Fix remote stack overflow and uninit read in Content-Type/-Encoding parsing

treathead() parsed the Content-Type and Content-Encoding response-header
values with a width-less sscanf("%s") into a 1100-byte stack scratch,
bounds-checking the result only after the copy. The receive path in
back_wait hands treathead a line buffer of up to 2000 bytes (htsback.c
rcvd[2048], binput cap 2000), so a hostile server sending a header value
longer than 1099 non-whitespace bytes overflowed the stack. Bound both
scans to the buffer (%1099s), matching the existing %31s/%255s idiom in
the request-line parsers.

The Content-Encoding branch also ignored the sscanf return, so an empty
value (e.g. "Content-Encoding:") left tempo uninitialized and then ran
strlen() over it: an uninitialized read that can also over-read past the
buffer. The Content-Type branch already guarded this; mirror it.

Found via the P3-3 static-analysis baseline. New -#test=headerlong plus
empty-value cases in 01_engine-header.test build the hostile inputs in
the handler (the CLI caps argument length); they abort on the pre-fix
binary under ASan (overflow) and MSan (uninitialized read).

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

.gitignore

@@ -34,9 +34,8 @@ Makefile
 *.so.*
 *.a
 
-# make dist output; dist/ holds httrack-gh-release staging artifacts.
+# make dist output.
 /httrack-*.tar.gz
-/dist/
 
 # Editor / autotools backup files.
 *~

fuzz/Makefile.am

@@ -1,7 +1,7 @@
 # libFuzzer harnesses; built only with --enable-fuzzers (requires clang).
 if FUZZERS
 noinst_PROGRAMS = fuzz-charset fuzz-meta fuzz-idna fuzz-entities \
-	fuzz-unescape fuzz-filters fuzz-url fuzz-header fuzz-cachendx
+	fuzz-unescape fuzz-filters fuzz-url
 endif
 
 AM_CPPFLAGS = \
@@ -23,8 +23,6 @@ fuzz_entities_SOURCES = fuzz-entities.c fuzz.h
 fuzz_unescape_SOURCES = fuzz-unescape.c fuzz.h
 fuzz_filters_SOURCES = fuzz-filters.c fuzz.h
 fuzz_url_SOURCES = fuzz-url.c fuzz.h
-fuzz_header_SOURCES = fuzz-header.c fuzz.h
-fuzz_cachendx_SOURCES = fuzz-cachendx.c fuzz.h
 
 # List corpus files explicitly: automake does not expand EXTRA_DIST globs.
 EXTRA_DIST = README.md run-fuzzers.sh \
@@ -37,8 +35,4 @@ EXTRA_DIST = README.md run-fuzzers.sh \
 	corpus/filters/filter.bin corpus/filters/filter-size.bin \
 	corpus/filters/regress-empty-subject-unique.bin \
 	corpus/url/http-url.txt corpus/url/relative-path.txt \
-	corpus/url/regress-file-empty-path.txt corpus/url/regress-long-path-abort.txt \
-	corpus/header/full-response.txt corpus/header/redirect.txt \
-	corpus/cachendx/new-format.txt corpus/cachendx/old-format.txt \
-	corpus/cachendx/regress-overadvance.bin \
-	corpus/cachendx/regress-truncated-entry.bin
+	corpus/url/regress-file-empty-path.txt corpus/url/regress-long-path-abort.txt

fuzz/corpus/cachendx/new-format.txt

@@ -1,7 +0,0 @@
-8
-CACHE-1.1
-28
-Mon, 01 Jan 2024 00:00:00 GMT
-www.example.com
-/index.html
-123

fuzz/corpus/cachendx/old-format.txt

@@ -1,5 +0,0 @@
-3
-1.0
-www.example.com
-/page
-5

fuzz/corpus/cachendx/regress-overadvance.bin

@@ -1,2 +0,0 @@
-32768
-CACHE-1.1

fuzz/corpus/cachendx/regress-truncated-entry.bin

@@ -1,5 +0,0 @@
-8
-CACHE-1.1
-1
-x
-www.example.com

fuzz/corpus/header/full-response.txt

@@ -1,10 +0,0 @@
-HTTP/1.1 200 OK
-Content-Type: text/html; charset=utf-8
-Content-Length: 1234
-Content-Encoding: gzip
-Last-Modified: Mon, 01 Jan 2024 00:00:00 GMT
-Etag: "abc"
-Location: http://example.com/x
-Set-Cookie: ID=42; path=/; domain=.example.com
-Content-Range: bytes 0-99/100
-Transfer-Encoding: chunked

fuzz/corpus/header/redirect.txt

@@ -1,3 +0,0 @@
-HTTP/1.0 301 Moved
-Location: /elsewhere
-Content-Disposition: attachment; filename="a.pdf"

fuzz/fuzz-cachendx.c

@@ -1,66 +0,0 @@
-/* ------------------------------------------------------------ */
-/*
-HTTrack Website Copier, Offline Browser for Windows and Unix
-Copyright (C) 2026 Xavier Roche and other contributors
-
-SPDX-License-Identifier: GPL-3.0-or-later
-
-This program is free software: you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation, either version 3 of the License, or
-(at your option) any later version.
-
-This program is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with this program. If not, see <http://www.gnu.org/licenses/>.
-
-Ethical use: we kindly ask that you NOT use this software to harvest email
-addresses or to collect any other private information about people. Doing so
-would dishonor our work and waste the many hours we have spent on it.
-
-Please visit our Website: http://www.httrack.com
-*/
-
-/* Fuzz the cache-index (.ndx) parser: a corrupt or truncated index must not
-   walk the length-prefixed cache_brstr/cache_binput scan past the buffer.
-   Mirrors the loader in cache_readex_new (htscache.c). */
-#include "fuzz.h"
-#include "htscache.h"
-#include "htslib.h"
-
-int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) {
-  char *buf = fuzz_strdup(data, size);
-  const char *const end = buf + size;
-  char firstline[256];
-  char *a = buf;
-
-  /* header: two length-prefixed fields (version, last-modified) */
-  a += cache_brstr(a, firstline, sizeof(firstline));
-  a += cache_brstr(a, firstline, sizeof(firstline));
-
-  /* body: newline-delimited host/file/position triples. The coucal insert
-     the real loader ends with is a separate library's concern; this harness
-     targets the length-prefixed scan that must stay inside the buffer. */
-  while (a != NULL && a < end) {
-    char BIGSTK line[HTS_URLMAXSIZE * 2];
-    char linepos[256];
-    int pos;
-
-    a = strchr(a + 1, '\n');
-    if (a == NULL)
-      break;
-    a++;
-    a += cache_binput(a, end, line, HTS_URLMAXSIZE);
-    a += cache_binput(a, end, line + strlen(line), HTS_URLMAXSIZE);
-    a += cache_binput(a, end, linepos, 200);
-    sscanf(linepos, "%d", &pos);
-    (void) pos;
-  }
-
-  freet(buf);
-  return 0;
-}

fuzz/fuzz-header.c

@@ -1,70 +0,0 @@
-/* ------------------------------------------------------------ */
-/*
-HTTrack Website Copier, Offline Browser for Windows and Unix
-Copyright (C) 2026 Xavier Roche and other contributors
-
-SPDX-License-Identifier: GPL-3.0-or-later
-
-This program is free software: you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation, either version 3 of the License, or
-(at your option) any later version.
-
-This program is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with this program. If not, see <http://www.gnu.org/licenses/>.
-
-Ethical use: we kindly ask that you NOT use this software to harvest email
-addresses or to collect any other private information about people. Doing so
-would dishonor our work and waste the many hours we have spent on it.
-
-Please visit our Website: http://www.httrack.com
-*/
-
-/* Fuzz the HTTP response-header parser (htslib.c): treatfirstline on the
-   status line, treathead on each following header. Both consume raw bytes
-   off the wire and copy fields into fixed htsblk buffers; treathead also
-   mutates its line in place and drives cookie parsing. */
-#include "fuzz.h"
-#include "htslib.h"
-#include "htsbauth.h"
-
-int LLVMFuzzerTestOneInput(const uint8_t *data, size_t size) {
-  char *buf = fuzz_strdup(data, size);
-  htsblk r;
-  t_cookie *cookie = calloct(1, sizeof(*cookie));
-  char *line = malloct(size + 1);
-  char *p = buf;
-  int first = 1;
-
-  memset(&r, 0, sizeof(r));
-  cookie->max_len = (int) sizeof(cookie->data);
-  r.location = malloct(HTS_URLMAXSIZE * 2);
-  r.location[0] = '\0';
-
-  /* feed one header line at a time, as the receive loop does */
-  while (p != NULL && *p != '\0') {
-    char *nl = strchr(p, '\n');
-    size_t n = (nl != NULL) ? (size_t) (nl - p) : strlen(p);
-
-    memcpy(line, p, n);
-    line[n] = '\0';
-    if (first) {
-      treatfirstline(&r, line);
-      first = 0;
-    } else {
-      treathead(cookie, "www.example.com", "/", &r, line);
-    }
-    p = (nl != NULL) ? nl + 1 : NULL;
-  }
-
-  freet(r.location);
-  freet(line);
-  freet(cookie);
-  freet(buf);
-  return 0;
-}

src/htscache.c

@@ -1938,17 +1938,15 @@ void cache_init(cache_back * cache, httrackp * opt) {
             char linepos[256];
             int pos;
 
-            const char *const end = cache->use + buffl;
-
-            while ((a != NULL) && (a < end)) {
+            while((a != NULL) && (a < (cache->use + buffl))) {
               a = strchr(a + 1, '\n');  /* start of line */
               if (a) {
                 a++;
                 /* read "host/file" */
-                a += cache_binput(a, end, line, HTS_URLMAXSIZE);
-                a += cache_binput(a, end, line + strlen(line), HTS_URLMAXSIZE);
+                a += binput(a, line, HTS_URLMAXSIZE);
+                a += binput(a, line + strlen(line), HTS_URLMAXSIZE);
                 /* read position */
-                a += cache_binput(a, end, linepos, 200);
+                a += binput(a, linepos, 200);
                 sscanf(linepos, "%d", &pos);
                 coucal_add(cache->hashtable, line, pos);
               }
@@ -2293,40 +2291,23 @@ int cache_brstr(char *adr, char *s, size_t s_size) {
   char buff[256 + 4];
 
   off = binput(adr, buff, 256);
-  /* binput stops at the buffer's terminating NUL; a value can only follow a
-     real line terminator, so never step past end-of-buffer. */
-  if (adr[off - 1] == '\0') {
-    s[0] = '\0';
-    return off - 1;
-  }
-  /* an empty/non-numeric field leaves i unset: treat as length 0 */
-  if (sscanf(buff, "%d", &i) != 1 || i < 0 || i > 32768)
+  adr += off;
+  sscanf(buff, "%d", &i);
+  if (i < 0 || i > 32768) /* guard a corrupt length */
     i = 0;
   if (i > 0) {
-    /* A corrupt/truncated cache may declare a length past the buffer end;
-       bound both the copy and the advance to the bytes actually present. */
-    const size_t avail = strnlen(adr + off, (size_t) i);
-    const size_t store = avail < s_size ? avail : s_size - 1;
+    /* copy at most s_size-1 bytes; advance past the full field regardless */
+    const size_t store = (size_t) i < s_size ? (size_t) i : s_size - 1;
 
-    memcpy(s, adr + off, store);
+    strncpy(s, adr, store);
     s[store] = '\0';
-    off += (int) avail;
   } else {
     s[0] = '\0';
   }
+  off += i;
   return off;
 }
 
-/* binput bounded to a NUL-terminated buffer: refuse to start a read at or
-   past `end`, so a prior over-advance can't walk a cache-index parse OOB. */
-int cache_binput(char *adr, const char *end, char *s, int max) {
-  if (adr >= end) {
-    s[0] = '\0';
-    return 0;
-  }
-  return binput(adr, s, max);
-}
-
 /* idem, mais en int */
 int cache_brint(char *adr, int *i) {
   char s[256];

src/htscache.h

@@ -97,8 +97,6 @@ int cache_readdata(cache_back * cache, const char *str1, const char *str2,
 void cache_rstr(FILE *fp, char *s, size_t s_size);
 char *cache_rstr_addr(FILE * fp);
 int cache_brstr(char *adr, char *s, size_t s_size);
-/* binput over a NUL-terminated buffer, bounded: no read starts at/past end. */
-int cache_binput(char *adr, const char *end, char *s, int max);
 int cache_brint(char *adr, int *i);
 void cache_rint(FILE * fp, int *i);
 void cache_rLLint(FILE * fp, LLint * i);

src/htscoremain.c

@@ -1953,20 +1953,18 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
                     char BIGSTK url[HTS_URLMAXSIZE * 2];
                     char linepos[256];
                     int pos;
-                    LLint cacheNdxLen = 0;
-                    char *cacheNdx = readfile2(
-                        fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                                StringBuff(opt->path_log), "hts-cache/new.ndx"),
-                        &cacheNdxLen);
+                    char *cacheNdx =
+                      readfile(fconcat
+                               (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
+                                "hts-cache/new.ndx"));
                     cache_init(&cache, opt);    /* load cache */
                     if (cacheNdx != NULL) {
                       char firstline[256];
                       char *a = cacheNdx;
-                      const char *const end = cacheNdx + cacheNdxLen;
 
                       a += cache_brstr(a, firstline, sizeof(firstline));
                       a += cache_brstr(a, firstline, sizeof(firstline));
-                      while (a != NULL && a < end) {
+                      while(a != NULL) {
                         a = strchr(a + 1, '\n');        /* start of line */
                         if (a) {
                           htsblk r;
@@ -1974,15 +1972,15 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
                           /* */
                           a++;
                           /* read "host/file" */
-                          a += cache_binput(a, end, afs.af.adr, HTS_URLMAXSIZE);
-                          a += cache_binput(a, end, afs.af.fil, HTS_URLMAXSIZE);
+                          a += binput(a, afs.af.adr, HTS_URLMAXSIZE);
+                          a += binput(a, afs.af.fil, HTS_URLMAXSIZE);
                           url[0] = '\0';
                           if (!link_has_authority(afs.af.adr))
                             strcatbuff(url, "http://");
                           strcatbuff(url, afs.af.adr);
                           strcatbuff(url, afs.af.fil);
                           /* read position */
-                          a += cache_binput(a, end, linepos, 200);
+                          a += binput(a, linepos, 200);
                           sscanf(linepos, "%d", &pos);
                           if (!hasFilter
                               || (strjoker(url, filter, NULL, NULL) != NULL)

src/htsselftest.c

@@ -46,7 +46,6 @@ Please visit our Website: http://www.httrack.com
 #include "htsdefines.h"
 #include "htslib.h"
 #include "htsparse.h"
-#include "htscache.h"
 #include "htscache_selftest.h"
 #include "htsdns_selftest.h"
 #include "htscharset.h"
@@ -1391,78 +1390,6 @@ static int st_cache(httrackp *opt, int argc, char **argv) {
   return err;
 }
 
-/* A corrupt cache index (.ndx) must not walk the length-prefixed scan past
-   the buffer. Checks the two primitives the loader is built from. */
-static int st_cacheindex(httrackp *opt, int argc, char **argv) {
-  int fail = 0;
-
-  (void) opt;
-  (void) argc;
-  (void) argv;
-
-  /* A length prefix that overstates the bytes present must bound the advance
-     to the buffer, not trust the declared length. */
-  {
-    static const char src[] = "32768\nCACHE-1.1";
-    const size_t len = sizeof(src) - 1;
-    char *buf = malloct(len + 1);
-    char s[256];
-    int off;
-
-    memcpy(buf, src, len + 1);
-    off = cache_brstr(buf, s, sizeof(s));
-    if (off > (int) len) {
-      printf("cacheindex: over-advance off=%d len=%d\n", off, (int) len);
-      fail = 1;
-    }
-    if (strcmp(s, "CACHE-1.1") != 0) {
-      printf("cacheindex: value=%s\n", s);
-      fail = 1;
-    }
-    freet(buf);
-  }
-
-  /* cache_binput reads a field while in bounds, but refuses one starting at
-     or past end-of-buffer. */
-  {
-    char buf[8] = "ab\ncd";
-    const char *const end = buf + 5;
-    char s[16];
-
-    if (cache_binput(buf, end, s, sizeof(s)) != 3 || strcmp(s, "ab") != 0)
-      fail = 1; /* normal read: "ab" then the '\n', 3 bytes consumed */
-    if (cache_binput(end, end, s, sizeof(s)) != 0 || s[0] != '\0')
-      fail = 1;
-  }
-
-  /* Drive the full loader scan over a truncated index: a declared length that
-     overshoots plus a half-written entry. ASan aborts here on the pre-fix
-     scan; the cursor must never leave the buffer. */
-  {
-    static const char src[] = "9\nCACHE-1.1\n99\nwww.example.com\n/a";
-    const size_t len = sizeof(src) - 1;
-    char *buf = malloct(len + 1);
-    const char *const end = buf + len;
-    char line[256];
-    char *a = buf;
-
-    memcpy(buf, src, len + 1);
-    a += cache_brstr(a, line, sizeof(line));
-    a += cache_brstr(a, line, sizeof(line));
-    while (a != NULL && a < end) {
-      a = strchr(a + 1, '\n');
-      if (a == NULL)
-        break;
-      a++;
-      a += cache_binput(a, end, line, sizeof(line));
-    }
-    freet(buf);
-  }
-
-  printf("cacheindex: %s\n", fail ? "FAIL" : "OK");
-  return fail;
-}
-
 static int st_cache_golden(httrackp *opt, int argc, char **argv) {
   int regen, err;
 
@@ -2286,8 +2213,6 @@ static const struct selftest_entry {
     {"sniff", "<content-type> <hex:..|text>", "MIME magic consistency",
      st_sniff},
     {"cache", "<dir>", "cache read/write round-trip self-test", st_cache},
-    {"cacheindex", "", "cache-index (.ndx) parse must stay in bounds",
-     st_cacheindex},
     {"cache-golden", "<dir> [regen]", "frozen cache-format read self-test",
      st_cache_golden},
     {"cache-writefail", "<dir>", "cache write-failure handling self-test",

tests/01_engine-cacheindex.test

@@ -1,10 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# Cache-index (.ndx) parse must stay inside the buffer on a corrupt length
-# prefix (ASan aborts here on the pre-fix binary; the OK check gates the
-# non-sanitized build too).
-
-test "$(httrack -O /dev/null -#test=cacheindex)" == 'cacheindex: OK'

tests/Makefile.am

@@ -38,7 +38,6 @@ TESTS = \
 	01_engine-filterdual.test \
 	01_engine-ftp-line.test \
 	01_engine-ftp-userpass.test \
-	01_engine-cacheindex.test \
 	01_engine-hashtable.test \
 	01_engine-header.test \
 	01_engine-idna.test \