Windows: enumerate long or non-ASCII directories via FindFirstFileW (#688)

* Windows: enumerate long or non-ASCII directories via FindFirstFileW

The opendir/readdir emulation was fully ANSI (FindFirstFileA, CP_ACP
cFileName), so it capped listing at MAX_PATH and mis-decoded a non-ASCII
path — the odd wrapper out in an engine that feeds UTF-8 paths and reads
d_name as UTF-8. The one live Windows caller, the end-of-crawl
hts-cache/ref cleanup, silently no-op'd at a long or non-ASCII project
root, leaking the temp ref/ directory into the finished mirror.

Route both through hts_pathToUCS2 (\\?\ prefixing, #684) and the wide
APIs, converting cFileName back to UTF-8. d_name now holds UTF-8, so
HTS_DIRENT_SIZE grows to 1024 to fit MAX_PATH's worst-case expansion; the
struct is _WIN32-only, no POSIX ABI impact. Self-test direnum drives it.

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

* Trim comments to one line each (review follow-up)

Condense the opendir/readdir and direnum-selftest comments per the
review's comment-conciseness pass; no behavior change.

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

---------

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

html/cmdguide.html

@@ -473,6 +473,15 @@ then:</small><br>
 lifts the built-in caps; use it only against infrastructure you are allowed to
 load.</small></p>
 
+<h4>Save a WARC archive of the crawl</h4>
+<p><tt>httrack https://example.com/ --warc --path mydir</tt><br>
+<small>Writes a standard WARC/1.1 file (<tt>httrack-&lt;timestamp&gt;.warc.gz</tt>) in
+the project folder alongside the browsable mirror, not instead of it. Set the name
+with <tt>--warc-file NAME</tt> and split a large crawl with <tt>--warc-max-size N</tt>;
+add <tt>--warc-cdx</tt> for a sorted CDXJ index, or <tt>--wacz</tt> to bundle the
+archive, index and pages into one WACZ for replay tools such as
+replayweb.page.</small></p>
+
 <h4>HTTrack as a fetch tool</h4>
 <p><tt>httrack --get https://host/file.bin --path tmp</tt><br>
 <small><tt>--get</tt> fetches one file with cache, index, depth, cookies and robots

html/httrack.man.html

@@ -90,35 +90,34 @@ offline browser : copy websites to a local directory</p>
 ] [ <b>-NN, --structure[=N]</b> ] [ <b>-%N,
 --delayed-type-check</b> ] [ <b>-%D,
 --cached-delayed-type-check</b> ] [ <b>-%M, --mime-html</b>
-] [ <b>-%r, --warc</b> ] [ <b>-LN, --long-names[=N]</b> ] [
-<b>-KN, --keep-links[=N]</b> ] [ <b>-x,
---replace-external</b> ] [ <b>-%x, --disable-passwords</b> ]
-[ <b>-%q, --include-query-string</b> ] [ <b>-%g,
---strip-query</b> ] [ <b>-o, --generate-errors</b> ] [
-<b>-X, --purge-old[=N]</b> ] [ <b>-%p, --preserve</b> ] [
-<b>-%T, --utf8-conversion</b> ] [ <b>-bN, --cookies[=N]</b>
-] [ <b>-%K, --cookies-file</b> ] [ <b>-%Y, --why</b> ] [
-<b>-u, --check-type[=N]</b> ] [ <b>-j, --parse-java[=N]</b>
-] [ <b>-sN, --robots[=N]</b> ] [ <b>-%h, --http-10</b> ] [
-<b>-%k, --keep-alive</b> ] [ <b>-%z,
---disable-compression</b> ] [ <b>-%B, --tolerant</b> ] [
-<b>-%s, --updatehack</b> ] [ <b>-%u, --urlhack</b> ] [
+] [ <b>-LN, --long-names[=N]</b> ] [ <b>-KN,
+--keep-links[=N]</b> ] [ <b>-x, --replace-external</b> ] [
+<b>-%x, --disable-passwords</b> ] [ <b>-%q,
+--include-query-string</b> ] [ <b>-%g, --strip-query</b> ] [
+<b>-o, --generate-errors</b> ] [ <b>-X, --purge-old[=N]</b>
+] [ <b>-%p, --preserve</b> ] [ <b>-%T, --utf8-conversion</b>
+] [ <b>-bN, --cookies[=N]</b> ] [ <b>-%K, --cookies-file</b>
+] [ <b>-%Y, --why</b> ] [ <b>-u, --check-type[=N]</b> ] [
+<b>-j, --parse-java[=N]</b> ] [ <b>-sN, --robots[=N]</b> ] [
+<b>-%h, --http-10</b> ] [ <b>-%k, --keep-alive</b> ] [
+<b>-%z, --disable-compression</b> ] [ <b>-%B, --tolerant</b>
+] [ <b>-%s, --updatehack</b> ] [ <b>-%u, --urlhack</b> ] [
 <b>-%A, --assume</b> ] [ <b>-@iN, --protocol[=N]</b> ] [
 <b>-%w, --disable-module</b> ] [ <b>-F, --user-agent</b> ] [
 <b>-%R, --referer</b> ] [ <b>-%E, --from</b> ] [ <b>-%F,
 --footer</b> ] [ <b>-%l, --language</b> ] [ <b>-%a,
 --accept</b> ] [ <b>-%X, --headers</b> ] [ <b>-C,
 --cache[=N]</b> ] [ <b>-k, --store-all-in-cache</b> ] [
-<b>-%n, --do-not-recatch</b> ] [ <b>-%v, --display</b> ] [
-<b>-Q, --do-not-log</b> ] [ <b>-q, --quiet</b> ] [ <b>-z,
---extra-log</b> ] [ <b>-Z, --debug-log</b> ] [ <b>-v,
---verbose</b> ] [ <b>-f, --file-log</b> ] [ <b>-f2,
---single-log</b> ] [ <b>-I, --index</b> ] [ <b>-%i,
---build-top-index</b> ] [ <b>-%I, --search-index</b> ] [
-<b>-pN, --priority[=N]</b> ] [ <b>-S, --stay-on-same-dir</b>
-] [ <b>-D, --can-go-down</b> ] [ <b>-U, --can-go-up</b> ] [
-<b>-B, --can-go-up-and-down</b> ] [ <b>-a,
---stay-on-same-address</b> ] [ <b>-d,
+<b>-%r, --warc</b> ] [ <b>-%n, --do-not-recatch</b> ] [
+<b>-%v, --display</b> ] [ <b>-Q, --do-not-log</b> ] [ <b>-q,
+--quiet</b> ] [ <b>-z, --extra-log</b> ] [ <b>-Z,
+--debug-log</b> ] [ <b>-v, --verbose</b> ] [ <b>-f,
+--file-log</b> ] [ <b>-f2, --single-log</b> ] [ <b>-I,
+--index</b> ] [ <b>-%i, --build-top-index</b> ] [ <b>-%I,
+--search-index</b> ] [ <b>-pN, --priority[=N]</b> ] [ <b>-S,
+--stay-on-same-dir</b> ] [ <b>-D, --can-go-down</b> ] [
+<b>-U, --can-go-up</b> ] [ <b>-B, --can-go-up-and-down</b> ]
+[ <b>-a, --stay-on-same-address</b> ] [ <b>-d,
 --stay-on-same-domain</b> ] [ <b>-l, --stay-on-same-tld</b>
 ] [ <b>-e, --go-everywhere</b> ] [ <b>-%H,
 --debug-headers</b> ] [ <b>-%!,
@@ -649,18 +648,6 @@ don&rsquo;t wait) (--cached-delayed-type-check)</p></td></tr>
 <td width="4%">
 
 
-<p>-%r</p></td>
-<td width="5%"></td>
-<td width="82%">
-
-
-<p>write an ISO-28500 WARC/1.1 archive; --warc-file NAME
-sets the output name (--warc)</p></td></tr>
-<tr valign="top" align="left">
-<td width="9%"></td>
-<td width="4%">
-
-
 <p>-%t</p></td>
 <td width="5%"></td>
 <td width="82%">
@@ -1142,6 +1129,20 @@ update before) (--cache[=N])</p></td></tr>
 <td width="4%">
 
 
+<p>-%r</p></td>
+<td width="5%"></td>
+<td width="82%">
+
+
+<p>write an ISO-28500 WARC/1.1 archive; --warc-file NAME
+sets the output name, --warc-max-size N rotates segments
+past N bytes, --warc-cdx also writes a sorted CDXJ index,
+--wacz packages it all as a WACZ file (--warc)</p></td></tr>
+<tr valign="top" align="left">
+<td width="9%"></td>
+<td width="4%">
+
+
 <p>-%n</p></td>
 <td width="5%"></td>
 <td width="82%">

man/httrack.1

@@ -3,7 +3,7 @@
 .\"
 .\" This file is generated by man/makeman.sh; do not edit by hand.
 .\" SPDX-License-Identifier: GPL-3.0-or-later
-.TH httrack 1 "22 July 2026" "httrack website copier"
+.TH httrack 1 "23 July 2026" "httrack website copier"
 .SH NAME
 httrack \- offline browser : copy websites to a local directory
 .SH SYNOPSIS
@@ -40,7 +40,6 @@ httrack \- offline browser : copy websites to a local directory
 [ \fB\-%N, \-\-delayed\-type\-check\fR ]
 [ \fB\-%D, \-\-cached\-delayed\-type\-check\fR ]
 [ \fB\-%M, \-\-mime\-html\fR ]
-[ \fB\-%r, \-\-warc\fR ]
 [ \fB\-LN, \-\-long\-names[=N]\fR ]
 [ \fB\-KN, \-\-keep\-links[=N]\fR ]
 [ \fB\-x, \-\-replace\-external\fR ]
@@ -75,6 +74,7 @@ httrack \- offline browser : copy websites to a local directory
 [ \fB\-%X, \-\-headers\fR ]
 [ \fB\-C, \-\-cache[=N]\fR ]
 [ \fB\-k, \-\-store\-all\-in\-cache\fR ]
+[ \fB\-%r, \-\-warc\fR ]
 [ \fB\-%n, \-\-do\-not\-recatch\fR ]
 [ \fB\-%v, \-\-display\fR ]
 [ \fB\-Q, \-\-do\-not\-log\fR ]
@@ -198,8 +198,6 @@ delayed type check, don't make any link test but wait for files download to star
 cached delayed type check, don't wait for remote type during updates, to speedup them (%D0 wait, * %D1 don't wait) (\-\-cached\-delayed\-type\-check)
 .IP \-%M
 generate a RFC MIME\-encapsulated full\-archive (.mht) (\-\-mime\-html)
-.IP \-%r
-write an ISO\-28500 WARC/1.1 archive; \-\-warc\-file NAME sets the output name (\-\-warc)
 .IP \-%t
 keep the original file extension, don't rewrite it from the MIME type (%t0 rewrite)
 .IP \-LN
@@ -279,6 +277,8 @@ additional HTTP header line (\-%X "X\-Magic: 42" (\-\-headers <param>)
 create/use a cache for updates and retries (C0 no cache,C1 cache is prioritary,* C2 test update before) (\-\-cache[=N])
 .IP \-k
 store all files in cache (not useful if files on disk) (\-\-store\-all\-in\-cache)
+.IP \-%r
+write an ISO\-28500 WARC/1.1 archive; \-\-warc\-file NAME sets the output name, \-\-warc\-max\-size N rotates segments past N bytes, \-\-warc\-cdx also writes a sorted CDXJ index, \-\-wacz packages it all as a WACZ file (\-\-warc)
 .IP \-%n
 do not re\-download locally erased files (\-\-do\-not\-recatch)
 .IP \-%v

src/htsalias.c

@@ -116,6 +116,13 @@ const char *hts_optalias[][4] = {
    "load extra cookies from a Netscape cookies.txt"},
   {"warc", "-%r", "single", "write an ISO-28500 WARC/1.1 archive of the crawl"},
   {"warc-file", "-%rf", "param1", "write a WARC archive to the given base name"},
+  {"warc-max-size", "-%rs", "param1",
+   "rotate the WARC archive once a segment passes N bytes (0: single file)"},
+  {"warc-cdx", "-%rc", "single",
+   "write a sorted CDXJ index next to the WARC archive"},
+  {"warc-cdxj", "-%rc", "single", ""},
+  {"wacz", "-%rz", "single",
+   "package the WARC archive, CDXJ index and pages as a WACZ file"},
   {"why", "-%Y", "param1",
    "explain which filter rule accepts or rejects a URL, then exit"},
   {"pause", "-%G", "param1",

src/htsback.c

@@ -750,9 +750,20 @@ int back_finalize(httrackp * opt, cache_back * cache, struct_back * sback,
                     fexist_utf8(back[p].url_sav))
                   filenote(&opt->state.strc, back[p].url_sav, NULL);
               }
+              /* Keep the compressed spool so the WARC record stores the body
+                 verbatim (Content-Encoding preserved) instead of unlinking it.
+               */
+              if (StringNotEmpty(opt->warc_file)) {
+                warc_adopt_rawspool(&back[p].r, back[p].tmpfile);
+                if (back[p].r.warc_rawpath != NULL)
+                  back[p].tmpfile =
+                      NULL; /* adopted: freed via warc_free_request */
+              }
               /* ensure that no remaining temporary file exists */
-              unlink(back[p].tmpfile);
-              back[p].tmpfile = NULL;
+              if (back[p].tmpfile != NULL) {
+                unlink(back[p].tmpfile);
+                back[p].tmpfile = NULL;
+              }
             }
             // stats
             HTS_STAT.total_packed += back[p].compressed_size;
@@ -1062,6 +1073,9 @@ void back_copy_static(const lien_back * src, lien_back * dst) {
   dst->r.headers = NULL;
   dst->r.warc_reqhdr = NULL;
   dst->r.warc_resphdr = NULL;
+  dst->r.warc_rawpath =
+      NULL; /* the spool stays owned by src (no double-unlink) */
+  dst->r.warc_truncated = 0;
   dst->r.out = NULL;
   dst->r.location = dst->location_buffer;
   dst->r.fp = NULL;
@@ -1127,6 +1141,8 @@ int back_unserialize(FILE * fp, lien_back ** dst) {
     (*dst)->r.out = NULL;
     (*dst)->r.warc_reqhdr = NULL;
     (*dst)->r.warc_resphdr = NULL;
+    (*dst)->r.warc_rawpath = NULL;
+    (*dst)->r.warc_truncated = 0;
     (*dst)->r.location = (*dst)->location_buffer;
     (*dst)->r.fp = NULL;
     (*dst)->r.soc = INVALID_SOCKET;
@@ -2519,6 +2535,19 @@ void back_wait(struct_back * sback, httrackp * opt, cache_back * cache,
 
     for (i = 0; i < (unsigned int) back_max; i++) {
       if (back[i].status > 0 && back[i].status < STATUS_FTP_TRANSFER) {
+        /* A cap-truncated body is deliberate, not broken: archive what arrived
+           with WARC-Truncated before the abort overwrites the slot's real 2xx
+           status. HTTrack still treats the slot as incomplete afterwards. */
+        if (StringNotEmpty(opt->warc_file) && back[i].r.statuscode > 0 &&
+            back[i].r.warc_resphdr != NULL && back[i].r.size > 0 &&
+            !(back[i].r.is_write && IS_DELAYED_EXT(back[i].url_sav))) {
+          if (back[i].r.is_write && back[i].r.out != NULL)
+            fflush(back[i].r.out);
+          back[i].r.warc_truncated = (limit == HTS_MIRROR_LIMIT_SIZE)
+                                         ? WARC_TRUNC_LENGTH
+                                         : WARC_TRUNC_TIME;
+          warc_write_backtransaction(opt, &back[i]);
+        }
         if (back[i].r.soc != INVALID_SOCKET) {
           deletehttp(&back[i].r);
         }

src/htscore.c

@@ -699,7 +699,7 @@ int httpmirror(char *url1, httrackp * opt) {
     int primary_len = 8192;
 
     if (StringNotEmpty(opt->filelist)) {
-      primary_len += max(0, fsize(StringBuff(opt->filelist)) * 2);
+      primary_len += max(0, fsize_utf8(StringBuff(opt->filelist)) * 2);
     }
     primary_len += (int) strlen(url1) * 2;
 
@@ -856,19 +856,19 @@ int httpmirror(char *url1, httrackp * opt) {
       char *filelist_buff = NULL;
       size_t filelist_sz = 0;
       const char *filelist_err = NULL; /* failure reason, NULL on success */
-      const LLint fs = fsize(StringBuff(opt->filelist));
+      const LLint fs = fsize_utf8(StringBuff(opt->filelist));
 
       if (fs < 0) {
         /* fsize() hides the cause; redo stat() for a precise errno (#49) */
-        struct stat st;
-        filelist_err = stat(StringBuff(opt->filelist), &st) != 0
+        STRUCT_STAT st;
+        filelist_err = STAT(StringBuff(opt->filelist), &st) != 0
                            ? strerror(errno)
                            : "not a regular file";
       } else if ((filelist_sz = llint_to_size_t(fs)) == (size_t) -1) {
         filelist_err = "file too large";
         filelist_sz = 0;
       } else {
-        FILE *fp = fopen(StringBuff(opt->filelist), "rb");
+        FILE *fp = FOPEN(StringBuff(opt->filelist), "rb");
 
         if (fp == NULL) {
           filelist_err = strerror(errno);
@@ -976,10 +976,9 @@ int httpmirror(char *url1, httrackp * opt) {
   }
   // statistiques
   if (opt->makestat) {
-    makestat_fp =
-      fopen(fconcat
-            (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-stats.txt"),
-            "wb");
+    makestat_fp = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                StringBuff(opt->path_log), "hts-stats.txt"),
+                        "wb");
     if (makestat_fp != NULL) {
       fprintf(makestat_fp, "HTTrack statistics report, every minutes" LF LF);
       fflush(makestat_fp);
@@ -987,10 +986,9 @@ int httpmirror(char *url1, httrackp * opt) {
   }
   // tracking -- débuggage
   if (opt->maketrack) {
-    maketrack_fp =
-      fopen(fconcat
-            (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-track.txt"),
-            "wb");
+    maketrack_fp = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                 StringBuff(opt->path_log), "hts-track.txt"),
+                         "wb");
     if (maketrack_fp != NULL) {
       fprintf(maketrack_fp, "HTTrack tracking report, every minutes" LF LF);
       fflush(maketrack_fp);
@@ -1639,7 +1637,8 @@ int httpmirror(char *url1, httrackp * opt) {
 
           /* Remove file if being processed */
           if (is_loaded_from_file) {
-            (void) unlink(fconv(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), savename()));
+            (void) UNLINK(
+                fconv(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), savename()));
             is_loaded_from_file = 0;
           }
 
@@ -1952,8 +1951,8 @@ int httpmirror(char *url1, httrackp * opt) {
            } else */
 
         /* External modules */
-        if (opt->parsejava && (opt->parsejava & HTSPARSE_NO_CLASS) == 0
-            && fexist(savename())) {
+        if (opt->parsejava && (opt->parsejava & HTSPARSE_NO_CLASS) == 0 &&
+            fexist_utf8(savename())) {
           char BIGSTK buff_err_msg[1024];
           htsmoduleStruct BIGSTK str;
 
@@ -1994,7 +1993,7 @@ int httpmirror(char *url1, httrackp * opt) {
       }                         // text/html ou autre
 
       /* Post-processing */
-      if (fexist(savename())) {
+      if (fexist_utf8(savename())) {
         usercommand(opt, 0, NULL, savename(), urladr(), urlfil());
       }
 
@@ -2085,18 +2084,16 @@ int httpmirror(char *url1, httrackp * opt) {
       //
       opt->state._hts_in_html_parsing = 3;
       //
-      old_lst =
-        fopen(fconcat
-              (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-               "hts-cache/old.lst"), "rb");
+      old_lst = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                              StringBuff(opt->path_log), "hts-cache/old.lst"),
+                      "rb");
       if (old_lst) {
-        const size_t sz = llint_to_size_t(
-            fsize(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                          StringBuff(opt->path_log), "hts-cache/new.lst")));
-        new_lst =
-          fopen(fconcat
-                (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                 "hts-cache/new.lst"), "rb");
+        const size_t sz = llint_to_size_t(fsize_utf8(
+            fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                    StringBuff(opt->path_log), "hts-cache/new.lst")));
+        new_lst = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                StringBuff(opt->path_log), "hts-cache/new.lst"),
+                        "rb");
         if (new_lst != NULL && sz != (size_t) -1) {
           /* +1 for the NUL below: new.lst is read raw, and the strstr()
              that follows needs a terminated C string. */
@@ -2116,9 +2113,9 @@ int httpmirror(char *url1, httrackp * opt) {
                   strcpybuff(file, StringBuff(opt->path_html));
                   strcatbuff(file, line + 1);
                   file[strlen(file) - 1] = '\0';
-                  if (fexist(file)) {   // toujours sur disque: virer
+                  if (fexist_utf8(file)) { // toujours sur disque: virer
                     hts_log_print(opt, LOG_INFO, "Purging %s", file);
-                    remove(file);
+                    UNLINK(file);
                     purge = 1;
                   }
                 }
@@ -2139,7 +2136,8 @@ int httpmirror(char *url1, httrackp * opt) {
 
                       strcpybuff(file, StringBuff(opt->path_html));
                       strcatbuff(file, line + 1);
-                      while((strnotempty(file)) && (rmdir(file) == 0)) {        // ok, éliminé (existait)
+                      while ((strnotempty(file)) &&
+                             (RMDIR(file) == 0)) { // ok, éliminé (existait)
                         purge = 1;
                         if (opt->log) {
                           hts_log_print(opt, LOG_INFO, "Purging directory %s/",
@@ -2962,8 +2960,8 @@ static void postprocess_file(httrackp *opt, const char *save, const char *adr,
     if (adr != NULL && strcmp(adr, "primary") == 0) {
       adr = NULL;
     }
-    if (save != NULL && opt != NULL && adr != NULL && adr[0]
-        && strnotempty(save) && fexist(save)) {
+    if (save != NULL && opt != NULL && adr != NULL && adr[0] &&
+        strnotempty(save) && fexist_utf8(save)) {
       const char *rsc_save = save;
       const char *rsc_fil = strrchr(fil, '/');
       int n;
@@ -2985,13 +2983,11 @@ static void postprocess_file(httrackp *opt, const char *save, const char *adr,
 
       if (!opt->state.mimehtml_created) {
         opt->state.mimefp =
-          fopen(fconcat
-                (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                StringBuff(opt->path_html), "index.mht"),
-                "wb");
-        (void) unlink(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-          StringBuff(opt->path_html),
-                              "index.eml"));
+            FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                          StringBuff(opt->path_html), "index.mht"),
+                  "wb");
+        (void) UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                              StringBuff(opt->path_html), "index.eml"));
 #ifndef _WIN32
         if (symlink("index.mht",
                     fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_html),
@@ -3632,6 +3628,9 @@ HTSEXT_API int copy_htsopt(const httrackp * from, httrackp * to) {
 
   if (StringNotEmpty(from->warc_file))
     StringCopyS(to->warc_file, from->warc_file);
+  to->warc_max_size = from->warc_max_size;
+  to->warc_cdx = from->warc_cdx;
+  to->warc_wacz = from->warc_wacz;
 
   if (from->pause_max_ms > 0) {
     to->pause_min_ms = from->pause_min_ms;

src/htscoremain.c

@@ -442,11 +442,10 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
           || (strnotempty(StringBuff(opt->path_html))))
         loops++;                // do not loop once again and do not include rc file (O option exists)
       else {
-        if ((!fexist
-             (fconcat
-              (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-              StringBuff(opt->path_log),
-               "hts-cache/doit.log"))) || (argv_url > 0)) {
+        if ((!fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                  StringBuff(opt->path_log),
+                                  "hts-cache/doit.log"))) ||
+            (argv_url > 0)) {
           if (!optinclude_file(
                   fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
                           StringBuff(opt->path_log), HTS_HTTRACKRC),
@@ -473,16 +472,12 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
   }                             // traiter -O
 
   /* load doit.log and insert in current command line */
-  if (fexist
-      (fconcat
-       (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-       StringBuff(opt->path_log), "hts-cache/doit.log"))
-      && (argv_url <= 0)) {
-    FILE *fp =
-      fopen(fconcat
-            (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-            StringBuff(opt->path_log),
-             "hts-cache/doit.log"), "rb");
+  if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                          StringBuff(opt->path_log), "hts-cache/doit.log")) &&
+      (argv_url <= 0)) {
+    FILE *fp = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/doit.log"),
+                     "rb");
     if (fp) {
       int insert_after = 1;     /* insérer après nom au début */
 
@@ -531,23 +526,16 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
 
   /* Interrupted mirror detected */
   if (!opt->quiet) {
-    if (fexist
-        (fconcat
-         (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-         StringBuff(opt->path_log),
-          "hts-in_progress.lock"))) {
+    if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                            StringBuff(opt->path_log),
+                            "hts-in_progress.lock"))) {
       /* Old cache */
-      if ((fexist
-           (fconcat
-            (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-            StringBuff(opt->path_log),
-             "hts-cache/old.dat")))
-          &&
-          (fexist
-           (fconcat
-            (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-            StringBuff(opt->path_log),
-             "hts-cache/old.ndx")))) {
+      if ((fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                               StringBuff(opt->path_log),
+                               "hts-cache/old.dat"))) &&
+          (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                               StringBuff(opt->path_log),
+                               "hts-cache/old.ndx")))) {
         if (opt->log != NULL) {
           fprintf(opt->log, "Warning!\n");
           fprintf(opt->log,
@@ -576,111 +564,82 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
         if (argv[i][1] == '-') {        // --xxx
           if ((strfield2(argv[i] + 2, "clean")) || (strfield2(argv[i] + 2, "tide"))) {  // nettoyer
             argv[i][1] = '\0';
-            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"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_html), "index.html")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_html),
-                      "index.html"));
+            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"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_html), "index.html")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_html), "index.html"));
             /* */
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/new.zip")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/new.zip"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/old.zip")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/old.zip"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/new.dat")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/new.dat"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/new.ndx")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/new.ndx"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/old.dat")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/old.dat"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/old.ndx")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/old.ndx"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/new.lst")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/new.lst"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/old.lst")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/old.lst"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/new.txt")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/new.txt"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/old.txt")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/old.txt"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-cache/doit.log")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-cache/doit.log"));
-            if (fexist
-                (fconcat
-                 (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                  "hts-in_progress.lock")))
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                      "hts-in_progress.lock"));
-            rmdir(fconcat
-                  (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-cache"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/new.zip")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/new.zip"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/old.zip")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/old.zip"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/new.dat")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/new.dat"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/new.ndx")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/new.ndx"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/old.dat")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/old.dat"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/old.ndx")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/old.ndx"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/new.lst")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/new.lst"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/old.lst")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/old.lst"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/new.txt")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/new.txt"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/old.txt")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/old.txt"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-cache/doit.log")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/doit.log"));
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-in_progress.lock")))
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log),
+                             "hts-in_progress.lock"));
+            RMDIR(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                          StringBuff(opt->path_log), "hts-cache"));
             //
           } else if (strfield2(argv[i] + 2, "catchurl")) {      // capture d'URL via proxy temporaire!
             argv_url = 1;       // forcer a passer les parametres
@@ -737,29 +696,27 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
 #endif
   if (argv_url == 0) {
     // Présence d'un cache, que faire?..
-    if ((fexist
-         (fconcat
-          (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), 
-          StringBuff(opt->path_log), "hts-cache/new.zip")))
-        ||
-        (fexist
-         (fconcat
-          (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), 
-          StringBuff(opt->path_log), "hts-cache/new.dat"))
-         &&
-         fexist(fconcat
-                (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                StringBuff(opt->path_log),
-                 "hts-cache/new.ndx")))
-      ) {                       // il existe déja un cache précédent.. renommer
-      if (fexist(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-        StringBuff(opt->path_log), "hts-cache/doit.log"))) {        // un cache est présent
+    if ((fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log),
+                             "hts-cache/new.zip"))) ||
+        (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-cache/new.dat")) &&
+         fexist_utf8(
+             fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                     StringBuff(opt->path_log),
+                     "hts-cache/new.ndx")))) { // il existe déja un cache
+                                               // précédent.. renommer
+      if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                              StringBuff(opt->path_log),
+                              "hts-cache/doit.log"))) { // un cache est présent
         if (x_argvblk != NULL) {
           int m;
 
           // établir mode - mode cache: 1 (cache valide) 2 (cache à vérifier)
-          if (fexist(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-            StringBuff(opt->path_log), "hts-in_progress.lock"))) {  // cache prioritaire
+          if (fexist_utf8(
+                  fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                          StringBuff(opt->path_log),
+                          "hts-in_progress.lock"))) { // cache prioritaire
             m = 1;
           } else {
             m = 2;
@@ -794,7 +751,7 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
           htsmain_free();
           return -1;
         }
-      } else {                  // log existe pas
+      } else { // log existe pas
         HTS_PANIC_PRINTF("A cache has been found, but no command line");
         printf
           ("Please launch httrack with proper parameters to reuse the cache\n");
@@ -802,7 +759,7 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
         return -1;
       }
 
-    } else {                    // aucune URL définie et pas de cache
+    } else { // aucune URL définie et pas de cache
       if (opt->quiet) {
         help(argv[0], !opt->quiet);
         htsmain_free();
@@ -817,26 +774,21 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
     }
   } else {                      // plus de 2 paramètres
     // un fichier log existe?
-    if (fexist(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                       StringBuff(opt->path_log),
-                       "hts-in_progress.lock"))) { // fichier lock?
+    if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                            StringBuff(opt->path_log),
+                            "hts-in_progress.lock"))) { // fichier lock?
 
       opt->cache = HTS_CACHE_PRIORITY; // cache prioritaire
       if (opt->quiet == 0) {
-        if ((fexist
-             (fconcat
-              (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-               "hts-cache/new.zip")))
-            ||
-            (fexist
-             (fconcat
-              (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-               "hts-cache/new.dat"))
-             &&
-             fexist(fconcat
-                    (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                     "hts-cache/new.ndx")))
-          ) {
+        if ((fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                 StringBuff(opt->path_log),
+                                 "hts-cache/new.zip"))) ||
+            (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                 StringBuff(opt->path_log),
+                                 "hts-cache/new.dat")) &&
+             fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                 StringBuff(opt->path_log),
+                                 "hts-cache/new.ndx")))) {
           HT_REQUEST_START;
           HT_PRINT("There is a lock-file in the directory ");
           HT_PRINT(StringBuff(opt->path_log));
@@ -850,24 +802,19 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
           }
         }
       }
-    } else if (fexist(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                              StringBuff(opt->path_html), "index.html"))) {
+    } else if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                   StringBuff(opt->path_html), "index.html"))) {
       opt->cache = HTS_CACHE_TEST_UPDATE;
       if (opt->quiet == 0) {
-        if ((fexist
-             (fconcat
-              (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-               "hts-cache/new.zip")))
-            ||
-            (fexist
-             (fconcat
-              (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-               "hts-cache/new.dat"))
-             &&
-             fexist(fconcat
-                    (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                     "hts-cache/new.ndx")))
-          ) {
+        if ((fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                 StringBuff(opt->path_log),
+                                 "hts-cache/new.zip"))) ||
+            (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                 StringBuff(opt->path_log),
+                                 "hts-cache/new.dat")) &&
+             fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                 StringBuff(opt->path_log),
+                                 "hts-cache/new.ndx")))) {
           HT_REQUEST_START;
           HT_PRINT
             ("There is an index.html and a hts-cache folder in the directory ");
@@ -1500,13 +1447,13 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
                   LLint fz;
 
                   na++;
-                  fz = fsize(argv[na]);
+                  fz = fsize_utf8(argv[na]);
                   if (fz < 0) {
                     HTS_PANIC_PRINTF("File url list could not be opened");
                     htsmain_free();
                     return -1;
                   } else {
-                    FILE *fp = fopen(argv[na], "rb");
+                    FILE *fp = FOPEN(argv[na], "rb");
 
                     if (fp != NULL) {
                       int cl = (int) strlen(url);
@@ -1806,6 +1753,34 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
                     return -1;
                   }
                   StringCopy(opt->warc_file, argv[na]);
+                } else if (*(com + 1) == 's') { // --warc-max-size N: rotation
+                  com++;
+                  if ((na + 1 >= argc) || (argv[na + 1][0] == '-')) {
+                    HTS_PANIC_PRINTF(
+                        "Option warc-max-size needs a blank space and a size");
+                    htsmain_free();
+                    return -1;
+                  }
+                  na++;
+                  { // reject non-numeric/negative/overflow; keep default 0
+                    // (single file)
+                    char *end;
+                    LLint v;
+                    errno = 0;
+                    v = strtoll(argv[na], &end, 10);
+                    if (isdigit((unsigned char) argv[na][0]) && *end == '\0' &&
+                        errno != ERANGE)
+                      opt->warc_max_size = v;
+                  }
+                } else if (*(com + 1) == 'c') { // --warc-cdx: sorted CDXJ index
+                  com++;
+                  opt->warc_cdx = 1;
+                } else if (*(com + 1) == 'z') { // --wacz: WACZ package
+                  com++;
+                  opt->warc_wacz = 1;
+                  opt->warc_cdx = 1; // WACZ embeds the CDXJ index
+                  if (!StringNotEmpty(opt->warc_file))
+                    StringCopy(opt->warc_file, WARC_AUTONAME);
                 } else { // --warc: auto-named archive under the output dir
                   StringCopy(opt->warc_file, WARC_AUTONAME);
                 }
@@ -2022,7 +1997,7 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
                                      /*cache */ &cache, /*hash */ NULL, /*ptr */
                                      0, /*numero_passe */ 0, /*mime_type */
                                      NULL) != -1) {
-                                  if (fexist(afs.save)) {
+                                  if (fexist_utf8(afs.save)) {
                                     fprintf(stdout, "Content-location: %s\r\n",
                                             afs.save);
                                   }
@@ -2120,18 +2095,16 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
                   uLong repaired = 0;
                   uLong repairedBytes = 0;
 
-                  if (fexist
-                      (fconcat
-                       (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                        "hts-cache/new.zip"))) {
+                  if (fexist_utf8(fconcat(
+                          OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                          StringBuff(opt->path_log), "hts-cache/new.zip"))) {
                     name =
                       fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
                               "hts-cache/new.zip");
-                  } else
-                    if (fexist
-                        (fconcat
-                         (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                          "hts-cache/old.zip"))) {
+                  } else if (fexist_utf8(fconcat(OPT_GET_BUFF(opt),
+                                                 OPT_GET_BUFF_SIZE(opt),
+                                                 StringBuff(opt->path_log),
+                                                 "hts-cache/old.zip"))) {
                     name =
                       fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
                               "hts-cache/old.zip");
@@ -2150,10 +2123,11 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
                        fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
                                "hts-cache/repair.tmp"), &repaired,
                        &repairedBytes) == Z_OK) {
-                    unlink(name);
-                    rename(fconcat
-                           (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                            "hts-cache/repair.zip"), name);
+                    UNLINK(name);
+                    RENAME(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                   StringBuff(opt->path_log),
+                                   "hts-cache/repair.zip"),
+                           name);
                     fprintf(stderr,
                             "Cache: %d bytes successfully recovered in %d entries\n",
                             (int) repairedBytes, (int) repaired);
@@ -2371,10 +2345,9 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
   hts_cache_reconcile(opt, CACHE_RECONCILE_INTERRUPTED);
   // Débuggage des en têtes
   if (_DEBUG_HEAD) {
-    ioinfo =
-      fopen(fconcat
-            (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-ioinfo.txt"),
-            "wb");
+    ioinfo = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                           StringBuff(opt->path_log), "hts-ioinfo.txt"),
+                   "wb");
   }
 
   {
@@ -2449,9 +2422,9 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
       /* readme for information purpose */
       {
         FILE *fp =
-          fopen(fconcat
-                (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                 "hts-cache/readme.txt"), "wb");
+            FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                          StringBuff(opt->path_log), "hts-cache/readme.txt"),
+                  "wb");
         if (fp) {
           fprintf(fp, "What's in this folder?" LF);
           fprintf(fp, "" LF);
@@ -2503,17 +2476,16 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
         int i;
 
 #ifdef _WIN32
-        mkdir(fconcat
-              (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-cache"));
+        hts_mkdir_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                               StringBuff(opt->path_log), "hts-cache"));
 #else
         mkdir(fconcat
               (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), "hts-cache"),
               HTS_PROTECT_FOLDER);
 #endif
-        fp =
-          fopen(fconcat
-                (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log),
-                 "hts-cache/doit.log"), "wb");
+        fp = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                           StringBuff(opt->path_log), "hts-cache/doit.log"),
+                   "wb");
         if (fp) {
           for(i = 0 + 1; i < argc; i++) {
             if (((strchr(argv[i], ' ') != NULL)
@@ -2558,7 +2530,7 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
         }
       }
       // petit message dans le lock
-      if ((fp = fopen(n_lock, "wb")) != NULL) {
+      if ((fp = FOPEN(n_lock, "wb")) != NULL) {
         int i;
 
         fprintf(fp, "Mirror in progress since %s .. please wait!" LF, t);
@@ -2718,16 +2690,15 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
           char *f = OPT_GET_BUFF(opt);
 
           sprintf(f, "%s/%s", CACHE_REFNAME, entry->d_name);
-          (void)
-            unlink(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), f));
+          (void) UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                StringBuff(opt->path_log), f));
         }
       }
       if (dir != NULL) {
         (void) closedir(dir);
       }
-      (void)
-        rmdir(fconcat
-              (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), StringBuff(opt->path_log), CACHE_REFNAME));
+      (void) RMDIR(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                           StringBuff(opt->path_log), CACHE_REFNAME));
     }
 
     /* Info for wrappers */
@@ -2755,7 +2726,7 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
       }
     }
     // supprimer lock
-    remove(n_lock);
+    UNLINK(n_lock);
   }
 
   if (x_argvblk)

src/htshelp.c

@@ -412,9 +412,9 @@ void help_catchurl(const char *dest_path) {
       do {
         snprintf(dest, sizeof(dest), "%s%s%d", dest_path, "hts-post", i);
         i++;
-      } while(fexist(dest));
+      } while (fexist_utf8(dest));
       {
-        FILE *fp = fopen(dest, "wb");
+        FILE *fp = FOPEN(dest, "wb");
 
         if (fp) {
           fwrite(data, strlen(data), 1, fp);
@@ -524,8 +524,6 @@ void help(const char *app, int more) {
   infomsg
     (" %D  cached delayed type check, don't wait for remote type during updates, to speedup them (%D0 wait, * %D1 don't wait)");
   infomsg(" %M  generate a RFC MIME-encapsulated full-archive (.mht)");
-  infomsg(" %r  write an ISO-28500 WARC/1.1 archive; --warc-file NAME sets the "
-          "output name");
   infomsg(" %t  keep the original file extension, don't rewrite it from the "
           "MIME type (%t0 rewrite)");
   infomsg
@@ -595,6 +593,10 @@ void help(const char *app, int more) {
   infomsg
     ("  C  create/use a cache for updates and retries (C0 no cache,C1 cache is prioritary,* C2 test update before)");
   infomsg("  k  store all files in cache (not useful if files on disk)");
+  infomsg(" %r  write an ISO-28500 WARC/1.1 archive; --warc-file NAME sets the "
+          "output name, --warc-max-size N rotates segments past N bytes, "
+          "--warc-cdx also writes a sorted CDXJ index, --wacz packages it all "
+          "as a WACZ file");
   infomsg(" %n  do not re-download locally erased files");
   infomsg
     (" %v  display on screen filenames downloaded (in realtime) - * %v1 short version - %v2 full animation");

src/htsindex.c

@@ -349,9 +349,13 @@ void index_finish(const char *indexpath, int mode) {
 
             // Write new file
             if (mode == 1)      // TEXT
-              fp = fopen(concat(catbuff, sizeof(catbuff), indexpath, "index.txt"), "wb");
+              fp = FOPEN(
+                  concat(catbuff, sizeof(catbuff), indexpath, "index.txt"),
+                  "wb");
             else                // HTML
-              fp = fopen(concat(catbuff, sizeof(catbuff), indexpath, "sindex.html"), "wb");
+              fp = FOPEN(
+                  concat(catbuff, sizeof(catbuff), indexpath, "sindex.html"),
+                  "wb");
             if (fp) {
               char current_word[KEYW_LEN + 32];
               char word[KEYW_LEN + 32];

src/htslib.c

@@ -6019,6 +6019,7 @@ HTSEXT_API httrackp *hts_create_opt(void) {
   StringCopy(opt->strip_query, "");
   StringCopy(opt->cookies_file, "");
   StringCopy(opt->warc_file, "");
+  opt->warc_max_size = 0; /* no rotation unless --warc-max-size sets it */
   StringCopy(opt->why_url, "");
   opt->pause_min_ms = 0;
   opt->pause_max_ms = 0;
@@ -6430,21 +6431,29 @@ HTSEXT_API int hts_resetvar(void) {
 #ifdef _WIN32
 
 typedef struct dirent dirent;
+static LPWSTR hts_pathToUCS2(const char *path);
+
 DIR *opendir(const char *name) {
   WIN32_FILE_ATTRIBUTE_DATA st;
   DIR *dir;
   size_t len;
   int i;
+  LPWSTR wname;
 
   if (name == NULL || *name == '\0') {
     errno = ENOENT;
     return NULL;
   }
-  if (!GetFileAttributesEx(name, GetFileExInfoStandard, &st)
-      || (st.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) == 0) {
+  // Wide \\?\ path: no MAX_PATH cap, no CP_ACP mis-decode (#133,#630).
+  wname = hts_pathToUCS2(name);
+  if (wname == NULL ||
+      !GetFileAttributesExW(wname, GetFileExInfoStandard, &st) ||
+      (st.dwFileAttributes & FILE_ATTRIBUTE_DIRECTORY) == 0) {
+    freet(wname);
     errno = ENOENT;
     return NULL;
   }
+  freet(wname);
   dir = calloc(sizeof(DIR), 1);
   if (dir == NULL) {
     errno = ENOMEM;
@@ -6464,19 +6473,29 @@ DIR *opendir(const char *name) {
 }
 
 struct dirent *readdir(DIR * dir) {
-  WIN32_FIND_DATAA find;
+  WIN32_FIND_DATAW find;
 
   if (dir->h == INVALID_HANDLE_VALUE) {
-    dir->h = FindFirstFileA(dir->name, &find);
+    // \\?\-prefix so a long/non-ASCII directory enumerates instead of ENOENT.
+    LPWSTR wname = hts_pathToUCS2(dir->name);
+
+    dir->h =
+        wname != NULL ? FindFirstFileW(wname, &find) : INVALID_HANDLE_VALUE;
+    freet(wname);
   } else {
-    if (!FindNextFile(dir->h, &find)) {
+    if (!FindNextFileW(dir->h, &find)) {
       FindClose(dir->h);
       dir->h = INVALID_HANDLE_VALUE;
     }
   }
   if (dir->h != INVALID_HANDLE_VALUE) {
+    char *u = hts_convertUCS2StringToUTF8(find.cFileName, -1);
+
     dir->entry.d_name[0] = 0;
-    strncat(dir->entry.d_name, find.cFileName, HTS_DIRENT_SIZE - 1);
+    if (u != NULL) {
+      strncat(dir->entry.d_name, u, HTS_DIRENT_SIZE - 1);
+      freet(u);
+    }
     return &dir->entry;
   }
   errno = ENOENT;
@@ -6509,9 +6528,61 @@ static void copyWchar(LPWSTR dest, const char *src) {
   dest[i] = '\0';
 }
 
+/* UTF-8 path -> UCS-2 for the _w* file APIs. At/above HTS_WIN_LONGPATH_MIN,
+   \\?\-prefix it via GetFullPathNameW to clear MAX_PATH (#133); else unchanged.
+   Any prefixing failure falls back to the plain converted path. */
+#define HTS_WIN_LONGPATH_MIN 240 /* stay clear of MAX_PATH (260) */
+
+static LPWSTR hts_pathToUCS2(const char *path) {
+  LPWSTR wpath = hts_convertUTF8StringToUCS2(path, (int) strlen(path), NULL);
+
+  if (wpath == NULL) {
+    return NULL;
+  }
+  const size_t len = wcslen(wpath);
+  // Already "\\?\" or "\\.\": don't re-prefix.
+  const int verbatim = len >= 4 && wpath[0] == L'\\' && wpath[1] == L'\\' &&
+                       (wpath[2] == L'?' || wpath[2] == L'.') &&
+                       wpath[3] == L'\\';
+
+  if (len < HTS_WIN_LONGPATH_MIN || verbatim) {
+    return wpath;
+  }
+
+  const DWORD need = GetFullPathNameW(wpath, 0, NULL, NULL); /* incl NUL */
+  LPWSTR full = need != 0 ? malloct((size_t) need * sizeof(WCHAR)) : NULL;
+
+  if (full == NULL) {
+    return wpath; /* fall back to the plain path */
+  }
+  const DWORD written = GetFullPathNameW(wpath, need, full, NULL);
+
+  if (written == 0 || written >= need || full[0] == L'\0') {
+    freet(full);
+    return wpath;
+  }
+
+  const int isUNC = full[0] == L'\\' && full[1] == L'\\';
+  // UNC "\\srv\share" -> "\\?\UNC\srv\share": the prefix subsumes the "\\".
+  const WCHAR *const pfx = isUNC ? L"\\\\?\\UNC\\" : L"\\\\?\\";
+  const WCHAR *const body = isUNC ? full + 2 : full;
+  const size_t pfxLen = wcslen(pfx), bodyLen = wcslen(body);
+  LPWSTR out = malloct((pfxLen + bodyLen + 1) * sizeof(WCHAR));
+
+  if (out == NULL) {
+    freet(full);
+    return wpath;
+  }
+  memcpybuff(out, pfx, pfxLen * sizeof(WCHAR));
+  memcpybuff(out + pfxLen, body, (bodyLen + 1) * sizeof(WCHAR));
+  freet(full);
+  freet(wpath);
+  return out;
+}
+
 FILE *hts_fopen_utf8(const char *path, const char *mode) {
   WCHAR wmode[32];
-  LPWSTR wpath = hts_convertUTF8StringToUCS2(path, (int) strlen(path), NULL);
+  LPWSTR wpath = hts_pathToUCS2(path);
 
   assertf(strlen(mode) < sizeof(wmode) / sizeof(WCHAR));
   copyWchar(wmode, mode);
@@ -6527,7 +6598,7 @@ FILE *hts_fopen_utf8(const char *path, const char *mode) {
 }
 
 int hts_stat_utf8(const char *path, STRUCT_STAT * buf) {
-  LPWSTR wpath = hts_convertUTF8StringToUCS2(path, (int) strlen(path), NULL);
+  LPWSTR wpath = hts_pathToUCS2(path);
 
   if (wpath != NULL) {
     const int result = _wstat64(wpath, buf);
@@ -6541,7 +6612,7 @@ int hts_stat_utf8(const char *path, STRUCT_STAT * buf) {
 }
 
 int hts_unlink_utf8(const char *path) {
-  LPWSTR wpath = hts_convertUTF8StringToUCS2(path, (int) strlen(path), NULL);
+  LPWSTR wpath = hts_pathToUCS2(path);
 
   if (wpath != NULL) {
     const int result = _wunlink(wpath);
@@ -6555,10 +6626,8 @@ int hts_unlink_utf8(const char *path) {
 }
 
 int hts_rename_utf8(const char *oldpath, const char *newpath) {
-  LPWSTR woldpath =
-    hts_convertUTF8StringToUCS2(oldpath, (int) strlen(oldpath), NULL);
-  LPWSTR wnewpath =
-    hts_convertUTF8StringToUCS2(newpath, (int) strlen(newpath), NULL);
+  LPWSTR woldpath = hts_pathToUCS2(oldpath);
+  LPWSTR wnewpath = hts_pathToUCS2(newpath);
   if (woldpath != NULL && wnewpath != NULL) {
     const int result = _wrename(woldpath, wnewpath);
 
@@ -6575,8 +6644,22 @@ int hts_rename_utf8(const char *oldpath, const char *newpath) {
   }
 }
 
+int hts_rmdir_utf8(const char *path) {
+  LPWSTR wpath = hts_pathToUCS2(path);
+
+  if (wpath != NULL) {
+    const int result = _wrmdir(wpath);
+
+    free(wpath);
+    return result;
+  } else {
+    // Fallback on conversion error.
+    return _rmdir(path);
+  }
+}
+
 int hts_mkdir_utf8(const char *path) {
-  LPWSTR wpath = hts_convertUTF8StringToUCS2(path, (int) strlen(path), NULL);
+  LPWSTR wpath = hts_pathToUCS2(path);
 
   if (wpath != NULL) {
     const int result = _wmkdir(wpath);
@@ -6591,7 +6674,7 @@ int hts_mkdir_utf8(const char *path) {
 
 HTSEXT_API int hts_utime_utf8(const char *path, const STRUCT_UTIMBUF * times) {
   STRUCT_UTIMBUF mtimes = *times;
-  LPWSTR wpath = hts_convertUTF8StringToUCS2(path, (int) strlen(path), NULL);
+  LPWSTR wpath = hts_pathToUCS2(path);
 
   if (wpath != NULL) {
     const int result = _wutime(wpath, &mtimes);

src/htslib.h

@@ -609,7 +609,9 @@ static HTS_UNUSED size_t llint_to_size_t(LLint o) {
 
 /* dirent() compatibility */
 #ifdef _WIN32
-#define HTS_DIRENT_SIZE 256
+/* Holds a UTF-8 d_name: MAX_PATH (260) UTF-16 units expand to <=3 bytes each.
+   Windows-only struct, ABI free to break. */
+#define HTS_DIRENT_SIZE 1024
 struct dirent {
   ino_t d_ino;                  /* ignored */
   off_t d_off;                  /* ignored */

src/htsname.c

@@ -47,6 +47,7 @@ Please visit our Website: http://www.httrack.com
 #include "htszlib.h"
 #endif
 #include <ctype.h>
+#include <limits.h>
 
 #define ADD_STANDARD_PATH \
     {  /* ajout nom */\
@@ -1475,15 +1476,37 @@ int url_savename(lien_adrfilsave *const afs,
                   sizeof(afs->save) - (size_t) (lastDot - afs->save));
     }
   }
-  // enforce 260-character path limit before inserting destination path
-  // note: 12 characters at least for WIN32, and 12 for ".99.delayed"
-  // (MSDN) "When using an API to create a directory, the specified path 
-  // cannot be so long that you cannot append an 8.3 file name 
-  // (that is, the directory name cannot exceed MAX_PATH minus 12)."
-#define HTS_MAX_PATH_LEN ( 260 - 12 - 12 )
+  // Cap the save path: the final parent+name is copied into a fixed buffer that
+  // aborts() on overflow (htssafe.h), so clamp every ceiling to fit it.
+#define HTS_SAVE_BUFSIZE (HTS_URLMAXSIZE * 2) /* sizeof(afs->save) */
+#define HTS_PATH_TAIL_RESERVE 64 /* collision suffix + ".delayed" + NUL */
+#ifdef _WIN32
+  // MAX_PATH minus 8.3 headroom (MSDN) minus the ".delayed" marker; raising it
+  // needs the engine to "\\?\"-prefix its paths, which is separate work.
+#define HTS_MAX_PATH_LEN (260 - 12 - 12)
+#define MAX_SEG_LEN 48
+#else
+  // #133: use the platform's own PATH_MAX/NAME_MAX (Linux/Android 4096, macOS
+  // 1024) rather than the far smaller Windows MAX_PATH.
+#ifdef PATH_MAX
+#define HTS_PATH_MAX_ PATH_MAX
+#else
+#define HTS_PATH_MAX_ 1024
+#endif
+#ifdef NAME_MAX
+#define HTS_NAME_MAX_ NAME_MAX
+#else
+#define HTS_NAME_MAX_ 255
+#endif
+#define HTS_MAX_PATH_LEN                                                       \
+  ((HTS_PATH_MAX_ - HTS_PATH_TAIL_RESERVE) <                                   \
+           (HTS_SAVE_BUFSIZE - HTS_PATH_TAIL_RESERVE)                          \
+       ? (HTS_PATH_MAX_ - HTS_PATH_TAIL_RESERVE)                               \
+       : (HTS_SAVE_BUFSIZE - HTS_PATH_TAIL_RESERVE))
+#define MAX_SEG_LEN (HTS_NAME_MAX_ > 64 ? HTS_NAME_MAX_ - 16 : HTS_NAME_MAX_)
+#endif
 #define MIN_LAST_SEG_RESERVE 12
 #define MAX_LAST_SEG_RESERVE 24
-#define MAX_SEG_LEN 48
   if (hts_stringLengthUTF8(afs->save) +
       hts_stringLengthUTF8(StringBuff(opt->path_html_utf8)) >=
       HTS_MAX_PATH_LEN) {
@@ -1493,7 +1516,7 @@ int url_savename(lien_adrfilsave *const afs,
     if (wsave != NULL) {
       const size_t parentLen =
         hts_stringLengthUTF8(StringBuff(opt->path_html_utf8));
-      // parent path length is not insane (otherwise, ignore and pick 200 as 
+      // parent path length is not insane (otherwise, ignore and pick 200 as
       // suffix length)
       const size_t maxLen =
         parentLen <
@@ -1584,9 +1607,37 @@ int url_savename(lien_adrfilsave *const afs,
     // Re-check again ending space or dot after cut (see bug #5)
     cleanEndingSpaceOrDot(afs->save);
   }
+  // The cut above counts UTF-8 codepoints, but parent+name lands in a fixed
+  // byte buffer that aborts() on overflow (htssafe.h). A multibyte name can
+  // pass the codepoint cap yet overflow in bytes, so hard-cut on a codepoint
+  // boundary to keep parent+name inside the buffer regardless (#133).
+  {
+    const size_t parentBytes = strlen(StringBuff(opt->path_html_utf8));
+    const size_t cap = HTS_SAVE_BUFSIZE - HTS_PATH_TAIL_RESERVE;
+    // Shrink only the name. A parent that alone fills the buffer is left to the
+    // existing prepend abort, not collapsed to an empty name that would collide
+    // across URLs and overrun the unbounded collision-suffix sprintf.
+    if (parentBytes < cap) {
+      size_t budget = cap - parentBytes;
+      if (strlen(afs->save) > budget) {
+        while (budget > 0 && ((unsigned char) afs->save[budget] & 0xC0) == 0x80)
+          budget--; // back off a continuation byte, never split a char
+        afs->save[budget] = '\0';
+        cleanEndingSpaceOrDot(afs->save);
+      }
+    }
+  }
 #undef MAX_UTF8_SEQ_CHARS
 #undef MIN_LAST_SEG_RESERVE
+#undef MAX_LAST_SEG_RESERVE
+#undef MAX_SEG_LEN
 #undef HTS_MAX_PATH_LEN
+#undef HTS_PATH_TAIL_RESERVE
+#undef HTS_SAVE_BUFSIZE
+#ifndef _WIN32
+#undef HTS_PATH_MAX_
+#undef HTS_NAME_MAX_
+#endif
 
   // chemin primaire éventuel A METTRE AVANT
   if (strnotempty(StringBuff(opt->path_html_utf8))) {

src/htsopt.h

@@ -541,6 +541,12 @@ struct httrackp {
                          and exit without crawling */
   String warc_file; /**< WARC output: WARC_AUTONAME for --warc, or the
                          --warc-file basename (appended at the tail: ABI) */
+  LLint warc_max_size;  /**< --warc-max-size: rotate the archive past this many
+                             bytes (<=0: single file). Tail: ABI */
+  hts_boolean warc_cdx; /**< --warc-cdx: write a sorted CDXJ index next to the
+                             archive. Tail: ABI */
+  hts_boolean warc_wacz; /**< --wacz: package archive+index+pages as a WACZ zip
+                              (implies --warc + --warc-cdx). Tail: ABI */
 };
 
 /* Running statistics for a mirror. */
@@ -667,6 +673,11 @@ struct htsblk {
   char *warc_reqhdr; /**< stashed raw request header block for WARC (or NULL) */
   char *
       warc_resphdr; /**< stashed raw response header block for WARC (or NULL) */
+  int warc_truncated; /**< WARC-Truncated reason for a cap-truncated body
+                           (WARC_TRUNC_*, 0=none). Tail: ABI */
+  char *warc_rawpath; /**< verbatim WARC: spooled compressed body path, or NULL
+                           (owns the file; unlinked on free). Tail: ABI */
+  LLint warc_rawsize; /**< byte length of warc_rawpath. Tail: ABI */
   /*char digest[32+2];   // md5 digest generated by the engine ("" if none) */
 };
 

src/htsparse.c

@@ -3656,7 +3656,7 @@ int hts_mirror_check_moved(htsmoduleStruct * str,
           !ref_existed;
       if (fexist_utf8(heap(ptr)->sav)) {
         had_partial = 1;
-        remove(heap(ptr)->sav);
+        UNLINK(heap(ptr)->sav);
       }
 
       // Re-get once, only if a partial existed and both Range triggers are
@@ -3862,18 +3862,13 @@ void hts_mirror_process_user_interaction(htsmoduleStruct * str,
     int do_pause = 0;
 
     // user pause lockfile : create hts-paused.lock --> HTTrack will be paused
-    if (fexist
-        (fconcat
-         (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-         StringBuff(opt->path_log), "hts-stop.lock"))) {
+    if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                            StringBuff(opt->path_log), "hts-stop.lock"))) {
       // remove lockfile
-      remove(fconcat
-             (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-             StringBuff(opt->path_log), "hts-stop.lock"));
-      if (!fexist
-          (fconcat
-           (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-           StringBuff(opt->path_log), "hts-stop.lock"))) {
+      UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                     StringBuff(opt->path_log), "hts-stop.lock"));
+      if (!fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                               StringBuff(opt->path_log), "hts-stop.lock"))) {
         do_pause = 1;
       }
     }
@@ -3902,11 +3897,9 @@ void hts_mirror_process_user_interaction(htsmoduleStruct * str,
         }
       }
       {
-        FILE *fp =
-          fopen(fconcat
-                (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                StringBuff(opt->path_log),
-                 "hts-paused.lock"), "wb");
+        FILE *fp = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                 StringBuff(opt->path_log), "hts-paused.lock"),
+                         "wb");
         if (fp) {
           fspc(NULL, fp, "info");       // dater
           fprintf(fp,
@@ -4222,18 +4215,17 @@ int hts_mirror_wait_for_next_file(htsmoduleStruct * str,
             int a = 0;
 
             *stre->last_info_shell_ = tl;
-            if (fexist(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),  StringBuff(opt->path_log), "hts-autopsy"))) { // débuggage: teste si le robot est vivant
+            if (fexist_utf8(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                    StringBuff(opt->path_log),
+                                    "hts-autopsy"))) { // débuggage: teste si le
+                                                       // robot est vivant
               // (oui je sais un robot vivant.. mais bon.. il a le droit de vivre lui aussi)
               // (libérons les robots esclaves de l'internet!)
-              remove(fconcat
-                     (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                     StringBuff(opt->path_log),
-                      "hts-autopsy"));
-              fp =
-                fopen(fconcat
-                      (OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                      StringBuff(opt->path_log),
-                       "hts-isalive"), "wb");
+              UNLINK(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                             StringBuff(opt->path_log), "hts-autopsy"));
+              fp = FOPEN(fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
+                                 StringBuff(opt->path_log), "hts-isalive"),
+                         "wb");
               a = 1;
             }
             if ((*stre->info_shell_) || a) {

src/htstools.c

@@ -1027,6 +1027,17 @@ HTSEXT_API int hts_buildtopindex(httrackp * opt, const char *path,
                     freet(category);
                     category = NULL;
                   }
+#ifdef _WIN32
+                  /* category is ANSI-codepage, doc is utf-8: convert (#216) */
+                  else {
+                    char *cat_utf8 = hts_convertStringSystemToUTF8(
+                        category, strlen(category));
+                    if (cat_utf8 != NULL) {
+                      freet(category);
+                      category = cat_utf8;
+                    }
+                  }
+#endif
                 }
               }
               if (category == NULL) {
@@ -1044,7 +1055,18 @@ HTSEXT_API int hts_buildtopindex(httrackp * opt, const char *path,
                   oldchain->next = chain;
                 }
                 chain->next = NULL;
+#ifdef _WIN32
+                /* name is ANSI-codepage, doc is utf-8: convert (#216) */
+                {
+                  const char *const name = hts_findgetname(h);
+                  char *name_utf8 =
+                      hts_convertStringSystemToUTF8(name, strlen(name));
+                  strcpybuff(chain->name, name_utf8 != NULL ? name_utf8 : name);
+                  freet(name_utf8);
+                }
+#else
                 strcpybuff(chain->name, hts_findgetname(h));
+#endif
                 chain->category = category;
                 chain->level = level;
               }

src/htswarc.h

@@ -45,6 +45,16 @@ extern "C" {
    under the project's output directory at open time. */
 #define WARC_AUTONAME "\001auto"
 
+/* htsblk.warc_truncated / WARC-Truncated reason tokens (ISO 28500 sec 5.13).
+   A cap-truncated body is still archived, tagged with why it was cut short. */
+#define WARC_TRUNC_NONE 0
+#define WARC_TRUNC_LENGTH 1     /* hit a size cap (-M mirror / -m per-file) */
+#define WARC_TRUNC_TIME 2       /* hit the mirror time cap (-E/--max-time) */
+#define WARC_TRUNC_DISCONNECT 3 /* connection dropped mid-body */
+
+/* WARC-Truncated token for a warc_truncated code, or NULL for none. */
+const char *warc_truncated_reason(int code);
+
 typedef struct warc_writer warc_writer;
 
 /* Stash the raw request header block (bstr.buffer) on r for the later WARC
@@ -58,6 +68,13 @@ void warc_stash_response(htsblk *r, const char *resphdr);
 /* Free both stashed header blocks (idempotent, NULL-safe). */
 void warc_free_request(htsblk *r);
 
+/* Adopt the de-chunked compressed spool at tmpfile_path onto
+   r->warc_rawpath/warc_rawsize (strdupt; frees any prior) so the WARC record
+   stores the body verbatim. No-op leaving warc_rawpath NULL when tmpfile_path
+   is empty or the spool is missing/empty (the record then stores the decoded
+   in-memory/on-disk body instead). */
+void warc_adopt_rawspool(htsblk *r, const char *tmpfile_path);
+
 /* Emit the request + response (or revisit) records for one finished
    transaction. Lazily opens the writer into opt->state.warc; a no-op (logged
    once) if the archive cannot be created. */
@@ -76,6 +93,10 @@ warc_writer *warc_open(httrackp *opt, const char *path);
 /* Flush, close and free the writer (NULL-safe). */
 void warc_close(warc_writer *w);
 
+/* SURT-canonicalize url into out[outsz] (the CDXJ sort key). Returns 0 on
+   success, -1 on error or truncation. Exposed for the -#test=warc-surt test. */
+int warc_surt(const char *url, char *out, size_t outsz);
+
 /* Write one transaction's request + response (or revisit) records.
    target_uri:  absolute URL fetched.
    ip:          numeric peer IP, or NULL/"" to omit.
@@ -84,12 +105,24 @@ void warc_close(warc_writer *w);
    body/body_len: decoded in-memory body, or NULL when on disk.
    body_path:   file re-read for the body when body==NULL (may be NULL).
    is_update_unchanged: nonzero for a 304 server-not-modified revisit.
+   truncated: a WARC_TRUNC_* reason to tag a cap-truncated body, else 0.
+   The body is stored verbatim: Content-Encoding is kept and Content-Length set
+   to body_len, so body/body_len must be the as-received (coded) bytes.
    Returns 0 on success, -1 on error. */
 int warc_write_transaction(warc_writer *w, const char *target_uri,
                            const char *ip, const char *req_hdr,
                            const char *resp_hdr, const char *body,
                            size_t body_len, const char *body_path,
-                           int statuscode, int is_update_unchanged);
+                           int statuscode, int is_update_unchanged,
+                           int truncated);
+
+/* Write one non-HTTP capture as a single WARC 'resource' record: the block is
+   the raw payload (no HTTP envelope), Content-Type is the payload's own MIME.
+   Used for ftp:// transfers. truncated is a WARC_TRUNC_* reason or 0.
+   Returns 0 on success, -1 on error. */
+int warc_write_resource(warc_writer *w, const char *target_uri, const char *ip,
+                        const char *content_type, const char *body,
+                        size_t body_len, const char *body_path, int truncated);
 
 #ifdef __cplusplus
 }

src/httrack-library.h

@@ -758,6 +758,9 @@ HTSEXT_API int hts_rename_utf8(const char *oldpath, const char *newpath);
 
 HTSEXT_API int hts_mkdir_utf8(const char *pathname);
 
+#define RMDIR hts_rmdir_utf8
+HTSEXT_API int hts_rmdir_utf8(const char *pathname);
+
 #define UTIME(A, B) hts_utime_utf8(A, B)
 
 typedef struct _utimbuf STRUCT_UTIMBUF;
@@ -772,6 +775,7 @@ typedef struct stat STRUCT_STAT;
 #define UNLINK unlink
 #define RENAME rename
 #define MKDIR(F) mkdir(F, HTS_ACCESS_FOLDER)
+#define RMDIR rmdir
 
 typedef struct utimbuf STRUCT_UTIMBUF;
 

tests/01_engine-direnum.test

@@ -0,0 +1,11 @@
+#!/bin/bash
+#
+
+set -euo pipefail
+
+# Drives -#test=direnum: enumerate a long+non-ASCII directory through the
+# opendir/readdir wrappers, checking each child round-trips as UTF-8 (#133,#630).
+dir=$(mktemp -d)
+trap 'rm -rf "$dir"' EXIT
+
+httrack -O /dev/null -#test=direnum "$dir" | grep -q "direnum:.*OK"

tests/01_engine-fsize.test

@@ -8,9 +8,14 @@ set -euo pipefail
 dir=$(mktemp -d)
 trap 'rm -rf "$dir"' EXIT
 
-out=$(httrack -#test=fsize "$dir")
+rc=0
+out=$(httrack -#test=fsize "$dir") || rc=$?
 echo "$out"
 
+# 77 = platform can't host the 5GB probe (EFBIG); skip, don't fail.
+[ "$rc" = 77 ] && exit 77
+[ "$rc" = 0 ] || exit "$rc"
+
 want="fsize: width=8,8 size=5368709120,5368709120 psize=5368709120 absent=-1"
 test "$out" == "$want" || {
     echo "FAIL: unexpected size report (want '$want')"

tests/01_engine-longpath-io.test

@@ -0,0 +1,11 @@
+#!/bin/bash
+#
+
+set -euo pipefail
+
+# Drives -#test=longpath: a >MAX_PATH round trip exercising hts_pathToUCS2's
+# \\?\ prefixing on Windows (#133); a positive control on POSIX.
+dir=$(mktemp -d)
+trap 'rm -rf "$dir"' EXIT
+
+httrack -O /dev/null -#test=longpath "$dir" | grep -q "longpath:.*OK"

tests/01_engine-mirror-io.test

@@ -0,0 +1,12 @@
+#!/bin/bash
+#
+
+set -euo pipefail
+
+# Drives -#test=mirrorio: rounds a file through a path that is both long
+# (>MAX_PATH) and non-ASCII, exercising the mirror I/O wrappers the engine's
+# raw file ops now route to on Windows (#133, #630). Positive control on POSIX.
+dir=$(mktemp -d)
+trap 'rm -rf "$dir"' EXIT
+
+httrack -O /dev/null -#test=mirrorio "$dir" | grep -q "mirrorio:.*OK"

tests/01_engine-savename.test

@@ -149,10 +149,19 @@ full '/page.php?id=3&sid=42' 'text/html' '/dev/null/www.example.com/page475b.htm
     strip=sid 'prior=www.example.com|/page.php?id=4|/dev/null/www.example.com/PAGE-PRIOR.html'
 
 # Hostile fils stay rooted under the mirror: ../ (raw or %2e-encoded) drops out,
-# control characters become spaces, oversized names cap at 210 chars (the cap
-# can chop the extension off entirely).
+# control characters become spaces.
 full '/../../etc/passwd' 'text/html' '/dev/null/www.example.com///etc/passwd.html'
 full '/%2e%2e/%2e%2e/etc/passwd' 'text/html' '/dev/null/www.example.com///etc/passwd.html'
 full '/x.php' 'application/pdf' '/dev/null/www.example.com///evil.exe' 'cdispo=../../evil.exe'
 name $'/evil\rname\t.php' 'text/html' 'evil name .html'
-name "/$(printf 'a%.0s' {1..300}).php" 'text/html' "$(printf 'a%.0s' {1..210})"
+
+# #133: oversized names are capped only on Windows (MAX_PATH, chopping the
+# extension); elsewhere the platform PATH_MAX leaves a 300-char name whole.
+case "$(uname -s 2>/dev/null)" in
+MINGW* | MSYS* | CYGWIN*)
+    name "/$(printf 'a%.0s' {1..300}).php" 'text/html' "$(printf 'a%.0s' {1..210})"
+    ;;
+*)
+    name "/$(printf 'a%.0s' {1..300}).php" 'text/html' "$(printf 'a%.0s' {1..300}).html"
+    ;;
+esac

tests/01_engine-warc-surt.test

@@ -0,0 +1,8 @@
+#!/bin/bash
+#
+
+set -euo pipefail
+
+# SURT canonicalization of the CDXJ sort key (--warc-cdx). Pure string work,
+# so it runs under the MSan-instrumented 01_engine glob.
+httrack -O /dev/null -#test=warc-surt | grep -q "warc-surt: OK"

tests/01_zlib-warc-cdx.test

@@ -0,0 +1,19 @@
+#!/bin/bash
+#
+
+set -euo pipefail
+
+# --warc-cdx CDXJ index over synthetic transactions: sorted, one line per
+# response/revisit/resource, each offset/length inflates to the right member.
+
+httrack_bin=$(cd "$(dirname "$(command -v httrack)")" && pwd)/httrack
+
+scratch=$(mktemp -d)
+trap 'rm -rf "$scratch"' EXIT
+
+out=$("$httrack_bin" -O /dev/null -#test=warc-cdx "$scratch/")
+echo "$out"
+case "$out" in
+*": OK") ;;
+*) exit 1 ;;
+esac

tests/01_zlib-warc-wacz.test

@@ -0,0 +1,26 @@
+#!/bin/bash
+#
+
+set -euo pipefail
+
+# --wacz packaging over synthetic transactions: the WACZ unzips in-process to
+# the fixed layout, every entry is ZIP STORE, each datapackage sha256 recomputes
+# from the stored bytes, and the digest chains datapackage.json. WACZ needs the
+# OpenSSL SHA-256 digests, so the self-test is absent on non-OpenSSL builds.
+
+httrack_bin=$(cd "$(dirname "$(command -v httrack)")" && pwd)/httrack
+
+if ! "$httrack_bin" -#test 2>&1 | grep -q '^  warc-wacz'; then
+    echo "warc-wacz self-test unavailable (build without OpenSSL); skipping"
+    exit 77
+fi
+
+scratch=$(mktemp -d)
+trap 'rm -rf "$scratch"' EXIT
+
+out=$("$httrack_bin" -O /dev/null -#test=warc-wacz "$scratch/")
+echo "$out"
+case "$out" in
+*": OK") ;;
+*) exit 1 ;;
+esac

tests/01_zlib-warc.test

@@ -4,16 +4,19 @@
 set -euo pipefail
 
 # WARC/1.1 writer self-test: framing, Content-Length, gzip members, round-trip,
-# and identical-payload-digest revisit dedup, over synthetic transactions.
+# revisit dedup, plus v1.1 WARC-Truncated, ftp resource records, and
+# --warc-max-size rotation, over synthetic transactions.
 
 httrack_bin=$(cd "$(dirname "$(command -v httrack)")" && pwd)/httrack
 
 scratch=$(mktemp -d)
 trap 'rm -rf "$scratch"' EXIT
 
-out=$("$httrack_bin" -O /dev/null -#test=warc "$scratch/")
-echo "$out"
-case "$out" in
-*": OK") ;;
-*) exit 1 ;;
-esac
+for t in warc warc-trunc warc-ftp warc-rotate warc-verbatim; do
+    out=$("$httrack_bin" -O /dev/null "-#test=$t" "$scratch/")
+    echo "$out"
+    case "$out" in
+    *": OK") ;;
+    *) exit 1 ;;
+    esac
+done

tests/67_engine-delayed-truncate.test

@@ -3,30 +3,68 @@
 
 set -euo pipefail
 
-# #623: url_savename enforces the 236-char path ceiling by cutting the tail of
-# the last segment, where the mandatory ".delayed" placeholder marker lives. A
-# cut marker fails IS_DELAYED_EXT, so back_delayed_rename never renames the file
-# to its final name and the download is lost. The marker must survive the cut.
-
-# statuscode=302 status=-1 = a redirect answer still downloading: no type is
-# resolved, so the name gets a ".<id>.delayed" placeholder (see 01_engine-savename).
-CEIL=236
+# #623: url_savename shortens an over-ceiling path by cutting the tail of the
+# last segment, where the mandatory ".<id>.delayed" placeholder lives. A cut
+# marker fails IS_DELAYED_EXT, so back_delayed_rename never renames the file and
+# the download is lost. The marker must survive the cut.
+#
+# The ceiling is the Windows MAX_PATH on Windows but the (far larger) save
+# buffer elsewhere (#133). Each CLI arg is capped at HTS_CDLMAXSIZE (1024), so
+# on POSIX we lengthen the output dir as well to overrun the ceiling.
 
 httrack_bin=$(cd "$(dirname "$(command -v httrack)")" && pwd)/httrack
 scratch=$(mktemp -d)
 trap 'rm -rf "$scratch"' EXIT
 cd "$scratch"
 
+hexseg=$(printf 'a1b2c3d4e5f60718%.0s' {1..8}) # 128 hex chars
+
+# engine ceiling is the Windows MAX_PATH on Windows, the save buffer elsewhere
+case "$(uname -s 2>/dev/null)" in
+MINGW* | MSYS* | CYGWIN*)
+    ceil=236
+    outdir=/dev/null
+    deep="/d1$(printf 'a%.0s' {1..40})/d2$(printf 'b%.0s' {1..40})"
+    deep="$deep/d3$(printf 'c%.0s' {1..40})/d4$(printf 'd%.0s' {1..40})"
+    long="$deep/$(printf 'z%.0s' {1..90})"
+    hexpath="$deep/$hexseg"
+    ;;
+*)
+    # overrun the save-buffer ceiling with a long dir + long name
+    ceil=$((1024 * 2 - 64)) # HTS_URLMAXSIZE*2 - HTS_PATH_TAIL_RESERVE
+    outdir="$scratch/$(printf 'D%.0s' $(seq 1 $((1000 - ${#scratch}))))"
+    long="/d1/$(printf 'z%.0s' $(seq 1 985))"
+    hexpath="/d1/$(printf 'y%.0s' $(seq 1 850))/$hexseg"
+    ;;
+esac
+
 run() {
-    "$httrack_bin" -O /dev/null -#test=savename "$@" | sed -n 's/^savename: //p'
+    "$httrack_bin" -O "$outdir" -#test=savename "$@" | sed -n 's/^savename: //p'
 }
 
-# A deep path ending in a long segment that overruns the ceiling.
-deep="/d1$(printf 'a%.0s' {1..40})/d2$(printf 'b%.0s' {1..40})"
-deep="$deep/d3$(printf 'c%.0s' {1..40})/d4$(printf 'd%.0s' {1..40})"
-long="$deep/$(printf 'z%.0s' {1..90})"
+# untruncated "outdir/host/name" is longer than the ceiling, so truncation must
+# fire; out then sits at or under it. Together these keep the marker check below
+# non-vacuous.
+check_truncated() {
+    local out=$1 fil=$2
+    local untrunc=$((${#outdir} + 1 + 15 + ${#fil})) # host = www.example.com
+    test "$untrunc" -gt "$ceil" ||
+        {
+            echo "FAIL: input too short to truncate ($untrunc <= $ceil)"
+            exit 1
+        }
+    test "${#out}" -le "$ceil" ||
+        {
+            echo "FAIL: truncated name ${#out} > $ceil ceiling: '$out'"
+            exit 1
+        }
+}
 
-out="$(run "$long" text/html statuscode=302 status=-1)"
+# statuscode=302 status=-1 = a redirect still downloading: no type is resolved,
+# so the name gets a ".<id>.delayed" placeholder (see 01_engine-savename).
+long_delayed="$long.ext"
+out="$(run "$long_delayed" text/html statuscode=302 status=-1)"
+check_truncated "$out" "$long_delayed"
 case "$out" in
 *.delayed) ;;
 *)
@@ -34,16 +72,11 @@ case "$out" in
     exit 1
     ;;
 esac
-test "${#out}" -le "$CEIL" || {
-    echo "FAIL: truncated name ${#out} > $CEIL ceiling: '$out'"
-    exit 1
-}
 
 # #133-style hashed name: an all-hex last segment must keep the marker too (the
 # ".<id>." tag is not mistaken for part of the hash).
-hexseg=$(printf 'a1b2c3d4e5f60718%.0s' {1..8})
-hexpath="/d1$(printf 'x%.0s' {1..30})/d2$(printf 'y%.0s' {1..30})/$hexseg"
 out="$(run "$hexpath" text/html statuscode=302 status=-1)"
+check_truncated "$out" "$hexpath"
 case "$out" in
 *.delayed) ;;
 *)
@@ -51,17 +84,10 @@ case "$out" in
     exit 1
     ;;
 esac
-test "${#out}" -le "$CEIL" || {
-    echo "FAIL: hashed name ${#out} > $CEIL ceiling: '$out'"
-    exit 1
-}
 
 # A non-delayed name of the same shape still truncates, with no marker to keep.
-out="$(run "$long.html" text/html)"
-test "${#out}" -le "$CEIL" || {
-    echo "FAIL: non-delayed name ${#out} > $CEIL ceiling: '$out'"
-    exit 1
-}
+out="$(run "$long_delayed.html" text/html)"
+check_truncated "$out" "$long_delayed.html"
 case "$out" in
 *.delayed)
     echo "FAIL: non-delayed name grew a .delayed marker: '$out'"

tests/74_local-warc-verbatim.test

@@ -0,0 +1,22 @@
+#!/bin/bash
+#
+# A --warc crawl stores compressed bodies verbatim (the default): the response
+# record keeps Content-Encoding: gzip and its stored bytes inflate back to the
+# served page. The validator's --verbatim mode is the differential gate:
+# inflate(stored) must equal the decoded body the server compressed.
+#
+# Two fixtures cover both adoption branches of back_finalize: page.html (text/html)
+# takes the in-memory branch; data.bin (application/octet-stream) is streamed to
+# disk, so it exercises the is_write direct-to-disk spool adoption.
+
+set -eu
+
+: "${top_srcdir:=..}"
+
+# decoded bodies served (gzip-coded) by route_warcgz_page and route_warcgz_data.
+export WARC_VALIDATE_BODY="warcgz/page.html=3c68746d6c3e3c626f64793e766572626174696d20677a6970207061676520666f72205741524320737472617465677920413c2f626f64793e3c2f68746d6c3e0a warcgz/data.bin=766572626174696d20677a6970206f637465742d73747265616d20626f647920666f722074686520574152432069735f777269746520706174680a"
+export WARC_VALIDATE_VERBATIM=1
+
+bash "$top_srcdir/tests/local-crawl.sh" --errors 0 --warc-validate \
+    --found 'warcgz/index.html' --found 'warcgz/page.html' --found 'warcgz/data.bin' \
+    httrack 'BASEURL/warcgz/index.html' --warc-file warc-out

tests/74_local-warc-wacz.test

@@ -0,0 +1,15 @@
+#!/bin/bash
+#
+# A --wacz crawl packages the WARC archive, its CDXJ index and a generated
+# pages.jsonl into a single WACZ file at crawl end. The stdlib validator is the
+# real gate: STORE-mode entries, the fixed layout, recomputed sha256 digests and
+# the datapackage-digest chain (plus py-wacz/pywb when importable). The crawl
+# skips cleanly on a build without OpenSSL (no conformant SHA-256 -> no package).
+
+set -eu
+
+: "${top_srcdir:=..}"
+
+bash "$top_srcdir/tests/local-crawl.sh" --errors 0 --wacz-validate \
+    --found 'mini304/index.html' --found 'mini304/page.html' \
+    httrack 'BASEURL/mini304/index.html' --warc-file warc-out --wacz

tests/75_engine-longpath-posix.test

@@ -0,0 +1,64 @@
+#!/bin/bash
+#
+
+set -euo pipefail
+
+# #133: the save-path ceiling used to be the Windows MAX_PATH (236 usable) on
+# every platform, needlessly hashing long names on Linux/macOS/Android. Off
+# Windows a name well under the platform PATH_MAX must now be kept whole.
+
+httrack_bin=$(cd "$(dirname "$(command -v httrack)")" && pwd)/httrack
+scratch=$(mktemp -d)
+trap 'rm -rf "$scratch"' EXIT
+cd "$scratch"
+
+# ~300 chars: over the old 236 ceiling, well under any POSIX PATH_MAX. The
+# distinctive 90-char last segment lets us tell "kept whole" from "hashed".
+seg="$(printf 'Z%.0s' {1..90})"
+deep="/a/$(printf 'b%.0s' {1..100})/c/$(printf 'd%.0s' {1..100})"
+out="$("$httrack_bin" -O /dev/null -#test=savename "$deep/$seg.html" text/html |
+    sed -n 's/^savename: //p')"
+
+case "$(uname -s 2>/dev/null)" in
+MINGW* | MSYS* | CYGWIN*)
+    # Windows still caps at MAX_PATH: the distinctive tail is cut
+    case "$out" in
+    *"$seg"*)
+        echo "FAIL: Windows should have shortened the long name: '$out'"
+        exit 1
+        ;;
+    esac
+    ;;
+*)
+    # POSIX: the full name survives, no hashing
+    case "$out" in
+    *"$seg"*) ;;
+    *)
+        echo "FAIL: long POSIX name was truncated: '$out'"
+        exit 1
+        ;;
+    esac
+    test "${#out}" -gt 236 ||
+        {
+            echo "FAIL: POSIX name not longer than the old ceiling: '$out'"
+            exit 1
+        }
+
+    # byte-safety: a multibyte name whose codepoint count is under the ceiling
+    # but whose bytes plus a long -O dir overflow the save buffer must be cut to
+    # fit, not abort the crawl (SIGABRT before the byte-cut guard).
+    bigdir="/x/$(printf 'D%.0s' $(seq 1 1015))"      # ~1020 bytes
+    cjk="/$(printf '\xe4\xb8\xad%.0s' $(seq 1 339))" # 1018 bytes, 339 codepoints
+    if ! mb="$("$httrack_bin" -O "$bigdir" -#test=savename "$cjk" text/html 2>/dev/null)"; then
+        echo "FAIL: aborted on a multibyte over-buffer name"
+        exit 1
+    fi
+    mb="${mb#savename: }"
+    if [ -z "$mb" ] || [ "${#mb}" -ge 2048 ]; then
+        echo "FAIL: multibyte name not bounded (len=${#mb})"
+        exit 1
+    fi
+    ;;
+esac
+
+echo "longpath-posix OK"

tests/Makefile.am

@@ -3,7 +3,7 @@
 # silently drop it from the dist tarball and break "make distcheck".
 EXTRA_DIST = $(TESTS) crawl-test.sh run-all-tests.sh check-network.sh \
 	proxy-https-server.py socks5-server.py proxy-connect-server.py \
-	proxytestlib.py tls-stall-server.py warc-validate.py \
+	proxytestlib.py tls-stall-server.py warc-validate.py wacz-validate.py \
 	local-crawl.sh local-server.py testlib.sh server.crt server.key \
 	server-root/simple/basic.html server-root/simple/link.html \
 	server-root/stripquery/index.html server-root/stripquery/a.html \
@@ -64,6 +64,9 @@ TESTS = \
 	01_engine-expandhome.test \
 	01_engine-fsize.test \
 	01_engine-redirect.test \
+	01_engine-longpath-io.test \
+	01_engine-mirror-io.test \
+	01_engine-direnum.test \
 	01_engine-relative.test \
 	01_engine-robots.test \
 	01_engine-savename.test \
@@ -78,9 +81,12 @@ TESTS = \
 	01_engine-unescape-bounds.test \
 	01_engine-useragent.test \
 	01_engine-version-macros.test \
+	01_engine-warc-surt.test \
 	01_engine-xfread.test \
 	01_zlib-acceptencoding.test \
 	01_zlib-warc.test \
+	01_zlib-warc-cdx.test \
+	01_zlib-warc-wacz.test \
 	01_zlib-contentcodings.test \
 	01_zlib-cache.test \
 	01_zlib-cache-corrupt.test \
@@ -157,6 +163,9 @@ TESTS = \
 	69_local-intl-logdir.test \
 	71_local-crange-repaircache.test \
 	72_watchdog-crawl.test \
-	73_local-warc.test
+	73_local-warc.test \
+	74_local-warc-wacz.test \
+	74_local-warc-verbatim.test \
+	75_engine-longpath-posix.test
 
 CLEANFILES = check-network_sh.cache

tests/local-crawl.sh

@@ -48,6 +48,7 @@ key="${testdir}/server.key"
 tls=
 verbose=
 warc_validate=
+wacz_validate=
 html_subdir=
 outdir_intl=
 rerun=
@@ -117,6 +118,8 @@ while test "$pos" -lt "$nargs"; do
     --rerun-dead) rerun_dead=1 ;; # re-run with the server stopped (cache rollback)
     # validate the produced .warc.gz (see the validation block near the end)
     --warc-validate) warc_validate=1 ;;
+    # validate the produced .wacz package (stdlib, plus py-wacz/pywb if present)
+    --wacz-validate) wacz_validate=1 ;;
     --no-purge)
         nopurge=1
         audit+=("--no-purge")
@@ -372,9 +375,13 @@ if test -n "$warc_validate"; then
     fresh="${tmpdir}/warc-pass1.gz"
     test -f "$fresh" || fresh="$warc"
     declare -a bodyargs=()
-    if test -n "${WARC_VALIDATE_BODY:-}"; then
-        bodyargs=(--expect-body-hex "$WARC_VALIDATE_BODY")
-    fi
+    # WARC_VALIDATE_BODY holds one or more whitespace-separated SUB=HEX specs.
+    for spec in ${WARC_VALIDATE_BODY:-}; do
+        bodyargs+=(--expect-body-hex "$spec")
+    done
+    # compressed asset: assert the stored (verbatim) body inflates to the served
+    # body and keeps Content-Encoding, instead of expecting a decoded body.
+    test -n "${WARC_VALIDATE_VERBATIM:-}" && bodyargs+=(--verbatim)
     info "validating fresh WARC (response bodies)"
     "$python" "$validator" "$(nativepath "$fresh")" "${bodyargs[@]}" >&2 ||
         die "fresh WARC validation failed"
@@ -398,6 +405,23 @@ if test -n "$warc_validate"; then
     fi
 fi
 
+# --- optional WACZ validation (--wacz) --------------------------------------
+if test -n "$wacz_validate"; then
+    wacz=$(find "$mirrorroot" -maxdepth 2 -name '*.wacz' 2>/dev/null | sort | tail -n1)
+    if test -z "$wacz"; then
+        # No package: only acceptable when the build lacks OpenSSL (SHA-256).
+        if grep -aqi "WACZ requires an OpenSSL" "${logroot}/hts-log.txt"; then
+            info "no .wacz produced (build without OpenSSL); skipping"
+            exit 77
+        fi
+        die "no .wacz file produced under $mirrorroot"
+    fi
+    validator=$(nativepath "${testdir}/wacz-validate.py")
+    info "validating WACZ package"
+    "$python" "$validator" "$(nativepath "$wacz")" >&2 || die "WACZ validation failed"
+    result "OK"
+fi
+
 # No crawl, even a cancelled one, may leave engine temporaries: .delayed (#107,
 # #483), or the .z/.u content-coding temps (#557).
 info "checking for leftover engine temporaries"

tests/local-server.py

@@ -635,6 +635,34 @@ class Handler(SimpleHTTPRequestHandler):
             extra_headers=[("Content-Encoding", "gzip")],
         )
 
+    # A gzip-coded HTML page whose decoded body is known, for the verbatim-WARC
+    # differential (stored compressed bytes must inflate to this).
+    WARCGZ_BODY = b"<html><body>verbatim gzip page for WARC strategy A</body></html>\n"
+
+    # A NON-html gzip-coded asset: HTTrack streams it straight to disk, so the
+    # verbatim spool adoption runs on the is_write (direct-to-disk) branch of
+    # back_finalize, not the in-memory branch route_warcgz_page exercises.
+    WARCGZ_BIN_BODY = b"verbatim gzip octet-stream body for the WARC is_write path\n"
+
+    def route_warcgz_index(self):
+        self.send_html(
+            '\t<a href="page.html">page</a>\n' '\t<a href="data.bin">data</a>\n'
+        )
+
+    def route_warcgz_page(self):
+        self.send_raw(
+            gzip.compress(self.WARCGZ_BODY),
+            "text/html",
+            extra_headers=[("Content-Encoding", "gzip")],
+        )
+
+    def route_warcgz_data(self):
+        self.send_raw(
+            gzip.compress(self.WARCGZ_BIN_BODY),
+            "application/octet-stream",
+            extra_headers=[("Content-Encoding", "gzip")],
+        )
+
     # --- content codings ---------------------------------------------------
     # Canned br/zstd bodies (no brotli/zstd module in the stdlib): both decode
     # to CODEC_BODY. Regenerate with the brotli/zstd CLIs over that string.
@@ -1542,6 +1570,9 @@ class Handler(SimpleHTTPRequestHandler):
         "/gated/index.php": route_gated_index,
         "/gated/secret.php": route_gated_secret,
         "/robots.txt": route_robots,
+        "/warcgz/index.html": route_warcgz_index,
+        "/warcgz/page.html": route_warcgz_page,
+        "/warcgz/data.bin": route_warcgz_data,
         "/codec/index.html": route_codec_index,
         "/codec/br.html": route_codec_br,
         "/codec/zstd.html": route_codec_zstd,

tests/wacz-validate.py

@@ -0,0 +1,95 @@
+#!/usr/bin/env python3
+# Validate a WACZ package with the stdlib only (no py-wacz needed): every entry
+# is ZIP STORE, the fixed layout is present, each datapackage resource hash and
+# size recomputes, the digest chains datapackage.json, and pages.jsonl carries
+# the json-pages-1.0 header. If py-wacz (`wacz validate`) is importable it runs
+# too; both gates must pass. Exit 0 = valid, nonzero = invalid.
+import sys
+import json
+import zipfile
+import hashlib
+
+
+def fail(msg):
+    print("wacz-validate: FAIL: %s" % msg, file=sys.stderr)
+    sys.exit(1)
+
+
+def main():
+    if len(sys.argv) < 2:
+        fail("usage: wacz-validate.py FILE.wacz")
+    path = sys.argv[1]
+    z = zipfile.ZipFile(path)
+    names = z.namelist()
+
+    for info in z.infolist():
+        if info.compress_type != zipfile.ZIP_STORED:
+            fail("%s is not STORE mode (%d)" % (info.filename, info.compress_type))
+
+    need_arc = any(
+        n.startswith("archive/") and n.endswith((".warc.gz", ".warc")) for n in names
+    )
+    for req, ok in (
+        ("archive/*.warc.gz", need_arc),
+        ("indexes/index.cdx", "indexes/index.cdx" in names),
+        ("pages/pages.jsonl", "pages/pages.jsonl" in names),
+        ("datapackage.json", "datapackage.json" in names),
+        ("datapackage-digest.json", "datapackage-digest.json" in names),
+    ):
+        if not ok:
+            fail("missing %s (entries: %s)" % (req, names))
+
+    lines = [ln for ln in z.read("pages/pages.jsonl").split(b"\n") if ln.strip()]
+    if not lines or json.loads(lines[0]).get("format") != "json-pages-1.0":
+        fail("pages.jsonl header is not json-pages-1.0: %r" % lines[:1])
+    body = [json.loads(ln) for ln in lines[1:]]
+    if not body:
+        fail("pages.jsonl has no page rows after the header")
+    for row in body:
+        if "url" not in row or "ts" not in row:
+            fail("pages.jsonl row missing url/ts: %r" % row)
+
+    dp = json.loads(z.read("datapackage.json"))
+    if dp.get("profile") != "data-package":
+        fail("profile != data-package: %r" % dp.get("profile"))
+    if dp.get("wacz_version") != "1.1.1":
+        fail("wacz_version != 1.1.1: %r" % dp.get("wacz_version"))
+    resources = dp.get("resources", [])
+    if not resources:
+        fail("datapackage has no resources")
+    for r in resources:
+        data = z.read(r["path"])
+        h = "sha256:" + hashlib.sha256(data).hexdigest()
+        if h != r["hash"]:
+            fail("%s hash %s != %s" % (r["path"], h, r["hash"]))
+        if len(data) != r["bytes"]:
+            fail("%s bytes %d != %d" % (r["path"], len(data), r["bytes"]))
+
+    dig = json.loads(z.read("datapackage-digest.json"))
+    want = "sha256:" + hashlib.sha256(z.read("datapackage.json")).hexdigest()
+    if dig.get("path") != "datapackage.json" or dig.get("hash") != want:
+        fail("digest chain broken: %r" % dig)
+
+    print(
+        "wacz-validate: OK (%d entries, %d resources, stdlib)"
+        % (len(names), len(resources))
+    )
+
+    # Optional stricter gate when py-wacz is present.
+    try:
+        import wacz  # noqa: F401
+    except Exception:
+        return
+    try:
+        from wacz.main import main as wacz_main  # type: ignore
+    except Exception:
+        return
+    print("wacz-validate: running py-wacz validate")
+    rc = wacz_main(["validate", "-f", path])
+    if rc not in (0, None):
+        fail("py-wacz validate returned %r" % rc)
+    print("wacz-validate: py-wacz OK")
+
+
+if __name__ == "__main__":
+    main()

tests/warc-validate.py

@@ -11,6 +11,11 @@
 #                             WARC-Payload-Digest matching sha1(body) when present
 #   --no-response-for SUB     the asset containing SUB must be a revisit: no
 #                             response may target it, and a revisit must
+#   --verbatim                compressed asset: --expect-body-hex instead keeps
+#                             Content-Encoding, checks the HTTP Content-Length is
+#                             the stored (compressed) length, asserts the stored
+#                             body inflates to HEX (the served plaintext), and
+#                             requires the payload digest when the file emits any
 import base64
 import hashlib
 import sys
@@ -61,14 +66,53 @@ def check_body(rec, http_hdr, body, sub, want):
             sys.exit("WARC-Payload-Digest mismatch for %s" % sub)
 
 
+def check_body_verbatim(rec, http_hdr, body, sub, want, digests_emitted):
+    """Verbatim: the stored body is the coded octets, Content-Encoding is kept,
+    the HTTP Content-Length equals the stored (compressed) length, inflating the
+    body yields the served plaintext, and the payload digest is over the coded
+    body. The differential: inflate(stored) == the body the server compressed."""
+    if b"Content-Encoding" not in http_hdr:
+        sys.exit("verbatim record for %s dropped Content-Encoding" % sub)
+    if b"Transfer-Encoding" in http_hdr:
+        sys.exit("verbatim record for %s kept Transfer-Encoding" % sub)
+    hcl = field(http_hdr, b"Content-Length")
+    if hcl is None or int(hcl) != len(body):
+        sys.exit("verbatim record for %s: HTTP Content-Length != stored body" % sub)
+    try:
+        decoded = zlib.decompress(body, zlib.MAX_WBITS | 16)
+    except Exception as exc:
+        sys.exit("verbatim record for %s: body did not inflate: %s" % (sub, exc))
+    if decoded != want:
+        sys.exit(
+            "verbatim decoded mismatch for %s: got %d bytes, expected %d"
+            % (sub, len(decoded), len(want))
+        )
+    pd = field(rec[: rec.find(b"\r\n\r\n")], b"WARC-Payload-Digest")
+    # Catch a regression that drops the digest on the verbatim path, but only
+    # when this file emits digests at all (an OpenSSL build; none otherwise).
+    if digests_emitted and pd is None:
+        sys.exit("verbatim record for %s: missing WARC-Payload-Digest" % sub)
+    if pd is not None and pd.startswith(b"sha1:"):
+        want_b32 = base64.b32encode(hashlib.sha1(body).digest()).decode("ascii")
+        if pd[5:].decode("ascii") != want_b32:
+            sys.exit("WARC-Payload-Digest (compressed) mismatch for %s" % sub)
+
+
 def main():
     argv = sys.argv[1:]
     expect_revisit = "--expect-revisit" in argv
+    verbatim = "--verbatim" in argv
     body_specs = [s.split("=", 1) for s in opt_values(argv, "--expect-body-hex")]
     no_resp = opt_values(argv, "--no-response-for")
     path = [a for a in argv if not a.startswith("--") and "=" not in a][0]
     data = open(path, "rb").read()
 
+    # digests are emitted only on an OpenSSL build; detect it once so --verbatim
+    # can require the payload digest exactly when the file carries any.
+    digests_emitted = any(
+        b"WARC-Payload-Digest" in r[: r.find(b"\r\n\r\n")] for r in records(data)
+    )
+
     total = revisits = responses = infos = 0
     body_hits = {sub: False for sub, _ in body_specs}
     revisit_hits = {sub: False for sub in no_resp}
@@ -106,7 +150,13 @@ def main():
             http_hdr, body = block[:bsep], block[bsep + 4 :]
             for sub, hexval in body_specs:
                 if sub.encode() in uri:
-                    check_body(rec, http_hdr, body, sub, bytes.fromhex(hexval))
+                    want = bytes.fromhex(hexval)
+                    if verbatim:
+                        check_body_verbatim(
+                            rec, http_hdr, body, sub, want, digests_emitted
+                        )
+                    else:
+                        check_body(rec, http_hdr, body, sub, want)
                     body_hits[sub] = True
         elif wtype == b"revisit":
             revisits += 1