slen) return 0; return strcmp(str + slen - suflen, suffix) == 0;} char *str_tolower(char *str){ for (char *p = str; *p; p++) *p = tolower((unsigned char)*p); return str;} char *str_trim(char *str){ char *end; while (isspace((unsigned char)*str)) str++; if (*str == '\0') return str; end = str + strlen(str) - 1; while (end > str && isspace((unsigned char)*end)) end--; end[1] = '\0'; return str;} char *url_get_domain(const char *url){ const char *start, *end; char *domain; /* Skip protocol */ if (str_starts_with(url, "https://")) start = url + 8; else if (str_starts_with(url, "http://")) start = url + 7; else if (str_starts_with(url, "//")) start = url + 2; else start = url; /* Find end of domain */ end = start; while (*end && *end != '/' && *end != ':' && *end != '?') end++; size_t len = end - start; domain = xmalloc(len + 1); memcpy(domain, start, len); domain[len] = '\0'; /* Strip leading www. (and similar) so www.example.com and example.com * are treated as the same domain. Critical for sites with mixed links. */ if (strncasecmp(domain, "www.", 4) == 0) { memmove(domain, domain + 4, strlen(domain) - 3); } return domain;} inturl_same_domain(const char *url1, const char *url2){ char *d1 = url_get_domain(url1); char *d2 = url_get_domain(url2); int same = strcasecmp(d1, d2) == 0; free(d1); free(d2); return same;} char *url_resolve(const char *base, const char *relative){ char *result; /* Already absolute */ if (str_starts_with(relative, "http://") || str_starts_with(relative, "https://") || str_starts_with(relative, "data:")) { return xstrdup(relative); } /* Protocol-relative */ if (str_starts_with(relative, "//")) { size_t len = 6 + strlen(relative); result = xmalloc(len + 1); snprintf(result, len + 1, "https:%s", relative); return result; } char *domain = url_get_domain(base); const char *proto = str_starts_with(base, "https://") ? "https://" : "http://"; /* Root-relative */ if (relative[0] == '/') { size_t len = strlen(proto) + strlen(domain) + strlen(relative); result = xmalloc(len + 1); snprintf(result, len + 1, "%s%s%s", proto, domain, relative); free(domain); return result; } /* Find base path */ const char *path_start; if (str_starts_with(base, "https://")) path_start = base + 8; else if (str_starts_with(base, "http://")) path_start = base + 7; else path_start = base; /* Skip domain */ while (*path_start && *path_start != '/') path_start++; /* Find last slash in path */ const char *last_slash = strrchr(path_start, '/'); if (!last_slash) last_slash = path_start; size_t base_len = last_slash - path_start + 1; size_t len = strlen(proto) + strlen(domain) + base_len + strlen(relative); result = xmalloc(len + 1); snprintf(result, len + 1, "%s%s%.*s%s", proto, domain, (int)base_len, path_start, relative); free(domain); return result;} char *get_mime_type(const char *url){ /* Strip query string */ char *copy = xstrdup(url); char *query = strchr(copy, '?'); if (query) *query = '\0'; str_tolower(copy); const char *mime = "application/octet-stream"; if (str_ends_with(copy, ".jpg") || str_ends_with(copy, ".jpeg")) mime = "image/jpeg"; else if (str_ends_with(copy, ".png")) mime = "image/png"; else if (str_ends_with(copy, ".gif")) mime = "image/gif"; else if (str_ends_with(copy, ".webp")) mime = "image/webp"; else if (str_ends_with(copy, ".svg")) mime = "image/svg+xml"; else if (str_ends_with(copy, ".ico")) mime = "image/x-icon"; else if (str_ends_with(copy, ".css")) mime = "text/css"; else if (str_ends_with(copy, ".js")) mime = "application/javascript"; else if (str_ends_with(copy, ".woff")) mime = "font/woff"; else if (str_ends_with(copy, ".woff2")) mime = "font/woff2"; else if (str_ends_with(copy, ".ttf")) mime = "font/ttf"; else if (str_ends_with(copy, ".otf")) mime = "font/otf"; else if (str_ends_with(copy, ".eot")) mime = "application/vnd.ms-fontobject"; free(copy); return xstrdup(mime);} char *sanitize_filename(const char *url){ char *domain = url_get_domain(url); size_t len = strlen(domain) + 32; char *filename = xmalloc(len); /* Replace dots with underscores */ for (char *p = domain; *p; p++) if (*p == '.') *p = '_'; snprintf(filename, len, "%s", domain); free(domain); return filename;} char *get_iso_date(void){ time_t t = time(NULL); struct tm *tm = gmtime(&t); char *buf = xmalloc(32); strftime(buf, 32, "%Y-%m-%dT%H:%M:%SZ", tm); return buf;} /* make_relative_path: given two paths from site root (like url_to_path results), * compute minimal relative reference from 'from' (current) to 'to' (target). * Handles going up with ../ and down. Assumes no .. or . in inputs. */char *make_relative_path(const char *from, const char *to){ if (!from || !to) return xstrdup(to ? to : ""); /* If identical (after norm), empty relative (or just filename if same) */ if (strcmp(from, to) == 0) { const char *last = strrchr(to, '/'); return xstrdup(last ? last + 1 : to); } /* Work on dir of 'from' only (current document's directory) */ char *from_dir = xstrdup(from); char *slash = strrchr(from_dir, '/'); if (slash) *slash = '\0'; else from_dir[0] = '\0'; /* Find common prefix length in from_dir and to */ const char *f = from_dir; const char *t = to; const char *last_match = NULL; while (*f && *t && *f == *t) { if (*f == '/') last_match = f; f++; t++; } if (*f == '\0' && (*t == '\0' || *t == '/')) last_match = f; /* full match or to under from_dir */ /* Count how many dirs up from from_dir's remaining */ int up = 0; const char *f_rem = last_match ? last_match + 1 : from_dir; if (*f_rem == '/') f_rem++; for (const char *p = f_rem; *p; p++) if (*p == '/') up++; /* from_dir may have had trailing after last_match, count its / */ /* Build result: ups * "../" + suffix of to after last_match */ const char *to_rem = last_match ? last_match + 1 : to; if (*to_rem == '/') to_rem++; size_t up_len = (size_t)up * 3; size_t rem_len = strlen(to_rem); char *result = xmalloc(up_len + rem_len + 1); result[0] = '\0'; for (int i = 0; i < up; i++) strcat(result, "../"); strcat(result, to_rem); /* If result empty, use the basename of to */ if (result[0] == '\0') { const char *b = strrchr(to, '/'); free(result); result = xstrdup(b ? b + 1 : to); } free(from_dir); return result;} /* * url_path - suckless helper: return pointer to the path part of a URL. * Used everywhere we used to manually skip protocol + domain. * Examples: * "https://example.com/foo/bar?x=1" -> "/foo/bar?x=1" * "http://www.example.com" -> "/" * "/already/relative" -> "/already/relative" */const char *url_path(const char *url){ const char *p; if (!url) return "/"; p = url; if (str_starts_with(p, "https://")) p += 8; else if (str_starts_with(p, "http://")) p += 7; else if (str_starts_with(p, "//")) p += 2; /* Skip domain (until /, ?, #, or end) */ while (*p && *p != '/' && *p != '?' && *p != '#') p++; if (*p == '\0' || *p == '?' || *p == '#') return "/"; return p; /* points at the leading '/' of the path */}
/* See LICENSE file for copyright and license details. */
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <stdarg.h>
#include <ctype.h>
#include <time.h>
#include "util.h"
static const char base64_table[] =
"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
void
die(const char *fmt, ...)
{
va_list ap;
va_start(ap, fmt);
vfprintf(stderr, fmt, ap);
va_end(ap);
if (fmt[0] && fmt[strlen(fmt)-1] == ':') {
fputc(' ', stderr);
perror(NULL);
} else {
fputc('\n', stderr);
}
exit(1);
}
void
warn(const char *fmt, ...)
{
va_list ap;
va_start(ap, fmt);
fprintf(stderr, "warning: ");
vfprintf(stderr, fmt, ap);
va_end(ap);
fputc('\n', stderr);
}
void *
xmalloc(size_t size)
{
void *p = malloc(size);
if (!p)
die("malloc:");
return p;
}
void *
xrealloc(void *ptr, size_t size)
{
void *p = realloc(ptr, size);
if (!p)
die("realloc:");
return p;
}
char *
xstrdup(const char *s)
{
char *p = strdup(s);
if (!p)
die("strdup:");
return p;
}
char *
base64_encode(const unsigned char *data, size_t input_len, size_t *output_len)
{
size_t olen = 4 * ((input_len + 2) / 3);
char *encoded = xmalloc(olen + 1);
size_t i, j;
for (i = 0, j = 0; i < input_len;) {
unsigned int a = i < input_len ? data[i++] : 0;
unsigned int b = i < input_len ? data[i++] : 0;
unsigned int c = i < input_len ? data[i++] : 0;
unsigned int triple = (a << 16) | (b << 8) | c;
encoded[j++] = base64_table[(triple >> 18) & 0x3F];
encoded[j++] = base64_table[(triple >> 12) & 0x3F];
encoded[j++] = base64_table[(triple >> 6) & 0x3F];
encoded[j++] = base64_table[triple & 0x3F];
}
/* Add padding */
size_t mod = input_len % 3;
if (mod) {
encoded[olen - 1] = '=';
if (mod == 1)
encoded[olen - 2] = '=';
}
encoded[olen] = '\0';
if (output_len)
*output_len = olen;
return encoded;
}
int
str_starts_with(const char *str, const char *prefix)
{
return strncmp(str, prefix, strlen(prefix)) == 0;
}
int
str_ends_with(const char *str, const char *suffix)
{
size_t slen = strlen(str);
size_t suflen = strlen(suffix);
if (suflen > slen)
return 0;
return strcmp(str + slen - suflen, suffix) == 0;
}
char *
str_tolower(char *str)
{
for (char *p = str; *p; p++)
*p = tolower((unsigned char)*p);
return str;
}
char *
str_trim(char *str)
{
char *end;
while (isspace((unsigned char)*str))
str++;
if (*str == '\0')
return str;
end = str + strlen(str) - 1;
while (end > str && isspace((unsigned char)*end))
end--;
end[1] = '\0';
return str;
}
char *
url_get_domain(const char *url)
{
const char *start, *end;
char *domain;
/* Skip protocol */
if (str_starts_with(url, "https://"))
start = url + 8;
else if (str_starts_with(url, "http://"))
start = url + 7;
else if (str_starts_with(url, "//"))
start = url + 2;
else
start = url;
/* Find end of domain */
end = start;
while (*end && *end != '/' && *end != ':' && *end != '?')
end++;
size_t len = end - start;
domain = xmalloc(len + 1);
memcpy(domain, start, len);
domain[len] = '\0';
/* Strip leading www. (and similar) so www.example.com and example.com
* are treated as the same domain. Critical for sites with mixed links. */
if (strncasecmp(domain, "www.", 4) == 0) {
memmove(domain, domain + 4, strlen(domain) - 3);
}
return domain;
}
int
url_same_domain(const char *url1, const char *url2)
{
char *d1 = url_get_domain(url1);
char *d2 = url_get_domain(url2);
int same = strcasecmp(d1, d2) == 0;
free(d1);
free(d2);
return same;
}
char *
url_resolve(const char *base, const char *relative)
{
char *result;
/* Already absolute */
if (str_starts_with(relative, "http://") ||
str_starts_with(relative, "https://") ||
str_starts_with(relative, "data:")) {
return xstrdup(relative);
}
/* Protocol-relative */
if (str_starts_with(relative, "//")) {
size_t len = 6 + strlen(relative);
result = xmalloc(len + 1);
snprintf(result, len + 1, "https:%s", relative);
return result;
}
char *domain = url_get_domain(base);
const char *proto = str_starts_with(base, "https://") ? "https://" : "http://";
/* Root-relative */
if (relative[0] == '/') {
size_t len = strlen(proto) + strlen(domain) + strlen(relative);
result = xmalloc(len + 1);
snprintf(result, len + 1, "%s%s%s", proto, domain, relative);
free(domain);
return result;
}
/* Find base path */
const char *path_start;
if (str_starts_with(base, "https://"))
path_start = base + 8;
else if (str_starts_with(base, "http://"))
path_start = base + 7;
else
path_start = base;
/* Skip domain */
while (*path_start && *path_start != '/')
path_start++;
/* Find last slash in path */
const char *last_slash = strrchr(path_start, '/');
if (!last_slash)
last_slash = path_start;
size_t base_len = last_slash - path_start + 1;
size_t len = strlen(proto) + strlen(domain) + base_len + strlen(relative);
result = xmalloc(len + 1);
snprintf(result, len + 1, "%s%s%.*s%s", proto, domain, (int)base_len, path_start, relative);
free(domain);
return result;
}
char *
get_mime_type(const char *url)
{
/* Strip query string */
char *copy = xstrdup(url);
char *query = strchr(copy, '?');
if (query)
*query = '\0';
str_tolower(copy);
const char *mime = "application/octet-stream";
if (str_ends_with(copy, ".jpg") || str_ends_with(copy, ".jpeg"))
mime = "image/jpeg";
else if (str_ends_with(copy, ".png"))
mime = "image/png";
else if (str_ends_with(copy, ".gif"))
mime = "image/gif";
else if (str_ends_with(copy, ".webp"))
mime = "image/webp";
else if (str_ends_with(copy, ".svg"))
mime = "image/svg+xml";
else if (str_ends_with(copy, ".ico"))
mime = "image/x-icon";
else if (str_ends_with(copy, ".css"))
mime = "text/css";
else if (str_ends_with(copy, ".js"))
mime = "application/javascript";
else if (str_ends_with(copy, ".woff"))
mime = "font/woff";
else if (str_ends_with(copy, ".woff2"))
mime = "font/woff2";
else if (str_ends_with(copy, ".ttf"))
mime = "font/ttf";
else if (str_ends_with(copy, ".otf"))
mime = "font/otf";
else if (str_ends_with(copy, ".eot"))
mime = "application/vnd.ms-fontobject";
free(copy);
return xstrdup(mime);
}
char *
sanitize_filename(const char *url)
{
char *domain = url_get_domain(url);
size_t len = strlen(domain) + 32;
char *filename = xmalloc(len);
/* Replace dots with underscores */
for (char *p = domain; *p; p++)
if (*p == '.')
*p = '_';
snprintf(filename, len, "%s", domain);
free(domain);
return filename;
}
char *
get_iso_date(void)
{
time_t t = time(NULL);
struct tm *tm = gmtime(&t);
char *buf = xmalloc(32);
strftime(buf, 32, "%Y-%m-%dT%H:%M:%SZ", tm);
return buf;
}
/* make_relative_path: given two paths from site root (like url_to_path results),
* compute minimal relative reference from 'from' (current) to 'to' (target).
* Handles going up with ../ and down. Assumes no .. or . in inputs.
*/
char *
make_relative_path(const char *from, const char *to)
{
if (!from || !to)
return xstrdup(to ? to : "");
/* If identical (after norm), empty relative (or just filename if same) */
if (strcmp(from, to) == 0) {
const char *last = strrchr(to, '/');
return xstrdup(last ? last + 1 : to);
}
/* Work on dir of 'from' only (current document's directory) */
char *from_dir = xstrdup(from);
char *slash = strrchr(from_dir, '/');
if (slash)
*slash = '\0';
else
from_dir[0] = '\0';
/* Find common prefix length in from_dir and to */
const char *f = from_dir;
const char *t = to;
const char *last_match = NULL;
while (*f && *t && *f == *t) {
if (*f == '/')
last_match = f;
f++;
t++;
}
if (*f == '\0' && (*t == '\0' || *t == '/'))
last_match = f; /* full match or to under from_dir */
/* Count how many dirs up from from_dir's remaining */
int up = 0;
const char *f_rem = last_match ? last_match + 1 : from_dir;
if (*f_rem == '/')
f_rem++;
for (const char *p = f_rem; *p; p++)
if (*p == '/')
up++;
/* from_dir may have had trailing after last_match, count its / */
/* Build result: ups * "../" + suffix of to after last_match */
const char *to_rem = last_match ? last_match + 1 : to;
if (*to_rem == '/')
to_rem++;
size_t up_len = (size_t)up * 3;
size_t rem_len = strlen(to_rem);
char *result = xmalloc(up_len + rem_len + 1);
result[0] = '\0';
for (int i = 0; i < up; i++)
strcat(result, "../");
strcat(result, to_rem);
/* If result empty, use the basename of to */
if (result[0] == '\0') {
const char *b = strrchr(to, '/');
free(result);
result = xstrdup(b ? b + 1 : to);
}
free(from_dir);
return result;
}
/*
* url_path - suckless helper: return pointer to the path part of a URL.
* Used everywhere we used to manually skip protocol + domain.
* Examples:
* "https://example.com/foo/bar?x=1" -> "/foo/bar?x=1"
* "http://www.example.com" -> "/"
* "/already/relative" -> "/already/relative"
*/
const char *
url_path(const char *url)
{
const char *p;
if (!url)
return "/";
p = url;
if (str_starts_with(p, "https://"))
p += 8;
else if (str_starts_with(p, "http://"))
p += 7;
else if (str_starts_with(p, "//"))
p += 2;
/* Skip domain (until /, ?, #, or end) */
while (*p && *p != '/' && *p != '?' && *p != '#')
p++;
if (*p == '\0' || *p == '?' || *p == '#')
return "/";
return p; /* points at the leading '/' of the path */
}