Fix three network-facing overflows in the FTP and Java parsers

get_ftp_line() copied a server reply byte-by-byte into a fixed char[1024]
with no index bound, so a hostile or MITM FTP server could smash the stack
with an over-long CRLF-less line. Bound the write and truncate.

The ftp:// userinfo parser copied "user:pass@" into user[256]/pass[256] with
two unbounded loops, overflowing from a long userinfo supplied by a hostile
ftp:// link. Extract the split into ftp_split_userpass(), which truncates
each field to fit.

The Java .class parser did calloc(header.count, sizeof(RESP_STRUCT)) on an
attacker-controlled u2 count, allocating ~68 MB per crafted class (DoS). Cap
the count to the file size (each constant-pool entry is at least one byte on
disk) via a new hts_count_fits() guard, and move the alloc/free to the
bounds-checked calloct/freet wrappers.

Self-tests: ftp-line drives get_ftp_line over a socketpair with a 4 KB reply,
ftp-userpass feeds an over-long userinfo, java exercises the count cap. The
first two abort under ASan on the pre-fix code.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Xavier Roche <roche@httrack.com>

configure.ac

@@ -63,16 +63,6 @@ AC_SUBST(LT_CV_OBJDIR,$lt_cv_objdir)
 # Export version info
 AC_SUBST(VERSION_INFO)
 
-# Versioned plugin name for dlopen() in hts_create_opt(); soname major is
-# libtool's current - age, so this tracks VERSION_INFO bumps automatically.
-HTS_SONAME_MAJOR=$((${VERSION_INFO%%:*} - ${VERSION_INFO##*:}))
-case "$host_os" in
-darwin*) HTS_LIBHTSJAVA_NAME="libhtsjava.$HTS_SONAME_MAJOR.dylib" ;;
-*) HTS_LIBHTSJAVA_NAME="libhtsjava.so.$HTS_SONAME_MAJOR" ;;
-esac
-AC_DEFINE_UNQUOTED([HTS_LIBHTSJAVA_NAME], ["$HTS_LIBHTSJAVA_NAME"],
-  [Versioned libhtsjava runtime name, derived from VERSION_INFO])
-
 ### Default CFLAGS
 DEFAULT_CFLAGS="-Wall -Wformat -Wformat-security \
 -Wmultichar -Wwrite-strings -Wcast-qual -Wcast-align \

src/htscatchurl.c

@@ -175,9 +175,7 @@ HTSEXT_API hts_boolean catch_url(T_SOC soc, char *url, char *method,
       //
       socinput(soc, line, 1000);
       if (strnotempty(line)) {
-        /* widths bound the caller buffers: method[32], url[HTS_URLMAXSIZE*2],
-           protocol[256] */
-        if (sscanf(line, "%31s %2047s %255s", method, url, protocol) == 3) {
+        if (sscanf(line, "%s %s %s", method, url, protocol) == 3) {
           lien_adrfil af;
 
           // méthode en majuscule

src/htsdefines.h

@@ -69,15 +69,11 @@ typedef struct t_hts_callbackarg t_hts_callbackarg;
 typedef struct t_hts_callbackarg t_hts_callbackarg;
 #endif
 
-/* Marks a symbol an external wrapper module exports back to the engine.
-   Must override -fvisibility=hidden on ELF, or dlopen()ed plugins (htsjava)
-   hide their own hts_plug()/hts_unplug() entry points. */
+/* Marks a symbol an external wrapper module exports back to the engine
+   (dllexport on Windows, nothing elsewhere). */
 #ifndef EXTERNAL_FUNCTION
 #ifdef _WIN32
 #define EXTERNAL_FUNCTION __declspec(dllexport)
-#elif ((defined(__GNUC__) && (__GNUC__ >= 4)) ||                               \
-       (defined(HAVE_VISIBILITY) && HAVE_VISIBILITY))
-#define EXTERNAL_FUNCTION __attribute__((visibility("default")))
 #else
 #define EXTERNAL_FUNCTION
 #endif

src/htsencoding.c

@@ -190,9 +190,9 @@ int hts_unescapeEntitiesWithCharset(const char *src, char *dest, const size_t ma
         }
       }
     }
-
-    /* reserve one byte for the trailing NUL written after the loop */
-    if (j + 1 >= max) {
+    
+    /* copy */
+    if (j + 1 > max) {
       /* overflow */
       return -1;
     }
@@ -300,11 +300,6 @@ int hts_unescapeUrlSpecial(const char *src, char *dest, const size_t max,
 
           /* Was the character read successfully ? */
           if (nRead == utfBufferSize) {
-            /* the 'continue' below skips the NUL-reserve guard: re-check */
-            if (utfBufferJ + utfBufferSize >= max) {
-              return -1;
-            }
-
             /* Rollback write position to sequence start write position */
             j = utfBufferJ;
 
@@ -319,8 +314,8 @@ int hts_unescapeUrlSpecial(const char *src, char *dest, const size_t max,
       }
     }
 
-    /* reserve one byte for the trailing NUL written after the loop */
-    if (j + 1 >= max) {
+    /* Check for overflow */
+    if (j + 1 > max) {
       return -1;
     }
 

src/htsftp.c

@@ -133,8 +133,6 @@ void ftp_split_userpass(const char *src, const char *end, char *user,
                         size_t user_size, char *pass, size_t pass_size) {
   size_t n = 0;
 
-  assertf(user_size > 0 && pass_size > 0); /* the size-1 math underflows on 0 */
-
   while (src[n] != '\0' && src[n] != ':') {
     if (n < user_size - 1)
       user[n] = src[n];

src/htsftp.h

@@ -71,8 +71,7 @@ int run_launch_ftp(FTPDownloadStruct * params);
 int send_line(T_SOC soc, const char *data);
 int get_ftp_line(T_SOC soc, char *line, size_t line_size, int timeout);
 /* Split a "user[:pass]@" prefix (end = jump_identification result) into
-   bounded, NUL-terminated user/pass buffers, truncating to fit.
-   Both sizes must be nonzero. */
+   bounded, NUL-terminated user/pass buffers, truncating to fit. */
 void ftp_split_userpass(const char *src, const char *end, char *user,
                         size_t user_size, char *pass, size_t pass_size);
 T_SOC get_datasocket(char *to_send, size_t to_send_size);

src/htsjava.h

@@ -33,19 +33,15 @@ Please visit our Website: http://www.httrack.com
 #ifndef HTSJAVA_DEFH
 #define HTSJAVA_DEFH
 
-#include <stdint.h>
-
 #ifndef HTS_DEF_FWSTRUCT_JAVA_HEADER
 #define HTS_DEF_FWSTRUCT_JAVA_HEADER
 typedef struct JAVA_HEADER JAVA_HEADER;
 #endif
-/* 10-byte on-disk .class header image, fread() directly: fields need exact
-   widths (LP64's 8-byte 'unsigned long' magic never matched 0xCAFEBABE). */
 struct JAVA_HEADER {
-  uint32_t magic;
-  uint16_t minor;
-  uint16_t major;
-  uint16_t count;
+  unsigned long int magic;
+  unsigned short int minor;
+  unsigned short int major;
+  unsigned short int count;
 };
 
 #ifndef HTS_DEF_FWSTRUCT_RESP_STRUCT

src/htslib.c

@@ -1149,8 +1149,7 @@ int http_sendhead(httrackp * opt, t_cookie * cookie, int mode,
           char BIGSTK protocol[256], url[HTS_URLMAXSIZE * 2], method[256];
 
           linput(fp, line, 1000);
-          /* widths bound method[256], url[HTS_URLMAXSIZE*2], protocol[256] */
-          if (sscanf(line, "%255s %2047s %255s", method, url, protocol) == 3) {
+          if (sscanf(line, "%s %s %s", method, url, protocol) == 3) {
             size_t ret;
             // selon que l'on a ou pas un proxy
             if (retour->req.proxy.active) {
@@ -6023,11 +6022,9 @@ HTSEXT_API httrackp *hts_create_opt(void) {
     "htsswf", "htsjava", "httrack-plugin", NULL
   };
 #else
-#ifndef HTS_LIBHTSJAVA_NAME
-#define HTS_LIBHTSJAVA_NAME "libhtsjava.so" /* non-autoconf fallback */
-#endif
-  static const char *defaultModules[] = {"libhtsswf.so.1", HTS_LIBHTSJAVA_NAME,
-                                         "httrack-plugin", NULL};
+  static const char *defaultModules[] = {
+    "libhtsswf.so.1", "libhtsjava.so.2", "httrack-plugin", NULL
+  };
 #endif
   httrackp *opt = malloc(sizeof(httrackp));
 

src/htsparse.c

@@ -604,14 +604,13 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
                         }
 
                         // Decode title with encoding
-                        if (str->page_charset_ != NULL &&
-                            *str->page_charset_ != '\0') {
-                          char *sUtf = hts_convertStringToUTF8(
-                              s, strlen(s), str->page_charset_);
+                        if (str->page_charset_ != NULL 
+                            && *str->page_charset_ != '\0') {
+                          char *const sUtf = 
+                            hts_convertStringToUTF8(s, strlen(s), str->page_charset_);
                           if (sUtf != NULL) {
-                            /* UTF-8 can expand past s[]; truncate to fit */
-                            snprintf(s, sizeof(s), "%s", sUtf);
-                            freet(sUtf);
+                            strcpy(s, sUtf);
+                            free(sUtf);
                           }
                         }
 

src/htsselftest.c

@@ -713,8 +713,7 @@ static int st_entities(httrackp *opt, int argc, char **argv) {
   }
   s = strdupt(argv[0]);
   enc = argc >= 2 ? argv[1] : "UTF-8";
-  if (s != NULL &&
-      hts_unescapeEntitiesWithCharset(s, s, strlen(s) + 1, enc) == 0) {
+  if (s != NULL && hts_unescapeEntitiesWithCharset(s, s, strlen(s), enc) == 0) {
     printf("%s\n", s);
     freet(s);
   } else {
@@ -723,34 +722,6 @@ static int st_entities(httrackp *opt, int argc, char **argv) {
   return 0;
 }
 
-/* The unescapers must reserve one byte for the trailing NUL: a 'max'-byte
-   dest holding 'max' output chars pre-fix wrote dest[max] (1-byte OOB, caught
-   by ASan). Both unescapeEntities and unescapeUrl share the guard. */
-static int st_unescape_bounds(httrackp *opt, int argc, char **argv) {
-  char dest[4];
-
-  (void) opt;
-  (void) argc;
-  (void) argv;
-  assertf(hts_unescapeEntities("abcd", dest, sizeof(dest)) == -1);
-  assertf(hts_unescapeUrl("abcd", dest, sizeof(dest)) == -1);
-  assertf(hts_unescapeEntities("abc", dest, sizeof(dest)) == 0);
-  assertf(strcmp(dest, "abc") == 0);
-  /* raw multi-byte UTF-8 flush path (bypasses the per-byte guard) */
-  assertf(hts_unescapeUrl("ab\xC3\xA9", dest, sizeof(dest)) == -1);
-  assertf(hts_unescapeUrl("a\xC3\xA9", dest, sizeof(dest)) == 0);
-  assertf(strcmp(dest, "a\xC3\xA9") == 0);
-  {
-    /* %xx-encoded flush path (utfBufferJ = lastJ rollback) */
-    char wide[8];
-
-    assertf(hts_unescapeUrl("%C3%A9", wide, sizeof(wide)) == 0);
-    assertf(strcmp(wide, "\xC3\xA9") == 0);
-  }
-  printf("unescape-bounds self-test OK\n");
-  return 0;
-}
-
 static int st_hashtable(httrackp *opt, int argc, char **argv) {
   char *snum;
   unsigned long count = 0;
@@ -1853,17 +1824,6 @@ static int st_ftpuser(httrackp *opt, int argc, char **argv) {
   ftp_split_userpass(in, in + 802, user, sizeof(user), pass, sizeof(pass));
   assertf(strlen(user) == sizeof(user) - 1);
   assertf(strlen(pass) == sizeof(pass) - 1);
-  {
-    /* tight sizes + guard byte catch an off-by-one the 256 case can't */
-    char ubuf[16], pbuf[16];
-
-    memset(ubuf, 'Z', sizeof(ubuf));
-    memset(pbuf, 'Z', sizeof(pbuf));
-    ftp_split_userpass(in, in + 802, ubuf, 8, pbuf, 8);
-    assertf(strcmp(ubuf, "uuuuuuu") == 0);
-    assertf(strcmp(pbuf, "ppppppp") == 0);
-    assertf(ubuf[8] == 'Z' && pbuf[8] == 'Z');
-  }
   printf("ftp-userpass self-test OK\n");
   return 0;
 }
@@ -1913,8 +1873,6 @@ static const struct selftest_entry {
     {"idna-decode", "<host>", "decode an IDNA/punycode hostname",
      st_idna_decode},
     {"entities", "<string> [encoding]", "unescape HTML entities", st_entities},
-    {"unescape-bounds", "", "unescapers reserve the NUL byte (no 1-byte OOB)",
-     st_unescape_bounds},
     {"hashtable", "<count|file>", "coucal hashtable stress test", st_hashtable},
     {"strsafe", "[overflow|overflow-buff [str]]", "bounded string-op self-test",
      st_strsafe},

tests/01_engine-unescape-bounds.test

@@ -1,7 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# Entity/URL unescapers reserve one byte for the trailing NUL (no 1-byte OOB).
-httrack -O /dev/null -#test=unescape-bounds run | grep -q "unescape-bounds self-test OK"

tests/31_local-javaclass.test

@@ -1,23 +0,0 @@
-#!/bin/bash
-# The java plugin must load (versioned dlopen name) and parse a .class
-# constant pool: a resource named only inside Foo.class gets crawled.
-set -e
-
-: "${top_srcdir:=..}"
-
-tmproot=$(mktemp -d)
-trap 'rm -rf "$tmproot"' EXIT
-mkdir "$tmproot/javaclass"
-
-cat >"$tmproot/javaclass/index.html" <<'EOF'
-<html><body><a href="Foo.class">applet</a></body></html>
-EOF
-printf 'GIF89a' >"$tmproot/javaclass/hello.gif"
-# magic/minor/major, count=2, one CONSTANT_Utf8 "hello.gif", class/superclass
-printf '\xCA\xFE\xBA\xBE\x00\x00\x00\x32\x00\x02\x01\x00\x09hello.gif\x00\x00\x00\x00' \
-    >"$tmproot/javaclass/Foo.class"
-
-bash "$top_srcdir/tests/local-crawl.sh" --root "$tmproot" --errors 0 \
-    --found 'javaclass/Foo.class' \
-    --found 'javaclass/hello.gif' \
-    httrack 'BASEURL/javaclass/index.html'

tests/Makefile.am

@@ -57,7 +57,6 @@ TESTS = \
 	01_engine-stripquery.test \
 	01_engine-strsafe.test \
 	01_engine-urlhack.test \
-	01_engine-unescape-bounds.test \
 	01_engine-useragent.test \
 	01_zlib-acceptencoding.test \
 	01_zlib-cache.test \
@@ -90,7 +89,6 @@ TESTS = \
 	27_local-cookies-file.test \
 	28_local-pause.test \
 	29_local-redirect-fragment.test \
-	30_local-fragment-link.test \
-	31_local-javaclass.test
+	30_local-fragment-link.test
 
 CLEANFILES = check-network_sh.cache