Hermit/tests/test_webui.c

/* WebUI: the socket/HTTP layer, the SSE framing, auth and the HTML
 * sanitizer. The full turn path is covered by scripts/e2e_webui.sh, which
 * runs a recorded stream through a real server. */
#ifndef _DEFAULT_SOURCE
#define _DEFAULT_SOURCE
#endif

#include "cst_http.h"
#include "cst_net.h"
#include "cst_util.h"
#include "cst_web_md.h"

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#include "cst_test_util.h"

#define CHECK(c)                                                               \
  do {                                                                         \
    if (!(c)) {                                                                \
      fprintf(stderr, "%s:%d: CHECK(%s) failed\n", __FILE__, __LINE__, #c);   \
      exit(1);                                                                 \
    }                                                                          \
  } while (0)

/* --- a tiny HTTP client --------------------------------------------------- */

static os_socket_t connect_to(int port) {
  char host[32];
  snprintf(host, sizeof(host), "127.0.0.1");
  os_socket_t s = os_net_connect(host, port, 3000);
  CHECK(s != OS_SOCKET_BAD);
  return s;
}

/* Sends data, reads until the peer closes (or 5 s pass). malloc'd. */
static char *roundtrip(int port, const char *req, size_t *out_len) {
  os_socket_t s = connect_to(port);
  CHECK(os_net_send_all(s, req, strlen(req), 3000) == 0);
  cst_buf_t b = {0};
  char buf[4096];
  for (;;) {
    long n = os_net_recv(s, buf, sizeof(buf), 5000);
    if (n <= 0)
      break;
    CHECK(cst_buf_append(&b, buf, (size_t)n) == 0);
  }
  os_net_close(s);
  *out_len = b.len;
  return cst_buf_steal(&b);
}

/* Same, but stops as soon as `needle` shows up. */

/* --- handlers ------------------------------------------------------------- */

static int h_hello(cst_http_conn_t *c, void *ud) {
  (void)ud;
  return cst_http_send_str(c, 200, "text/plain", "world");
}

static int h_secret(cst_http_conn_t *c, void *ud) {
  (void)ud;
  return cst_http_send_str(c, 200, "text/plain", "secret");
}

static int h_echo(cst_http_conn_t *c, void *ud) {
  (void)ud;
  const cst_http_req_t *r = cst_http_request(c);
  char head[64];
  snprintf(head, sizeof(head), "%zu:", r->body_len);
  cst_buf_t b = {0};
  cst_buf_puts(&b, head);
  cst_buf_append(&b, r->body ? r->body : "", r->body_len);
  int rc = cst_http_send(c, 200, "text/plain", b.buf, b.len);
  cst_buf_free(&b);
  return rc;
}

static int h_events(cst_http_conn_t *c, void *ud) {
  (void)ud;
  CHECK(cst_http_stream_begin(c, "text/event-stream") == 0);
  CHECK(cst_http_stream_write(c, "event: one\ndata: {\"n\":1}\n\n", 26) == 0);
  CHECK(cst_http_stream_write(c, "event: two\ndata: {\"n\":2}\n\n", 26) == 0);
  CHECK(cst_http_stream_end(c) == 0);
  return 0;
}

static int auth_bearer(const cst_http_req_t *req, void *ud) {
  const char *want = ud;
  if (!req->authorization)
    return -1;
  if (strncmp(req->authorization, "Bearer ", 7) != 0)
    return -1;
  return strcmp(req->authorization + 7, want) == 0 ? 0 : -1;
}

static void test_http(void) {
  static const cst_http_route_t routes[] = {
      {"GET", "/hello", h_hello, NULL, 0, 0},
      {"GET", "/secret", h_secret, NULL, 1, 0},
      {"GET", "/events", h_events, NULL, 0, 0},
      {"POST", "/echo", h_echo, NULL, 0, 4096},
  };
  cst_http_opts_t o;
  memset(&o, 0, sizeof(o));
  o.host = "127.0.0.1";
  o.port = 0;
  o.max_body = 1024 * 1024;
  o.routes = routes;
  o.nroutes = (int)(sizeof(routes) / sizeof(routes[0]));
  o.auth = auth_bearer;
  o.auth_ud = (void *)"s3cret";
  char err[256] = "";
  cst_http_server_t *s = cst_http_start(&o, err, sizeof(err));
  CHECK(s != NULL);
  int port = cst_http_port(s);
  CHECK(port > 0);

  size_t n = 0;
  char *resp = roundtrip(port, "GET /hello HTTP/1.1\r\nHost: x\r\n"
                               "Connection: close\r\n\r\n", &n);
  CHECK(strstr(resp, "200 OK") && strstr(resp, "world"));
  free(resp);

  /* keep-alive: two requests, one connection */
  resp = roundtrip(port, "GET /hello HTTP/1.1\r\nHost: x\r\n\r\n"
                         "GET /hello HTTP/1.1\r\nHost: x\r\n"
                         "Connection: close\r\n\r\n", &n);
  CHECK(strstr(resp, "200 OK"));
  CHECK(strstr(resp, "200 OK") != strrchr(resp, '2')); /* twice */
  {
    int hits = 0;
    for (const char *p = resp; (p = strstr(p, "200 OK")); p++)
      hits++;
    CHECK(hits == 2);
  }
  free(resp);

  /* auth: without, then with the token */
  resp = roundtrip(port, "GET /secret HTTP/1.1\r\nHost: x\r\n"
                         "Connection: close\r\n\r\n", &n);
  CHECK(strstr(resp, "401"));
  free(resp);
  resp = roundtrip(port, "GET /secret HTTP/1.1\r\nHost: x\r\n"
                         "Authorization: Bearer s3cret\r\n"
                         "Connection: close\r\n\r\n", &n);
  CHECK(strstr(resp, "200") && strstr(resp, "secret"));
  free(resp);

  /* unknown route */
  resp = roundtrip(port, "GET /nope HTTP/1.1\r\nHost: x\r\n"
                         "Connection: close\r\n\r\n", &n);
  CHECK(strstr(resp, "404"));
  free(resp);
  /* traversal is rejected before any handler runs */
  resp = roundtrip(port, "GET /../etc/passwd HTTP/1.1\r\nHost: x\r\n"
                         "Connection: close\r\n\r\n", &n);
  CHECK(strstr(resp, "400") || strstr(resp, "404"));
  free(resp);

  /* a body round trip, and the per-route cap */
  resp = roundtrip(port, "POST /echo HTTP/1.1\r\nHost: x\r\n"
                         "Content-Length: 3\r\nConnection: close\r\n\r\nabc", &n);
  CHECK(strstr(resp, "200") && strstr(resp, "3:abc"));
  free(resp);
  {
    cst_buf_t b = {0};
    cst_buf_puts(&b, "POST /echo HTTP/1.1\r\nHost: x\r\nContent-Length: 9999\r\n"
                     "Connection: close\r\n\r\n");
    for (int i = 0; i < 9999; i++)
      cst_buf_puts(&b, "a");
    resp = roundtrip(port, b.buf, &n);
    CHECK(strstr(resp, "413"));
    free(resp);
    cst_buf_free(&b);
  }

  /* chunked streaming: the client must see the decoded frames */
  resp = roundtrip(port, "GET /events HTTP/1.1\r\nHost: x\r\n"
                         "Connection: close\r\n\r\n", &n);
  CHECK(strstr(resp, "Transfer-Encoding: chunked"));
  CHECK(strstr(resp, "event: one") && strstr(resp, "event: two"));
  CHECK(strstr(resp, "\r\n0\r\n\r\n")); /* terminator */
  free(resp);

  cst_http_stop(s);
  cst_http_free(s);
}

/* --- query parsing -------------------------------------------------------- */

static void test_query(void) {
  cst_http_req_t r;
  memset(&r, 0, sizeof(r));
  snprintf(r.query, sizeof(r.query), "token=a%%20b&since=17&x=");
  char buf[64];
  CHECK(cst_http_query(&r, "token", buf, sizeof(buf)) == 0);
  CHECK(strcmp(buf, "a b") == 0);
  CHECK(cst_http_query(&r, "since", buf, sizeof(buf)) == 0);
  CHECK(strcmp(buf, "17") == 0);
  CHECK(cst_http_query(&r, "nope", buf, sizeof(buf)) != 0);
}

/* --- markdown + sanitizer ------------------------------------------------- */

static cst_buf_t g_md_out;

static int collect_md(const char *html, size_t len, void *ud) {
  (void)ud;
  return cst_buf_append(&g_md_out, html, len) == 0 ? 0 : 1;
}

static void test_sanitize(void) {
  char *html = cst_web_md_html("**hi** [x](javascript:alert(1)) "
                               "![i](/api/attach/img-1.png)",
                               strlen("**hi** [x](javascript:alert(1)) "
                                      "![i](/api/attach/img-1.png)"));
  CHECK(html != NULL);
  CHECK(strstr(html, "<strong>hi</strong>"));
  CHECK(strstr(html, "javascript:") == NULL);
  CHECK(strstr(html, "/api/attach/img-1.png"));
  free(html);

  /* raw markup that sashidown would pass through gets stripped */
  const char *evil = "<p>ok</p><script>alert(1)</script>"
                     "<img src=\"http://evil.example/x.png\" alt=\"a\">"
                     "<a href=\"//evil.example\">l</a>";
  html = cst_web_md_sanitize(evil, strlen(evil));
  CHECK(html != NULL);
  CHECK(strstr(html, "ok"));
  CHECK(strstr(html, "<script") == NULL);
  CHECK(strstr(html, "evil.example/x.png") == NULL);
  CHECK(strstr(html, "//evil.example") == NULL);
  CHECK(strstr(html, "<a") == NULL); /* an unusable href drops the tag */
  free(html);

  /* a quote inside a single-quoted or bare value cannot open a new attribute */
  const char *quoted = "<a href='http://x\" onmouseover=\"alert(1)'>q</a>"
                       "<a href=http://y\"onmouseover=alert(2)//>r</a>";
  html = cst_web_md_sanitize(quoted, strlen(quoted));
  CHECK(html != NULL);
  CHECK(strstr(html, "\" onmouseover") == NULL);
  CHECK(strstr(html, "\"onmouseover") == NULL);
  CHECK(strstr(html, "&quot;") != NULL);
  free(html);

  /* streaming: a javascript: link is dropped, a split tag never leaks */
  {
    cst_buf_free(&g_md_out);
    const char *src = "see [x](javascript:alert(1)) and **ok**";
    cst_web_md_t *md = cst_web_md_new(collect_md, NULL);
    CHECK(md != NULL);
    CHECK(cst_web_md_feed(md, src, 4) == 0); /* a deliberately split tag */
    CHECK(cst_web_md_feed(md, src + 4, strlen(src) - 4) == 0);
    CHECK(cst_web_md_finish(md) == 0);
    cst_web_md_free(md);
    const char *out = g_md_out.buf ? g_md_out.buf : "";
    CHECK(strstr(out, "javascript:") == NULL);
    CHECK(strstr(out, "href") == NULL);
    CHECK(strstr(out, "see"));
    CHECK(strstr(out, "<strong>ok</strong>") != NULL);
    cst_buf_free(&g_md_out);
  }
  {
    /* and the same text fed one byte at a time still comes out whole */
    cst_buf_free(&g_md_out);
    const char *md_src = "**bold** and [ok](https://example.com)";
    cst_web_md_t *md = cst_web_md_new(collect_md, NULL);
    CHECK(md != NULL);
    for (size_t i = 0; i < strlen(md_src); i++)
      CHECK(cst_web_md_feed(md, md_src + i, 1) == 0);
    CHECK(cst_web_md_finish(md) == 0);
    cst_web_md_free(md);
    CHECK(strstr(g_md_out.buf, "<strong>bold</strong>"));
    CHECK(strstr(g_md_out.buf, "https://example.com"));
    CHECK(strstr(g_md_out.buf, "noopener") != NULL);
    cst_buf_free(&g_md_out);
  }
}

/* Streamed deltas cut anywhere must reach the browser as whole characters. */
static void test_utf8_hold(void) {
  const char *src = "思考：先读 main.c，再改 😀 emoji。";
  size_t len = strlen(src);
  for (size_t step = 1; step <= 5; step++) {
    cst_utf8_hold_t h = {0};
    cst_buf_t out = {0};
    for (size_t i = 0; i < len; i += step) {
      size_t n = len - i < step ? len - i : step;
      size_t got = 0;
      char *piece = cst_utf8_hold_feed(&h, src + i, n, &got);
      CHECK(piece != NULL);
      CHECK(h.n < 4);
      CHECK(cst_utf8_valid(piece, got));
      CHECK(cst_buf_append(&out, piece, got) == 0);
      free(piece);
    }
    CHECK(h.n == 0);
    CHECK(out.len == len && memcmp(out.buf, src, len) == 0);
    cst_buf_free(&out);
  }
  /* a stray continuation byte is not held forever */
  cst_utf8_hold_t h = {0};
  size_t got = 0;
  char *piece = cst_utf8_hold_feed(&h, "\x80\x80\x80\x80", 4, &got);
  CHECK(piece && got == 4 && h.n == 0);
  free(piece);
}

int main(void) {
  CHECK(os_net_init() == 0);
  test_utf8_hold();
  test_query();
  test_sanitize();
  test_http();
  printf("test_webui: OK\n");
  return 0;
}