Bound test 19's dead-address connects so it is fast on macOS

19_local-connect-fallback pins dead loopback IPs (127.0.0.2/.3) before the live
one to exercise address fallback. On Linux those refuse instantly (all of
127.0.0.0/8 is loopback); macOS has only 127.0.0.1, so the connects block to
--timeout, and default retries replay the whole crawl, so the exhaustion case
took ~2 minutes. It passed but dominated the macOS suite once the rest runs in
parallel. Drop --timeout to 5s (bounds the last candidate's stall; the earlier
ones already fall back at min(timeout,10)) and set --retries=0 (no ×3 replay).
Linux is unchanged: the refused connects never reach the timeout.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Signed-off-by: Xavier Roche <roche@httrack.com>

src/htsback.c

@@ -3540,22 +3540,20 @@ void back_wait(struct_back * sback, httrackp * opt, cache_back * cache,
                   if (back[i].r.statuscode == 406) {    // 'Not Acceptable'
                     back[i].r.statuscode = HTTP_OK;
                   }
-                  // 'do not erase already downloaded file'
-                  // on an updated file
-                  // with an error : consider a 304 error
-                  if (!opt->delete_old) {
-                    if (HTTP_IS_ERROR(back[i].r.statuscode) && back[i].is_update
-                        && !back[i].testmode) {
-                      if (back[i].url_sav[0] && fexist_utf8(back[i].url_sav)) {
-                        hts_log_print(opt, LOG_DEBUG,
-                                      "Error ignored %d (%s) because of 'no purge' option for %s%s",
-                                      back[i].r.statuscode, back[i].r.msg,
-                                      back[i].url_adr, back[i].url_fil);
-                        back[i].r.statuscode = HTTP_NOT_MODIFIED;
-                        deletehttp(&back[i].r);
-                        back[i].r.soc = INVALID_SOCKET;
-                      }
-                    }
+                  // On update, keep the good copy on error (mask as 304); skip
+                  // resume paths so a stale-partial 416 still re-fetches.
+                  if (HTTP_IS_ERROR(back[i].r.statuscode) &&
+                      back[i].is_update && !back[i].testmode &&
+                      back[i].range_req_size == 0 && back[i].url_sav[0] &&
+                      fexist_utf8(back[i].url_sav)) {
+                    hts_log_print(opt, LOG_DEBUG,
+                                  "Error %d (%s) ignored on update, keeping "
+                                  "existing copy: %s%s",
+                                  back[i].r.statuscode, back[i].r.msg,
+                                  back[i].url_adr, back[i].url_fil);
+                    back[i].r.statuscode = HTTP_NOT_MODIFIED;
+                    deletehttp(&back[i].r);
+                    back[i].r.soc = INVALID_SOCKET;
                   }
                   // Various hacks to limit re-transfers when updating a mirror
                   // Force update if same size detected

src/htsglobal.h

@@ -56,6 +56,10 @@ Please visit our Website: http://www.httrack.com
 // Platform detection (sizes, feature macros)
 #include "htsconfig.h"
 
+// Fixed-width integer types + PRI* format macros for the LLint/TStamp typedefs
+#include <stdint.h>
+#include <inttypes.h>
+
 // WIN32 types
 #ifdef _WIN32
 #ifndef SIZEOF_LONG
@@ -111,29 +115,6 @@ Please visit our Website: http://www.httrack.com
 #define HTS_DO_NOT_USE_UID
 #endif
 
-#ifndef HTS_LONGLONG
-#ifdef SIZEOF_LONG_LONG
-#if SIZEOF_LONG_LONG == 8
-#define HTS_LONGLONG 1
-#endif
-#endif
-
-#ifndef HTS_LONGLONG
-#ifdef __sun
-#define HTS_LONGLONG 0
-#endif
-#ifdef __osf__
-#define HTS_LONGLONG 0
-#endif
-#ifdef __linux
-#define HTS_LONGLONG 1
-#endif
-#ifdef _WIN32
-#define HTS_LONGLONG 1
-#endif
-#endif
-#endif
-
 #ifdef DLLIB
 #define HTS_DLOPEN 1
 #else
@@ -326,66 +307,11 @@ typedef int hts_tristate;
 #define HTS_DEPRECATED(msg)
 #endif
 
-#ifndef HTS_LONGLONG
-#ifdef HTS_NO_64_BIT
-#define HTS_LONGLONG 0
-#else
-#define HTS_LONGLONG 1
-#endif
-#endif
-
-/* Wide integer types, chosen per platform.
-   LLint:  signed 64-bit counter for byte counts and large sizes (falls back to
-           plain int where 64-bit is unavailable).
-   TStamp: timestamp/duration in the same width (a double in the no-64-bit
-           fallback).
-   LLintP: the printf conversion for an LLint. */
-#if HTS_LONGLONG
-#ifdef LLINT_FORMAT
-typedef LLINT_TYPE LLint;
-
-typedef LLINT_TYPE TStamp;
-
-#define LLintP LLINT_FORMAT
-#else
-
-#ifdef _WIN32
-typedef __int64 LLint;
-
-typedef __int64 TStamp;
-
-#define LLintP "%I64d"
-/* x32/ILP32 sets __x86_64__ but long is 32-bit; exclude it so LLint stays
- * 64-bit. */
-#elif (defined(_LP64) || defined(__x86_64__) || defined(__powerpc64__) ||      \
-       defined(__64BIT__)) &&                                                  \
-    !defined(__ILP32__)
-
-typedef long int LLint;
-
-typedef long int TStamp;
-
-#define LLintP "%ld"
-#else
-typedef long long int LLint;
-
-typedef long long int TStamp;
-
-#define LLintP "%lld"
-#endif
-
-#endif /* HTS_LONGLONG */
-
-/* Claiming long-long support must yield a real 64-bit LLint (x32 regressed:
-   __x86_64__ set but long is 32-bit). Compile-time trip, portable to C90. */
-typedef char hts_assert_llint_is_64bit[sizeof(LLint) == 8 ? 1 : -1];
-
-#else
-typedef int LLint;
-
-#define LLintP "%d"
-typedef double TStamp;
-#endif
+/* LLint/TStamp: signed exact-width 64-bit; -1 is a sentinel engine-wide. */
+typedef int64_t LLint;
+typedef int64_t TStamp;
+/* Full printf conversion, '%' included (PRId64 has none): "X: " LLintP. */
+#define LLintP "%" PRId64
 
 /* Integer type for file offsets/sizes passed to the C library. Widens to
    LLint (with HTS_FSEEKO for fseeko/ftello) under large-file support, plain

src/htslib.c

@@ -3038,12 +3038,11 @@ HTSEXT_API char *int2char(strc_int2bytes2 * strc, int n) {
 /* See http://physics.nist.gov/cuu/Units/binary.html */
 #define ToLLint(a) ((LLint)(a))
 #define ToLLintKiB (ToLLint(1024))
-#define ToLLintMiB (ToLLintKiB*ToLLintKiB)
-#ifdef HTS_LONGLONG
-#define ToLLintGiB (ToLLintKiB*ToLLintKiB*ToLLintKiB)
-#define ToLLintTiB (ToLLintKiB*ToLLintKiB*ToLLintKiB*ToLLintKiB)
-#define ToLLintPiB (ToLLintKiB*ToLLintKiB*ToLLintKiB*ToLLintKiB*ToLLintKiB)
-#endif
+#define ToLLintMiB (ToLLintKiB * ToLLintKiB)
+#define ToLLintGiB (ToLLintKiB * ToLLintKiB * ToLLintKiB)
+#define ToLLintTiB (ToLLintKiB * ToLLintKiB * ToLLintKiB * ToLLintKiB)
+#define ToLLintPiB                                                             \
+  (ToLLintKiB * ToLLintKiB * ToLLintKiB * ToLLintKiB * ToLLintKiB)
 HTSEXT_API char **int2bytes2(strc_int2bytes2 * strc, LLint n) {
   if (n < ToLLintKiB) {
     sprintf(strc->buff1, "%d", (int) (LLint) n);
@@ -3052,9 +3051,7 @@ HTSEXT_API char **int2bytes2(strc_int2bytes2 * strc, LLint n) {
     sprintf(strc->buff1, "%d,%02d", (int) ((LLint) (n / ToLLintKiB)),
             (int) ((LLint) ((n % ToLLintKiB) * 100) / ToLLintKiB));
     strcpybuff(strc->buff2, "KiB");
-  }
-#ifdef HTS_LONGLONG
-  else if (n < ToLLintGiB) {
+  } else if (n < ToLLintGiB) {
     sprintf(strc->buff1, "%d,%02d", (int) ((LLint) (n / (ToLLintMiB))),
             (int) ((LLint) (((n % (ToLLintMiB)) * 100) / (ToLLintMiB))));
     strcpybuff(strc->buff2, "MiB");
@@ -3071,13 +3068,6 @@ HTSEXT_API char **int2bytes2(strc_int2bytes2 * strc, LLint n) {
             (int) ((LLint) (((n % (ToLLintPiB)) * 100) / (ToLLintPiB))));
     strcpybuff(strc->buff2, "PiB");
   }
-#else
-  else {
-    sprintf(strc->buff1, "%d,%02d", (int) ((LLint) (n / (ToLLintMiB))),
-            (int) ((LLint) (((n % (ToLLintMiB)) * 100) / (ToLLintMiB))));
-    strcpybuff(strc->buff2, "MiB");
-  }
-#endif
   strc->buffadr[0] = strc->buff1;
   strc->buffadr[1] = strc->buff2;
   return strc->buffadr;

tests/19_local-connect-fallback.test

@@ -66,9 +66,12 @@ test -n "$port" || {
 }
 
 out="$tmpdir/crawl"
+# macOS has only 127.0.0.1, so the dead 127.0.0.x connects block to the timeout
+# instead of refusing instantly (as on Linux); a small --timeout + --retries=0
+# bound the stall without changing what the fallback exercises.
 HTTRACK_DEBUG_RESOLVE="deadhost:127.0.0.2,127.0.0.1" \
     httrack "http://deadhost:$port/simple/basic.html" -O "$out" \
-    -c1 --robots=0 --timeout=30 --quiet -Z >"$tmpdir/log" 2>&1
+    -c1 --robots=0 --timeout=5 --retries=0 --quiet -Z >"$tmpdir/log" 2>&1
 
 log="$out/hts-log.txt"
 
@@ -92,12 +95,12 @@ test -f "$out/deadhost_$port/simple/basic.html" || {
     exit 1
 }
 
-# every address refused: the slot exhausts the list, then errors out (the
-# harness timeout would catch a hang/loop; refused connects are instant)
+# every address dead: the slot exhausts the list, then errors out (the harness
+# timeout would catch a hang/loop)
 out2="$tmpdir/crawl2"
 HTTRACK_DEBUG_RESOLVE="alldead:127.0.0.2,127.0.0.3" \
     httrack "http://alldead:$port/simple/basic.html" -O "$out2" \
-    -c1 --robots=0 --timeout=30 --quiet -Z >"$tmpdir/log2" 2>&1
+    -c1 --robots=0 --timeout=5 --retries=0 --quiet -Z >"$tmpdir/log2" 2>&1
 log2="$out2/hts-log.txt"
 
 grep -q "trying next address" "$log2" || {

tests/31_local-javaclass.test

@@ -17,7 +17,10 @@ printf 'GIF89a' >"$tmproot/javaclass/hello.gif"
 printf '\xCA\xFE\xBA\xBE\x00\x00\x00\x32\x00\x02\x01\x00\x09hello.gif\x00\x00\x00\x00' \
     >"$tmproot/javaclass/Foo.class"
 
+# --log-found first: a plugin-load failure (wrong soname, shadowing system
+# libhtsjava) trips here on the banner suffix, not as an opaque "not found".
 bash "$top_srcdir/tests/local-crawl.sh" --root "$tmproot" --errors 0 \
+    --log-found '\+libhtsjava' \
     --found 'javaclass/Foo.class' \
     --found 'javaclass/hello.gif' \
     httrack 'BASEURL/javaclass/index.html'

tests/36_local-bigcrawl.test

@@ -9,7 +9,7 @@ set -euo pipefail
 : "${top_srcdir:=..}"
 
 bash "$top_srcdir/tests/local-crawl.sh" --rerun \
-    --errors 4 --files 361 \
+    --errors-content 4 --files 361 \
     --found 'big/p/95.html' \
     --found 'big/a/d1/d2/d3/d4/d5/d6/d7/d8/deep.png' \
     --found 'big/a/f2-2x.png' \

tests/44_local-update-errormask.test

@@ -0,0 +1,13 @@
+#!/bin/bash
+# Issue #176: on --update a page that 200'd on the first crawl but now 403s must
+# keep its good local copy, not be overwritten by the error body nor purged.
+set -euo pipefail
+
+: "${top_srcdir:=..}"
+
+bash "$top_srcdir/tests/local-crawl.sh" --rerun \
+    --found 'errmask/keep.dat' \
+    --file-min-bytes 'errmask/keep.dat' 1024 \
+    --file-matches 'errmask/keep.dat' '^KEEP' \
+    --file-not-matches 'errmask/keep.dat' 'error 403' \
+    httrack 'BASEURL/errmask/index.html'

tests/Makefile.am

@@ -112,6 +112,7 @@ TESTS = \
 	40_local-why.test \
 	41_local-utf8-link.test \
 	42_local-maxsize-slow.test \
-	43_local-update-truncate.test
+	43_local-update-truncate.test \
+	44_local-update-errormask.test
 
 CLEANFILES = check-network_sh.cache

tests/local-crawl.sh

@@ -14,11 +14,13 @@
 # Usage:
 #   bash local-crawl.sh [--tls] [--root DIR] [--cookie NAME=VALUE ...] \
 #       [--rerun-args 'ARGS'] \
-#       --errors N --files N --found PATH ... --directory PATH ... \
+#       --errors N --errors-content N --files N --found PATH ... --directory PATH ... \
 #       --log-found REGEX ... --log-not-found REGEX ... \
 #       --file-matches PATH REGEX ... --file-not-matches PATH REGEX ... \
 #       --file-min-bytes PATH N --max-mirror-bytes N \
 #       httrack BASEURL/some/path [httrack-args...]
+# --errors counts every "Error:" log line; --errors-content drops transient
+# network failures (codes -2..-6) that flake on busy loopback under -c8.
 # --log-found/--log-not-found grep (ERE) the crawl's hts-log.txt.
 # --max/--min-mirror-bytes bound the mirrored content bytes (host root).
 # --file-matches/--file-not-matches grep (ERE) a mirrored file (PATH under the
@@ -131,7 +133,7 @@ while test "$pos" -lt "$nargs"; do
         pos=$((pos + 1))
         rerun_args="${args[$pos]}"
         ;;
-    --errors | --files)
+    --errors | --errors-content | --files)
         audit+=("${args[$pos]}" "${args[$((pos + 1))]}")
         pos=$((pos + 1))
         ;;
@@ -334,6 +336,14 @@ while test "$i" -lt "${#audit[@]}"; do
         assert_equals "checking errors" "${audit[$i]}" \
             "$(grep -iEc "^[0-9:]*[[:space:]]Error:" "${out}/hts-log.txt")"
         ;;
+    --errors-content)
+        i=$((i + 1))
+        total=$(grep -icE "^[0-9:]*[[:space:]]Error:" "${out}/hts-log.txt")
+        # transient network failures (statuscode -2..-6) flake on busy loopback;
+        # the code parens are followed by " at link" or " after N retries at link"
+        transient=$(grep -cE '\(-[2-6]\) (at link|after )' "${out}/hts-log.txt" || true)
+        assert_equals "checking content errors" "${audit[$i]}" "$((total - transient))"
+        ;;
     --files)
         i=$((i + 1))
         nFiles=$(grep -E "^HTTrack Website Copier/[^ ]* mirror complete in " "${out}/hts-log.txt" |

tests/local-server.py

@@ -934,6 +934,33 @@ class Handler(SimpleHTTPRequestHandler):
     def route_mini304_page(self):
         self.big_send(b"<html><body>tiny cacheable page</body></html>\n", "text/html")
 
+    # --- /errmask/: issue #176 — a page that 200'd on the first crawl but 403s
+    # on the update fetch must keep its good copy, not be overwritten nor purged.
+    ERRMASK_GOOD = b"KEEP" + b"." * 1020  # 1024 B distinctive non-HTML body
+    ERRMASK_ERR = b"<html><body>error 403</body></html>\n"
+
+    def route_errmask_index(self):
+        self.send_html('\t<a href="keep.dat">keep</a>\n')
+
+    def route_errmask_keep(self):
+        # First crawl (no validator) gets the 1024 B body + Last-Modified; the
+        # update sends a conditional and gets a 403 error page.
+        if self.headers.get("If-Modified-Since") or self.headers.get("If-None-Match"):
+            self.send_response(403, "Forbidden")
+            self.send_header("Content-Type", "text/html")
+            self.send_header("Content-Length", str(len(self.ERRMASK_ERR)))
+            self.end_headers()
+            if self.command != "HEAD":
+                self.wfile.write(self.ERRMASK_ERR)
+            return
+        self.send_response(200)
+        self.send_header("Content-Type", "application/octet-stream")
+        self.send_header("Last-Modified", BIG_LASTMOD)
+        self.send_header("Content-Length", str(len(self.ERRMASK_GOOD)))
+        self.end_headers()
+        if self.command != "HEAD":
+            self.wfile.write(self.ERRMASK_GOOD)
+
     # --- delayed-type degenerate paths (issues #5/#107) --------------------
     def route_delayed_index(self):
         self.send_html(
@@ -1177,6 +1204,8 @@ class Handler(SimpleHTTPRequestHandler):
         "/redir/target.html": route_redir_target,
         "/mini304/index.html": route_mini304_index,
         "/mini304/page.html": route_mini304_page,
+        "/errmask/index.html": route_errmask_index,
+        "/errmask/keep.dat": route_errmask_keep,
     }
 
     # --- /big/ seeded pseudo-site ------------------------------------------