Merge origin/master into fix-623-delayed-marker

Resolve tests/Makefile.am: keep both 66_engine-port80-strip (from #632) and
67_engine-delayed-truncate.

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

.github/dependabot.yml

@@ -1,8 +0,0 @@
-version: 2
-updates:
-  # Bumps builtin-baseline in src/vcpkg.json so the pinned OpenSSL/zlib can't rot.
-  # windows-build.yml validates each bump; the pin makes what ships reproducible.
-  - package-ecosystem: vcpkg
-    directory: /src
-    schedule:
-      interval: weekly

.github/workflows/windows-build.yml

@@ -44,20 +44,6 @@ jobs:
         shell: pwsh
         run: vcpkg integrate install
 
-      # The runner image's vcpkg checkout is pinned to some commit; our manifest's
-      # builtin-baseline is usually newer, so `git show <baseline>:versions/...`
-      # fails until that commit is local. Fetch exactly the pinned baseline (read
-      # from the manifest, so Dependabot bumps need no workflow edit).
-      - name: Fetch the pinned vcpkg baseline
-        shell: pwsh
-        run: |
-          $ErrorActionPreference = "Stop"
-          $root = if ($env:VCPKG_INSTALLATION_ROOT) { $env:VCPKG_INSTALLATION_ROOT } else { "C:\vcpkg" }
-          $baseline = (Get-Content src/vcpkg.json -Raw | ConvertFrom-Json).'builtin-baseline'
-          if (-not $baseline) { throw "no builtin-baseline in src/vcpkg.json" }
-          git -C $root fetch --no-tags origin $baseline
-          if ($LASTEXITCODE -ne 0) { throw "could not fetch vcpkg baseline $baseline" }
-
       # httrack and webhttrack carry a ProjectReference to libhttrack, so building
       # them builds it first; proxytrack is standalone.
       - name: Build
@@ -202,7 +188,7 @@ jobs:
 
           # Every gate here exits 77, so an all-skipped suite would report green having
           # tested nothing: pin the skips, and floor the passes in case the glob empties.
-          expected_skips=" 48_local-crange-memresume.test 71_local-crange-repaircache.test"  # pending #581
+          expected_skips=" 48_local-crange-memresume.test"  # pending #581
           [ "$pass" -ge 90 ] || { echo "::error::only $pass tests passed ($skip skipped)"; exit 1; }
           [ "$skipped" = "$expected_skips" ] || { echo "::error::unexpected skips:$skipped"; exit 1; }
           [ "$fail" -eq 0 ] || { echo "::error::failing:$failed"; exit 1; }

AGENTS.md

@@ -19,13 +19,6 @@ the operational checklist: toolchain, invariants, and how to ship a change.
   (`request_queue_size`) so macOS/BSD don't drop connections under a parallel
   `-c16` bigcrawl the way Python's default backlog of 5 did.
   Or run `sh build.sh` to do bootstrap + configure + make in one shot.
-- A `tests/NN_*.test` runs only if listed in `tests/Makefile.am`'s `TESTS`; an
-  unregistered file is silently skipped.
-- `make check` prepends the build's `src/` to `PATH`, but a hand-run `.test` does
-  not — an installed `/usr/bin/httrack` then shadows your build. Run via `make
-  check`, or `PATH="<bld>/src:$PATH"` for a manual run.
-- Give new `.test` scripts `set -e`: the older ones predate the rule, so several
-  `local-crawl.sh` calls with no `set -e` report PASS on any non-last failure.
 
 ## Hard invariants
 - **Generated autotools files are NOT in git.** `configure`, every
@@ -38,32 +31,15 @@ the operational checklist: toolchain, invariants, and how to ship a change.
 - **Format only changed lines** with `git clang-format` (clang-format 19). Never
   reformat untouched code: the engine was formatted by an old tool and won't
   round-trip.
-- **Byte-safe edits.** A few tracked files carry raw ISO-8859-1 high bytes
-  (French comments): `src/htsconcat.c`, `lang/*.txt`, `html/contact.html`, and
-  the `fuzz/corpus/*` vectors. Edit those byte-wise (`perl -0pi`, `sed`), not
-  through a tool that re-encodes to UTF-8 and corrupts them. The rest of the tree
-  is UTF-8 and safe to edit normally.
+- **Byte-safe edits.** Files with raw high bytes are ISO-8859-1 (French
+  comments). Edit them byte-wise (`perl -0pi`, `sed`), not through a tool that
+  re-encodes to UTF-8 and corrupts them.
 
 ## Security (HTTrack parses hostile input off the network)
 - Bounds-check every copy. Overflow-safe form: put the untrusted value alone,
   `untrusted < limit - controlled` — never `controlled + untrusted < limit`,
   which can wrap and pass.
 
-## C conventions
-- **Use the `*t` allocator wrappers, never raw libc** (`htssafe.h`):
-  `malloct`/`calloct`/`realloct`/`freet`/`strdupt`, in test and selftest code
-  too. `freet` NULLs its (lvalue) argument and tolerates NULL; `calloct(n, sz)`
-  keeps calloc's arg order. Only exception: storing or calling a libc symbol
-  itself (e.g. a resolver-backend function pointer).
-- **Exported API is `HTSEXT_API`.** Everything else is hidden by
-  `-fvisibility=hidden` and free to change (check with `nm -D --defined-only
-  libhttrack.so`). Touching an installed-header struct (see `DevIncludes_DATA` in
-  `src/Makefile.am`) or an exported signature is an ABI break — flag and discuss,
-  bump the soname, and prefer keeping the old entry point beside a new one.
-- **Windows ABI is free to break, POSIX is not.** The Windows DLL ships next to
-  the exe with no soname contract, so a `_WIN32`-only ABI change needs no
-  deprecation dance; POSIX/ELF keeps the flag-discuss-bump rules.
-
 ## Code & prose
 - Be terse. Comment the why, in English; translate French comments you touch.
 - Strip AI tells from prose (em-dash overuse, rule-of-three, filler, vague

src/htscharset.c

@@ -373,8 +373,7 @@ char *hts_convertStringCPFromUTF8(const char *s, size_t size, UINT cp) {
   return NULL;
 }
 
-HTSEXT_API char *hts_convertStringToUTF8(const char *s, size_t size,
-                                         const char *charset) {
+char *hts_convertStringToUTF8(const char *s, size_t size, const char *charset) {
   const UINT cp = hts_getCodepage(charset);
 
   return hts_convertStringCPToUTF8(s, size, cp);
@@ -555,8 +554,7 @@ static char *hts_convertStringCharset(const char *s, size_t size,
   return NULL;
 }
 
-HTSEXT_API char *hts_convertStringToUTF8(const char *s, size_t size,
-                                         const char *charset) {
+char *hts_convertStringToUTF8(const char *s, size_t size, const char *charset) {
   /* Empty string ? */
   if (size == 0) {
     return strdup("");

src/htscharset.h

@@ -51,8 +51,8 @@ typedef unsigned int hts_UCS4;
  * Convert the string "s" from charset "charset" to UTF-8.
  * Return NULL upon error.
  **/
-HTSEXT_API char *hts_convertStringToUTF8(const char *s, size_t size,
-                                         const char *charset);
+extern char *hts_convertStringToUTF8(const char *s, size_t size,
+                                     const char *charset);
 
 /**
  * Convert the string "s" from UTF-8 to charset "charset".

src/htscoremain.c

@@ -2368,28 +2368,30 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
       opt->errlog = stderr;
     } else if (httrack_logmode >= 2) {
       // deux fichiers log
-      // path_log holds UTF-8 bytes (argv is UTF-8): the ANSI file calls would
-      // read them as the codepage and drop the logs into a mangled twin (#630).
-      structcheck_utf8(StringBuff(opt->path_log));
-      if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                              StringBuff(opt->path_log), "hts-log.txt")))
-        UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                       StringBuff(opt->path_log), "hts-log.txt"));
-      if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                              StringBuff(opt->path_log), "hts-err.txt")))
-        UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                       StringBuff(opt->path_log), "hts-err.txt"));
+      structcheck(StringBuff(opt->path_log));
+      if (fexist
+          (fconcat
+           (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-log.txt")))
+        remove(fconcat
+               (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-log.txt"));
+      if (fexist
+          (fconcat
+           (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-err.txt")))
+        remove(fconcat
+               (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-err.txt"));
 
       /* Check FS directory structure created */
-      structcheck_utf8(StringBuff(opt->path_log));
+      structcheck(StringBuff(opt->path_log));
 
-      opt->log = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                               StringBuff(opt->path_log), "hts-log.txt"),
-                       "w");
+      opt->log =
+        fopen(fconcat
+              (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-log.txt"),
+              "w");
       if (httrack_logmode == 2)
-        opt->errlog = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                                    StringBuff(opt->path_log), "hts-err.txt"),
-                            "w");
+        opt->errlog =
+          fopen(fconcat
+                (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-err.txt"),
+                "w");
       else
         opt->errlog = opt->log;
       if (opt->log == NULL) {

src/htsselftest.c

@@ -2146,52 +2146,6 @@ static int st_cache_corrupt(httrackp *opt, int argc, char **argv) {
   return err;
 }
 
-/* Drives unzRepair over a damaged local file header whose CRC field's high
-   16-bit word has bit 15 set. Before the READ_32 fix that shifted an int and
-   overflowed, so UBSan aborts here; after it, repair recovers the one entry. */
-static int st_zip_repair_shift(httrackp *opt, int argc, char **argv) {
-  static const unsigned char zip[] = {
-      0x50, 0x4b, 0x03, 0x04, /* local file header signature */
-      0x14, 0x00,             /* version needed */
-      0x00, 0x00,             /* general purpose flag */
-      0x00, 0x00,             /* method */
-      0x00, 0x00,             /* time */
-      0x00, 0x00,             /* date */
-      0x00, 0x00, 0xe8, 0x8a, /* crc: high word 0x8ae8, bit 15 set */
-      0x00, 0x00, 0x00, 0x00, /* compressed size */
-      0x00, 0x00, 0x00, 0x00, /* uncompressed size */
-      0x01, 0x00,             /* filename length */
-      0x00, 0x00,             /* extra field length */
-      0x61                    /* filename "a" */
-  };
-  char in[HTS_URLMAXSIZE], out[HTS_URLMAXSIZE], tmp[HTS_URLMAXSIZE];
-  uLong nrec = 0, bytes = 0;
-  FILE *fp;
-  int err;
-
-  (void) opt;
-  if (argc < 1) {
-    fprintf(stderr, "zip-repair-shift: needs a directory\n");
-    return 1;
-  }
-  snprintf(in, sizeof(in), "%s/damaged.zip", argv[0]);
-  snprintf(out, sizeof(out), "%s/repair.zip", argv[0]);
-  snprintf(tmp, sizeof(tmp), "%s/repair.tmp", argv[0]);
-  fp = fopen(in, "wb");
-  if (fp == NULL || fwrite(zip, 1, sizeof(zip), fp) != sizeof(zip)) {
-    if (fp != NULL)
-      fclose(fp);
-    fprintf(stderr, "zip-repair-shift: cannot write %s\n", in);
-    return 1;
-  }
-  fclose(fp);
-  err = unzRepair(in, out, tmp, &nrec, &bytes);
-  printf("zip-repair-shift: %s (recovered %lu entr%s)\n",
-         (err == Z_OK && nrec == 1) ? "OK" : "FAIL", (unsigned long) nrec,
-         nrec == 1 ? "y" : "ies");
-  return (err == Z_OK && nrec == 1) ? 0 : 1;
-}
-
 static int st_cache_legacy(httrackp *opt, int argc, char **argv) {
   int err;
 
@@ -3299,9 +3253,6 @@ static const struct selftest_entry {
      st_cache_legacy},
     {"cache-corrupt", "<dir>", "cache read-side corruption self-test",
      st_cache_corrupt},
-    {"zip-repair-shift", "<dir>",
-     "cache zip-repair header read must not overflow a signed shift",
-     st_zip_repair_shift},
     {"dns", "", "DNS resolver/cache self-test", st_dns},
     {"dnstimeout", "", "a slow DNS resolve is bounded and holds no lock",
      st_dnstimeout},

src/htsserver.c

@@ -40,7 +40,6 @@ Please visit our Website: http://www.httrack.com
 
 #include "htsnet.h"
 #include "htslib.h"
-#include "htscharset.h"
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -750,14 +749,7 @@ int smallserver(T_SOC soc, char *url, char *method, char *data, char *path) {
                            RUN THE SERVER
                          */
                         if (strcmp((char *) adrcd, "start") == 0) {
-                          /* POST body is in the form's charset, not the
-                             UTF-8 argv the engine now assumes (#629). */
-                          char *const cmdl = (char *) adr + p;
-                          char *cmdlUtf8 = hts_convertStringToUTF8(
-                              cmdl, strlen(cmdl), LANGSEL("LANGUAGE_CHARSET"));
-
-                          webhttrack_main(cmdlUtf8 != NULL ? cmdlUtf8 : cmdl);
-                          freet(cmdlUtf8);
+                          webhttrack_main((char *) adr + p);
                         } else {
                           commandRunning = 0;
                           commandEnd = 1;

src/htsweb.c

@@ -65,7 +65,6 @@ Please visit our Website: http://www.httrack.com
 #include "htsserver.h"
 #include "htsurlport.h"
 #include "htsweb.h"
-#include "htscharset.h"
 
 #if USE_BEGINTHREAD==0
 #error fatal: no threads support
@@ -157,7 +156,6 @@ int main(int argc, char *argv[]) {
   printf("Initializing the server..\n");
 
 #ifdef _WIN32
-  hts_argv_utf8(&argc, &argv);
   {
     WORD wVersionRequested;     // requested version WinSock API
     WSADATA wsadata;            // Windows Sockets API data

src/minizip/mztools.c

@@ -14,7 +14,7 @@
 
 #define READ_8(adr)  ((unsigned char)*(adr))
 #define READ_16(adr) ( READ_8(adr) | (READ_8(adr+1) << 8) )
-#define READ_32(adr) ((uLong) READ_16(adr) | ((uLong) READ_16((adr) + 2) << 16))
+#define READ_32(adr) ( READ_16(adr) | (READ_16((adr)+2) << 16) )
 
 #define WRITE_8(buff, n) do { \
   *((unsigned char*)(buff)) = (unsigned char) ((n) & 0xff); \

src/vcpkg.json

@@ -1,12 +1,12 @@
 {
-  "$schema": "https://raw.githubusercontent.com/microsoft/vcpkg-tool/main/docs/vcpkg.schema.json",
+  "$schema":
+      "https://raw.githubusercontent.com/microsoft/vcpkg-tool/main/docs/vcpkg.schema.json",
   "name": "libhttrack",
   "version-string": "3.49",
-  "builtin-baseline": "cd61e1e26a038e82d6550a3ebbe0fbbfe7da78e3",
   "dependencies": [
     "brotli",
     "openssl",
     "zlib",
     "zstd"
   ]
-}
+}

tests/01_engine-xfread.test

@@ -9,23 +9,21 @@ set -euo pipefail
 # drives both refusal paths and the accept path.
 
 out="$(httrack -O /dev/null -#test=xfread-limit)"
-# Match with here-strings, not `echo | grep -q`: under pipefail the early grep
-# exit SIGPIPEs echo and fails the pipeline even when the pattern matched.
 for case in bylen bygrow; do
-    grep -q "${case}: refused=1 adr=null msg=In-memory content too large" <<<"$out" || {
+    echo "$out" | grep -q "${case}: refused=1 adr=null msg=In-memory content too large" || {
         echo "FAIL ${case}: $out"
         exit 1
     }
 done
 
 # Exactly INT32_MAX must be refused too (the reallocs add 1).
-grep -q 'boundary: msg=In-memory content too large' <<<"$out" || {
+echo "$out" | grep -q 'boundary: msg=In-memory content too large' || {
     echo "FAIL boundary: $out"
     exit 1
 }
 
 # The guard must NOT fire for a legitimate small size.
-if grep -q 'accept: msg=In-memory content too large' <<<"$out"; then
+if echo "$out" | grep -q 'accept: msg=In-memory content too large'; then
     echo "FAIL accept (guard fired on a legit size): $out"
     exit 1
 fi

tests/01_zlib-repair-shift.test

@@ -1,21 +0,0 @@
-#!/bin/bash
-#
-# Keep this POSIX-portable: the harness runs it via $(BASH), which is a plain
-# POSIX /bin/sh on some platforms (e.g. macOS), so avoid bashisms and GNU-only
-# tool flags despite the #!/bin/bash above.
-
-# unzRepair header read must not overflow a signed shift (-#test=zip-repair-shift
-# <dir>). A damaged local file header whose CRC high word has bit 15 set made
-# READ_32 shift an int past INT_MAX; UBSan aborts before the fix casts to uLong.
-
-set -eu
-
-dir=$(mktemp -d)
-trap 'rm -rf "$dir"' EXIT
-
-out=$(httrack -#test=zip-repair-shift "$dir")
-
-printf '%s\n' "$out" | grep -qx "zip-repair-shift: OK (recovered 1 entry)" || {
-    echo "expected 'zip-repair-shift: OK (recovered 1 entry)', got: $out" >&2
-    exit 1
-}

tests/68_webhttrack-outdir-charset.test

@@ -1,90 +0,0 @@
-#!/bin/bash
-#
-# webhttrack's POST body arrives in the form's declared charset, not UTF-8, yet
-# the engine now assumes UTF-8 argv. A non-ASCII -O output dir submitted through
-# the web UI must land under the UTF-8 directory, not an ISO-8859-1 twin (#629).
-# Drives the real htsserver over HTTP; the default (English) form is served
-# ISO-8859-1, so 'café' travels as the single byte 0xE9. The crawl target is a
-# dead port: only the -O directory name is under test, and htsserver creates it
-# from the decoded path before any fetch.
-
-set -euo pipefail
-
-testdir=$(cd "$(dirname "$0")" && pwd)
-distdir=${top_srcdir:-$(cd "${testdir}/.." && pwd)}
-distdir=$(cd "${distdir}" && pwd)
-# shellcheck source=tests/testlib.sh
-. "${testdir}/testlib.sh"
-
-fail() {
-    echo "FAIL: $*" >&2
-    exit 1
-}
-
-command -v htsserver >/dev/null || fail "no htsserver in PATH"
-python=$(find_python) || {
-    echo "python3 not found; skipping" >&2
-    exit 77
-}
-
-work=$(mktemp -d "${TMPDIR:-/tmp}/webhttrack_charset.XXXXXX") || fail "no tmpdir"
-srvlog=$(mktemp)
-srv=
-cleanup() {
-    # htsserver keeps SIGTERM ignored across its exec, so only -9 reaps it.
-    test -z "${srv}" || kill -9 "${srv}" 2>/dev/null || true
-    wait "${srv}" 2>/dev/null || true # absorb bash's async "Killed" notice
-    rm -rf "${work}" "${srvlog}"
-}
-trap cleanup EXIT HUP INT QUIT PIPE TERM
-
-# webhttrack server on a pre-picked port; an isolated HOME keeps a stray
-# ~/.httrack.ini out of it.
-sport=$("${python}" -c 'import socket
-s = socket.socket()
-s.bind(("127.0.0.1", 0))
-print(s.getsockname()[1])
-s.close()')
-(
-    trap '' TERM TTOU
-    export HOME="${work}"
-    exec htsserver "${distdir}/" --port "${sport}" >"${srvlog}" 2>&1
-) &
-srv=$!
-for _ in $(seq 1 40); do
-    url=$(sed -n 's/^URL=//p' "${srvlog}") && test -n "${url}" && break
-    kill -0 "${srv}" 2>/dev/null || break
-    sleep 0.25
-done
-test -n "${url:-}" || fail "htsserver did not start: $(cat "${srvlog}")"
-
-# Post a "start" whose -O dir is 'café' in the form's ISO-8859-1 charset.
-"${python}" - "${url}" "${work}" <<'PY'
-import sys, urllib.parse, urllib.request
-
-url, work = sys.argv[1], sys.argv[2]
-outdir = work + "/caf\xe9"  # 'café' as the single byte the browser would send
-# Port 1 refuses at once: only the decoded -O dir matters, not the fetch.
-cmd = "httrack --quiet --robots=0 http://127.0.0.1:1/x.html -O " + outdir
-fields = [("path", work), ("projname", "proj"), ("command_do", "start"),
-          ("winprofile", "x"), ("command", cmd)]
-body = "&".join("%s=%s" % (k, urllib.parse.quote(v, safe="", encoding="latin-1"))
-                for k, v in fields)
-req = urllib.request.Request(url + "step4.html", data=body.encode("latin-1"),
-                             method="POST")
-urllib.request.urlopen(req, timeout=20).read()
-PY
-
-# The crawl runs inside htsserver; wait for the -O dir, then check it chose the
-# UTF-8 name and not the ISO-8859-1 twin.
-utf8dir="${work}/café"
-latin1dir="${work}/caf"$'\xe9'
-for _ in $(seq 1 80); do
-    test -e "${utf8dir}" && break
-    kill -0 "${srv}" 2>/dev/null || break
-    sleep 0.25
-done
-test -e "${utf8dir}" || fail "-O dir never landed as UTF-8 café/: $(find "${work}" -maxdepth 2 2>/dev/null)"
-test ! -e "${latin1dir}" || fail "-O dir created as the ISO-8859-1 twin instead"
-
-echo "PASS"

tests/69_local-intl-logdir.test

@@ -1,15 +0,0 @@
-#!/bin/bash
-#
-# A single non-ASCII -O sets both path_html and path_log to "café" (#630). The
-# logs (hts-log.txt/hts-err.txt) must land there, not in an ANSI-mangled twin:
-# on Windows path_log holds UTF-8 bytes the raw file calls read as the codepage.
-# The --errors/--files/--log-found audits all grep logroot=café/hts-log.txt, so
-# a log written to the twin fails them. POSIX has no twin, so this bites on the
-# Windows CI leg (like test 64). The cache twin is a separate, larger fix.
-
-: "${top_srcdir:=..}"
-
-bash "$top_srcdir/tests/local-crawl.sh" --outdir-intl 'café' --errors 0 --files 5 \
-    --found 'simple/basic.html' \
-    --log-found 'mirror complete in' \
-    httrack 'BASEURL/simple/basic.html'

tests/71_local-crange-repaircache.test

@@ -1,134 +0,0 @@
-#!/bin/bash
-# #581: an interrupted mirror can leave a truncated new.zip that pass 2 must
-# repair before resuming. The repaired-cache resume then hits the same hostile
-# 206 as test 48, so restart-whole must still drop the partial and refetch the
-# whole file. Pass 1 leaves a partial + temp-ref; we truncate new.zip past its
-# last local entry (dropping the central directory) so unzOpen fails and the
-# repair path runs; pass 2 resumes, rejects the range, and refetches whole.
-set -u
-
-: "${top_srcdir:=..}"
-testdir=$(cd "$(dirname "$0")" && pwd)
-# shellcheck source=tests/testlib.sh
-. "${testdir}/testlib.sh"
-server=$(nativepath "${testdir}/local-server.py")
-root=$(nativepath "${testdir}/server-root")
-
-python=$(find_python) || ! echo "python3 not found; skipping" >&2 || exit 77
-
-# On Windows the pass-1 interrupt is a hard kill (MSYS can't signal a native
-# exe) and the restart-whole path fails on the repaired cache (#581) -- the very
-# bug this exercises; skip until the engine fix lands.
-if is_windows; then
-    echo "Windows: restart-whole fails on a repaired cache (#581), skipping"
-    exit 77
-fi
-
-tmpdir=$(mktemp -d "${TMPDIR:-/tmp}/httrack_crangerep.XXXXXX") || exit 1
-serverpid=
-crawlpid=
-cleanup() {
-    test -n "$crawlpid" && kill -9 "$crawlpid" 2>/dev/null
-    stop_server "$serverpid"
-    rm -rf "$tmpdir"
-}
-trap cleanup EXIT HUP INT QUIT PIPE TERM
-
-serverlog="${tmpdir}/server.log"
-: >"$serverlog"
-"$python" "$server" --root "$root" >"$serverlog" 2>&1 &
-serverpid=$!
-port=
-for _ in $(seq 1 50); do
-    line=$(head -n1 "$serverlog" 2>/dev/null)
-    if test "${line%% *}" == "PORT"; then
-        port="${line#PORT }"
-        break
-    fi
-    kill -0 "$serverpid" 2>/dev/null || {
-        echo "server exited early: $(cat "$serverlog")"
-        exit 1
-    }
-    sleep 0.1
-done
-test -n "$port" || {
-    echo "could not discover server port"
-    exit 1
-}
-base="http://127.0.0.1:${port}"
-
-which httrack >/dev/null || {
-    echo "could not find httrack"
-    exit 1
-}
-out="${tmpdir}/crawl"
-mkdir "$out"
-common=(-O "$out" --quiet --disable-security-limits --robots=0 --timeout=30 --retries=1 -c1)
-refdir="${out}/hts-cache/ref"
-newzip="${out}/hts-cache/new.zip"
-
-# --- pass 1: crawl, interrupt once the blob download is underway -------------
-printf '[pass 1: interrupt mid-download] ..\t'
-httrack "${common[@]}" "${base}/crange206mem/index.html" >"${tmpdir}/log1" 2>&1 &
-crawlpid=$!
-for _ in $(seq 1 300); do
-    test -n "$(find "$refdir" -name '*.ref' 2>/dev/null)" && break
-    kill -0 "$crawlpid" 2>/dev/null || break
-    sleep 0.1
-done
-sleep 0.3
-kill -TERM "$crawlpid" 2>/dev/null
-wait "$crawlpid" 2>/dev/null
-crawlpid=
-test -n "$(find "$refdir" -name '*.ref' 2>/dev/null)" || {
-    echo "FAIL: no temp-ref survived pass 1; cannot drive the resume"
-    exit 1
-}
-echo "OK (temp-ref present)"
-
-# --- damage the cache: drop the central directory a hard kill never wrote ----
-printf '[damage new.zip -> forces repair] ..\t'
-"$python" - "$newzip" <<'PY' || exit 1
-import os, sys
-path = sys.argv[1]
-data = open(path, "rb").read()
-cut = data.find(b"PK\x01\x02")  # first central-directory header
-if cut <= 0:
-    sys.exit("no central directory to drop in %s" % path)
-os.truncate(path, cut)
-PY
-echo "OK"
-
-# --- pass 2: --continue -> repair -> resume -> hostile 206 -> refetch whole ---
-printf '[pass 2: repair, reject range, refetch] ..\t'
-rc=0
-httrack "${common[@]}" --continue "${base}/crange206mem/index.html" >"${tmpdir}/log2" 2>&1 || rc=$?
-test "$rc" -eq 0 || {
-    echo "FAIL: httrack exited $rc"
-    cat "${tmpdir}/log2" >&2
-    exit 1
-}
-echo "OK (terminated)"
-
-# The repair path must actually have run, else this is just a copy of test 48.
-printf '[cache repair fired] ..\t'
-grep -q 'damaged cache' "${out}/hts-log.txt" 2>/dev/null || {
-    echo "FAIL: repair path did not run; damage was ineffective"
-    exit 1
-}
-echo "OK"
-
-blob=$(find "$out" -name blob.bin 2>/dev/null | head -1)
-full=6008 # len(CRANGE206_BODY) = len("CR206DAT") + 6000
-
-printf '[file recovered whole] ..\t'
-test -s "$blob" || {
-    echo "FAIL: blob.bin missing after the repaired-cache resume"
-    exit 1
-}
-got=$(wc -c <"$blob")
-test "$got" -eq "$full" || {
-    echo "FAIL: blob.bin is ${got} bytes, expected ${full}"
-    exit 1
-}
-echo "OK (${got} bytes)"

tests/Makefile.am

@@ -84,7 +84,6 @@ TESTS = \
 	01_zlib-cache-legacy.test \
 	01_zlib-cache-golden.test \
 	01_zlib-cache-writefail.test \
-	01_zlib-repair-shift.test \
 	01_zlib-savename-cached.test \
 	02_manpage-regen.test \
 	02_update-cache.test \
@@ -149,9 +148,6 @@ TESTS = \
 	64_local-intl-outdir.test \
 	65_port-siblings.test \
 	66_engine-port80-strip.test \
-	67_engine-delayed-truncate.test \
-	68_webhttrack-outdir-charset.test \
-	69_local-intl-logdir.test \
-	71_local-crange-repaircache.test
+	67_engine-delayed-truncate.test
 
 CLEANFILES = check-network_sh.cache

tests/local-crawl.sh

@@ -48,7 +48,6 @@ key="${testdir}/server.key"
 tls=
 verbose=
 html_subdir=
-outdir_intl=
 rerun=
 rerun_args=
 rerun_dead=
@@ -141,13 +140,6 @@ while test "$pos" -lt "$nargs"; do
         pos=$((pos + 1))
         html_subdir="${args[$pos]}"
         ;;
-    --outdir-intl)
-        # Single non-ASCII -O "$out/NAME": path_html AND path_log are NAME, so
-        # the logs (and the harness reads of them) go through the non-ASCII path
-        # (#630). Distinct from --html-subdir, which keeps path_log ASCII.
-        pos=$((pos + 1))
-        outdir_intl="${args[$pos]}"
-        ;;
     --errors | --errors-content | --files)
         audit+=("${args[$pos]}" "${args[$((pos + 1))]}")
         pos=$((pos + 1))
@@ -224,18 +216,13 @@ test -n "$ver" || die "could not run httrack"
 out="${tmpdir}/crawl"
 mkdir "$out" || die "could not create $out"
 # path_html holds the mirror + index; path_log holds hts-cache/hts-log.txt.
-# Default: both are "$out". --html-subdir moves path_html to "$out/NAME" while
-# path_log (logroot) stays "$out"; --outdir-intl moves both to "$out/NAME".
+# Default: both are "$out". With --html-subdir, path_html becomes "$out/NAME"
+# (the mirror root the audits inspect) while path_log stays "$out".
 mirrorroot="$out"
-logroot="$out"
 odir="$out"
 if test -n "$html_subdir"; then
     mirrorroot="${out}/${html_subdir}"
     odir="${mirrorroot},${out}"
-elif test -n "$outdir_intl"; then
-    mirrorroot="${out}/${outdir_intl}"
-    logroot="$mirrorroot"
-    odir="$mirrorroot"
 fi
 # Localhost is fast; disable the rate/bandwidth safety limits but keep a
 # max-time backstop so a hang cannot wedge the suite.
@@ -254,7 +241,7 @@ test "$crawlres" -eq 0 || ! result "httrack exited $crawlres" || {
     exit 1
 }
 result "OK"
-grep -iE "^[0-9:]*[[:space:]]Error:" "${logroot}/hts-log.txt" >&2
+grep -iE "^[0-9:]*[[:space:]]Error:" "${out}/hts-log.txt" >&2
 
 # --- optional second pass: re-mirror into the same dir (cache/update path) ----
 if test -n "$rerun"; then
@@ -273,7 +260,7 @@ if test -n "$rerun"; then
     # The update summary reports "files updated"; a fresh crawl never does. Assert
     # it so a regression that bypasses the cache (re-crawls fresh) can't pass.
     info "checking update used the cache"
-    if grep -aqE "mirror complete in .*files updated" "${logroot}/hts-log.txt"; then
+    if grep -aqE "mirror complete in .*files updated" "${out}/hts-log.txt"; then
         result "OK"
     else
         result "update pass did not report cache activity"
@@ -305,7 +292,7 @@ if test -n "$rerun_dead"; then
     zip="${out}/hts-cache/new.zip"
     test -s "$zip" || die "no cache was written by the first pass"
     cp "$zip" "${tmpdir}/cache-before.zip"
-    cp "${logroot}/hts-log.txt" "${tmpdir}/log-before.txt"
+    cp "${out}/hts-log.txt" "${tmpdir}/log-before.txt"
     stop_server "$serverpid"
     serverpid=
     info "re-running httrack against the stopped server"
@@ -318,7 +305,7 @@ if test -n "$rerun_dead"; then
     # The dead pass must have gone through the no-data rollback, not bailed out
     # before the mirror loop (which would leave the cache trivially untouched).
     info "checking the dead pass hit the rollback"
-    if grep -aq "No data seems to have been transferred" "${logroot}/hts-log.txt"; then
+    if grep -aq "No data seems to have been transferred" "${out}/hts-log.txt"; then
         result "OK"
     else
         result "rollback notice not found in hts-log.txt"
@@ -333,7 +320,7 @@ if test -n "$rerun_dead"; then
         exit 1
     fi
     # Audits below describe the healthy crawl, not the dead pass.
-    cp "${tmpdir}/log-before.txt" "${logroot}/hts-log.txt"
+    cp "${tmpdir}/log-before.txt" "${out}/hts-log.txt"
 fi
 
 # --- discover the single host root (127.0.0.1_<port> or 127.0.0.1) -----------
@@ -363,19 +350,19 @@ while test "$i" -lt "${#audit[@]}"; do
     --errors)
         i=$((i + 1))
         assert_equals "checking errors" "${audit[$i]}" \
-            "$(grep -iEc "^[0-9:]*[[:space:]]Error:" "${logroot}/hts-log.txt")"
+            "$(grep -iEc "^[0-9:]*[[:space:]]Error:" "${out}/hts-log.txt")"
         ;;
     --errors-content)
         i=$((i + 1))
-        total=$(grep -icE "^[0-9:]*[[:space:]]Error:" "${logroot}/hts-log.txt")
+        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 )' "${logroot}/hts-log.txt" || true)
+        transient=$(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 " "${logroot}/hts-log.txt" |
+        nFiles=$(grep -E "^HTTrack Website Copier/[^ ]* mirror complete in " "${out}/hts-log.txt" |
             sed -e 's/.*[[:space:]]\([^ ]*\)[[:space:]]files written.*/\1/g')
         assert_equals "checking files" "${audit[$i]}" "$nFiles"
         ;;
@@ -406,7 +393,7 @@ while test "$i" -lt "${#audit[@]}"; do
     --log-found)
         i=$((i + 1))
         info "checking log matches ${audit[$i]}"
-        if grep -aqE "${audit[$i]}" "${logroot}/hts-log.txt"; then result "OK"; else
+        if grep -aqE "${audit[$i]}" "${out}/hts-log.txt"; then result "OK"; else
             result "not in log"
             exit 1
         fi
@@ -414,7 +401,7 @@ while test "$i" -lt "${#audit[@]}"; do
     --log-not-found)
         i=$((i + 1))
         info "checking log lacks ${audit[$i]}"
-        if grep -aqE "${audit[$i]}" "${logroot}/hts-log.txt"; then
+        if grep -aqE "${audit[$i]}" "${out}/hts-log.txt"; then
             result "present in log"
             exit 1
         else result "OK"; fi