Fix macro-hygiene defects in htsstrings.h (dup define, double-eval)

StringSubRW was defined twice (the second under a redundant comment); drop the duplicate. StringCatN and StringSetLength each evaluated their SIZE argument twice, so a side-effecting argument would run twice and a wrapping expression could clamp inconsistently. Capture SIZE once into a local, the way StringCopyN already does. StringSetLength keeps a signed local so the "negative means strlen(buffer_)" contract is preserved. No current call site passes a side-effecting SIZE, so the change is behavior-preserving; the strsafe self-test now passes a (counter++, V) argument and asserts a single evaluation, which fails on the old macros. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Signed-off-by: Xavier Roche <roche@httrack.com>
2026-07-05 08:34:10 +03:00 · 2026-06-29 20:16:30 +02:00
42 changed files with 417 additions and 1459 deletions
--- a/configure.ac
+++ b/configure.ac
@@ -63,16 +63,6 @@ AC_SUBST(LT_CV_OBJDIR,$lt_cv_objdir)
 # Export version info
 AC_SUBST(VERSION_INFO)

-# Versioned plugin name for dlopen() in hts_create_opt(); soname major is
-# libtool's current - age, so this tracks VERSION_INFO bumps automatically.
-HTS_SONAME_MAJOR=$((${VERSION_INFO%%:*} - ${VERSION_INFO##*:}))
-case "$host_os" in
-darwin*) HTS_LIBHTSJAVA_NAME="libhtsjava.$HTS_SONAME_MAJOR.dylib" ;;
-*) HTS_LIBHTSJAVA_NAME="libhtsjava.so.$HTS_SONAME_MAJOR" ;;
-esac
-AC_DEFINE_UNQUOTED([HTS_LIBHTSJAVA_NAME], ["$HTS_LIBHTSJAVA_NAME"],
-  [Versioned libhtsjava runtime name, derived from VERSION_INFO])
-
 ### Default CFLAGS
 DEFAULT_CFLAGS="-Wall -Wformat -Wformat-security \
 -Wmultichar -Wwrite-strings -Wcast-qual -Wcast-align \
--- a/debian/control
+++ b/debian/control
@@ -1,8 +1,9 @@
 Source: httrack
 Section: web
+Priority: optional
 Maintainer: Xavier Roche <roche@httrack.com>
 Standards-Version: 4.7.4
-Build-Depends: debhelper-compat (= 14), autoconf, autoconf-archive, automake, libtool, zlib1g-dev, libssl-dev
+Build-Depends: debhelper-compat (= 13), autoconf, autoconf-archive, automake, libtool, zlib1g-dev, libssl-dev
 Rules-Requires-Root: no
 Homepage: http://www.httrack.com
 Vcs-Git: https://github.com/xroche/httrack.git
--- a/debian/source/lintian-overrides
+++ b/debian/source/lintian-overrides
@@ -1,6 +1,4 @@
-# Maintainer uploads sign the changelog as xavier@debian.org while the control
-# Maintainer is roche@httrack.com; lintian reads the address mismatch as an NMU.
-httrack source: no-nmu-in-changelog
+httrack source: changelog-should-mention-nmu
 httrack source: source-nmu-has-incorrect-version-number

 # The bundled HTML pages are the genuine upstream documentation taken from
--- a/debian/upstream/metadata
+++ b/debian/upstream/metadata
@@ -1,6 +0,0 @@
---
-Repository: https://github.com/xroche/httrack.git
-Repository-Browse: https://github.com/xroche/httrack
-Bug-Database: https://github.com/xroche/httrack/issues
-Bug-Submit: https://github.com/xroche/httrack/issues/new
-Contact: Xavier Roche <roche@httrack.com>
--- a/src/htsback.c
+++ b/src/htsback.c
@@ -2237,13 +2237,12 @@ int host_wait(httrackp * opt, lien_back * back) {

 static int slot_can_be_cleaned(const lien_back * back) {
  return (back->status == STATUS_READY) // ready
-         /* Check autoclean */
-         && (!back->locked)   // not held by hts_wait_delayed (name pending)
-         && (!back->testmode) // not test mode
-         && (strnotempty(back->url_sav))     // filename exists
-         && (HTTP_IS_OK(back->r.statuscode)) // HTTP "OK"
-         && (back->r.size >= 0)              // size>=0
-      ;
+    /* Check autoclean */
+    && (!back->testmode)        // not test mode
+    && (strnotempty(back->url_sav))     // filename exists
+    && (HTTP_IS_OK(back->r.statuscode)) // HTTP "OK"
+    && (back->r.size >= 0)      // size>=0
+    ;
 }

 static int slot_can_be_finalized(httrackp * opt, const lien_back * back) {
@@ -2892,10 +2891,10 @@ void back_wait(struct_back * sback, httrackp * opt, cache_back * cache,
          // range size hack old location

 #if HTS_DIRECTDISK
-          // Shortcut: store the file directly on disk when possible,
-          // sparing memory
-          if (back[i].status &&
-              !back[i].locked) { // name still pending when locked
+          // Court-circuit:
+          // Peut-on stocker le fichier directement sur disque?
+          // Ahh que ca serait vachement mieux et que ahh que la mémoire vous dit merci!
+          if (back[i].status) {
            if (back[i].r.is_write == 0) {      // mode mémoire
              if (back[i].r.adr == NULL) {      // rien n'a été écrit
                if (!back[i].testmode) {        // pas mode test
@@ -3961,12 +3960,8 @@ void back_wait(struct_back * sback, httrackp * opt, cache_back * cache,
                              && (back[i].r.adr = (char *) malloct(2))) {
                            back[i].r.adr[0] = 0;
                          }
-                          /* locked = name pending; the waiter finalizes after
-                             patching url_sav (else: cached as .delayed, #5) */
-                          if (!back[i].locked) {
-                            hts_log_print(opt, LOG_TRACE, "finalizing empty");
-                            back_finalize(opt, cache, sback, i);
-                          }
+                          hts_log_print(opt, LOG_TRACE, "finalizing empty");
+                          back_finalize(opt, cache, sback, i);
                        } else if (!back[i].r.is_chunk) {       // pas de chunk
                          //if (back[i].r.http11!=2) {    // pas de chunk
                          back[i].is_chunk = 0;
--- a/src/htscatchurl.c
+++ b/src/htscatchurl.c
@@ -175,9 +175,7 @@ HTSEXT_API hts_boolean catch_url(T_SOC soc, char *url, char *method,
      //
      socinput(soc, line, 1000);
      if (strnotempty(line)) {
-        /* widths bound the caller buffers: method[32], url[HTS_URLMAXSIZE*2],
-           protocol[256] */
-        if (sscanf(line, "%31s %2047s %255s", method, url, protocol) == 3) {
+        if (sscanf(line, "%s %s %s", method, url, protocol) == 3) {
          lien_adrfil af;

          // méthode en majuscule
--- a/src/htscore.c
+++ b/src/htscore.c
@@ -406,106 +406,29 @@ void hts_invalidate_link(httrackp * opt, int lpos) {
  opt->liens[lpos]->pass2 = -1;
 }

-// Write the makeindex footer (refresh meta when makeindex_links==1), close
-// the file, then run usercommand.
-void hts_finish_makeindex(httrackp *opt, int *makeindex_done,
-                          FILE **makeindex_fp, int makeindex_links,
-                          const char *makeindex_firstlink,
-                          const char *template_footer, const char *adr,
-                          const char *fil) {
-  if (!*makeindex_done) {
-    if (*makeindex_fp) {
-      char BIGSTK tempo[1024];
-      if (makeindex_links == 1) {
-        char BIGSTK link_escaped[HTS_URLMAXSIZE * 2];
-        escape_uri_utf(makeindex_firstlink, link_escaped, sizeof(link_escaped));
-        snprintf(tempo, sizeof(tempo),
-                 "<meta HTTP-EQUIV=\"Refresh\" CONTENT=\"0; URL=%s\">" CRLF,
-                 link_escaped);
-      } else
-        tempo[0] = '\0';
-      hts_template_format(*makeindex_fp, template_footer,
-                          "<!-- Mirror and index made by HTTrack Website "
-                          "Copier/" HTTRACK_VERSION " " HTTRACK_AFF_AUTHORS
-                          " -->",
-                          tempo, /* EOF */ NULL);
-      fflush(*makeindex_fp);
-      fclose(*makeindex_fp);
-      *makeindex_fp = NULL;
-      usercommand(opt, 0, NULL,
-                  fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),
-                          StringBuff(opt->path_html_utf8), "index.html"),
-                  adr, fil);
-    }
-  }
-  *makeindex_done = 1;
-}

-/* Flush the parsed HTML output buffer to disk, skipping the rewrite when the
- * on-disk MD5 is unchanged. */
-void hts_finish_html_file(httrackp *opt, cache_back *cache, htsblk *r,
-                          FILE **fp, const char *ht_buff, size_t ht_len,
-                          const char *adr, const char *fil, const char *save) {
-  char digest[32 + 2];
-  off_t fsize_old =
-      fsize(fconv(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt), save));
-  int ok = 0;
-
-  digest[0] = '\0';
-  domd5mem(ht_buff, ht_len, digest, 1);
-  if (fsize_old == (off_t) ht_len) {
-    int mlen = 0;
-    char *mbuff;
-
-    cache_readdata(cache, "//[HTML-MD5]//", save, &mbuff, &mlen);
-    if (mlen)
-      mbuff[mlen] = '\0';
-    if ((mlen == 32) && (strcmp(((mbuff != NULL) ? mbuff : ""), digest) == 0)) {
-      ok = 1;
-      hts_log_print(opt, LOG_DEBUG, "File not re-written (md5): %s", save);
-    }
-    freet(mbuff);
-  }
-  if (!ok) {
-    file_notify(opt, adr, fil, save, 1, 1, r->notmodified);
-    *fp = filecreate(&opt->state.strc, save);
-    if (*fp) {
-      if (ht_len > 0 && fwrite(ht_buff, 1, ht_len, *fp) != ht_len) {
-        int fcheck = check_fatal_io_errno();
-
-        if (fcheck)
-          opt->state.exit_xh = -1;
-        if (opt->log) {
-          hts_log_print(opt, LOG_ERROR | LOG_ERRNO,
-                        "Unable to write HTML file %s", save);
-          if (fcheck)
-            hts_log_print(opt, LOG_ERROR, "* * Fatal write error, giving up");
-        }
-      }
-      fclose(*fp);
-      *fp = NULL;
-      if (strnotempty(r->lastmodified))
-        set_filetime_rfc822(save, r->lastmodified);
-    } else {
-      int fcheck = check_fatal_io_errno();
-
-      if (fcheck) {
-        hts_log_print(opt, LOG_ERROR,
-                      "Mirror aborted: disk full or filesystem problems");
-        opt->state.exit_xh = -1;
-      }
-      hts_log_print(opt, LOG_ERROR | LOG_ERRNO, "Unable to save file %s", save);
-      if (fcheck)
-        hts_log_print(opt, LOG_ERROR, "* * Fatal write error, giving up");
-    }
-  } else {
-    file_notify(opt, adr, fil, save, 0, 0, r->notmodified);
-    filenote(&opt->state.strc, save, NULL);
-  }
-  if (cache->ndx)
-    cache_writedata(cache->ndx, cache->dat, "//[HTML-MD5]//", save, digest,
-                    (int) strlen(digest));
-}
+#define HT_INDEX_END do { \
+if (!makeindex_done) { \
+if (makeindex_fp) { \
+  char BIGSTK tempo[1024]; \
+  if (makeindex_links == 1) { \
+    char BIGSTK link_escaped[HTS_URLMAXSIZE*2]; \
+    escape_uri_utf(makeindex_firstlink, link_escaped, sizeof(link_escaped)); \
+    snprintf(tempo,sizeof(tempo),"<meta HTTP-EQUIV=\"Refresh\" CONTENT=\"0; URL=%s\">"CRLF, link_escaped); \
+  } else \
+    tempo[0]='\0'; \
+    hts_template_format(makeindex_fp,template_footer, \
+    "<!-- Mirror and index made by HTTrack Website Copier/"HTTRACK_VERSION" "HTTRACK_AFF_AUTHORS" -->", \
+    tempo, /* EOF */ NULL \
+    ); \
+  fflush(makeindex_fp); \
+  fclose(makeindex_fp);  /* à ne pas oublier sinon on passe une nuit blanche */  \
+  makeindex_fp=NULL; \
+  usercommand(opt,0,NULL,fconcat(OPT_GET_BUFF(opt),OPT_GET_BUFF_SIZE(opt),StringBuff(opt->path_html_utf8),"index.html"),"","");  \
+} \
+} \
+makeindex_done=1;    /* ok c'est fait */  \
+} while(0)

 /* does it look like XML ? (SVG et al.) */
 static int look_like_xml(const char *s) {
@@ -2121,8 +2044,7 @@ int httpmirror(char *url1, httrackp * opt) {
  /*
     Ensure the index is being closed
   */
-  hts_finish_makeindex(opt, &makeindex_done, &makeindex_fp, makeindex_links,
-                       makeindex_firstlink, template_footer, "", "");
+  HT_INDEX_END;

  /* 
     updating-a-remotely-deteted-website hack
--- a/src/htscore.h
+++ b/src/htscore.h
@@ -362,20 +362,6 @@ void usercommand(httrackp * opt, int exe, const char *cmd, const char *file,

 void usercommand_exe(const char *cmd, const char *file);

-// Finish the makeindex index.html (footer + refresh meta), run usercommand.
-// Updates *makeindex_done/*makeindex_fp in place; adr/fil are the mode strings.
-void hts_finish_makeindex(httrackp *opt, int *makeindex_done,
-                          FILE **makeindex_fp, int makeindex_links,
-                          const char *makeindex_firstlink,
-                          const char *template_footer, const char *adr,
-                          const char *fil);
-
-// Flush ht_buff[0..ht_len] to save on disk (skip if MD5 unchanged); *fp
-// closed+NULLed on write. Precondition: ht_len>0.
-void hts_finish_html_file(httrackp *opt, cache_back *cache, htsblk *r,
-                          FILE **fp, const char *ht_buff, size_t ht_len,
-                          const char *adr, const char *fil, const char *save);
-
 int filters_init(char ***ptrfilters, int maxfilter, int filterinc);

 int fspc(httrackp * opt, FILE * fp, const char *type);
@@ -484,8 +470,4 @@ void voidf(void);
 /* HTML marker comment marking where the top index is spliced. */
 #define HTS_TOPINDEX "TOP_INDEX_HTTRACK"

-/* Worst-case byte expansion HT_ADD_HTMLESCAPED* must reserve per escaper. */
-#define HTS_HTMLESCAPE_MAXEXP 5      /* escape_for_html_print: '&'->"&amp;" */
-#define HTS_HTMLESCAPE_FULL_MAXEXP 6 /* _full: high byte->"&#xHH;" */
-
 #endif
--- a/src/htsdefines.h
+++ b/src/htsdefines.h
@@ -69,15 +69,11 @@ typedef struct t_hts_callbackarg t_hts_callbackarg;
 typedef struct t_hts_callbackarg t_hts_callbackarg;
 #endif

-/* Marks a symbol an external wrapper module exports back to the engine.
-   Must override -fvisibility=hidden on ELF, or dlopen()ed plugins (htsjava)
-   hide their own hts_plug()/hts_unplug() entry points. */
+/* Marks a symbol an external wrapper module exports back to the engine
+   (dllexport on Windows, nothing elsewhere). */
 #ifndef EXTERNAL_FUNCTION
 #ifdef _WIN32
 #define EXTERNAL_FUNCTION __declspec(dllexport)
-#elif ((defined(__GNUC__) && (__GNUC__ >= 4)) ||                               \
-       (defined(HAVE_VISIBILITY) && HAVE_VISIBILITY))
-#define EXTERNAL_FUNCTION __attribute__((visibility("default")))
 #else
 #define EXTERNAL_FUNCTION
 #endif
--- a/src/htsencoding.c
+++ b/src/htsencoding.c
@@ -190,9 +190,9 @@ int hts_unescapeEntitiesWithCharset(const char *src, char *dest, const size_t ma
        }
      }
    }
-
-    /* reserve one byte for the trailing NUL written after the loop */
-    if (j + 1 >= max) {
+    
+    /* copy */
+    if (j + 1 > max) {
      /* overflow */
      return -1;
    }
@@ -300,11 +300,6 @@ int hts_unescapeUrlSpecial(const char *src, char *dest, const size_t max,

          /* Was the character read successfully ? */
          if (nRead == utfBufferSize) {
-            /* the 'continue' below skips the NUL-reserve guard: re-check */
-            if (utfBufferJ + utfBufferSize >= max) {
-              return -1;
-            }
-
            /* Rollback write position to sequence start write position */
            j = utfBufferJ;

@@ -319,8 +314,8 @@ int hts_unescapeUrlSpecial(const char *src, char *dest, const size_t max,
      }
    }

-    /* reserve one byte for the trailing NUL written after the loop */
-    if (j + 1 >= max) {
+    /* Check for overflow */
+    if (j + 1 > max) {
      return -1;
    }

--- a/src/htsftp.c
+++ b/src/htsftp.c
@@ -128,33 +128,6 @@ void launch_ftp(FTPDownloadStruct * params) {
  return 0; \
  }

-/* Bounded split of a hostile-URL "user[:pass]@" prefix (see htsftp.h). */
-void ftp_split_userpass(const char *src, const char *end, char *user,
-                        size_t user_size, char *pass, size_t pass_size) {
-  size_t n = 0;
-
-  assertf(user_size > 0 && pass_size > 0); /* the size-1 math underflows on 0 */
-
-  while (src[n] != '\0' && src[n] != ':') {
-    if (n < user_size - 1)
-      user[n] = src[n];
-    n++;
-  }
-  user[n < user_size ? n : user_size - 1] = '\0';
-  pass[0] = '\0';
-  if (src[n] == ':') { // password follows the colon
-    const size_t base = n + 1;
-    size_t k = 0;
-
-    while (&src[base + k + 1] < end && src[base + k] != '\0') {
-      if (k < pass_size - 1)
-        pass[k] = src[base + k];
-      k++;
-    }
-    pass[k < pass_size ? k : pass_size - 1] = '\0';
-  }
-}
-
 // la véritable fonction une fois lancées les routines thread/fork
 int run_launch_ftp(FTPDownloadStruct * pStruct) {
  lien_back *back = pStruct->pBack;
@@ -200,7 +173,24 @@ int run_launch_ftp(FTPDownloadStruct * pStruct) {
  while(*real_adr == '/')
    real_adr++;                 // sauter /
  if ((adr = jump_identification(real_adr)) != real_adr) {      // user
-    ftp_split_userpass(real_adr, adr, user, sizeof(user), pass, sizeof(pass));
+    int i = -1;
+
+    pass[0] = '\0';
+    do {
+      i++;
+      user[i] = real_adr[i];
+    } while((real_adr[i] != ':') && (real_adr[i]));
+    user[i] = '\0';
+    if (real_adr[i] == ':') {   // pass
+      int j = -1;
+
+      i++;                      // oui on saute aussi le :
+      do {
+        j++;
+        pass[j] = real_adr[i + j];
+      } while(((&real_adr[i + j + 1]) < adr) && (real_adr[i + j]));
+      pass[j] = '\0';
+    }
  }
  // Calculer RETR <nom>
  {
@@ -994,8 +984,8 @@ int get_ftp_line(T_SOC soc, char *ptrline, size_t line_size, int timeout) {
      //case 0: break;    // pas encore --> erreur (on attend)!
    case 1:
      HTS_STAT.HTS_TOTAL_RECV += 1;     // compter flux entrant
-      if ((b != 10) && (b != 13) && (i < (int) sizeof(data) - 1))
-        data[i++] = b; // truncate hostile over-long reply lines
+      if ((b != 10) && (b != 13))
+        data[i++] = b;
      break;
    default:
      if (ptrline)
--- a/src/htsftp.h
+++ b/src/htsftp.h
@@ -70,11 +70,6 @@ int back_launch_ftp(FTPDownloadStruct * params);
 int run_launch_ftp(FTPDownloadStruct * params);
 int send_line(T_SOC soc, const char *data);
 int get_ftp_line(T_SOC soc, char *line, size_t line_size, int timeout);
-/* Split a "user[:pass]@" prefix (end = jump_identification result) into
-   bounded, NUL-terminated user/pass buffers, truncating to fit.
-   Both sizes must be nonzero. */
-void ftp_split_userpass(const char *src, const char *end, char *user,
-                        size_t user_size, char *pass, size_t pass_size);
 T_SOC get_datasocket(char *to_send, size_t to_send_size);
 int stop_ftp(lien_back * back);
 char *linejmp(char *line);
--- a/src/htsjava.c
+++ b/src/htsjava.c
@@ -63,9 +63,6 @@ Please visit our Website: http://www.httrack.com
 /* This file */
 #include "htsjava.h"

-/* calloct/freet wrappers */
-#include "htssafe.h"
-
 static int reverse_endian(void) {
  int endian = 1;

@@ -207,16 +204,7 @@ static int hts_parse_java(t_hts_callbackarg * carg, httrackp * opt,
        return 0;
      }

-      /* A constant-pool entry is >= 1 byte on disk; reject a count exceeding
-         the file size (hostile .class ~68 MB alloc DoS). */
-      if (!hts_count_fits(header.count, (LLint) fsize(file))) {
-        fclose(fpout);
-        sprintf(str->err_msg,
-                "Invalid constant pool count %u (file len " LLintP ")",
-                (unsigned) header.count, (LLint) fsize(file));
-        return 0;
-      }
-      tab = (RESP_STRUCT *) calloct(header.count, sizeof(RESP_STRUCT));
+      tab = (RESP_STRUCT *) calloc(header.count, sizeof(RESP_STRUCT));
      if (!tab) {
        sprintf(str->err_msg, "Unable to alloc %d bytes",
                (int) sizeof(RESP_STRUCT));
@@ -242,7 +230,7 @@ static int hts_parse_java(t_hts_callbackarg * carg, httrackp * opt,
          } else {              // ++ une erreur est survenue!
            if (strnotempty(str->err_msg) == 0)
              strcpy(str->err_msg, "Internal readtable error");
-            freet(tab);
+            free(tab);
            if (fpout) {
              fclose(fpout);
              fpout = NULL;
@@ -300,7 +288,7 @@ static int hts_parse_java(t_hts_callbackarg * carg, httrackp * opt,
 #if JAVADEBUG
      printf("end\n");
 #endif
-      freet(tab);
+      free(tab);
      if (fpout) {
        fclose(fpout);
        fpout = NULL;
--- a/src/htsjava.h
+++ b/src/htsjava.h
@@ -33,19 +33,15 @@ Please visit our Website: http://www.httrack.com
 #ifndef HTSJAVA_DEFH
 #define HTSJAVA_DEFH

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

 #ifndef HTS_DEF_FWSTRUCT_RESP_STRUCT
--- a/src/htslib.c
+++ b/src/htslib.c
@@ -1149,8 +1149,7 @@ int http_sendhead(httrackp * opt, t_cookie * cookie, int mode,
          char BIGSTK protocol[256], url[HTS_URLMAXSIZE * 2], method[256];

          linput(fp, line, 1000);
-          /* widths bound method[256], url[HTS_URLMAXSIZE*2], protocol[256] */
-          if (sscanf(line, "%255s %2047s %255s", method, url, protocol) == 3) {
+          if (sscanf(line, "%s %s %s", method, url, protocol) == 3) {
            size_t ret;
            // selon que l'on a ou pas un proxy
            if (retour->req.proxy.active) {
@@ -4132,32 +4131,24 @@ DECLARE_APPEND_ESCAPE_VERSION(escape_uri)

 #undef DECLARE_APPEND_ESCAPE_VERSION

-// In-place escaping: copy dest aside, then escape that copy back into dest.
-typedef size_t (*escape_fn_t)(const char *src, char *dest, size_t size);
-
-static size_t inplace_escape(char *const dest, const size_t size,
-                             escape_fn_t escape) {
-  char buffer[256];
-  const size_t len = strnlen(dest, size);
-  const int in_buffer = len + 1 < sizeof(buffer);
-  char *src = in_buffer ? buffer : malloct(len + 1);
-  size_t ret;
-  assertf(src != NULL);
-  assertf(len < size);
-  memcpy(src, dest, len + 1);
-  ret = escape(src, dest, size);
-  if (!in_buffer) {
-    freet(src);
-  }
-  return ret;
-}
-
-// Thin exported wrappers binding inplace_escape() to each escaper (ABI).
+// Same as above, but in-place
 #undef DECLARE_INPLACE_ESCAPE_VERSION
-#define DECLARE_INPLACE_ESCAPE_VERSION(NAME)                                   \
-  HTSEXT_API size_t inplace_##NAME(char *const dest, const size_t size) {      \
-    return inplace_escape(dest, size, NAME);                                   \
-  }
+#define DECLARE_INPLACE_ESCAPE_VERSION(NAME) \
+HTSEXT_API size_t inplace_ ##NAME(char *const dest, const size_t size) { \
+  char buffer[256]; \
+  const size_t len = strnlen(dest, size); \
+  const int in_buffer = len + 1 < sizeof(buffer); \
+  char *src = in_buffer ? buffer : malloct(len + 1); \
+  size_t ret; \
+  assertf(src != NULL); \
+  assertf(len < size); \
+  memcpy(src, dest, len + 1); \
+  ret = NAME(src, dest, size); \
+  if (!in_buffer) { \
+    freet(src); \
+  } \
+  return ret; \
+}

 DECLARE_INPLACE_ESCAPE_VERSION(escape_in_url)
 DECLARE_INPLACE_ESCAPE_VERSION(escape_spc_url)
@@ -6023,11 +6014,9 @@ HTSEXT_API httrackp *hts_create_opt(void) {
    "htsswf", "htsjava", "httrack-plugin", NULL
  };
 #else
-#ifndef HTS_LIBHTSJAVA_NAME
-#define HTS_LIBHTSJAVA_NAME "libhtsjava.so" /* non-autoconf fallback */
-#endif
-  static const char *defaultModules[] = {"libhtsswf.so.1", HTS_LIBHTSJAVA_NAME,
-                                         "httrack-plugin", NULL};
+  static const char *defaultModules[] = {
+    "libhtsswf.so.1", "libhtsjava.so.2", "httrack-plugin", NULL
+  };
 #endif
  httrackp *opt = malloc(sizeof(httrackp));

--- a/src/htsname.c
+++ b/src/htsname.c
@@ -138,66 +138,37 @@ static void cleanEndingSpaceOrDot(char *s) {
  }
 }

-/* Wire Content-Type vs URL extension: a patchable wire type wins over an
-   unspecific ext, the HTS_UNKNOWN_MIME sentinel keeps a specific non-HTML ext
-   (#267 guard), a declared disagreement is CONTESTED. Sentinel and verdict
-   ride the cache, so updates stay consistent. */
-typedef enum wire_verdict {
-  WIRE_KEEPS_EXT,
-  WIRE_WINS,
-  WIRE_CONTESTED
-} wire_verdict;
-
-static wire_verdict wire_ext_verdict(httrackp *opt, const char *wiremime,
-                                     const char *file, char *urlmime,
-                                     size_t urlmime_size) {
-  if (may_unknown2(opt, wiremime, file))
-    return WIRE_KEEPS_EXT; /* type kept verbatim (keep-list / bogus-multiple) */
-  urlmime[0] = '\0';
-  /* type implied by the URL extension, only when confidently known (flag 0) */
-  if (!get_httptype_sized(opt, urlmime, urlmime_size, file, 0))
-    return WIRE_WINS; /* URL ext implies no known type */
-  if (strfield2(wiremime, urlmime))
-    return WIRE_KEEPS_EXT; /* agreement (no .htm->.html churn) */
-  if (!is_hypertext_mime(opt, urlmime, file) &&
-      strfield2(wiremime, HTS_UNKNOWN_MIME))
-    return WIRE_KEEPS_EXT; /* no declared type */
-  return WIRE_CONTESTED;
-}
-
+/* Should the wire Content-Type override the URL's own extension when naming the
+   saved file? True when the type is patchable (may_unknown2) and either the URL
+   extension implies no specific type or the server declared a disagreeing one.
+   A URL extension mapping to a specific non-HTML type is kept only when the
+   server declared NO type (the HTS_UNKNOWN_MIME sentinel; the #267 mangle
+   guard): a typeless .png stays .png, but a .pdf explicitly served as text/html
+   is named .html. The sentinel rides the cache, so updates stay consistent. */
 static int wire_patches_ext(httrackp *opt, const char *wiremime,
                            const char *file) {
  char urlmime[256];

-  switch (wire_ext_verdict(opt, wiremime, file, urlmime, sizeof(urlmime))) {
-  case WIRE_KEEPS_EXT:
+  if (may_unknown2(opt, wiremime, file))
+    return 0; /* type kept verbatim (keep-list / bogus-multiple) */
+  urlmime[0] = '\0';
+  /* type implied by the URL extension, only when confidently known (flag 0) */
+  if (!get_httptype_sized(opt, urlmime, sizeof(urlmime), file, 0))
+    return 1; /* URL ext implies no known type: trust the wire type */
+  if (strfield2(wiremime, urlmime))
+    return 0; /* wire agrees with the ext: keep it (no .htm->.html churn) */
+  /* wire disagrees with a specific non-HTML URL ext. Keep the ext only when
+     the server declared no type (the sentinel); an explicitly declared type,
+     even text/html, is trusted, so a binary-looking URL that really serves
+     HTML (login/error interstitial, soft-404) is named .html. */
+  if (!is_hypertext_mime(opt, urlmime, file) &&
+      strfield2(wiremime, HTS_UNKNOWN_MIME))
    return 0;
-  case WIRE_WINS:
-    return 1;
-  case WIRE_CONTESTED:
-    break; /* no content evidence is consulted today: trust the wire */
-  }
  return 1;
 }

-/* Wire-metadata name change: a Content-Disposition filename wins (returns 2),
-   else the declared type's ext when wire_patches_ext() allows (returns 1),
-   else 0. ext receives the new extension or replacement filename. */
-static int resolve_extension(httrackp *opt, const char *cdispo,
-                             const char *contenttype, const char *fil,
-                             char *ext, size_t ext_size) {
-  if (strnotempty(cdispo)) {
-    strlcpybuff(ext, cdispo, ext_size);
-    return 2;
-  }
-  if (wire_patches_ext(opt, contenttype, fil) &&
-      give_mimext(ext, ext_size, contenttype))
-    return 1;
-  return 0;
-}
-
-// Build the local save name (save) from adr/fil; renames on collision
-// (e.g. INDEX.HTML vs index.html).
+// forme le nom du fichier à sauver (save) à partir de fil et adr
+// système intelligent, qui renomme en cas de besoin (exemple: deux INDEX.HTML et index.html)
 int url_savename(lien_adrfilsave *const afs,
                 lien_adrfil *const former,
                 const char *referer_adr, const char *referer_fil, 
@@ -434,23 +405,45 @@ int url_savename(lien_adrfilsave *const afs,

  // si option check_type activée
  if (is_html < 0 && opt->check_type && !ext_chg) {
+    int ishtest = 0;
+
    if (protocol != PROTOCOL_FILE
        && protocol != PROTOCOL_FTP
      ) {
      // tester type avec requète HEAD si on ne connait pas le type du fichier
      if (!((opt->check_type == 1) && (fil[strlen(fil) - 1] == '/')))   // slash doit être html?
        if (opt->savename_delayed == HTS_SAVENAME_DELAYED_HARD ||
-            ishtml(opt, fil) < 0) { // unsure whether it's html or a file
+            (ishtest = ishtml(opt, fil)) <
+                0) { // unsure whether it's html or a file
          // lire dans le cache
          htsblk r = cache_read_including_broken(opt, cache, adr, fil); // test uniquement

-          if (r.statuscode != -1) { // cache entry read OK
+          if (r.statuscode != -1) {     // pas d'erreur de lecture cache
+            char s[32];
+
+            s[0] = '\0';
            hts_log_print(opt, LOG_DEBUG, "Testing link type (from cache) %s%s",
                          adr_complete, fil_complete);
            if (!HTTP_IS_REDIRECT(r.statuscode)) {
-              ext_chg = resolve_extension(opt, r.cdispo, r.contenttype, fil,
-                                          ext, sizeof(ext));
+              if (strnotempty(r.cdispo)) {        /* filename given */
+                ext_chg = 2;      /* change filename */
+                strcpybuff(ext, r.cdispo);
+              } else if (wire_patches_ext(opt, r.contenttype, fil)) {
+                if (give_mimext(s, sizeof(s),
+                                r.contenttype)) { // recognized extension
+                  ext_chg = 1;
+                  strcpybuff(ext, s);
+                }
+              }
            }
+#ifdef DEFAULT_BIN_EXT
+            // no extension and potentially bogus
+            else if (ishtest == -2) {
+              ext_chg = 1;
+              strcpybuff(ext, DEFAULT_BIN_EXT + 1);
+            }
+#endif
+            //
          } else if (opt->savename_delayed != HTS_SAVENAME_DELAYED_HARD &&
                     is_userknowntype(opt, fil)) { /* PATCH BY BRIAN SCHRÖDER.
                              Lookup mimetype not only by extension,
@@ -474,11 +467,22 @@ int url_savename(lien_adrfilsave *const afs,
          // fail later
          else if (opt->savename_delayed != HTS_SAVENAME_DELAYED_NONE &&
                   !opt->state.stop) {
-            // Check if the file is ready in backing.
+            // Check if the file is ready in backing. We basically take the same logic as later.
+            // FIXME: we should cleanup and factorize this unholy mess
            if (headers != NULL && headers->status >= 0 && !is_redirect) {
-              ext_chg = resolve_extension(opt, headers->r.cdispo,
-                                          headers->r.contenttype,
-                                          headers->url_fil, ext, sizeof(ext));
+              if (strnotempty(headers->r.cdispo)) {        /* filename given */
+                ext_chg = 2;      /* change filename */
+                strcpybuff(ext, headers->r.cdispo);
+              } else if (wire_patches_ext(opt, headers->r.contenttype,
+                                          headers->url_fil)) {
+                char s[16];
+                if (give_mimext(
+                        s, sizeof(s),
+                        headers->r.contenttype)) { // recognized extension
+                  ext_chg = 1;
+                  strcpybuff(ext, s);
+                }
+              }
            }
            else if (mime_type != NULL) {
              ext[0] = '\0';
@@ -496,6 +500,13 @@ int url_savename(lien_adrfilsave *const afs,
                  if (!may_unknown2(opt, mime_type, fil)) {
                    ext_chg = 1;
                  }
+#ifdef DEFAULT_BIN_EXT
+                  // no extension and potentially bogus
+                  else if (ishtml(opt, fil) == -2) {
+                    ext_chg = 1;
+                    strcpybuff(ext, DEFAULT_BIN_EXT + 1);
+                  }
+#endif
                } else {
                  ext_chg = 0;
                }
@@ -685,10 +696,30 @@ int url_savename(lien_adrfilsave *const afs,
                    // libérer emplacement backing
                  }

-                  // no error: change the type?
-                  ext_chg = resolve_extension(
-                      opt, back[b].r.cdispo, back[b].r.contenttype,
-                      back[b].url_fil, ext, sizeof(ext));
+                  {             // pas d'erreur, changer type?
+                    char s[16];
+
+                    s[0] = '\0';
+                    if (strnotempty(back[b].r.cdispo)) {        /* filename given */
+                      ext_chg = 2;      /* change filename */
+                      strcpybuff(ext, back[b].r.cdispo);
+                    } else if (wire_patches_ext(opt, back[b].r.contenttype,
+                                                back[b].url_fil)) {
+                      if (give_mimext(
+                              s, sizeof(s),
+                              back[b].r.contenttype)) { // recognized extension
+                        ext_chg = 1;
+                        strcpybuff(ext, s);
+                      }
+                    }
+#ifdef DEFAULT_BIN_EXT
+                    // no extension and potentially bogus
+                    else if (ishtest == -2) {
+                      ext_chg = 1;
+                      strcpybuff(ext, DEFAULT_BIN_EXT + 1);
+                    }
+#endif
+                  }
                }
                // FIN Si non déplacé, forcer type?

--- a/src/htsparse.c
+++ b/src/htsparse.c
@@ -77,14 +77,13 @@ Please visit our Website: http://www.httrack.com
 /** Append to the output buffer the string 'A'. **/
 #define HT_ADD(A) TypedArrayAppend(output_buffer, A, strlen(A))

-/* clang-format off: an edit realigns all backslashes, churning the macro. */
-/* clang-format off */
-/** Append 'A' to the output buffer, html-escaped; FACTOR = max byte expansion. **/
-#define HT_ADD_HTMLESCAPED_ANY(A, FUNCTION, FACTOR) do { \
+/** Append to the output buffer the string 'A', html-escaped. **/
+#define HT_ADD_HTMLESCAPED_ANY(A, FUNCTION) do { \
  if ((opt->getmode & 1) != 0 && ptr>0) { \
    const char *const str_ = (A); \
    size_t size_; \
-    TypedArrayEnsureRoom(output_buffer, strlen(str_) * (FACTOR) + 1024); \
+    /* &amp; is the maximum expansion */ \
+    TypedArrayEnsureRoom(output_buffer, strlen(str_) * 5 + 1024); \
    size_ = FUNCTION(str_, &TypedArrayTail(output_buffer), \
                     TypedArrayRoom(output_buffer)); \
    TypedArraySize(output_buffer) += size_; \
@@ -92,113 +91,188 @@ Please visit our Website: http://www.httrack.com
 } while(0)

 /** Append to the output buffer the string 'A', html-escaped for &. **/
-#define HT_ADD_HTMLESCAPED(A) \
-  HT_ADD_HTMLESCAPED_ANY(A, escape_for_html_print, HTS_HTMLESCAPE_MAXEXP)
+#define HT_ADD_HTMLESCAPED(A) HT_ADD_HTMLESCAPED_ANY(A, escape_for_html_print)

 /**
- * Append to the output buffer the string 'A', html-escaped for & and
+ * Append to the output buffer the string 'A', html-escaped for & and 
 * high chars.
 **/
-#define HT_ADD_HTMLESCAPED_FULL(A) \
-  HT_ADD_HTMLESCAPED_ANY(A, escape_for_html_print_full, HTS_HTMLESCAPE_FULL_MAXEXP)
-/* clang-format on */
+#define HT_ADD_HTMLESCAPED_FULL(A) HT_ADD_HTMLESCAPED_ANY(A, escape_for_html_print_full)

 // does nothing
 #define XH_uninit do {} while(0)

+#define HT_ADD_END { \
+  int ok=0;\
+  if (TypedArraySize(output_buffer) != 0) { \
+    const size_t ht_len = TypedArraySize(output_buffer); \
+    const char *const ht_buff = TypedArrayElts(output_buffer); \
+    char digest[32+2];\
+    off_t fsize_old = fsize(fconv(OPT_GET_BUFF(opt),OPT_GET_BUFF_SIZE(opt),savename()));\
+    digest[0] = '\0';\
+    domd5mem(TypedArrayElts(output_buffer), ht_len, digest, 1);\
+    if (fsize_old == (off_t) ht_len) { \
+      int mlen = 0;\
+      char* mbuff;\
+      cache_readdata(cache,"//[HTML-MD5]//",savename(),&mbuff,&mlen);\
+      if (mlen) \
+        mbuff[mlen]='\0';\
+      if ((mlen == 32) && (strcmp(((mbuff!=NULL)?mbuff:""),digest)==0)) {\
+        ok=1;\
+        hts_log_print(opt, LOG_DEBUG, "File not re-written (md5): %s",savename());\
+      } else {\
+        ok=0;\
+      } \
+    }\
+    if (!ok) { \
+      file_notify(opt,urladr(), urlfil(), savename(), 1, 1, r->notmodified); \
+      fp=filecreate(&opt->state.strc, savename()); \
+      if (fp) { \
+        if (ht_len>0) {\
+        if (fwrite(ht_buff,1,ht_len,fp) != ht_len) { \
+          int fcheck;\
+          if ((fcheck=check_fatal_io_errno())) {\
+            opt->state.exit_xh=-1;\
+          }\
+          if (opt->log) {   \
+            hts_log_print(opt, LOG_ERROR | LOG_ERRNO, "Unable to write HTML file %s", savename());\
+            if (fcheck) {\
+              hts_log_print(opt, LOG_ERROR, "* * Fatal write error, giving up");\
+            }\
+          }\
+        }\
+        }\
+        fclose(fp); fp=NULL; \
+        if (strnotempty(r->lastmodified)) \
+        set_filetime_rfc822(savename(),r->lastmodified); \
+      } else {\
+        int fcheck;\
+        if ((fcheck=check_fatal_io_errno())) {\
+  				hts_log_print(opt, LOG_ERROR, "Mirror aborted: disk full or filesystem problems"); \
+          opt->state.exit_xh=-1;\
+        }\
+        hts_log_print(opt, LOG_ERROR | LOG_ERRNO, "Unable to save file %s", savename());\
+        if (fcheck) {\
+          hts_log_print(opt, LOG_ERROR, "* * Fatal write error, giving up");\
+        }\
+      }\
+    } else {\
+      file_notify(opt,urladr(), urlfil(), savename(), 0, 0, r->notmodified); \
+      filenote(&opt->state.strc, savename(),NULL); \
+    }\
+    if (cache->ndx)\
+      cache_writedata(cache->ndx,cache->dat,"//[HTML-MD5]//",savename(),digest,(int)strlen(digest));\
+  } \
+  TypedArrayFree(output_buffer); \
+}
 #define HT_ADD_FOP

-/* Mutable extended-context fields: one source of truth so the DEFINE/SET/SAVE
-   load and store lists can't drift apart. */
-/* clang-format off */
-#define ENGINE_MUTABLE_FIELDS(X) \
-  X(int, error, stre->error_) \
-  X(int, store_errpage, stre->store_errpage_) \
-  X(int, makeindex_done, stre->makeindex_done_) \
-  X(FILE *, makeindex_fp, stre->makeindex_fp_) \
-  X(int, makeindex_links, stre->makeindex_links_) \
-  X(LLint, stat_fragment, stre->stat_fragment_)
-
-#define ENGINE_FIELD_DECLARE(type, name, src) type name = *(src);
-#define ENGINE_FIELD_LOAD(type, name, src) name = *(src);
-#define ENGINE_FIELD_STORE(type, name, src) *(src) = name;
+// COPY IN HTSCORE.C
+#define HT_INDEX_END do { \
+  if (!makeindex_done) { \
+  if (makeindex_fp) { \
+  char BIGSTK tempo[1024]; \
+  if (makeindex_links == 1) { \
+  char BIGSTK link_escaped[HTS_URLMAXSIZE*2]; \
+  escape_uri_utf(makeindex_firstlink, link_escaped, sizeof(link_escaped)); \
+  snprintf(tempo,sizeof(tempo),"<meta HTTP-EQUIV=\"Refresh\" CONTENT=\"0; URL=%s\">"CRLF,link_escaped); \
+  } else \
+  tempo[0]='\0'; \
+  hts_template_format(makeindex_fp,template_footer, \
+  "<!-- Mirror and index made by HTTrack Website Copier/"HTTRACK_VERSION" "HTTRACK_AFF_AUTHORS" -->", \
+  tempo, /* EOF */ NULL \
+  ); \
+  fflush(makeindex_fp); \
+  fclose(makeindex_fp);  /* à ne pas oublier sinon on passe une nuit blanche */  \
+  makeindex_fp=NULL; \
+  usercommand(opt,0,NULL,fconcat(OPT_GET_BUFF(opt), OPT_GET_BUFF_SIZE(opt),  StringBuff(opt->path_html_utf8),"index.html"),"primary","primary");  \
+  } \
+  } \
+  makeindex_done=1;    /* ok c'est fait */  \
+} while(0)

 #define ENGINE_DEFINE_CONTEXT() \
  ENGINE_DEFINE_CONTEXT_BASE(); \
+  /* */ \
  htsblk* const r HTS_UNUSED = stre->r_; \
  hash_struct* const hash HTS_UNUSED = stre->hash_; \
  char* const codebase HTS_UNUSED = stre->codebase; \
  char* const base HTS_UNUSED = stre->base; \
+  /* */ \
  const char * const template_header HTS_UNUSED = stre->template_header_; \
  const char * const template_body HTS_UNUSED = stre->template_body_; \
  const char * const template_footer HTS_UNUSED = stre->template_footer_; \
+  /* */ \
  HTS_UNUSED char* const makeindex_firstlink = stre->makeindex_firstlink_; \
-  ENGINE_MUTABLE_FIELDS(ENGINE_FIELD_DECLARE) \
-  /* load-once (kept out of SET/SAVE): re-reading would reset the throttle */ \
+  /* */ \
+  /* */ \
+  int error = * stre->error_; \
+  int store_errpage = * stre->store_errpage_; \
+  /* */ \
+  int makeindex_done = *stre->makeindex_done_; \
+  FILE* makeindex_fp = *stre->makeindex_fp_; \
+  int makeindex_links = *stre->makeindex_links_; \
+  /* */ \
+  LLint stat_fragment = *stre->stat_fragment_; \
  HTS_UNUSED TStamp makestat_time = stre->makestat_time; \
  HTS_UNUSED FILE* makestat_fp = stre->makestat_fp

 #define ENGINE_SET_CONTEXT() \
  ENGINE_SET_CONTEXT_BASE(); \
-  ENGINE_MUTABLE_FIELDS(ENGINE_FIELD_LOAD)
+  /* */ \
+  error = * stre->error_; \
+  store_errpage = * stre->store_errpage_; \
+  /* */ \
+  makeindex_done = *stre->makeindex_done_; \
+  makeindex_fp = *stre->makeindex_fp_; \
+  makeindex_links = *stre->makeindex_links_; \
+  /* */ \
+  stat_fragment = *stre->stat_fragment_; \
+  makestat_time = stre->makestat_time; \
+  makestat_fp = stre->makestat_fp

 #define ENGINE_LOAD_CONTEXT() \
  ENGINE_DEFINE_CONTEXT()

 #define ENGINE_SAVE_CONTEXT() \
  ENGINE_SAVE_CONTEXT_BASE(); \
-  ENGINE_MUTABLE_FIELDS(ENGINE_FIELD_STORE)
-/* clang-format on */
+  /* */ \
+  * stre->error_ = error; \
+  * stre->store_errpage_ = store_errpage; \
+  /* */ \
+  *stre->makeindex_done_ = makeindex_done; \
+  *stre->makeindex_fp_ = makeindex_fp; \
+  *stre->makeindex_links_ = makeindex_links; \
+  /* */ \
+  *stre->stat_fragment_ = stat_fragment

+#define _FILTERS     (*opt->filters.filters)
+#define _FILTERS_PTR (opt->filters.filptr)
 #define _ROBOTS      ((robots_wizard*)opt->robotsptr)

-/* JS-detection automaton states; INSCRIPT_DEFAULT is the synthetic "any other
-   char" column of the transition table. */
-typedef enum {
-  INSCRIPT_START = 0,
-  INSCRIPT_ANTISLASH,
-  INSCRIPT_INQUOTE,
-  INSCRIPT_INQUOTE2,
-  INSCRIPT_SLASH,
-  INSCRIPT_SLASHSLASH,
-  INSCRIPT_COMMENT,
-  INSCRIPT_COMMENT2,
-  INSCRIPT_ANTISLASH_IN_QUOTE,
-  INSCRIPT_ANTISLASH_IN_QUOTE2,
-  INSCRIPT_DEFAULT = 256
-} INSCRIPT;
+/* Apply current *adr character for the script automate */
+#define AUTOMATE_LOOKUP_CURRENT_ADR() do { \
+  if (inscript) { \
+  int new_state_pos; \
+  new_state_pos=inscript_state[inscript_state_pos][(unsigned char)*html]; \
+  if (new_state_pos < 0) { \
+  new_state_pos=inscript_state[inscript_state_pos][INSCRIPT_DEFAULT]; \
+  } \
+  assertf(new_state_pos >= 0); \
+  assertf(new_state_pos*sizeof(inscript_state[0]) < sizeof(inscript_state)); \
+  inscript_state_pos=new_state_pos; \
+  } \
+} while(0)

-#define INSCRIPT_NSTATES 10 /* rows in the transition table */
-
-/* Live view of the parser's automaton locals, set up once so the helpers below
-   can drive it without capturing them by lexical scope. */
-typedef struct {
-  const int *inscript;             /* nonzero while inside a script body */
-  const signed char (*table)[257]; /* [INSCRIPT_NSTATES][257] transitions */
-  INSCRIPT *pos;                   /* current state */
-  const char **html;               /* parse cursor */
-} script_automate;
-
-/* Feed the current *html byte to the automaton. No-op outside a script body. */
-static void hts_automate_lookup(const script_automate *aut) {
-  if (*aut->inscript) {
-    int next = aut->table[*aut->pos][(unsigned char) **aut->html];
-    if (next < 0) {
-      next = aut->table[*aut->pos][INSCRIPT_DEFAULT];
-    }
-    assertf(next >= 0 && next < INSCRIPT_NSTATES);
-    *aut->pos = (INSCRIPT) next;
-  }
-}
-
-/* Advance the cursor by 'steps' bytes, feeding each to the automaton. */
-static void hts_automate_increment(const script_automate *aut, int steps) {
-  while (steps > 0) {
-    (*aut->html)++;
-    hts_automate_lookup(aut);
-    steps--;
-  }
-}
+/* Increment current pointer to 'steps' characters, modifying automate if necessary */
+#define INCREMENT_CURRENT_ADR(steps) do { \
+  int steps__ = (int) ( steps ); \
+  while(steps__ > 0) { \
+  html++; \
+  AUTOMATE_LOOKUP_CURRENT_ADR(); \
+  steps__ --; \
+  } \
+} while(0)

 /* Percent-encode the angle brackets of a string so it is safe to embed inside
   an HTML comment (the default footer) or any other HTML context. A URL holding
@@ -343,7 +417,20 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
      int incomment = 0;        // dans un <!--
      int inscript = 0;         // dans un scipt pour applets javascript)
      int inscript_locked = 0;  // in locked script (ie. js file)
-      signed char inscript_state[INSCRIPT_NSTATES][257];
+      signed char inscript_state[10][257];
+      typedef enum {
+        INSCRIPT_START = 0,
+        INSCRIPT_ANTISLASH,
+        INSCRIPT_INQUOTE,
+        INSCRIPT_INQUOTE2,
+        INSCRIPT_SLASH,
+        INSCRIPT_SLASHSLASH,
+        INSCRIPT_COMMENT,
+        INSCRIPT_COMMENT2,
+        INSCRIPT_ANTISLASH_IN_QUOTE,
+        INSCRIPT_ANTISLASH_IN_QUOTE2,
+        INSCRIPT_DEFAULT = 256
+      } INSCRIPT;
      INSCRIPT inscript_state_pos = INSCRIPT_START;
      const char *inscript_name = NULL;       // script tag name
      int inscript_tag = 0;     // on est dans un <body onLoad="... terminé par >
@@ -404,8 +491,6 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
      inscript_state[INSCRIPT_COMMENT2]['*'] = INSCRIPT_COMMENT2;
      inscript_state[INSCRIPT_ANTISLASH_IN_QUOTE][INSCRIPT_DEFAULT] = INSCRIPT_INQUOTE; /* #8: escape in '' */
      inscript_state[INSCRIPT_ANTISLASH_IN_QUOTE2][INSCRIPT_DEFAULT] = INSCRIPT_INQUOTE2;       /* #9: escape in "" */
-      const script_automate saut = {&inscript, inscript_state,
-                                    &inscript_state_pos, &html};

      /* Primary list or URLs */
      if (ptr == 0) {
@@ -604,14 +689,13 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
                        }

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

@@ -625,9 +709,7 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
              }

            } else if (heap(ptr)->depth < opt->depth) {        // on a sauté level1+1 et level1
-              hts_finish_makeindex(opt, &makeindex_done, &makeindex_fp,
-                                   makeindex_links, makeindex_firstlink,
-                                   template_footer, "primary", "primary");
+              HT_INDEX_END;
            }
          }                     // if (opt->makeindex)
        }
@@ -845,7 +927,7 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
            }

            /* automate */
-            hts_automate_lookup(&saut);
+            AUTOMATE_LOOKUP_CURRENT_ADR();

            // Note:
            // Certaines pages ne respectent pas le html
@@ -1761,7 +1843,7 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
            }
            // sauter espaces
            // adr+=p;
-            hts_automate_increment(&saut, p);
+            INCREMENT_CURRENT_ADR(p);
            while((is_space(*html)
                   || (inscriptgen && html[0] == '\\' && is_space(html[1])
                   )
@@ -1776,7 +1858,7 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
                }
              // puis quitter
              // html++;    // sauter les espaces, "" et cie
-              hts_automate_increment(&saut, 1);
+              INCREMENT_CURRENT_ADR(1);
            }

            /* Stop at \n (LF) if primary links or link lists */
@@ -1791,7 +1873,7 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
              if (*html == '\\') {
                if ((*(html + 1) == '\'') || (*(html + 1) == '"')) {      // \" ou \'
                  // html+=2;    // sauter
-                  hts_automate_increment(&saut, 2);
+                  INCREMENT_CURRENT_ADR(2);
                }
              }
            }
@@ -1839,7 +1921,7 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
            if (srcset_p) {
              while(html < r->adr + r->size
                    && (is_realspace(*html) || *html == ','))
-                hts_automate_increment(&saut, 1);
+                INCREMENT_CURRENT_ADR(1);
            }
            eadr = html;

@@ -3299,7 +3381,7 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {

            assertf(eadr - html >= 0);   // Should not go back
            if (eadr > html) {
-              hts_automate_increment(&saut, (int) (eadr - 1 - html));
+              INCREMENT_CURRENT_ADR(eadr - 1 - html);
            }
            // adr=eadr-1;  // ** sauter

@@ -3318,8 +3400,7 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
                  q++;          // skip whitespace and empty candidates
                if (q < endp && *q != '\0' && *q != ',' && *q != quote
                    && *q != '<' && *q != '>' && (unsigned char) *q >= 32) {
-                  hts_automate_increment(
-                      &saut, (int) (q - html)); // keep the automate in sync
+                  INCREMENT_CURRENT_ADR(q - html);   // keep the automate in sync
                  ok = 1;
                  goto srcset_next;
                }
@@ -3459,12 +3540,7 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
        }

        /* Flush and save to disk */
-        if (TypedArraySize(output_buffer) != 0) {
-          hts_finish_html_file(
-              opt, cache, r, &fp, TypedArrayElts(output_buffer),
-              TypedArraySize(output_buffer), urladr(), urlfil(), savename());
-        }
-        TypedArrayFree(output_buffer);
+        HT_ADD_END;             // achever
      }
      //
      //
@@ -3489,24 +3565,6 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
  return 0;
 }

-/* Mirror the savename to tell whether a redirect saves to the same file (#159);
- * contract in htsparse.h. */
-hts_boolean hts_redirect_same_savefile(httrackp *opt, const char *cur_adr,
-                                       const char *cur_fil,
-                                       const char *moved_adr,
-                                       const char *moved_fil) {
-  const int norm_slash = opt->urlhack && !opt->no_slash_dedup;
-  const int norm_query = opt->urlhack && !opt->no_query_dedup;
-  char BIGSTK n_fil[HTS_URLMAXSIZE * 2], pn_fil[HTS_URLMAXSIZE * 2];
-
-  if (strcasecmp(jump_identification_const(moved_adr),
-                 jump_identification_const(cur_adr)) != 0)
-    return HTS_FALSE;
-  fil_normalized_filtered_ex(moved_fil, n_fil, NULL, norm_slash, norm_query);
-  fil_normalized_filtered_ex(cur_fil, pn_fil, NULL, norm_slash, norm_query);
-  return strcasecmp(n_fil, pn_fil) == 0;
-}
-
 /*
 Check 301, 302, .. statuscodes (moved)
 */
@@ -3552,9 +3610,36 @@ int hts_mirror_check_moved(htsmoduleStruct * str,
        if ((reponse =
             ident_url_relatif(mov_url, urladr(), urlfil(), moved)) >= 0) {
          int set_prio_to = 0;  // pas de priotité fixéd par wizard
-          // A same-file alias redirect must be followed, not stubbed (#159).
-          const hts_boolean same_savefile = hts_redirect_same_savefile(
-              opt, urladr(), urlfil(), moved->adr, moved->fil);
+
+          // check whether URLHack is harmless or not (per the effective
+          // sub-flags)
+          if (opt->urlhack && (!opt->no_www_dedup || !opt->no_slash_dedup ||
+                               !opt->no_query_dedup)) {
+            const int norm_host = !opt->no_www_dedup;
+            const int norm_slash = !opt->no_slash_dedup;
+            const int norm_query = !opt->no_query_dedup;
+            char BIGSTK n_adr[HTS_URLMAXSIZE * 2], n_fil[HTS_URLMAXSIZE * 2];
+            char BIGSTK pn_adr[HTS_URLMAXSIZE * 2], pn_fil[HTS_URLMAXSIZE * 2];
+
+            strlcpybuff(n_adr,
+                        norm_host ? jump_normalized_const(moved->adr)
+                                  : jump_identification_const(moved->adr),
+                        sizeof(n_adr));
+            strlcpybuff(pn_adr,
+                        norm_host ? jump_normalized_const(urladr())
+                                  : jump_identification_const(urladr()),
+                        sizeof(pn_adr));
+            fil_normalized_filtered_ex(moved->fil, n_fil, NULL, norm_slash,
+                                       norm_query);
+            fil_normalized_filtered_ex(urlfil(), pn_fil, NULL, norm_slash,
+                                       norm_query);
+            if (strcasecmp(n_adr, pn_adr) == 0
+                && strcasecmp(n_fil, pn_fil) == 0) {
+              hts_log_print(opt, LOG_WARNING,
+                            "Redirected link is identical because of 'URL Hack' option: %s%s and %s%s",
+                            urladr(), urlfil(), moved->adr, moved->fil);
+            }
+          }
          //if (ident_url_absolute(mov_url,moved->adr,moved->fil)!=-1) {    // ok URL reconnue
          // c'est (en gros) la même URL..
          // si c'est un problème de casse dans le host c'est que le serveur est buggé
@@ -3582,17 +3667,7 @@ int hts_mirror_check_moved(htsmoduleStruct * str,
                hts_log_print(opt, LOG_DEBUG, "moved link accepted: %s%s",
                              moved->adr, moved->fil);
              }
-            } else if (same_savefile) {
-              // A stub would point at itself; follow the redirect instead.
-              if (hts_acceptlink(opt, ptr, moved->adr, moved->fil, NULL, NULL,
-                                 &set_prio_to, NULL) != 1) {
-                get_it = 1;
-                hts_log_print(opt, LOG_WARNING,
-                              "Redirect to a same-file alias, fetching real "
-                              "content: %s%s -> %s%s",
-                              urladr(), urlfil(), moved->adr, moved->fil);
-              }
-            } /* sinon traité normalement */
+            }                   /* sinon traité normalement */
          }

          //if ((strfield2(moved->adr,urladr())!=0) && (strfield2(moved->fil,urlfil())!=0)) {  // identique à casse près
@@ -3615,11 +3690,7 @@ int hts_mirror_check_moved(htsmoduleStruct * str,
                   heap(heap(ptr)->precedent)->adr,
                   heap(heap(ptr)->precedent)->fil, opt,
                   sback, cache, hash, ptr, numero_passe, NULL) != -1) {
-                // Same-file alias: the reserved name is the invalidated source,
-                // so record anyway.
-                if (same_savefile ||
-                    hash_read(hash, savedmoved.save, NULL,
-                              HASH_STRUCT_FILENAME) < 0) { // n'existe pas déja
+                if (hash_read(hash, savedmoved.save, NULL, HASH_STRUCT_FILENAME) < 0) {   // n'existe pas déja
                  // enregistrer lien avec SAV IDENTIQUE
                  if (hts_record_link(opt, moved->adr, moved->fil, heap(ptr)->sav, "", "", NULL)) {
                    // mode test?
@@ -3643,6 +3714,7 @@ int hts_mirror_check_moved(htsmoduleStruct * str,
                                "moving %s to an existing file %s",
                                heap(ptr)->fil, urlfil());
                }
+
              }
            }

@@ -4845,9 +4917,6 @@ int hts_wait_delayed(htsmoduleStruct * str, lien_adrfilsave *afs,

          /* Still have a back reference */
          if (b >= 0) {
-            /* Patch destination filename for direct-to-disk mode, BEFORE any
-               finalize: it records and caches the entry under url_sav */
-            strcpybuff(back[b].url_sav, afs->save);
            /* Finalize now as we have the type */
            if (back[b].status == STATUS_READY) {
              if (!back[b].finalized) {
@@ -4855,6 +4924,8 @@ int hts_wait_delayed(htsmoduleStruct * str, lien_adrfilsave *afs,
                back_finalize(opt, cache, sback, b);
              }
            }
+            /* Patch destination filename for direct-to-disk mode */
+            strcpybuff(back[b].url_sav, afs->save);
          }

        }                       // b >= 0
--- a/src/htsparse.h
+++ b/src/htsparse.h
@@ -116,19 +116,6 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre);
 int hts_mirror_check_moved(htsmoduleStruct * str,
                           htsmoduleStructExtended * stre);

-/*
-  Non-zero if a redirect (cur_adr,cur_fil)->(moved_adr,moved_fil) saves to the
-  same local file, so it must be followed rather than turned into a
-  self-pointing "moved" stub (#159). Mirrors the savename: scheme+userinfo
-  stripped, www kept (www dedup is the crawl layer's job), path
-  slash/query-normalized per the URL-hack flags. Not hash_url_equals: that keys
-  on the dedup hash, which folds www and never collapses http<->https.
-*/
-hts_boolean hts_redirect_same_savefile(httrackp *opt, const char *cur_adr,
-                                       const char *cur_fil,
-                                       const char *moved_adr,
-                                       const char *moved_fil);
-
 /*
  Process user intercations: pause, add link, delete link..
 */
--- a/src/htssafe.h
+++ b/src/htssafe.h
@@ -456,13 +456,6 @@ static HTS_INLINE HTS_UNUSED const char *htsbuff_str(const htsbuff *b) {
  return b->buf;
 }

-/** True if 'count' records of >= 1 byte each fit in 'available' bytes; guards
-    an attacker-controlled count driving a large allocation. */
-static HTS_INLINE HTS_UNUSED hts_boolean hts_count_fits(size_t count,
-                                                        LLint available) {
-  return (available >= 0 && (LLint) count <= available) ? HTS_TRUE : HTS_FALSE;
-}
-
 /* Thin aliases over the libc allocator/memcpy (historical "t" suffix); no
   added bounds checking. freet() also NULLs the freed pointer and tolerates
   NULL. memcpybuff() despite the name is a raw memcpy: the caller owns the
--- a/src/htsselftest.c
+++ b/src/htsselftest.c
@@ -45,12 +45,10 @@ Please visit our Website: http://www.httrack.com
 #include "htscore.h"
 #include "htsdefines.h"
 #include "htslib.h"
-#include "htsparse.h"
 #include "htscache_selftest.h"
 #include "htsdns_selftest.h"
 #include "htscharset.h"
 #include "htsencoding.h"
-#include "htsftp.h"
 #include "htsmd5.h"
 #if HTS_USEZLIB
 #include "htszlib.h"
@@ -62,10 +60,6 @@ Please visit our Website: http://www.httrack.com
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
-#ifndef _WIN32
-#include <sys/socket.h>
-#include <unistd.h>
-#endif

 /* very minimalistic internal tests */
 static void basic_selftests(void) {
@@ -713,8 +707,7 @@ static int st_entities(httrackp *opt, int argc, char **argv) {
  }
  s = strdupt(argv[0]);
  enc = argc >= 2 ? argv[1] : "UTF-8";
-  if (s != NULL &&
-      hts_unescapeEntitiesWithCharset(s, s, strlen(s) + 1, enc) == 0) {
+  if (s != NULL && hts_unescapeEntitiesWithCharset(s, s, strlen(s), enc) == 0) {
    printf("%s\n", s);
    freet(s);
  } else {
@@ -723,34 +716,6 @@ static int st_entities(httrackp *opt, int argc, char **argv) {
  return 0;
 }

-/* The unescapers must reserve one byte for the trailing NUL: a 'max'-byte
-   dest holding 'max' output chars pre-fix wrote dest[max] (1-byte OOB, caught
-   by ASan). Both unescapeEntities and unescapeUrl share the guard. */
-static int st_unescape_bounds(httrackp *opt, int argc, char **argv) {
-  char dest[4];
-
-  (void) opt;
-  (void) argc;
-  (void) argv;
-  assertf(hts_unescapeEntities("abcd", dest, sizeof(dest)) == -1);
-  assertf(hts_unescapeUrl("abcd", dest, sizeof(dest)) == -1);
-  assertf(hts_unescapeEntities("abc", dest, sizeof(dest)) == 0);
-  assertf(strcmp(dest, "abc") == 0);
-  /* raw multi-byte UTF-8 flush path (bypasses the per-byte guard) */
-  assertf(hts_unescapeUrl("ab\xC3\xA9", dest, sizeof(dest)) == -1);
-  assertf(hts_unescapeUrl("a\xC3\xA9", dest, sizeof(dest)) == 0);
-  assertf(strcmp(dest, "a\xC3\xA9") == 0);
-  {
-    /* %xx-encoded flush path (utfBufferJ = lastJ rollback) */
-    char wide[8];
-
-    assertf(hts_unescapeUrl("%C3%A9", wide, sizeof(wide)) == 0);
-    assertf(strcmp(wide, "\xC3\xA9") == 0);
-  }
-  printf("unescape-bounds self-test OK\n");
-  return 0;
-}
-
 static int st_hashtable(httrackp *opt, int argc, char **argv) {
  char *snum;
  unsigned long count = 0;
@@ -1093,202 +1058,35 @@ static int st_resolve(httrackp *opt, int argc, char **argv) {
  return 0;
 }

-/* Extra args are key=value: adr= cdispo= statuscode= status= strip= urlhack=
-   no-www= no-slash= no-query= n83= type=, plus repeatable prior=adr|fil|sav
-   registering an already-crawled link (dedup/collision paths). */
-/* Parse raw response-header lines and print the naming-relevant fields. */
-static int st_header(httrackp *opt, int argc, char **argv) {
-  htsblk r;
-  int i;
-
-  (void) opt;
-  if (argc < 1) {
-    fprintf(stderr, "header: needs at least one raw header line\n");
-    return 1;
-  }
-  memset(&r, 0, sizeof(r));
-  for (i = 0; i < argc; i++) {
-    char BIGSTK line[HTS_URLMAXSIZE * 2];
-
-    strcpybuff(line, argv[i]);
-    treathead(NULL, "www.example.com", "/", &r, line);
-  }
-  printf("contenttype=%s cdispo=%s\n", r.contenttype, r.cdispo);
-  return 0;
-}
-
-/* Decode a body argument ("hex:FFD8.." or literal text) into buf. */
-static size_t st_decode_body(const char *arg, char *buf, size_t size) {
-  size_t n = 0;
-
-  if (strncmp(arg, "hex:", 4) == 0) {
-    const char *s = arg + 4;
-
-    for (; s[0] != '\0' && s[1] != '\0' && n + 1 < size; s += 2) {
-      unsigned int byte;
-
-      if (sscanf(s, "%2x", &byte) != 1)
-        break;
-      buf[n++] = (char) byte;
-    }
-  } else {
-    n = strlen(arg);
-    if (n >= size)
-      n = size - 1;
-    memcpy(buf, arg, n);
-  }
-  buf[n] = '\0';
-  return n;
-}
-
 static int st_savename(httrackp *opt, int argc, char **argv) {
  lien_adrfilsave afs;
  cache_back cache;
  struct_back *sback;
  hash_struct hash;
  lien_back headers;
-  const char *adr = "www.example.com";
-  const char *cdispo = NULL;
-  const char *body = NULL;
-  const char *cached = NULL;
-  const char *bodyfile = "st-savename-body.tmp";
-  int statuscode = HTTP_OK, status = 0;
-  int i;

  if (argc < 2) {
    fprintf(stderr, "savename: needs a fil and a content-type\n");
    return 1;
  }
-  /* knobs first: hash_init and the prior links depend on them */
-  for (i = 2; i < argc; i++) {
-    const char *const a = argv[i];
-
-    if (strncmp(a, "adr=", 4) == 0)
-      adr = a + 4;
-    else if (strncmp(a, "cdispo=", 7) == 0)
-      cdispo = a + 7;
-    else if (strncmp(a, "statuscode=", 11) == 0)
-      statuscode = atoi(a + 11);
-    else if (strncmp(a, "status=", 7) == 0)
-      status = atoi(a + 7);
-    else if (strncmp(a, "strip=", 6) == 0)
-      StringCopy(opt->strip_query, a + 6);
-    else if (strncmp(a, "urlhack=", 8) == 0)
-      opt->urlhack = atoi(a + 8) ? HTS_TRUE : HTS_FALSE;
-    else if (strncmp(a, "no-www=", 7) == 0)
-      opt->no_www_dedup = atoi(a + 7) ? HTS_TRUE : HTS_FALSE;
-    else if (strncmp(a, "no-slash=", 9) == 0)
-      opt->no_slash_dedup = atoi(a + 9) ? HTS_TRUE : HTS_FALSE;
-    else if (strncmp(a, "no-query=", 9) == 0)
-      opt->no_query_dedup = atoi(a + 9) ? HTS_TRUE : HTS_FALSE;
-    else if (strncmp(a, "n83=", 4) == 0)
-      opt->savename_83 = atoi(a + 4);
-    else if (strncmp(a, "type=", 5) == 0)
-      opt->savename_type = atoi(a + 5);
-    else if (strncmp(a, "body=", 5) == 0)
-      body = a + 5;
-    else if (strncmp(a, "cached=", 7) == 0)
-      cached = a + 7;
-    else if (strncmp(a, "prior=", 6) != 0) {
-      fprintf(stderr, "savename: unknown arg '%s'\n", a);
-      return 1;
-    }
-  }
  memset(&afs, 0, sizeof(afs));
-  strcpybuff(afs.af.adr, adr);
+  strcpybuff(afs.af.adr, "www.example.com");
  strcpybuff(afs.af.fil, argv[0]);

  memset(&cache, 0, sizeof(cache));
-  if (cached != NULL) { /* cached=<content-type>|<save name> */
-    char *dup = strdupt(cached);
-    char *const sep = strchr(dup, '|');
-    char locbuf[64] = "";
-    htsblk cr;
-
-    if (sep == NULL) {
-      fprintf(stderr, "savename: cached needs ctype|save\n");
-      return 1;
-    }
-    *sep = '\0';
-    /* one-entry cache in cwd, reopened read-only; body is PNG magic on
-       purpose: naming must not depend on stored content */
-    StringCopy(opt->path_log, "");
-    cache.type = 1;
-    cache.log = cache.errlog = stderr;
-    cache.hashtable = coucal_new(0);
-    cache_init(&cache, opt);
-    hts_init_htsblk(&cr);
-    cr.statuscode = HTTP_OK;
-    strcpybuff(cr.msg, "OK");
-    strcpybuff(cr.contenttype, dup);
-    cr.location = locbuf;
-    cr.adr = strdupt("\x89PNG\r\n\x1a\n");
-    cr.size = 8;
-    cache_add(opt, &cache, &cr, adr, argv[0], sep + 1, 1, NULL);
-    freet(cr.adr);
-    if (cache.zipOutput != NULL) {
-      zipClose(cache.zipOutput, NULL);
-      cache.zipOutput = NULL;
-    }
-    memset(&cache, 0, sizeof(cache));
-    cache.type = 1;
-    cache.log = cache.errlog = stderr;
-    cache.hashtable = coucal_new(0);
-    cache.ro = 1;
-    cache_init(&cache, opt);
-    freet(dup);
-  } else {
-    cache.hashtable = (void *) coucal_new(0);
-  }
+  cache.hashtable = (void *) coucal_new(0);

  sback = back_new(opt, opt->maxsoc * 32 + 1024);
-  /* same wiring as hts_mirror (htscore.c) */
  hash_init(opt, &hash, opt->urlhack);
-  hash.liens = (const lien_url *const *const *) &opt->liens;
-  opt->hash = &hash;
-  hts_record_init(opt);
-
-  for (i = 2; i < argc; i++) {
-    if (strncmp(argv[i], "prior=", 6) == 0) {
-      char *dup = strdupt(argv[i] + 6);
-      char *const p1 = strchr(dup, '|');
-      char *const p2 = p1 != NULL ? strchr(p1 + 1, '|') : NULL;
-
-      if (p2 == NULL) {
-        fprintf(stderr, "savename: prior needs adr|fil|sav\n");
-        return 1;
-      }
-      *p1 = *p2 = '\0';
-      if (!hts_record_link(opt, dup, p1 + 1, p2 + 1, "", "", NULL))
-        return 1;
-      freet(dup);
-    }
-  }

  memset(&headers, 0, sizeof(headers));
-  headers.status = status;
-  headers.r.statuscode = statuscode;
+  headers.status = 0;
+  headers.r.statuscode = HTTP_OK;
  strcpybuff(headers.r.contenttype, argv[1]);
-  if (cdispo != NULL)
-    strcpybuff(headers.r.cdispo, cdispo);
  strcpybuff(headers.url_fil, argv[0]);
-  if (body != NULL) { /* leading body bytes, exposed via url_sav */
-    char BIGSTK data[1024];
-    const size_t n = st_decode_body(body, data, sizeof(data));
-    FILE *const fp = fopen(bodyfile, "wb");
-
-    if (fp == NULL || fwrite(data, 1, n, fp) != n) {
-      fprintf(stderr, "savename: can not write %s\n", bodyfile);
-      return 1;
-    }
-    fclose(fp);
-    strcpybuff(headers.url_sav, bodyfile);
-  }

  url_savename(&afs, NULL, NULL, NULL, opt, sback, &cache, &hash, 0, 0,
               &headers);
-  if (body != NULL)
-    (void) UNLINK(bodyfile);
  printf("savename: %s\n", afs.save);
  return 0;
 }
@@ -1542,165 +1340,6 @@ static int st_urlhack(httrackp *opt, int argc, char **argv) {
  return 0;
 }

-/* #159: hts_redirect_same_savefile decides whether a redirect is a same-file
- * alias. */
-static int st_redirect_samefile(httrackp *opt, int argc, char **argv) {
-  (void) argc;
-  (void) argv;
-#define SAME(aa, fa, ab, fb) hts_redirect_same_savefile(opt, aa, fa, ab, fb)
-  /* scheme and userinfo collapse (the #159 case); a different path does not */
-  assertf(SAME("http://foo.com", "/a/b", "https://foo.com", "/a/b"));
-  assertf(SAME("http://user@foo.com", "/a", "http://foo.com", "/a"));
-  assertf(!SAME("http://foo.com", "/a", "http://foo.com", "/b"));
-  /* www stays distinct here; the crawl's dedup layer folds www, not this helper
-   */
-  opt->urlhack = HTS_TRUE;
-  opt->no_www_dedup = opt->no_slash_dedup = opt->no_query_dedup = HTS_FALSE;
-  assertf(!SAME("http://www.foo.com", "/a", "http://foo.com", "/a"));
-  /* slash/query fold only when the dedup flag is on */
-  assertf(SAME("https://foo.com", "/a//b", "http://foo.com", "/a/b"));
-  assertf(
-      SAME("https://foo.com", "/p?b=2&a=1", "http://foo.com", "/p?a=1&b=2"));
-  opt->no_slash_dedup = opt->no_query_dedup = HTS_TRUE;
-  assertf(!SAME("https://foo.com", "/a//b", "http://foo.com", "/a/b"));
-  assertf(
-      !SAME("https://foo.com", "/p?b=2&a=1", "http://foo.com", "/p?a=1&b=2"));
-  /* but a pure scheme alias still collapses regardless of dedup opt-outs */
-  assertf(SAME("http://foo.com", "/a/b", "https://foo.com", "/a/b"));
-  opt->no_slash_dedup = opt->no_query_dedup = HTS_FALSE;
-#undef SAME
-  printf("redirect-samefile self-test OK\n");
-  return 0;
-}
-
-// hts_finish_makeindex writes the footer, emits the refresh meta only when
-// makeindex_links==1, and clears *fp / sets *done. argv[0] is a writable dir.
-static int st_makeindex(httrackp *opt, int argc, char **argv) {
-  char path[HTS_URLMAXSIZE];
-  char buf[4096];
-  FILE *fp;
-  size_t n;
-  int done;
-
-  assertf(argc >= 1);
-  snprintf(path, sizeof(path), "%s/index.html", argv[0]);
-
-  /* single first link: footer + a refresh meta carrying the escaped URL */
-  done = 0;
-  fp = fopen(path, "wb");
-  assertf(fp != NULL);
-  hts_finish_makeindex(opt, &done, &fp, 1, "http://example.com/a b", "%s%s", "",
-                       "");
-  assertf(fp == NULL); /* the function closed and cleared it */
-  assertf(done != 0);
-  fp = fopen(path, "rb");
-  assertf(fp != NULL);
-  n = fread(buf, 1, sizeof(buf) - 1, fp);
-  fclose(fp);
-  buf[n] = '\0';
-  assertf(strstr(buf, "Mirror and index made by HTTrack") != NULL);
-  assertf(strstr(buf, "Refresh") != NULL);
-  assertf(strstr(buf, "example.com") != NULL);
-
-  /* no single link: footer only, no refresh meta */
-  done = 0;
-  fp = fopen(path, "wb");
-  assertf(fp != NULL);
-  hts_finish_makeindex(opt, &done, &fp, 0, NULL, "%s%s", "", "");
-  assertf(fp == NULL);
-  assertf(done != 0);
-  fp = fopen(path, "rb");
-  assertf(fp != NULL);
-  n = fread(buf, 1, sizeof(buf) - 1, fp);
-  fclose(fp);
-  buf[n] = '\0';
-  assertf(strstr(buf, "Mirror and index made by HTTrack") != NULL);
-  assertf(strstr(buf, "Refresh") == NULL);
-
-  UNLINK(path);
-  printf("makeindex self-test OK\n");
-  return 0;
-}
-
-/* Each inplace_escape_*() must equal escape_*() on a copy. */
-static int st_inplace_escape(httrackp *opt, int argc, char **argv) {
-  /* >255 bytes forces the helper's malloct path, not the stack buffer */
-  static char longstr[600];
-  static const char *const samples[] = {
-      "",          "abc",           "a b/c?d=e&f", "h\x8ello w\x94rld",
-      "a%b\"c<d>", "/path to/file", longstr};
-  static size_t (*const inplace[])(char *, size_t) = {
-      inplace_escape_in_url, inplace_escape_spc_url, inplace_escape_uri_utf,
-      inplace_escape_check_url, inplace_escape_uri};
-  static size_t (*const plain[])(const char *, char *, size_t) = {
-      escape_in_url, escape_spc_url, escape_uri_utf, escape_check_url,
-      escape_uri};
-  size_t i, f;
-
-  (void) opt;
-  (void) argc;
-  (void) argv;
-
-  memset(longstr, 'a', sizeof(longstr) - 1);
-  for (f = 0; f < sizeof(inplace) / sizeof(inplace[0]); f++) {
-    for (i = 0; i < sizeof(samples) / sizeof(samples[0]); i++) {
-      char ref[4096], work[4096];
-      size_t rret, iret;
-      rret = plain[f](samples[i], ref, sizeof(ref));
-      strcpybuff(work, samples[i]);
-      iret = inplace[f](work, sizeof(work));
-      assertf(iret == rret);
-      assertf(strcmp(work, ref) == 0);
-    }
-  }
-  printf("inplace-escape self-test OK\n");
-  return 0;
-}
-
-/* Pin HTS_HTMLESCAPE*_MAXEXP to each escaper's true max byte expansion. */
-static int st_escape_room(httrackp *opt, int argc, char **argv) {
-  /* N > 1023: where 6n outgrows the old 5n+1024 reservation */
-  enum { N = 2000 };
-
-  char *src = malloct(N + 1);
-  char *dst;
-  size_t room, got;
-  (void) opt;
-  (void) argc;
-  (void) argv;
-
-  /* _full worst case: a high byte expands to "&#xHH;" (6 bytes) */
-  memset(src, 0xE9, N);
-  src[N] = '\0';
-  room = (size_t) N * HTS_HTMLESCAPE_FULL_MAXEXP + 1024;
-  dst = malloct(room);
-  got = escape_for_html_print_full(src, dst, room);
-  assertf(got == (size_t) N * HTS_HTMLESCAPE_FULL_MAXEXP);
-  assertf(strlen(dst) == got);
-  freet(dst);
-
-  /* one factor short overflows (returns size), truncating the page: the bug */
-  room = (size_t) N * (HTS_HTMLESCAPE_FULL_MAXEXP - 1) + 1024;
-  dst = malloct(room);
-  got = escape_for_html_print_full(src, dst, room);
-  assertf(got == room);
-  freet(dst);
-
-  /* plain escaper worst case: '&' -> "&amp;" (5); high bytes stay verbatim */
-  memset(src, '&', N);
-  src[N] = '\0';
-  room = (size_t) N * HTS_HTMLESCAPE_MAXEXP + 1024;
-  dst = malloct(room);
-  got = escape_for_html_print(src, dst, room);
-  assertf(got == (size_t) N * HTS_HTMLESCAPE_MAXEXP);
-  assertf(strlen(dst) == got);
-  freet(dst);
-
-  freet(src);
-  printf("escape-room self-test OK\n");
-  return 0;
-}
-
 /* Default User-Agent: honest HTTrack token, no resurrected Windows 98. */
 static int st_useragent(httrackp *opt, int argc, char **argv) {
  const char *ua = StringBuff(opt->user_agent);
@@ -1970,86 +1609,6 @@ static int st_robots(httrackp *opt, int argc, char **argv) {
  return 0;
 }

-/* get_ftp_line must bound a hostile, CRLF-less reply into its internal
-   1024-byte buffer; ASan turns the pre-fix overflow into an abort here. */
-#ifndef _WIN32
-static int st_ftpline(httrackp *opt, int argc, char **argv) {
-  int sv[2];
-  char line[2048];
-  char flood[4096];
-
-  (void) opt;
-  (void) argc;
-  (void) argv;
-  memset(flood, 'x', sizeof(flood));
-  assertf(socketpair(AF_UNIX, SOCK_STREAM, 0, sv) == 0);
-  assertf(write(sv[1], "220 ", 4) == 4); // valid 3-digit code
-  assertf(write(sv[1], flood, sizeof(flood)) == (ssize_t) sizeof(flood));
-  assertf(write(sv[1], "\r\n", 2) == 2); // end the line so we return
-  close(sv[1]);
-  line[0] = '\0';
-  get_ftp_line(sv[0], line, sizeof(line), 5);
-  close(sv[0]);
-  printf("ftp-line self-test OK (bounded %d-byte reply)\n",
-         (int) sizeof(flood));
-  return 0;
-}
-#endif
-
-/* ftp_split_userpass: well-formed split, plus a hostile over-long userinfo
-   that pre-fix overran user[256]/pass[256]. */
-static int st_ftpuser(httrackp *opt, int argc, char **argv) {
-  char user[256], pass[256];
-  char in[1200];
-
-  (void) opt;
-  (void) argc;
-  (void) argv;
-  {
-    const char ok[] = "bob:secret@host/f"; // '@' at index 10
-
-    ftp_split_userpass(ok, ok + 11, user, sizeof(user), pass, sizeof(pass));
-    assertf(strcmp(user, "bob") == 0);
-    assertf(strcmp(pass, "secret") == 0);
-  }
-  memset(in, 'u', 400);
-  in[400] = ':';
-  memset(in + 401, 'p', 400);
-  in[801] = '@';
-  in[802] = '\0';
-  ftp_split_userpass(in, in + 802, user, sizeof(user), pass, sizeof(pass));
-  assertf(strlen(user) == sizeof(user) - 1);
-  assertf(strlen(pass) == sizeof(pass) - 1);
-  {
-    /* tight sizes + guard byte catch an off-by-one the 256 case can't */
-    char ubuf[16], pbuf[16];
-
-    memset(ubuf, 'Z', sizeof(ubuf));
-    memset(pbuf, 'Z', sizeof(pbuf));
-    ftp_split_userpass(in, in + 802, ubuf, 8, pbuf, 8);
-    assertf(strcmp(ubuf, "uuuuuuu") == 0);
-    assertf(strcmp(pbuf, "ppppppp") == 0);
-    assertf(ubuf[8] == 'Z' && pbuf[8] == 'Z');
-  }
-  printf("ftp-userpass self-test OK\n");
-  return 0;
-}
-
-/* hts_count_fits caps the .class constant-pool entry count to the file size,
-   rejecting the ~68 MB-per-file calloc DoS. */
-static int st_java(httrackp *opt, int argc, char **argv) {
-  (void) opt;
-  (void) argc;
-  (void) argv;
-  assertf(hts_count_fits(10, 1000) == HTS_TRUE);
-  assertf(hts_count_fits(0, 10) == HTS_TRUE);
-  assertf(hts_count_fits(65535, 10) == HTS_FALSE);
-  assertf(hts_count_fits(1, 0) == HTS_FALSE);
-  assertf(hts_count_fits(1, -1) == HTS_FALSE);
-  printf("java constant-pool cap self-test OK\n");
-  return 0;
-}
-
 /* ------------------------------------------------------------ */
 /* Registry: name -> handler, with a usage hint and a one-line description. */
 /* ------------------------------------------------------------ */
@@ -2070,8 +1629,6 @@ static const struct selftest_entry {
     st_stripquery},
    {"urlhack", "", "-%u url-hack sub-flag (www/slash/query) self-test",
     st_urlhack},
-    {"redirect-samefile", "", "same-file redirect detection self-test (#159)",
-     st_redirect_samefile},
    {"mime", "<filename>", "MIME type for a filename", st_mime},
    {"charset", "<charset> <string>",
     "convert a string to UTF-8 from a charset", st_charset},
@@ -2080,8 +1637,6 @@ static const struct selftest_entry {
    {"idna-decode", "<host>", "decode an IDNA/punycode hostname",
     st_idna_decode},
    {"entities", "<string> [encoding]", "unescape HTML entities", st_entities},
-    {"unescape-bounds", "", "unescapers reserve the NUL byte (no 1-byte OOB)",
-     st_unescape_bounds},
    {"hashtable", "<count|file>", "coucal hashtable stress test", st_hashtable},
    {"strsafe", "[overflow|overflow-buff [str]]", "bounded string-op self-test",
     st_strsafe},
@@ -2091,10 +1646,8 @@ static const struct selftest_entry {
     st_relative},
    {"resolve", "<link> <adr> <fil>", "resolve a link against an origin",
     st_resolve},
-    {"header", "<raw-header-line> ...", "response header-line parsing",
-     st_header},
-    {"savename", "<fil> <content-type> [key=value ...]",
-     "local save-name for a URL", st_savename},
+    {"savename", "<fil> <content-type>", "local save-name for a URL",
+     st_savename},
    {"cache", "<dir>", "cache read/write round-trip self-test", st_cache},
    {"cache-golden", "<dir> [regen]", "frozen cache-format read self-test",
     st_cache_golden},
@@ -2103,23 +1656,11 @@ static const struct selftest_entry {
    {"dns", "", "DNS resolver/cache self-test", st_dns},
    {"cookies", "", "cookie request-header self-test", st_cookies},
    {"useragent", "", "default User-Agent self-test", st_useragent},
-    {"makeindex", "[dir]", "hts_finish_makeindex footer/refresh self-test",
-     st_makeindex},
-    {"inplace-escape", "", "inplace_escape_* vs escape_* equivalence self-test",
-     st_inplace_escape},
-    {"escape-room", "", "HT_ADD_HTMLESCAPED* reservation-factor self-test",
-     st_escape_room},
    {"status", "", "HTTP status code -> reason phrase self-test", st_status},
    {"acceptencoding", "[dir]",
     "Accept-Encoding advertises gzip+deflate, both decode", st_acceptencoding},
    {"robots", "", "robots.txt RFC 9309 Allow/Disallow precedence self-test",
     st_robots},
-#ifndef _WIN32
-    {"ftp-line", "", "get_ftp_line bounds a hostile FTP reply line",
-     st_ftpline},
-#endif
-    {"ftp-userpass", "", "ftp_split_userpass bounds URL userinfo", st_ftpuser},
-    {"java", "", "java .class constant-pool count cap self-test", st_java},
 };

 static void list_selftests(void) {
--- a/tests/01_engine-escape-room.test
+++ b/tests/01_engine-escape-room.test
@@ -1,7 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# HT_ADD_HTMLESCAPED* must reserve the escaper's worst case (6 for _full).
-httrack -O /dev/null -#test=escape-room run | grep -q "escape-room self-test OK"
--- a/tests/01_engine-ftp-line.test
+++ b/tests/01_engine-ftp-line.test
@@ -1,7 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# get_ftp_line bounds a hostile CRLF-less FTP reply into its 1024-byte buffer.
-httrack -O /dev/null -#test=ftp-line run | grep -q "ftp-line self-test OK"
--- a/tests/01_engine-ftp-userpass.test
+++ b/tests/01_engine-ftp-userpass.test
@@ -1,7 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# ftp_split_userpass bounds an over-long user:pass@ from a hostile ftp:// URL.
-httrack -O /dev/null -#test=ftp-userpass run | grep -q "ftp-userpass self-test OK"
--- a/tests/01_engine-header.test
+++ b/tests/01_engine-header.test
@@ -1,29 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# Response header-line parsing (treathead via -#test=header <raw-line> ...).
-# Isolates the wire layer from url_savename, which strips traversal on its own.
-
-hdr() {
-    local want="$1"
-    shift
-    out="$(httrack -O /dev/null -#test=header "$@" | grep '^contenttype=')"
-    test "$out" == "$want" || {
-        echo "FAIL: $* -> '$out' (want '$want')"
-        exit 1
-    }
-}
-
-hdr 'contenttype=application/pdf cdispo=' 'Content-Type: application/pdf'
-
-# filename= is honored quoted or bare.
-hdr 'contenttype= cdispo=report.pdf' \
-    'Content-Disposition: attachment; filename="report.pdf"'
-hdr 'contenttype= cdispo=report.pdf' \
-    'Content-Disposition: attachment; filename=report.pdf'
-
-# Path components in the filename are dropped on the wire (RFC 2616).
-hdr 'contenttype= cdispo=evil.pdf' \
-    'Content-Disposition: attachment; filename="../../evil.pdf"'
--- a/tests/01_engine-inplace-escape.test
+++ b/tests/01_engine-inplace-escape.test
@@ -1,7 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# inplace_escape_*() must match escape_*() on a copy: guards the shared helper.
-httrack -O /dev/null -#test=inplace-escape run | grep -q "inplace-escape self-test OK"
--- a/tests/01_engine-java.test
+++ b/tests/01_engine-java.test
@@ -1,7 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# .class constant-pool count is capped to the file size (calloc DoS).
-httrack -O /dev/null -#test=java run | grep -q "java constant-pool cap self-test OK"
--- a/tests/01_engine-makeindex.test
+++ b/tests/01_engine-makeindex.test
@@ -1,12 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# hts_finish_makeindex writes the footer and gates the refresh meta on a single
-# first link (guards the macro->function extraction).
-dir=$(mktemp -d)
-trap 'rm -rf "$dir"' EXIT
-
-httrack -O /dev/null -#test=makeindex "$dir" run |
-    grep -q "makeindex self-test OK"
--- a/tests/01_engine-redirect.test
+++ b/tests/01_engine-redirect.test
@@ -1,9 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# #159: a redirect to a same-file alias (http<->https, user@host, ..) must be
-# followed through, not turned into a self-pointing "moved" stub. The decision
-# helper is exercised by the engine self-test.
-httrack -O /dev/null -#test=redirect-samefile run | grep -q "redirect-samefile self-test OK"
--- a/tests/01_engine-savename.test
+++ b/tests/01_engine-savename.test
@@ -3,38 +3,13 @@

 set -euo pipefail

-# Local save-name resolution (url_savename via -#test=savename <fil> <content-type> [key=value ...]).
-# name() asserts on the basename, full() on the whole path; prior= registers an
-# already-crawled link whose sav is rooted under the -O path (/dev/null here).
-
-# resolve httrack before cd: make check puts a RELATIVE ../src on PATH
-httrack_bin=$(cd "$(dirname "$(command -v httrack)")" && pwd)/httrack
-
-# scratch dir: body= and cached= write temp files (st-savename-body.tmp, hts-cache/)
-scratch=$(mktemp -d)
-trap 'rm -rf "$scratch"' EXIT
-cd "$scratch"
-
-run() {
-    "$httrack_bin" -O /dev/null -#test=savename "$@" | sed -n 's/^savename: //p'
-}
+# Local save-name extension resolution (url_savename via -#test=savename <fil> <content-type>).
+# Asserts on the basename of "savename: <path>".

 name() {
-    local fil="$1" ctype="$2" want="$3"
-    shift 3
-    out="$(run "$fil" "$ctype" "$@")"
-    test "${out##*/}" == "$want" || {
-        echo "FAIL: '$fil' '$ctype' $* -> '$out' (want '$want')"
-        exit 1
-    }
-}
-
-full() {
-    local fil="$1" ctype="$2" want="$3"
-    shift 3
-    out="$(run "$fil" "$ctype" "$@")"
-    test "$out" == "$want" || {
-        echo "FAIL: '$fil' '$ctype' $* -> '$out' (want '$want')"
+    out="$(httrack -O /dev/null -#test=savename "$1" "$2" | sed -n 's/^savename: //p')"
+    test "${out##*/}" == "$3" || {
+        echo "FAIL: '$1' '$2' -> '$out' (want '$3')"
        exit 1
    }
 }
@@ -64,96 +39,3 @@ name '/types/data.json' 'application/json' 'data.json'

 # Agreeing type must not rewrite the extension's casing (no strip-and-reappend).
 name '/x.JPG' 'image/jpeg' 'x.JPG'
-
-# A Content-Disposition filename replaces the URL name outright.
-name '/x.php' 'application/pdf' 'report.pdf' cdispo=report.pdf
-name '/download' 'text/html' 'setup.exe' cdispo=setup.exe
-
-# Reserved characters in a hostile Content-Disposition name are sanitized.
-name '/x.php' 'application/pdf' 'set_up.exe' 'cdispo=set:up.exe'
-
-# The md5-of-query suffix lands inside a Content-Disposition name too.
-name '/x.php?id=1' 'application/pdf' 'report681a.pdf' cdispo=report.pdf
-
-# Still-downloading path (status=-1): mime drives the ext, cdispo is ignored
-# there (the deliberately unfolded 4th resolve_extension variant).
-name '/x.pdf' 'text/html' 'x.html' status=-1
-name '/x.html' 'text/html' 'x.html' status=-1
-name '/x.php' 'application/pdf' 'x.pdf' status=-1 cdispo=report.pdf
-
-# Contested type (wire disagrees with a specific ext): the wire is trusted and
-# body bytes are not consulted; pinned so a content-based tie-break shows up
-# as an explicit flip of these rows.
-name '/photo.jpg' 'image/png' 'photo.png' body=hex:FFD8FFE000104A46
-name '/photo.jpg' 'image/png' 'photo.png' body=hex:89504E470D0A1A0A
-name '/photo.jpg' 'image/png' 'photo.png'
-name '/doc.pdf' 'text/html' 'doc.html' body=hex:255044462D312E34
-name '/doc.pdf' 'text/html' 'doc.html' 'body=<html><body>soft 404</body></html>'
-name '/style.css' 'image/png' 'style.png' 'body=body { }'
-
-# A redirect answer resolves nothing: delayed placeholder name.
-name '/x.php' 'text/html' 'x.0.delayed' statuscode=301
-
-# Root and query-only URLs get index + the md5-of-query suffix.
-name '/' 'text/html' 'index.html'
-name '/?a=1' 'text/html' 'index3872.html'
-
-# Same URL crawled before: reuse its sav verbatim (case preserved).
-full '/X.PHP' 'text/html' 'www.example.com/CASE.HTML' \
-    'prior=www.example.com|/X.PHP|www.example.com/CASE.HTML'
-
-# Another URL owns the name: collision suffix -2, then -3, case-insensitively.
-name '/x.php' 'text/html' 'x-2.html' \
-    'prior=www.example.com|/other.html|/dev/null/www.example.com/x.html'
-name '/x.php' 'text/html' 'x-3.html' \
-    'prior=www.example.com|/o1.html|/dev/null/www.example.com/x.html' \
-    'prior=www.example.com|/o2.html|/dev/null/www.example.com/x-2.html'
-name '/INDEX.HTML' 'text/html' 'INDEX-2.HTML' \
-    'prior=www.example.com|/index.html|/dev/null/www.example.com/index.html'
-
-# Same basename in another directory is NOT a collision.
-name '/x.php' 'text/html' 'x.html' \
-    'prior=www.example.com|/sub/x.html|/dev/null/www.example.com/sub/x.html'
-
-# 8-3 modes: DOS truncates every component to 8+3, ISO9660 level 2 to 31.
-full '/directory-long/verylongfilename.html' 'text/html' \
-    '/dev/null/EXAMPLE/DIRECTOR/VERYLONG.HTM' n83=1
-full '/directory-long/verylongfilename.html' 'text/html' \
-    '/dev/null/EXAMPLE_C/DIRECTORY_LONG/VERYLONGFILENAME.HTM' n83=2
-name '/verylongfilename.php' 'text/html' 'VERYLO-2.HTM' n83=1 \
-    'prior=www.example.com|/other.html|/dev/null/EXAMPLE/VERYLONG.HTM'
-
-# urlhack dedup (#271): // collapse and www-strip map to the prior link's sav;
-# the per-feature negatives opt out and take a fresh name.
-full '/a//b.php' 'text/html' '/dev/null/www.example.com/a/PRIOR.html' \
-    'prior=www.example.com|/a/b.php|/dev/null/www.example.com/a/PRIOR.html'
-full '/a//b.php' 'text/html' '/dev/null/www.example.com/a/b.html' no-slash=1 \
-    'prior=www.example.com|/a/b.php|/dev/null/www.example.com/a/PRIOR.html'
-full '/w.php' 'text/html' '/dev/null/www.example.com/W-PRIOR.html' adr=example.com \
-    'prior=www.example.com|/w.php|/dev/null/www.example.com/W-PRIOR.html'
-full '/w.php' 'text/html' '/dev/null/example.com/w.html' adr=example.com no-www=1 \
-    'prior=www.example.com|/w.php|/dev/null/www.example.com/W-PRIOR.html'
-
-# Distinct URLs must stay distinct under urlhack (no over-normalization).
-full '/a//b.php' 'text/html' '/dev/null/www.example.com/a/b.html' \
-    'prior=www.example.com|/a/c.php|/dev/null/www.example.com/a/C-PRIOR.html'
-
-# --strip-query (#112): stripped key dedups onto the prior sav; without the
-# option the same URLs stay distinct.
-full '/page.php?id=3&sid=42' 'text/html' '/dev/null/www.example.com/PAGE-PRIOR.html' \
-    strip=sid 'prior=www.example.com|/page.php?id=3|/dev/null/www.example.com/PAGE-PRIOR.html'
-full '/page.php?id=3&sid=42' 'text/html' '/dev/null/www.example.com/page475b.html' \
-    'prior=www.example.com|/page.php?id=3|/dev/null/www.example.com/PAGE-PRIOR.html'
-
-# A kept key that differs must still block the dedup (no over-stripping).
-full '/page.php?id=3&sid=42' 'text/html' '/dev/null/www.example.com/page475b.html' \
-    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).
-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' "$(printf 'a%.0s' {1..210})"
--- a/tests/01_engine-unescape-bounds.test
+++ b/tests/01_engine-unescape-bounds.test
@@ -1,7 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# Entity/URL unescapers reserve one byte for the trailing NUL (no 1-byte OOB).
-httrack -O /dev/null -#test=unescape-bounds run | grep -q "unescape-bounds self-test OK"
--- a/tests/01_zlib-savename-cached.test
+++ b/tests/01_zlib-savename-cached.test
@@ -1,33 +0,0 @@
-#!/bin/bash
-#
-
-set -euo pipefail
-
-# Update-run naming from a real cache entry (-#test=savename cached=<ctype>|<save>).
-# Named 01_zlib-*: the cache writer needs zlib, which the MSan job can't run.
-
-# resolve httrack before cd: make check puts a RELATIVE ../src on PATH
-httrack_bin=$(cd "$(dirname "$(command -v httrack)")" && pwd)/httrack
-
-scratch=$(mktemp -d)
-trap 'rm -rf "$scratch"' EXIT
-cd "$scratch"
-
-name() {
-    local fil="$1" ctype="$2" want="$3"
-    shift 3
-    out="$("$httrack_bin" -O /dev/null -#test=savename "$fil" "$ctype" "$@" | sed -n 's/^savename: //p')"
-    test "${out##*/}" == "$want" || {
-        echo "FAIL: '$fil' '$ctype' $* -> '$out' (want '$want')"
-        exit 1
-    }
-}
-
-# Names are re-derived from the stored headers on every run: neither the
-# recorded save name nor the cached body bytes change the verdict (pinned).
-name '/photo.jpg' 'image/png' 'photo.png' 'cached=image/png|www.example.com/photo.jpg'
-name '/photo.jpg' 'image/png' 'photo.png' 'cached=image/png|www.example.com/photo.png'
-name '/photo.jpg' 'image/jpeg' 'photo.jpg' 'cached=image/jpeg|www.example.com/photo.png'
-name '/style.css' 'image/png' 'style.png' 'cached=image/png|www.example.com/style.css'
-# agreement keeps the URL ext verbatim (.jpeg), never canonicalized to .jpg
-name '/photo.jpeg' 'image/jpeg' 'photo.jpeg' 'cached=image/jpeg|www.example.com/photo.jpeg'
--- a/tests/15_local-types.test
+++ b/tests/15_local-types.test
@@ -15,10 +15,6 @@ bash "$top_srcdir/tests/local-crawl.sh" --errors 0 \
    --found 'types/photo.png' \
    --found 'types/doc.pdf' \
    --found 'types/lie.html' --not-found 'types/lie.png' \
-    --found 'types/wrongtype.png' --not-found 'types/wrongtype.jpg' \
-    --found 'types/bigtype.png' --not-found 'types/bigtype.jpg' \
-    --found 'types/packed.png' --not-found 'types/packed.jpg' \
-    --found 'types/mutant.png' --not-found 'types/mutant.jpg' \
    --found 'types/report.html' --not-found 'types/report.pdf' \
    --found 'types/page.htm' --not-found 'types/page.html' \
    --found 'types/script.js' \
--- a/tests/18_local-update.test
+++ b/tests/18_local-update.test
@@ -12,7 +12,4 @@ bash "$top_srcdir/tests/local-crawl.sh" --errors 0 --rerun \
    --found 'types/report.html' --not-found 'types/report.pdf' \
    --found 'types/notype.png' --not-found 'types/notype.html' \
    --found 'types/lie.html' \
-    --found 'types/wrongtype.png' --not-found 'types/wrongtype.jpg' \
-    --found 'types/packed.png' --not-found 'types/packed.jpg' \
-    --found 'types/mutant.png' --not-found 'types/mutant.jpg' \
    httrack 'BASEURL/types/index.html'
--- a/tests/30_local-fragment-link.test
+++ b/tests/30_local-fragment-link.test
@@ -1,13 +0,0 @@
-#!/bin/bash
-# Issue #279: an anchored link (target.html#sec, quoted or bare) fetches the
-# target with the fragment dropped (strict server 400s on a '#' in the request)
-# but keeps it in the rewritten local link so the anchor still works.
-set -e
-
-: "${top_srcdir:=..}"
-
-bash "$top_srcdir/tests/local-crawl.sh" --errors 0 \
-    --found 'fraglink/target.html' \
-    --file-matches 'fraglink/index.html' 'href=target\.html#sec' \
-    --file-matches 'fraglink/index.html' 'href="target\.html#sec2"' \
-    httrack 'BASEURL/fraglink/index.html'
--- a/tests/31_local-javaclass.test
+++ b/tests/31_local-javaclass.test
@@ -1,23 +0,0 @@
-#!/bin/bash
-# The java plugin must load (versioned dlopen name) and parse a .class
-# constant pool: a resource named only inside Foo.class gets crawled.
-set -e
-
-: "${top_srcdir:=..}"
-
-tmproot=$(mktemp -d)
-trap 'rm -rf "$tmproot"' EXIT
-mkdir "$tmproot/javaclass"
-
-cat >"$tmproot/javaclass/index.html" <<'EOF'
-<html><body><a href="Foo.class">applet</a></body></html>
-EOF
-printf 'GIF89a' >"$tmproot/javaclass/hello.gif"
-# magic/minor/major, count=2, one CONSTANT_Utf8 "hello.gif", class/superclass
-printf '\xCA\xFE\xBA\xBE\x00\x00\x00\x32\x00\x02\x01\x00\x09hello.gif\x00\x00\x00\x00' \
-    >"$tmproot/javaclass/Foo.class"
-
-bash "$top_srcdir/tests/local-crawl.sh" --root "$tmproot" --errors 0 \
-    --found 'javaclass/Foo.class' \
-    --found 'javaclass/hello.gif' \
-    httrack 'BASEURL/javaclass/index.html'
--- a/tests/32_local-cdispo.test
+++ b/tests/32_local-cdispo.test
@@ -1,17 +0,0 @@
-#!/bin/bash
-#
-# Content-Disposition names the saved file: the attachment filename replaces
-# the URL-derived name, and a traversal filename is reduced to its last
-# component, inside the mirror.
-
-set -euo pipefail
-
-: "${top_srcdir:=..}"
-
-bash "$top_srcdir/tests/local-crawl.sh" --errors 0 \
-    --found 'cdispo/report.pdf' \
-    --file-matches 'cdispo/report.pdf' '%PDF' \
-    --not-found 'cdispo/fetch.pdf' \
-    --found 'cdispo/evil.pdf' \
-    --not-found 'evil.pdf' \
-    httrack 'BASEURL/cdispo/index.html'
--- a/tests/33_local-delayed.test
+++ b/tests/33_local-delayed.test
@@ -1,20 +0,0 @@
-#!/bin/bash
-#
-# Degenerate delayed-type paths (#5/#107 family): redirects that never resolve
-# a name must drop cleanly -- no .delayed leftovers (audited by local-crawl.sh),
-# no "bogus state" cache warnings, resolvable links still land correctly.
-
-set -euo pipefail
-
-: "${top_srcdir:=..}"
-
-bash "$top_srcdir/tests/local-crawl.sh" --rerun --errors 0 \
-    --found 'delayed/real.pdf' \
-    --file-matches 'delayed/real.pdf' '%PDF' \
-    --found 'delayed/notype.bin.html' \
-    --found 'delayed/empty.html' \
-    --not-found 'delayed/noloc.html' \
-    --not-found 'delayed/selfloop.html' \
-    --not-found 'delayed/chain9.pdf' \
-    --log-not-found 'bogus state' \
-    httrack 'BASEURL/delayed/index.html'
--- a/tests/Makefile.am
+++ b/tests/Makefile.am
@@ -6,7 +6,6 @@ EXTRA_DIST = $(TESTS) crawl-test.sh run-all-tests.sh check-network.sh \
 	local-crawl.sh local-server.py server.crt server.key \
 	server-root/simple/basic.html server-root/simple/link.html \
 	server-root/stripquery/index.html server-root/stripquery/a.html \
-	server-root/fraglink/index.html server-root/fraglink/target.html \
 	fixtures/cache-golden/hts-cache/new.zip

 TESTS_ENVIRONMENT =
@@ -35,20 +34,12 @@ TESTS = \
 	01_engine-entities.test \
 	01_engine-filelist.test \
 	01_engine-filter.test \
-	01_engine-ftp-line.test \
-	01_engine-ftp-userpass.test \
 	01_engine-hashtable.test \
-	01_engine-header.test \
 	01_engine-idna.test \
-	01_engine-escape-room.test \
-	01_engine-inplace-escape.test \
-	01_engine-java.test \
-	01_engine-makeindex.test \
 	01_engine-mime.test \
 	01_engine-parse.test \
 	01_engine-pause.test \
 	01_engine-rcfile.test \
-	01_engine-redirect.test \
 	01_engine-relative.test \
 	01_engine-robots.test \
 	01_engine-savename.test \
@@ -58,13 +49,11 @@ TESTS = \
 	01_engine-stripquery.test \
 	01_engine-strsafe.test \
 	01_engine-urlhack.test \
-	01_engine-unescape-bounds.test \
 	01_engine-useragent.test \
 	01_zlib-acceptencoding.test \
 	01_zlib-cache.test \
 	01_zlib-cache-golden.test \
 	01_zlib-cache-writefail.test \
-	01_zlib-savename-cached.test \
 	02_manpage-regen.test \
 	02_update-cache.test \
 	10_crawl-simple.test \
@@ -91,10 +80,6 @@ TESTS = \
 	26_local-strip-query.test \
 	27_local-cookies-file.test \
 	28_local-pause.test \
-	29_local-redirect-fragment.test \
-	30_local-fragment-link.test \
-	31_local-javaclass.test \
-	32_local-cdispo.test \
-	33_local-delayed.test
+	29_local-redirect-fragment.test

 CLEANFILES = check-network_sh.cache
--- a/tests/local-crawl.sh
+++ b/tests/local-crawl.sh
@@ -15,11 +15,8 @@
 #   bash local-crawl.sh [--tls] [--root DIR] [--cookie NAME=VALUE ...] \
 #       --errors N --files N --found PATH ... --directory PATH ... \
 #       --log-found REGEX ... --log-not-found REGEX ... \
-#       --file-matches PATH REGEX ... --file-not-matches PATH REGEX ... \
 #       httrack BASEURL/some/path [httrack-args...]
 # --log-found/--log-not-found grep (ERE) the crawl's hts-log.txt.
-# --file-matches/--file-not-matches grep (ERE) a mirrored file (PATH under the
-# host root), to assert rewritten link/content survived the crawl.
 # --cookie writes a Netscape cookies.txt (scoped to the discovered host:port,
 # which the ephemeral port forces into the cookie domain) and passes it to
 # httrack via --cookies-file, to exercise preloaded cookies.
@@ -124,10 +121,6 @@ while test "$pos" -lt "$nargs"; do
        audit+=("${args[$pos]}" "${args[$((pos + 1))]}")
        pos=$((pos + 1))
        ;;
-    --file-matches | --file-not-matches)
-        audit+=("${args[$pos]}" "${args[$((pos + 1))]}" "${args[$((pos + 2))]}")
-        pos=$((pos + 2))
-        ;;
    httrack)
        pos=$((pos + 1))
        break
@@ -246,14 +239,6 @@ done
 test -n "$hostroot" || die "could not find host root under $out"
 debug "host root: $hostroot"

-# A completed crawl must leave no .delayed temporaries (issue #107)
-info "checking for leftover .delayed files"
-leftovers=$(find "$out" -name '*.delayed' 2>/dev/null | head -5)
-if test -z "$leftovers"; then result "OK"; else
-    result "leftover: $leftovers"
-    exit 1
-fi
-
 # --- audit -------------------------------------------------------------------
 i=0
 while test "$i" -lt "${#audit[@]}"; do
@@ -309,24 +294,6 @@ while test "$i" -lt "${#audit[@]}"; do
            exit 1
        else result "OK"; fi
        ;;
-    --file-matches)
-        path="${audit[$((i + 1))]}"
-        i=$((i + 2))
-        info "checking ${path} matches ${audit[$i]}"
-        if grep -aqE "${audit[$i]}" "${hostroot}/${path}"; then result "OK"; else
-            result "no match"
-            exit 1
-        fi
-        ;;
-    --file-not-matches)
-        path="${audit[$((i + 1))]}"
-        i=$((i + 2))
-        info "checking ${path} lacks ${audit[$i]}"
-        if grep -aqE "${audit[$i]}" "${hostroot}/${path}"; then
-            result "matched"
-            exit 1
-        else result "OK"; fi
-        ;;
    esac
    i=$((i + 1))
 done
--- a/tests/local-server.py
+++ b/tests/local-server.py
@@ -14,7 +14,6 @@ stdlib only (http.server + ssl) -- no new build or runtime dependency.
 """

 import argparse
-import gzip
 import os
 import time
 from http.server import SimpleHTTPRequestHandler, ThreadingHTTPServer
@@ -135,14 +134,12 @@ class Handler(SimpleHTTPRequestHandler):

    # --- type/extension matrix (issue #267 family) -------------------------

-    def send_raw(self, body, content_type, extra_headers=()):
+    def send_raw(self, body, content_type):
        """Send a raw body with an explicit Content-Type, or none at all when
        content_type is None (to observe httrack's typeless-file naming)."""
        self.send_response(200)
        if content_type is not None:
            self.send_header("Content-Type", content_type)
-        for name, value in extra_headers:
-            self.send_header(name, value)
        self.send_header("Content-Length", str(len(body)))
        self.end_headers()
        if self.command != "HEAD":
@@ -151,8 +148,6 @@ class Handler(SimpleHTTPRequestHandler):
    # Fake-binary blobs for the image/pdf/typeless cases.
    FAKE_PNG = b"\x89PNG\r\n\x1a\n" + b"\x00" * 64
    FAKE_PDF = b"%PDF-1.4\n" + b"\x00" * 64
-    FAKE_JPEG = b"\xff\xd8\xff\xe0" + b"\x00" * 64
-    BIG_JPEG = b"\xff\xd8\xff\xe0" + bytes(range(256)) * 64  # > sniff window

    # path -> (body, content_type); None sends no header, "" sends an empty
    # Content-Type value (no usable type, must be treated like None).
@@ -164,8 +159,6 @@ class Handler(SimpleHTTPRequestHandler):
        "/types/notype.pdf": (FAKE_PDF, None),
        "/types/emptyct.png": (FAKE_PNG, ""),
        "/types/lie.png": (FAKE_PNG, "text/html"),
-        "/types/wrongtype.jpg": (FAKE_JPEG, "image/png"),
-        "/types/bigtype.jpg": (BIG_JPEG, "image/png"),
        "/types/report.pdf": (b"<html><body>real page</body></html>", "text/html"),
        "/types/page.htm": (b"<html><body>htm page</body></html>", "text/html"),
        "/types/script.js": (b"var x = 1;\n", "application/javascript"),
@@ -183,10 +176,6 @@ class Handler(SimpleHTTPRequestHandler):
            '\t<a href="notype.pdf">notypepdf</a>\n'
            '\t<img src="emptyct.png" />\n'
            '\t<img src="lie.png" />\n'
-            '\t<img src="wrongtype.jpg" />\n'
-            '\t<img src="bigtype.jpg" />\n'
-            '\t<img src="mutant.jpg" />\n'
-            '\t<img src="packed.jpg" />\n'
            '\t<a href="report.pdf">report</a>\n'
            '\t<a href="page.htm">htm</a>\n'
            '\t<script src="script.js"></script>\n'
@@ -201,25 +190,6 @@ class Handler(SimpleHTTPRequestHandler):
        body, ctype = self.TYPE_MATRIX[path]
        self.send_raw(body, ctype)

-    # content changes between crawls: run 1 sniffs JPEG, the update pass must
-    # keep the run-1 name (recorded verdict) even though the body is now PNG
-    MUTANT_SEEN = set()
-
-    def route_types_mutant(self):
-        path = urlsplit(self.path).path
-        body = self.FAKE_PNG if path in self.MUTANT_SEEN else self.FAKE_JPEG
-        if self.command != "HEAD":
-            self.MUTANT_SEEN.add(path)
-        self.send_raw(body, "image/png")
-
-    # gzip on the wire: the sniff must see the decoded body, not the stream
-    def route_types_packed(self):
-        self.send_raw(
-            gzip.compress(self.FAKE_JPEG),
-            "image/png",
-            extra_headers=[("Content-Encoding", "gzip")],
-        )
-
    # --- MIME-type exclusion abort (issue #58) -----------------------------
    # A -mime:application/pdf filter must abort the transfer once the header
    # arrives, not download the whole body and discard it.
@@ -384,27 +354,6 @@ class Handler(SimpleHTTPRequestHandler):
        if self.command != "HEAD":
            self.wfile.write(body)

-    # Content-Disposition naming: the attachment filename replaces the
-    # URL-derived name; path components in it are stripped (RFC 2616).
-    CDISPO_NAMES = {
-        "/cdispo/fetch.php": "report.pdf",
-        "/cdispo/evil.php": "../../evil.pdf",
-    }
-
-    def route_cdispo_index(self):
-        self.send_html(
-            '\t<a href="fetch.php">report</a>\n' '\t<a href="evil.php">evil</a>\n'
-        )
-
-    def route_cdispo(self):
-        filename = self.CDISPO_NAMES[urlsplit(self.path).path]
-        cdispo = 'attachment; filename="%s"' % filename
-        self.send_raw(
-            self.FAKE_PDF,
-            "application/pdf",
-            extra_headers=[("Content-Disposition", cdispo)],
-        )
-
    # 302 whose Location carries a #fragment (#204): the fragment is a UA anchor
    # that must be dropped before the target is fetched. A leaked '#' reaches the
    # strict-server guard below and 400s.
@@ -420,50 +369,6 @@ class Handler(SimpleHTTPRequestHandler):
    def route_redir_target(self):
        self.send_raw(b"<html><body>redirect target</body></html>\n", "text/html")

-    # --- delayed-type degenerate paths (issues #5/#107) --------------------
-    def route_delayed_index(self):
-        self.send_html(
-            '\t<a href="noloc.php">noloc</a>\n'
-            '\t<a href="selfloop.php">selfloop</a>\n'
-            '\t<a href="chain1.php">chain</a>\n'
-            '\t<a href="redir.php">redir</a>\n'
-            '\t<a href="notype.bin">notype</a>\n'
-            '\t<a href="empty.php">empty</a>\n'
-        )
-
-    def send_redirect(self, location):
-        self.send_response(302, "Found")
-        if location is not None:
-            self.send_header("Location", location)
-        self.send_header("Content-Length", "0")
-        self.end_headers()
-
-    def route_delayed_noloc(self):
-        self.send_redirect(None)  # 302 without Location: name never resolves
-
-    def route_delayed_selfloop(self):
-        self.send_redirect("selfloop.php")
-
-    def route_delayed_chain(self):
-        # chain1..chain9: one more hop than the type-check redirect budget
-        n = int(urlsplit(self.path).path.rsplit("chain", 1)[1].split(".")[0])
-        if n < 9:
-            self.send_redirect("chain%d.php" % (n + 1))
-        else:
-            self.send_raw(self.FAKE_PDF, "application/pdf")
-
-    def route_delayed_redir(self):
-        self.send_redirect("real.pdf")
-
-    def route_delayed_realpdf(self):
-        self.send_raw(self.FAKE_PDF, "application/pdf")
-
-    def route_delayed_notype(self):
-        self.send_raw(self.FAKE_PDF, None)
-
-    def route_delayed_empty(self):
-        self.send_raw(b"", "text/html")  # 200 + Content-Length: 0
-
    ROUTES = {
        "/cookies/entrance.php": route_entrance,
        "/cookies/second.php": route_second,
@@ -479,10 +384,6 @@ class Handler(SimpleHTTPRequestHandler):
        "/types/notype.pdf": route_types,
        "/types/emptyct.png": route_types,
        "/types/lie.png": route_types,
-        "/types/wrongtype.jpg": route_types,
-        "/types/bigtype.jpg": route_types,
-        "/types/mutant.jpg": route_types_mutant,
-        "/types/packed.jpg": route_types_packed,
        "/types/report.pdf": route_types,
        "/types/page.htm": route_types,
        "/types/script.js": route_types,
@@ -505,25 +406,6 @@ class Handler(SimpleHTTPRequestHandler):
        "/mimex/index.html": route_mimex_index,
        "/mimex/blob.pdf": route_mimex_blob,
        "/mimex/real.html": route_mimex_real,
-        "/cdispo/index.html": route_cdispo_index,
-        "/cdispo/fetch.php": route_cdispo,
-        "/cdispo/evil.php": route_cdispo,
-        "/delayed/index.html": route_delayed_index,
-        "/delayed/noloc.php": route_delayed_noloc,
-        "/delayed/selfloop.php": route_delayed_selfloop,
-        "/delayed/redir.php": route_delayed_redir,
-        "/delayed/real.pdf": route_delayed_realpdf,
-        "/delayed/notype.bin": route_delayed_notype,
-        "/delayed/empty.php": route_delayed_empty,
-        "/delayed/chain1.php": route_delayed_chain,
-        "/delayed/chain2.php": route_delayed_chain,
-        "/delayed/chain3.php": route_delayed_chain,
-        "/delayed/chain4.php": route_delayed_chain,
-        "/delayed/chain5.php": route_delayed_chain,
-        "/delayed/chain6.php": route_delayed_chain,
-        "/delayed/chain7.php": route_delayed_chain,
-        "/delayed/chain8.php": route_delayed_chain,
-        "/delayed/chain9.php": route_delayed_chain,
        "/redir/index.html": route_redir_index,
        "/redir/go.php": route_redir_go,
        "/redir/target.html": route_redir_target,
--- a/tests/server-root/fraglink/index.html
+++ b/tests/server-root/fraglink/index.html
@@ -1,4 +0,0 @@
-<html><body>
-<a href=target.html#sec>unquoted fragment link</a>
-<a href="target.html#sec2">quoted fragment link</a>
-</body></html>
--- a/tests/server-root/fraglink/target.html
+++ b/tests/server-root/fraglink/target.html
@@ -1 +0,0 @@
-<html><body><a name="sec"></a><a name="sec2"></a>target</body></html>
				`@@ -1 +0,0 @@`
				`<html><body><a name="sec"></a><a name="sec2"></a>target</body></html>`