Merge pull request #406 from xroche/debian/libhttrack3-rename

debian: rename libhttrack2 to libhttrack3 to follow the SONAME

.clang-format

@@ -16,6 +16,7 @@ BasedOnStyle: LLVM
 SpaceAfterCStyleCast: true   # "(int) x", overwhelmingly dominant (542 vs 7)
 SortIncludes: false          # C include order can be significant; never reorder
 IncludeBlocks: Preserve      # do not merge/reflow include groups
+SeparateDefinitionBlocks: Always  # blank line between definitions (readability)
 
 # Stated explicitly for robustness against base-style drift (these match LLVM):
 IndentWidth: 2

.github/workflows/ci.yml

@@ -320,6 +320,21 @@ jobs:
   lint:
     name: lint (shellcheck, shfmt)
     runs-on: ubuntu-24.04
+    # Every tracked shell script; the globs expand at run time. Kept here so the
+    # shellcheck and shfmt steps below cannot drift apart.
+    env:
+      SHELL_SCRIPTS: >-
+        .githooks/pre-commit
+        bootstrap
+        build.sh
+        html/div/search.sh
+        man/makeman.sh
+        src/htsbasiccharsets.sh
+        src/htsentities.sh
+        src/webhttrack
+        tests/*.sh
+        tests/*.test
+        tools/mkdeb.sh
     steps:
       - uses: actions/checkout@v6
 
@@ -332,12 +347,11 @@ jobs:
           sudo apt-get install -y --no-install-recommends shellcheck shfmt
           shfmt --version
 
-      # Lint the scripts we maintain; the legacy scripts are a separate cleanup.
       - name: shellcheck
-        run: shellcheck man/makeman.sh tools/mkdeb.sh .githooks/pre-commit tests/*.test tests/check-network.sh
+        run: shellcheck $SHELL_SCRIPTS
 
       - name: shfmt
-        run: shfmt -d -i 4 man/makeman.sh tools/mkdeb.sh .githooks/pre-commit
+        run: shfmt -d -i 4 $SHELL_SCRIPTS
 
   # Check clang-format on CHANGED LINES ONLY. The engine predates clang-format
   # (it was shaped by an old Visual Studio formatter) and does not round-trip,

debian/changelog

@@ -1,6 +1,22 @@
+httrack (3.49.8-2) unstable; urgency=medium
+
+  * Rename libhttrack2 to libhttrack3 to follow the SONAME, which the 3.49.8
+    ABI bump moved to libhttrack.so.3 (package-name-doesnt-match-sonames). In
+    3.49.8-1 the libhttrack2.files glob still matched .so.2, so the runtime
+    libraries fell through into the httrack package and libhttrack2 shipped no
+    library. The new .files uses a .so.3* wildcard so a future SONAME bump no
+    longer silently misplaces the libraries. New binary package, via NEW.
+  * Drop the stale debian/libhttrack-swf1.files: the swf module is no longer
+    built and no libhttrack-swf1 package exists.
+
+ -- Xavier Roche <xavier@debian.org>  Sat, 20 Jun 2026 14:42:13 +0200
+
 httrack (3.49.8-1) unstable; urgency=medium
 
-  * New upstream release.
+  * New upstream release: HTTPS-proxy CONNECT tunnelling and wider srcset
+    parsing, a batch of crawler and parser fixes (CSS @import, xmlns
+    namespaces, relative paths, RFC 6265 cookies), and security hardening of
+    the parser and of buffer copies throughout the engine.
   * Drop the OpenSSL linking exception from the license: OpenSSL 3.0+ is
     Apache-2.0 and GPL-compatible, so it is no longer needed. httrack is now
     plain GPL-3.0-or-later. Updated debian/copyright accordingly.
@@ -14,7 +30,7 @@ httrack (3.49.8-1) unstable; urgency=medium
     the QA debcheck page. Depend on firefox-esr | chromium | www-browser
     instead.
 
- -- Xavier Roche <xavier@debian.org>  Sun, 07 Jun 2026 14:29:24 +0200
+ -- Xavier Roche <xavier@debian.org>  Sat, 20 Jun 2026 13:02:08 +0200
 
 httrack (3.49.7-2) unstable; urgency=medium
 

debian/control

@@ -58,13 +58,13 @@ Description: webhttrack common files
  This package is the common files of webhttrack, website copier and
  mirroring utility
 
-Package: libhttrack2
+Package: libhttrack3
 Architecture: any
 Multi-Arch: same
 Section: libs
-Replaces: libhttrack1
-Conflicts: libhttrack1
 Depends: ${misc:Depends}, ${shlibs:Depends}
+Replaces: libhttrack2, httrack (<< 3.49.8-2~)
+Breaks: libhttrack2, httrack (<< 3.49.8-2~)
 Description: Httrack website copier library
  This package is the library part of httrack, website copier and mirroring
  utility

debian/libhttrack-swf1.files

@@ -1,2 +0,0 @@
-usr/lib/*/libhtsswf.so.1.0.0
-usr/lib/*/libhtsswf.so.1

debian/libhttrack2.files

@@ -1,5 +0,0 @@
-usr/lib/*/libhttrack.so.2.0.49
-usr/lib/*/libhttrack.so.2
-usr/lib/*/libhtsjava.so.2.0.49
-usr/lib/*/libhtsjava.so.2
-usr/share/httrack/templates

debian/libhttrack2.lintian-overrides

@@ -1,3 +0,0 @@
-# The shared libraries ship without a versioned symbols control file (ABI is
-# tracked via the SONAME and a strict =version dependency, see debian/rules).
-libhttrack2: no-symbols-control-file usr/lib/*

debian/libhttrack3.files

@@ -0,0 +1,3 @@
+usr/lib/*/libhttrack.so.3*
+usr/lib/*/libhtsjava.so.3*
+usr/share/httrack/templates

debian/libhttrack3.lintian-overrides

@@ -0,0 +1,3 @@
+# The shared libraries ship without a versioned symbols control file (ABI is
+# tracked via the SONAME plus a >= upstream-version dependency, see debian/rules).
+libhttrack3: no-symbols-control-file usr/lib/*

debian/rules

@@ -135,7 +135,7 @@ binary-arch: build install
 	dh_makeshlibs -a -X/usr/lib/$(DEB_HOST_MULTIARCH)/httrack/libtest --version-info
 	dh_installdeb -a
 	# we depend on the current version (ABI may change)
-	dh_shlibdeps -a -ldebian/libhttrack2/usr/lib/$(DEB_HOST_MULTIARCH)
+	dh_shlibdeps -a -ldebian/libhttrack3/usr/lib/$(DEB_HOST_MULTIARCH)
 	dh_gencontrol -a
 	dh_md5sums -a
 	dh_builddeb -a

history.txt

@@ -5,12 +5,31 @@ HTTrack Website Copier release history:
 This file lists all changes and fixes that have been made for HTTrack
 
 3.49-8
++ New: tunnel HTTPS downloads through the configured HTTP proxy via CONNECT (#85)
++ New: parse every candidate URL in <img> and <source> srcset lists (#326)
 + Changed: dropped the obsolete OpenSSL linking exception (OpenSSL 3.0+ is Apache-2.0 and GPL-compatible); httrack is now plain GPLv3-or-later
-+ Fixed: link libhtsjava and the libtest examples directly against libc
++ Fixed: several out-of-bounds reads in the HTML/CSS parser on hostile input (#94, #396)
++ Fixed: stored XSS via an unescaped URL in the generated page footer (#165)
++ Fixed: hardened buffer copies throughout the engine against overflow
++ Fixed: capture conditional CSS @import URLs (#94)
++ Fixed: don't crawl xmlns namespace declarations as links (#191)
++ Fixed: don't mistake the method argument of XMLHttpRequest.open for a URL (#218)
++ Fixed: percent-encode parentheses when rewriting CSS url() targets (#163)
++ Fixed: collapse ../ in file:// URLs and widen relative-link handling (#137, #162)
++ Fixed: drop the obsolete $Version/$Path attributes from the request Cookie header, per RFC 6265 (#151)
++ Fixed: keep empty quoted arguments when reloading doit.log for --update/--continue (#106)
++ Fixed: raise the User-Agent and custom-header length limits (#152)
++ Fixed: abort on a long log path (lock-file buffer too small) (#183)
++ Fixed: race in lazy mutex initialization (#297)
++ Fixed: sub-second mtime precision when comparing local files on POSIX (#383)
++ Fixed: modernize OpenSSL TLS initialization for the 3.x to 4.x transition (#308)
 + Fixed: in-place changes made by the postprocess callback were not applied (Roman Sęk)
 + Fixed: "preffered" typo in the help text and man page (yosinn1-blip)
 + Fixed: corrections and updates of the Russian translation (German Aizek)
 + Fixed: corrections and updates of the Danish translation (scootergrisen)
++ Fixed: link libhtsjava and the libtest examples directly against libc
++ New: documented the public library API headers and typed the option fields as named enums
++ Fixed: numerous build, packaging, CI and test-coverage improvements (out-of-tree builds, sanitizer/distcheck CI, shell and Python linting, AppStream metainfo)
 
 3.49-7
 + Fixed: keep generated config.h architecture-independent (Debian #1133728)

html/div/search.sh

@@ -1,8 +1,7 @@
-
 #!/bin/sh
 
 # Simple indexing test using HTTrack
-# A "real" script/program would use advanced search, and 
+# A "real" script/program would use advanced search, and
 # use dichotomy to find the word in the index.txt file
 # This script is really basic and NOT optimized, and
 # should not be used for professional purpose :)
@@ -11,50 +10,49 @@ TESTSITE="http://localhost/"
 
 # Create an index if necessary
 if ! test -f "index.txt"; then
-	echo "Building the index .."
-	rm -rf test
-	httrack --display "$TESTSITE"  -%I -O test
-	mv test/index.txt ./
+    echo "Building the index .."
+    rm -rf test
+    httrack --display "$TESTSITE" -%I -O test
+    mv test/index.txt ./
 fi
 
 # Convert crlf to lf
-if test "`head index.txt -n 1 | tr '\r' '#' | grep -c '#'`" = "1"; then
-	echo "Converting index to Unix LF style (not CR/LF) .."
-	mv -f index.txt index.txt.old
-	cat index.txt.old|tr -d '\r' > index.txt
+if test "$(head index.txt -n 1 | tr '\r' '#' | grep -c '#')" = "1"; then
+    echo "Converting index to Unix LF style (not CR/LF) .."
+    mv -f index.txt index.txt.old
+    tr -d '\r' <index.txt.old >index.txt
 fi
 
 keyword=-
 while test -n "$keyword"; do
-	printf "Enter a keyword: "
-	read keyword
+    printf "Enter a keyword: "
+    read -r keyword
 
-	if test -n "$keyword"; then
-		FOUNDK="`grep -niE \"^$keyword\" index.txt`"
+    if test -n "$keyword"; then
+        FOUNDK="$(grep -niE "^$keyword" index.txt)"
 
-		if test -n "$FOUNDK"; then	
-			if ! test `echo "$FOUNDK"|wc -l` = "1"; then
-				# Multiple matches
-				printf "Found multiple keywords: "
-				echo "$FOUNDK"|cut -f2 -d':'|tr '\n' ' '
-				echo ""
-				echo "Use keyword$ to find only one"
-			else
-				# One match
-				N=`echo "$FOUNDK"|cut -f1 -d':'`
-				PM=`tail +$N index.txt|grep -nE "\("|head -n 1`
-				if ! echo "$PM"|grep "ignored">/dev/null; then
-					M=`echo $PM|cut -f1 -d':'`
-					echo "Found in:"
-					cat index.txt | tail "+$N" | head -n "$M" | grep -E "[0-9]* " | cut -f2 -d' '
-				else
-					echo "keyword ignored (too many hits)"
-				fi
-					fi
-		else
-			echo "not found"
-		fi
+        if test -n "$FOUNDK"; then
+            if ! test "$(echo "$FOUNDK" | wc -l)" = "1"; then
+                # Multiple matches
+                printf "Found multiple keywords: "
+                echo "$FOUNDK" | cut -f2 -d':' | tr '\n' ' '
+                echo ""
+                echo "Use keyword$ to find only one"
+            else
+                # One match
+                N=$(echo "$FOUNDK" | cut -f1 -d':')
+                PM=$(tail "+$N" index.txt | grep -nE "\(" | head -n 1)
+                if ! echo "$PM" | grep "ignored" >/dev/null; then
+                    M=$(echo "$PM" | cut -f1 -d':')
+                    echo "Found in:"
+                    tail "+$N" index.txt | head -n "$M" | grep -E "[0-9]* " | cut -f2 -d' '
+                else
+                    echo "keyword ignored (too many hits)"
+                fi
+            fi
+        else
+            echo "not found"
+        fi
 
-	fi
+    fi
 done
-

src/htsarrays.h

@@ -48,112 +48,115 @@ Please visit our Website: http://www.httrack.com
 /* Abort (with the failed byte count) when a growth allocation fails. The
    array macros never return an out-of-memory error; they assert and abort. */
 static void hts_record_assert_memory_failed(const size_t size) {
-  fprintf(stderr, "memory allocation failed (%lu bytes)", \
-          (long int) size); \
-  assertf(! "memory allocation failed"); \
+  fprintf(stderr, "memory allocation failed (%lu bytes)", (long int) size);
+  assertf(!"memory allocation failed");
 }
 
 /** Dynamic array of T elements. **/
-#define TypedArray(T) \
-  struct {            \
-    /** Elements. **/ \
-    union {           \
-      /** Typed. **/  \
-      T* elts;        \
-      /** Opaque. **/ \
-      void* ptr;      \
-    } data;           \
-    /** Count. **/    \
-    size_t size;      \
-    /** Capacity. **/ \
-    size_t capa;      \
+#define TypedArray(T)                                                          \
+  struct {                                                                     \
+    /** Elements. **/                                                          \
+    union {                                                                    \
+      /** Typed. **/                                                           \
+      T *elts;                                                                 \
+      /** Opaque. **/                                                          \
+      void *ptr;                                                               \
+    } data;                                                                    \
+    /** Count. **/                                                             \
+    size_t size;                                                               \
+    /** Capacity. **/                                                          \
+    size_t capa;                                                               \
   }
 
 /** Initializer for an empty array (no backing store, size and capacity 0). **/
-#define EMPTY_TYPED_ARRAY { { NULL }, 0, 0 }
+#define EMPTY_TYPED_ARRAY {{NULL}, 0, 0}
 
 /** Array size, in elements. **/
-#define TypedArraySize(A)   ((A).size)
+#define TypedArraySize(A) ((A).size)
 
 /** Array capacity, in elements. **/
-#define TypedArrayCapa(A)   ((A).capa)
+#define TypedArrayCapa(A) ((A).capa)
 
 /**
- * Remaining free space, in elements. 
+ * Remaining free space, in elements.
  * Macro, first element evaluated multiple times.
  **/
-#define TypedArrayRoom(A)   ( TypedArrayCapa(A) - TypedArraySize(A) )
+#define TypedArrayRoom(A) (TypedArrayCapa(A) - TypedArraySize(A))
 
 /** Array elements, of type T*. **/
-#define TypedArrayElts(A)   ((A).data.elts)
+#define TypedArrayElts(A) ((A).data.elts)
 
 /** Array pointer, of type void*. **/
-#define TypedArrayPtr(A)    ((A).data.ptr)
+#define TypedArrayPtr(A) ((A).data.ptr)
 
 /** Size of T. **/
-#define TypedArrayWidth(A)  (sizeof(*TypedArrayElts(A)))
+#define TypedArrayWidth(A) (sizeof(*TypedArrayElts(A)))
 
 /** Nth element of the array, as an lvalue. No bounds check; N must be
     < TypedArraySize(A). **/
 #define TypedArrayNth(A, N) (TypedArrayElts(A)[N])
 
 /**
- * Tail of the array (outside the array). 
+ * Tail of the array (outside the array).
  * The returned pointer points to the beginning of TypedArrayRoom(A)
  * free elements.
  **/
 #define TypedArrayTail(A) (TypedArrayNth(A, TypedArraySize(A)))
 
 /**
-  * Ensure at least 'ROOM' elements can be put in the remaining space.
-  * After a call to this macro, TypedArrayRoom(A) is guaranteed to be at 
-  * least equal to 'ROOM'.
-  **/
-#define TypedArrayEnsureRoom(A, ROOM) do { \
-  const size_t room_ = (ROOM); \
-  while (TypedArrayRoom(A) < room_) { \
-    TypedArrayCapa(A) = TypedArrayCapa(A) < 16 ? 16 : TypedArrayCapa(A) * 2; \
-  } \
-  TypedArrayPtr(A) = realloc(TypedArrayPtr(A), \
-                             TypedArrayCapa(A)*TypedArrayWidth(A)); \
-  if (TypedArrayPtr(A) == NULL) { \
-    hts_record_assert_memory_failed(TypedArrayCapa(A)*TypedArrayWidth(A)); \
-  } \
-} while(0)
+ * Ensure at least 'ROOM' elements can be put in the remaining space.
+ * After a call to this macro, TypedArrayRoom(A) is guaranteed to be at
+ * least equal to 'ROOM'.
+ **/
+#define TypedArrayEnsureRoom(A, ROOM)                                          \
+  do {                                                                         \
+    const size_t room_ = (ROOM);                                               \
+    while (TypedArrayRoom(A) < room_) {                                        \
+      TypedArrayCapa(A) = TypedArrayCapa(A) < 16 ? 16 : TypedArrayCapa(A) * 2; \
+    }                                                                          \
+    TypedArrayPtr(A) =                                                         \
+        realloc(TypedArrayPtr(A), TypedArrayCapa(A) * TypedArrayWidth(A));     \
+    if (TypedArrayPtr(A) == NULL) {                                            \
+      hts_record_assert_memory_failed(TypedArrayCapa(A) * TypedArrayWidth(A)); \
+    }                                                                          \
+  } while (0)
 
 /** Add an element. Macro, first element evaluated multiple times. **/
-#define TypedArrayAdd(A, E) do { \
-  TypedArrayEnsureRoom(A, 1); \
-  assertf(TypedArraySize(A) < TypedArrayCapa(A)); \
-  TypedArrayTail(A) = (E); \
-  TypedArraySize(A)++; \
-} while(0)
+#define TypedArrayAdd(A, E)                                                    \
+  do {                                                                         \
+    TypedArrayEnsureRoom(A, 1);                                                \
+    assertf(TypedArraySize(A) < TypedArrayCapa(A));                            \
+    TypedArrayTail(A) = (E);                                                   \
+    TypedArraySize(A)++;                                                       \
+  } while (0)
 
 /**
- * Add 'COUNT' elements from 'PTR'. 
+ * Add 'COUNT' elements from 'PTR'.
  * Macro, first element evaluated multiple times.
  **/
-#define TypedArrayAppend(A, PTR, COUNT) do { \
-  const size_t count_ = (COUNT); \
-  /* This 1-case is to benefit from type safety. */ \
-  if (count_ == 1) { \
-    TypedArrayAdd(A, *(PTR)); \
-  } else { \
-    const void *const source_ = (PTR); \
-    TypedArrayEnsureRoom(A, count_); \
-    assertf(count_ <= TypedArrayRoom(A)); \
-    memcpy(&TypedArrayTail(A), source_, count_ * TypedArrayWidth(A)); \
-    TypedArraySize(A) += count_; \
-  } \
-} while(0)
+#define TypedArrayAppend(A, PTR, COUNT)                                        \
+  do {                                                                         \
+    const size_t count_ = (COUNT);                                             \
+    /* This 1-case is to benefit from type safety. */                          \
+    if (count_ == 1) {                                                         \
+      TypedArrayAdd(A, *(PTR));                                                \
+    } else {                                                                   \
+      const void *const source_ = (PTR);                                       \
+      TypedArrayEnsureRoom(A, count_);                                         \
+      assertf(count_ <= TypedArrayRoom(A));                                    \
+      memcpy(&TypedArrayTail(A), source_, count_ *TypedArrayWidth(A));         \
+      TypedArraySize(A) += count_;                                             \
+    }                                                                          \
+  } while (0)
 
 /** Clear an array, freeing memory and clearing size and capacity. **/
-#define TypedArrayFree(A) do { \
-  if (TypedArrayPtr(A) != NULL) { \
-    TypedArrayCapa(A) = TypedArraySize(A) = 0; \
-    free(TypedArrayPtr(A)); \
-    TypedArrayPtr(A) = NULL; \
-  } \
-} while(0)
+#define TypedArrayFree(A)                                                      \
+  do {                                                                         \
+    if (TypedArrayPtr(A) != NULL) {                                            \
+      TypedArrayCapa(A) = TypedArraySize(A) = 0;                               \
+      free(TypedArrayPtr(A));                                                  \
+      TypedArrayPtr(A) = NULL;                                                 \
+    }                                                                          \
+  } while (0)
 
 #endif

src/htsbasenet.h

@@ -41,7 +41,7 @@ Please visit our Website: http://www.httrack.com
 
 #ifdef _WIN32
 
-#if HTS_INET6==0
+#if HTS_INET6 == 0
 #include <winsock2.h>
 #else
 
@@ -49,13 +49,14 @@ Please visit our Website: http://www.httrack.com
 #define WIN32_LEAN_AND_MEAN
 // KB955045 (http://support.microsoft.com/kb/955045)
 // To execute an application using this function on earlier versions of Windows
-// (Windows 2000, Windows NT, and Windows Me/98/95), then it is mandatary to #include Ws2tcpip.h
-// and also Wspiapi.h. When the Wspiapi.h header file is included, the 'getaddrinfo' function is
-// #defined to the 'WspiapiGetAddrInfo' inline function in Wspiapi.h. 
+// (Windows 2000, Windows NT, and Windows Me/98/95), then it is mandatary to
+// #include Ws2tcpip.h and also Wspiapi.h. When the Wspiapi.h header file is
+// included, the 'getaddrinfo' function is #defined to the 'WspiapiGetAddrInfo'
+// inline function in Wspiapi.h.
 #include <ws2tcpip.h>
 #include <Wspiapi.h>
-//#include <winsock2.h>
-//#include <tpipv6.h>
+// #include <winsock2.h>
+// #include <tpipv6.h>
 
 #endif
 

src/htsbasiccharsets.sh

@@ -3,57 +3,59 @@
 
 # Change this to download files
 if false; then
-echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-*.TXT" | lftp
-echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/PC/CP*.TXT" | lftp
-echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP*.TXT" | lftp
-echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/EBCDIC/CP*.TXT" | lftp
-echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MISC/CP*.TXT" | lftp
-echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MISC/KOI8*.TXT" | lftp
-rm -f CP932.TXT CP936.TXT CP949.TXT CP950.TXT
+    echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/ISO8859/8859-*.TXT" | lftp
+    echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/PC/CP*.TXT" | lftp
+    echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP*.TXT" | lftp
+    echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/EBCDIC/CP*.TXT" | lftp
+    echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MISC/CP*.TXT" | lftp
+    echo "mget ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/MISC/KOI8*.TXT" | lftp
+    rm -f CP932.TXT CP936.TXT CP949.TXT CP950.TXT
 fi
 
 # Produce code
-printf "/** GENERATED FILE ($0), DO NOT EDIT **/\n\n"
-for i in *.TXT ; do
-  echo "processing $i" >&2
-  grep -vE "^(#|$)" $i | grep -E "^0x" | sed -e 's/[[:space:]]/ /g' | cut -f1,2 -d' ' | \
-  (
-    unset arr
-    while read LINE ; do
-      from=$[$(echo $LINE | cut -f1 -d' ')]
-      if ! test -n "$from"; then
-        echo "error with $i" >&2
-        exit 1
-      elif test $from -ge 256; then
-        echo "out-of-range ($LINE) with $i" >&2
-        exit 1
-      fi
-      to=$(echo $LINE | cut -f2 -d' ') 
-      arr[$from]=$to
-    done
-    name=$(echo $i | tr 'A-Z' 'a-z' | tr '-' '_' | sed -e 's/\.txt//' -e 's/8859/iso_8859/')
-    printf "/* Table for $i */\nstatic const hts_UCS4 table_${name}[256] = {\n  "
-    i=0
-    while test "$i" -lt 256; do
-      if test "$i" -gt 0; then
-        printf ", "
-        if test $[${i}%8] -eq 0; then
-          printf "\n  "
-        fi
-      fi
-      value=${arr[$i]:-0}
-      printf "0x%04x" $value
-      i=$[${i}+1]
-    done
-    printf " };\n\n"
-  )
-  echo "processed $i" >&2
+printf '/** GENERATED FILE (%s), DO NOT EDIT **/\n\n' "$0"
+for i in *.TXT; do
+    echo "processing $i" >&2
+    grep -vE "^(#|$)" "$i" | grep -E "^0x" | sed -e 's/[[:space:]]/ /g' | cut -f1,2 -d' ' |
+        (
+            unset arr
+            while read -r LINE; do
+                from=$(($(echo "$LINE" | cut -f1 -d' ')))
+                if ! test -n "$from"; then
+                    echo "error with $i" >&2
+                    exit 1
+                elif test $from -ge 256; then
+                    echo "out-of-range ($LINE) with $i" >&2
+                    exit 1
+                fi
+                to=$(echo "$LINE" | cut -f2 -d' ')
+                arr[from]=$to
+            done
+            # shellcheck disable=SC2018,SC2019 # charset filenames are ASCII; keep C-locale A-Z/a-z
+            name=$(echo "$i" | tr 'A-Z' 'a-z' | tr '-' '_' | sed -e 's/\.txt//' -e 's/8859/iso_8859/')
+            printf '/* Table for %s */\nstatic const hts_UCS4 table_%s[256] = {\n  ' "$i" "$name"
+            idx=0
+            while test "$idx" -lt 256; do
+                if test "$idx" -gt 0; then
+                    printf ", "
+                    if test $((idx % 8)) -eq 0; then
+                        printf "\n  "
+                    fi
+                fi
+                value=${arr[$idx]:-0}
+                printf "0x%04x" "$value"
+                idx=$((idx + 1))
+            done
+            printf " };\n\n"
+        )
+    echo "processed $i" >&2
 done
 
 # Indexes
 printf "static const struct {\n  const char *name;\n  const hts_UCS4 *table;\n} table_mappings[] = {\n"
-for i in *.TXT ; do
-  name=$(echo $i | tr 'A-Z' 'a-z' | tr '-' '_' | sed -e 's/\.txt//' -e 's/8859/iso_8859/')
-  printf "  { \"$(echo $name | tr -d '_')\", table_${name} },\n"
+for i in *.TXT; do
+    # shellcheck disable=SC2018,SC2019 # charset filenames are ASCII; keep C-locale A-Z/a-z
+    name=$(echo "$i" | tr 'A-Z' 'a-z' | tr '-' '_' | sed -e 's/\.txt//' -e 's/8859/iso_8859/')
+    printf '  { "%s", table_%s },\n' "$(echo "$name" | tr -d '_')" "$name"
 done
 printf "  { NULL, NULL }\n};\n"

src/htsbauth.h

@@ -68,14 +68,15 @@ struct t_cookie {
 #ifdef HTS_INTERNAL_BYTECODE
 
 /* cookies */
-int cookie_add(t_cookie * cookie, const  char *cook_name, const  char *cook_value,
-               const  char *domain, const  char *path);
+int cookie_add(t_cookie *cookie, const char *cook_name, const char *cook_value,
+               const char *domain, const char *path);
 
-int cookie_del(t_cookie * cookie, const char *cook_name, const char *domain, const char *path);
+int cookie_del(t_cookie *cookie, const char *cook_name, const char *domain,
+               const char *path);
 
-int cookie_load(t_cookie * cookie, const char *path, const char *name);
+int cookie_load(t_cookie *cookie, const char *path, const char *name);
 
-int cookie_save(t_cookie * cookie, const char *name);
+int cookie_save(t_cookie *cookie, const char *name);
 
 void cookie_insert(char *s, size_t s_size, const char *ins);
 
@@ -83,7 +84,8 @@ void cookie_delete(char *s, size_t s_size, size_t pos);
 
 const char *cookie_get(char *buffer, const char *cookie_base, int param);
 
-char *cookie_find(char *s, const char *cook_name, const char *domain, const char *path);
+char *cookie_find(char *s, const char *cook_name, const char *domain,
+                  const char *path);
 
 char *cookie_nextfield(char *a);
 
@@ -92,12 +94,13 @@ char *cookie_nextfield(char *a);
 /** Register credentials (auth = base-64 user:pass) for the prefix derived from
     adr (host) and fil (path). No-op returning 0 if cookie is NULL, allocation
     fails, or a matching prefix is already stored; returns 1 on insertion. */
-int bauth_add(t_cookie * cookie, const char *adr, const char *fil, const char *auth);
+int bauth_add(t_cookie *cookie, const char *adr, const char *fil,
+              const char *auth);
 
 /** Return the stored base-64 credentials whose prefix matches adr+fil, or NULL
     if none (or cookie is NULL). Returned pointer aliases the jar's bauth_chain;
     caller must not free it. */
-char *bauth_check(t_cookie * cookie, const char *adr, const char *fil);
+char *bauth_check(t_cookie *cookie, const char *adr, const char *fil);
 
 /** Build the auth lookup key (host + path, query string stripped, truncated at
     the last '/') from adr and fil into prefix; returns prefix. Caller must

src/htsconfig.h

@@ -52,12 +52,12 @@ Please visit our Website: http://www.httrack.com
 #define DEFAULT_FTP "index.txt"
 
 // extension par défaut pour fichiers n'en ayant pas
-#define DEFAULT_EXT       ".html"
+#define DEFAULT_EXT ".html"
 #define DEFAULT_EXT_SHORT ".htm"
-//#define DEFAULT_BIN_EXT   ".bin"
-//#define DEFAULT_BIN_EXT_SHORT ".bin"
-//#define DEFAULT_EXT       ".txt"
-//#define DEFAULT_EXT_SHORT ".txt"
+// #define DEFAULT_BIN_EXT   ".bin"
+// #define DEFAULT_BIN_EXT_SHORT ".bin"
+// #define DEFAULT_EXT       ".txt"
+// #define DEFAULT_EXT_SHORT ".txt"
 
 // éviter les /nul, /con..
 #define HTS_OVERRIDE_DOS_FOLDERS 1
@@ -87,7 +87,8 @@ Please visit our Website: http://www.httrack.com
 // fast cache (build hash table)
 #define HTS_FAST_CACHE 1
 
-// le > peut être considéré comme un tag de fermeture de commentaire (<!-- > est valide)
+// le > peut être considéré comme un tag de fermeture de commentaire (<!-- > est
+// valide)
 #define GT_ENDS_COMMENT 1
 
 // always adds a '/' at the end if a '~' is encountered (/~smith -> /~smith/)
@@ -97,7 +98,8 @@ Please visit our Website: http://www.httrack.com
 #define HTS_STRIP_DOUBLE_SLASH 0
 
 // case-sensitive pour les dossiers et fichiers (0/1)
-// [normalement 1, mais pose des problèmes (url malformée par exemple) et n'est pas très utile..
+// [normalement 1, mais pose des problèmes (url malformée par exemple) et n'est
+// pas très utile..
 // ..et pas bcp respecté]
 // REMOVED
 // #define HTS_CASSE 0

src/htscoremain.c

@@ -2787,6 +2787,47 @@ static int hts_main_internal(int argc, char **argv, httrackp * opt) {
                   return 0;
                 }
                 break;
+              case 'l': /* lienrelatif: relative link from curr_fil to link */
+                if (na + 2 >= argc) {
+                  HTS_PANIC_PRINTF(
+                      "Option #l needs a link and a current-file path");
+                  printf(
+                      "Example: '-#l' 'host/dir/img.gif' 'host/dir/p.html'\n");
+                  htsmain_free();
+                  return -1;
+                } else {
+                  char s[HTS_URLMAXSIZE * 2];
+
+                  if (lienrelatif(s, sizeof(s), argv[na + 1], argv[na + 2]) ==
+                      0)
+                    printf("relative=%s\n", s);
+                  else
+                    printf("relative=<ERROR>\n");
+                  htsmain_free();
+                  return 0;
+                }
+                break;
+              case 'i': /* ident_url_relatif: resolve a link -> adr/fil */
+                if (na + 3 >= argc) {
+                  HTS_PANIC_PRINTF(
+                      "Option #i needs a link, an origin address and file");
+                  printf("Example: '-#i' '../img.gif' 'www.foo.com' "
+                         "'/d/p.html'\n");
+                  htsmain_free();
+                  return -1;
+                } else {
+                  lien_adrfil af;
+                  const int r = ident_url_relatif(argv[na + 1], argv[na + 2],
+                                                  argv[na + 3], &af);
+
+                  if (r == 0)
+                    printf("adr=%s fil=%s\n", af.adr, af.fil);
+                  else
+                    printf("error=%d\n", r);
+                  htsmain_free();
+                  return 0;
+                }
+                break;
               case '2':        // mimedefs
                 if (na + 1 >= argc) {
                   HTS_PANIC_PRINTF("Option #2 needs to be followed by an URL");

src/htsdefines.h

@@ -82,9 +82,9 @@ typedef struct t_hts_callbackarg t_hts_callbackarg;
 /* Entry points of a --wrapper plug-in: hts_plug(opt, argv) is called once to
    install the wrapper (argv is the wrapper's argument string), hts_unplug(opt)
    once to tear it down. Both return non-zero on success. */
-typedef int (*t_hts_plug) (httrackp * opt, const char *argv);
+typedef int (*t_hts_plug)(httrackp *opt, const char *argv);
 
-typedef int (*t_hts_unplug) (httrackp * opt);
+typedef int (*t_hts_unplug)(httrackp *opt);
 
 /* Engine callback prototypes. Each is one hook the engine fires at a defined
    point of a mirror; a wrapper installs the ones it cares about in the
@@ -92,27 +92,27 @@ typedef int (*t_hts_unplug) (httrackp * opt);
    returns are 1 to continue/accept, 0 to abort/refuse unless noted. */
 
 /* Called once when the wrapper is installed; allocate per-run state here. */
-typedef void (*t_hts_htmlcheck_init) (t_hts_callbackarg * carg);
+typedef void (*t_hts_htmlcheck_init)(t_hts_callbackarg *carg);
 
 /* Called once when the wrapper is removed; release per-run state here. */
-typedef void (*t_hts_htmlcheck_uninit) (t_hts_callbackarg * carg);
+typedef void (*t_hts_htmlcheck_uninit)(t_hts_callbackarg *carg);
 
 /* Fired at the start of a mirror, after options are parsed. */
-typedef int (*t_hts_htmlcheck_start) (t_hts_callbackarg * carg, httrackp * opt);
+typedef int (*t_hts_htmlcheck_start)(t_hts_callbackarg *carg, httrackp *opt);
 
 /* Fired at the end of a mirror. */
-typedef int (*t_hts_htmlcheck_end) (t_hts_callbackarg * carg, httrackp * opt);
+typedef int (*t_hts_htmlcheck_end)(t_hts_callbackarg *carg, httrackp *opt);
 
 /* Fired while options are being changed, to validate or adjust them. */
-typedef int (*t_hts_htmlcheck_chopt) (t_hts_callbackarg * carg, httrackp * opt);
+typedef int (*t_hts_htmlcheck_chopt)(t_hts_callbackarg *carg, httrackp *opt);
 
 /* Rewrite hook over an in-memory page: the html and len arguments point at the
    buffer and its length (the callback may reallocate and resize it),
    url_adresse and url_fichier name it. */
-typedef int (*t_hts_htmlcheck_process) (t_hts_callbackarg * carg,
-                                        httrackp * opt, char **html, int *len,
-                                        const char *url_adresse,
-                                        const char *url_fichier);
+typedef int (*t_hts_htmlcheck_process)(t_hts_callbackarg *carg, httrackp *opt,
+                                       char **html, int *len,
+                                       const char *url_adresse,
+                                       const char *url_fichier);
 
 /* Same shape as process, run before HTML parsing. */
 typedef t_hts_htmlcheck_process t_hts_htmlcheck_preprocess;
@@ -121,113 +121,111 @@ typedef t_hts_htmlcheck_process t_hts_htmlcheck_preprocess;
 typedef t_hts_htmlcheck_process t_hts_htmlcheck_postprocess;
 
 /* Inspect a page (read-only html/len) without rewriting it. */
-typedef int (*t_hts_htmlcheck_check_html) (t_hts_callbackarg * carg,
-                                           httrackp * opt, char *html, int len,
-                                           const char *url_adresse,
-                                           const char *url_fichier);
+typedef int (*t_hts_htmlcheck_check_html)(t_hts_callbackarg *carg,
+                                          httrackp *opt, char *html, int len,
+                                          const char *url_adresse,
+                                          const char *url_fichier);
 
 /* Answer an engine query identified by 'question'; returns the answer string
    (owned by the callback, must stay valid until the next call). */
-typedef const char *(*t_hts_htmlcheck_query) (t_hts_callbackarg * carg,
-                                              httrackp * opt,
-                                              const char *question);
+typedef const char *(*t_hts_htmlcheck_query)(t_hts_callbackarg *carg,
+                                             httrackp *opt,
+                                             const char *question);
 
 /* Second query channel, same contract as query. */
-typedef const char *(*t_hts_htmlcheck_query2) (t_hts_callbackarg * carg,
-                                               httrackp * opt,
-                                               const char *question);
+typedef const char *(*t_hts_htmlcheck_query2)(t_hts_callbackarg *carg,
+                                              httrackp *opt,
+                                              const char *question);
 
 /* Third query channel, same contract as query. */
-typedef const char *(*t_hts_htmlcheck_query3) (t_hts_callbackarg * carg,
-                                               httrackp * opt,
-                                               const char *question);
+typedef const char *(*t_hts_htmlcheck_query3)(t_hts_callbackarg *carg,
+                                              httrackp *opt,
+                                              const char *question);
 
 /* Per-tick progress hook: 'back' is the transfer slot array of 'back_max'
    entries, back_index the active one; lien_tot/lien_ntot and stats report
    queue size and running totals, stat_time the elapsed time. */
-typedef int (*t_hts_htmlcheck_loop) (t_hts_callbackarg * carg, httrackp * opt,
-                                     lien_back * back, int back_max,
-                                     int back_index, int lien_tot,
-                                     int lien_ntot, int stat_time,
-                                     hts_stat_struct * stats);
+typedef int (*t_hts_htmlcheck_loop)(t_hts_callbackarg *carg, httrackp *opt,
+                                    lien_back *back, int back_max,
+                                    int back_index, int lien_tot, int lien_ntot,
+                                    int stat_time, hts_stat_struct *stats);
 
 /* Veto a link (adr host, fil path) after its transfer; status is the result.
    Return 0 to drop the link. */
-typedef int (*t_hts_htmlcheck_check_link) (t_hts_callbackarg * carg,
-                                           httrackp * opt, const char *adr,
-                                           const char *fil, int status);
+typedef int (*t_hts_htmlcheck_check_link)(t_hts_callbackarg *carg,
+                                          httrackp *opt, const char *adr,
+                                          const char *fil, int status);
 
 /* Veto a link by its MIME type before download; return 0 to skip it. */
-typedef int (*t_hts_htmlcheck_check_mime) (t_hts_callbackarg * carg,
-                                           httrackp * opt, const char *adr,
-                                           const char *fil, const char *mime,
-                                           int status);
+typedef int (*t_hts_htmlcheck_check_mime)(t_hts_callbackarg *carg,
+                                          httrackp *opt, const char *adr,
+                                          const char *fil, const char *mime,
+                                          int status);
 
 /* Fired when the mirror pauses, waiting on 'lockfile' to be removed. */
-typedef void (*t_hts_htmlcheck_pause) (t_hts_callbackarg * carg, httrackp * opt,
-                                       const char *lockfile);
+typedef void (*t_hts_htmlcheck_pause)(t_hts_callbackarg *carg, httrackp *opt,
+                                      const char *lockfile);
 
 /* Fired after a file is written to disk; 'file' is the local path. */
-typedef void (*t_hts_htmlcheck_filesave) (t_hts_callbackarg * carg,
-                                          httrackp * opt, const char *file);
+typedef void (*t_hts_htmlcheck_filesave)(t_hts_callbackarg *carg, httrackp *opt,
+                                         const char *file);
 
 /* Richer file-saved notification: source host/filename, local path, and flags
    telling whether the file is new, modified, or left unchanged. */
-typedef void (*t_hts_htmlcheck_filesave2) (t_hts_callbackarg * carg,
-                                           httrackp * opt, const char *hostname,
-                                           const char *filename,
-                                           const char *localfile, int is_new,
-                                           int is_modified, int not_updated);
+typedef void (*t_hts_htmlcheck_filesave2)(t_hts_callbackarg *carg,
+                                          httrackp *opt, const char *hostname,
+                                          const char *filename,
+                                          const char *localfile, int is_new,
+                                          int is_modified, int not_updated);
 
 /* Fired for each link parsed out of a page; 'link' may be edited in place. */
-typedef int (*t_hts_htmlcheck_linkdetected) (t_hts_callbackarg * carg,
-                                             httrackp * opt, char *link);
+typedef int (*t_hts_htmlcheck_linkdetected)(t_hts_callbackarg *carg,
+                                            httrackp *opt, char *link);
 
 /* As linkdetected, plus tag_start, the markup the link was found in. */
-typedef int (*t_hts_htmlcheck_linkdetected2) (t_hts_callbackarg * carg,
-                                              httrackp * opt, char *link,
-                                              const char *tag_start);
+typedef int (*t_hts_htmlcheck_linkdetected2)(t_hts_callbackarg *carg,
+                                             httrackp *opt, char *link,
+                                             const char *tag_start);
 
 /* Fired on each transfer-status change of slot 'back'. */
-typedef int (*t_hts_htmlcheck_xfrstatus) (t_hts_callbackarg * carg,
-                                          httrackp * opt, lien_back * back);
+typedef int (*t_hts_htmlcheck_xfrstatus)(t_hts_callbackarg *carg, httrackp *opt,
+                                         lien_back *back);
 
 /* Choose the local save path for a URL; write it into 'save'. adr/fil name the
    target, referer_adr/referer_fil the page that linked it. */
-typedef int (*t_hts_htmlcheck_savename) (t_hts_callbackarg * carg,
-                                         httrackp * opt,
-                                         const char *adr_complete,
-                                         const char *fil_complete,
-                                         const char *referer_adr,
-                                         const char *referer_fil, char *save);
+typedef int (*t_hts_htmlcheck_savename)(t_hts_callbackarg *carg, httrackp *opt,
+                                        const char *adr_complete,
+                                        const char *fil_complete,
+                                        const char *referer_adr,
+                                        const char *referer_fil, char *save);
 
 /* Extended save-name hook, same signature as savename. */
 typedef t_hts_htmlcheck_savename t_hts_htmlcheck_extsavename;
 
 /* Inspect or edit the outgoing request headers in 'buff' before they are sent.
  */
-typedef int (*t_hts_htmlcheck_sendhead) (t_hts_callbackarg * carg,
-                                         httrackp * opt, char *buff,
-                                         const char *adr, const char *fil,
-                                         const char *referer_adr,
-                                         const char *referer_fil,
-                                         htsblk * outgoing);
+typedef int (*t_hts_htmlcheck_sendhead)(t_hts_callbackarg *carg, httrackp *opt,
+                                        char *buff, const char *adr,
+                                        const char *fil,
+                                        const char *referer_adr,
+                                        const char *referer_fil,
+                                        htsblk *outgoing);
 
 /* Inspect the incoming response headers in 'buff' after they are received. */
-typedef int (*t_hts_htmlcheck_receivehead) (t_hts_callbackarg * carg,
-                                            httrackp * opt, char *buff,
-                                            const char *adr, const char *fil,
-                                            const char *referer_adr,
-                                            const char *referer_fil,
-                                            htsblk * incoming);
+typedef int (*t_hts_htmlcheck_receivehead)(t_hts_callbackarg *carg,
+                                           httrackp *opt, char *buff,
+                                           const char *adr, const char *fil,
+                                           const char *referer_adr,
+                                           const char *referer_fil,
+                                           htsblk *incoming);
 
 /* External parser module hooks: detect claims a document type (return 1 to
    take it), parse then extracts its links. 'str' carries the document. */
-typedef int (*t_hts_htmlcheck_detect) (t_hts_callbackarg * carg, httrackp * opt,
-                                       htsmoduleStruct * str);
+typedef int (*t_hts_htmlcheck_detect)(t_hts_callbackarg *carg, httrackp *opt,
+                                      htsmoduleStruct *str);
 
-typedef int (*t_hts_htmlcheck_parse) (t_hts_callbackarg * carg, httrackp * opt,
-                                      htsmoduleStruct * str);
+typedef int (*t_hts_htmlcheck_parse)(t_hts_callbackarg *carg, httrackp *opt,
+                                     htsmoduleStruct *str);
 
 /* Callbacks */
 #ifndef HTS_DEF_FWSTRUCT_t_hts_htmlcheck_callbacks
@@ -237,10 +235,10 @@ typedef struct t_hts_htmlcheck_callbacks t_hts_htmlcheck_callbacks;
 
 /* Declares one named callback slot: its function pointer (typed
    t_hts_htmlcheck_<NAME>) paired with the carg passed to it. */
-#define DEFCALLBACK(NAME)         \
-  struct NAME {                   \
-    t_hts_htmlcheck_ ##NAME fun;  \
-    t_hts_callbackarg *carg;      \
+#define DEFCALLBACK(NAME)                                                      \
+  struct NAME {                                                                \
+    t_hts_htmlcheck_##NAME fun;                                                \
+    t_hts_callbackarg *carg;                                                   \
   } NAME
 
 /* Generic, type-erased callback slot used where the hook type is opaque. */
@@ -324,18 +322,18 @@ extern const t_hts_htmlcheck_callbacks default_callbacks;
 /* Internal helpers for building an HTTP request/response into the engine's
    scratch buffer (opt->state.HTbuff): START resets it, PRINT appends; the
    PANIC variant records a fatal error message. */
-#define HT_PRINT(A) strcatbuff(opt->state.HTbuff,A);
+#define HT_PRINT(A) strcatbuff(opt->state.HTbuff, A);
 
-#define HT_REQUEST_START opt->state.HTbuff[0]='\0';
+#define HT_REQUEST_START opt->state.HTbuff[0] = '\0';
 
 #define HT_REQUEST_END
-#define HTT_REQUEST_START opt->state.HTbuff[0]='\0';
+#define HTT_REQUEST_START opt->state.HTbuff[0] = '\0';
 
 #define HTT_REQUEST_END
-#define HTS_REQUEST_START opt->state.HTbuff[0]='\0';
+#define HTS_REQUEST_START opt->state.HTbuff[0] = '\0';
 
 #define HTS_REQUEST_END
-#define HTS_PANIC_PRINTF(S) strcpybuff(opt->state._hts_errmsg,S);
+#define HTS_PANIC_PRINTF(S) strcpybuff(opt->state._hts_errmsg, S);
 
 #endif
 

src/htsentities.sh

@@ -33,43 +33,43 @@ EOF
         else
             GET "${url}"
         fi
-    ) \
-        | grep -E '^<!ENTITY [a-zA-Z0-9_]' \
-        | sed \
-        -e 's/<!ENTITY //' -e "s/[[:space:]][[:space:]]*/ /g" \
-        -e 's/-->$//' \
-        -e 's/\([^ ]*\) CDATA "&#\([^\"]*\);" -- \(.*\)/\1 \2 \3/'\
-| ( \
-        read A
-        while test -n "$A"; do
-            ent="${A%% *}"
-            code=$(echo "$A"|cut -f2 -d' ')
-            # compute hash
-            hash=0
-            i=0
-            a=1664525
-            c=1013904223
-            m="$[1 << 32]"
-            while test "$i" -lt ${#ent}; do
-                d="$(echo -n "${ent:${i}:1}"|hexdump -v -e '/1 "%d"')"
-                hash="$[((${hash}*${a})%(${m})+${d}+${c})%(${m})]"
-                i=$[${i}+1]
-            done
-            echo -e "    /* $A */"
-            echo -e "  case ${hash}u:"
-            echo -e "    if (len == ${#ent} /* && strncmp(ent, \"${ent}\") == 0 */) {"
-            echo -e "      return ${code};"
-            echo -e "    }"
-            echo -e "    break;"
+    ) |
+        grep -E '^<!ENTITY [a-zA-Z0-9_]' |
+        sed \
+            -e 's/<!ENTITY //' -e "s/[[:space:]][[:space:]]*/ /g" \
+            -e 's/-->$//' \
+            -e 's/\([^ ]*\) CDATA "&#\([^\"]*\);" -- \(.*\)/\1 \2 \3/' |
+        (
+            read -r A
+            while test -n "$A"; do
+                ent="${A%% *}"
+                code=$(echo "$A" | cut -f2 -d' ')
+                # compute hash
+                hash=0
+                i=0
+                a=1664525
+                c=1013904223
+                m="$((1 << 32))"
+                while test "$i" -lt ${#ent}; do
+                    d="$(echo -n "${ent:${i}:1}" | hexdump -v -e '/1 "%d"')"
+                    hash="$((((hash * a) % (m) + d + c) % (m)))"
+                    i=$((i + 1))
+                done
+                echo -e "    /* $A */"
+                echo -e "  case ${hash}u:"
+                echo -e "    if (len == ${#ent} /* && strncmp(ent, \"${ent}\") == 0 */) {"
+                echo -e "      return ${code};"
+                echo -e "    }"
+                echo -e "    break;"
 
-            # next
-            read A
-        done
-    )
+                # next
+                read -r A
+            done
+        )
     cat <<EOF
   }
   /* unknown */
   return -1;
 }
 EOF
-) > ${dest}
+) >${dest}

src/htsglobal.h

@@ -43,10 +43,10 @@ Please visit our Website: http://www.httrack.com
    configure.ac, decoupled from these). VERSION is the display form, VERSIONID
    the dotted numeric form, AFF_VERSION the short form shown in footers,
    LIB_VERSION the data/cache format generation. */
-#define HTTRACK_VERSION      "3.49-8"
-#define HTTRACK_VERSIONID    "3.49.8"
-#define HTTRACK_AFF_VERSION  "3.x"
-#define HTTRACK_LIB_VERSION  "2.0"
+#define HTTRACK_VERSION "3.49-8"
+#define HTTRACK_VERSIONID "3.49.8"
+#define HTTRACK_AFF_VERSION "3.x"
+#define HTTRACK_LIB_VERSION "2.0"
 
 #ifndef HTS_NOINCLUDES
 #include <stdio.h>
@@ -71,11 +71,11 @@ Please visit our Website: http://www.httrack.com
    varargs starting at arg. */
 #ifndef HTS_UNUSED
 #ifdef __GNUC__
-#define HTS_UNUSED __attribute__ ((unused))
+#define HTS_UNUSED __attribute__((unused))
 
-#define HTS_STATIC static __attribute__ ((unused))
+#define HTS_STATIC static __attribute__((unused))
 
-#define HTS_PRINTF_FUN(fmt, arg) __attribute__ ((format (printf, fmt, arg)))
+#define HTS_PRINTF_FUN(fmt, arg) __attribute__((format(printf, fmt, arg)))
 #else
 #define HTS_UNUSED
 #define HTS_STATIC static
@@ -113,7 +113,7 @@ Please visit our Website: http://www.httrack.com
 
 #ifndef HTS_LONGLONG
 #ifdef SIZEOF_LONG_LONG
-#if SIZEOF_LONG_LONG==8
+#if SIZEOF_LONG_LONG == 8
 #define HTS_LONGLONG 1
 #endif
 #endif
@@ -204,12 +204,12 @@ Please visit our Website: http://www.httrack.com
 #endif
 
 #define HTS_HTTRACKRC ".httrackrc"
-#define HTS_HTTRACKCNF HTS_ETCPATH"/httrack.conf"
+#define HTS_HTTRACKCNF HTS_ETCPATH "/httrack.conf"
 
 #ifdef DATADIR
-#define HTS_HTTRACKDIR DATADIR"/httrack/"
+#define HTS_HTTRACKDIR DATADIR "/httrack/"
 #else
-#define HTS_HTTRACKDIR HTS_PREFIX"/share/httrack/"
+#define HTS_HTTRACKDIR HTS_PREFIX "/share/httrack/"
 #endif
 
 #endif
@@ -226,12 +226,17 @@ Please visit our Website: http://www.httrack.com
 
 /* Copyright (C) 1998 Xavier Roche and other contributors */
 #define HTTRACK_AFF_AUTHORS "[XR&CO'2014]"
-#define HTS_DEFAULT_FOOTER "<!-- Mirrored from %s%s by HTTrack Website Copier/" HTTRACK_AFF_VERSION " " HTTRACK_AFF_AUTHORS ", %s -->"
+#define HTS_DEFAULT_FOOTER                                                     \
+  "<!-- Mirrored from %s%s by HTTrack Website Copier/" HTTRACK_AFF_VERSION     \
+  " " HTTRACK_AFF_AUTHORS ", %s -->"
 #define HTTRACK_WEB "http://www.httrack.com"
-#define HTS_UPDATE_WEBSITE "http://www.httrack.com/update.php3?Product=HTTrack&Version=" HTTRACK_VERSIONID "&VersionStr=" HTTRACK_VERSION "&Platform=%d&Language=%s"
+#define HTS_UPDATE_WEBSITE                                                     \
+  "http://www.httrack.com/"                                                    \
+  "update.php3?Product=HTTrack&Version=" HTTRACK_VERSIONID                     \
+  "&VersionStr=" HTTRACK_VERSION "&Platform=%d&Language=%s"
 
 #define H_CRLF "\x0d\x0a"
-#define CRLF   "\x0d\x0a"
+#define CRLF "\x0d\x0a"
 #ifdef _WIN32
 #define LF "\x0d\x0a"
 #else
@@ -247,13 +252,14 @@ Please visit our Website: http://www.httrack.com
    return type stays compatible with the int it replaces. */
 #ifndef HTS_DEF_DEFSTRUCT_hts_boolean
 #define HTS_DEF_DEFSTRUCT_hts_boolean
+
 typedef enum hts_boolean { HTS_FALSE = 0, HTS_TRUE = 1 } hts_boolean;
 #endif
 
 /* Larger/smaller of two values. Macros: arguments are evaluated twice. */
-#define maximum(A,B) ( (A) > (B) ? (A) : (B) )
+#define maximum(A, B) ((A) > (B) ? (A) : (B))
 
-#define minimum(A,B) ( (A) < (B) ? (A) : (B) )
+#define minimum(A, B) ((A) < (B) ? (A) : (B))
 
 /* True when A is a non-NULL, non-empty string. */
 #define strnotempty(A) (((A) != NULL && (A)[0] != '\0'))
@@ -278,10 +284,10 @@ typedef enum hts_boolean { HTS_FALSE = 0, HTS_TRUE = 1 } hts_boolean;
 #endif
 #else
 /* See <http://gcc.gnu.org/wiki/Visibility> */
-#if ( ( defined(__GNUC__) && ( __GNUC__ >= 4 ) ) \
-      || ( defined(HAVE_VISIBILITY) && HAVE_VISIBILITY ) )
+#if ((defined(__GNUC__) && (__GNUC__ >= 4)) ||                                 \
+     (defined(HAVE_VISIBILITY) && HAVE_VISIBILITY))
 
-#define HTSEXT_API __attribute__ ((visibility ("default")))
+#define HTSEXT_API __attribute__((visibility("default")))
 #else
 #define HTSEXT_API
 #endif
@@ -335,8 +341,8 @@ typedef __int64 LLint;
 typedef __int64 TStamp;
 
 #define LLintP "%I64d"
-#elif (defined(_LP64) || defined(__x86_64__) \
-       || defined(__powerpc64__) || defined(__64BIT__))
+#elif (defined(_LP64) || defined(__x86_64__) || defined(__powerpc64__) ||      \
+       defined(__64BIT__))
 
 typedef long int LLint;
 
@@ -400,16 +406,17 @@ typedef int T_SOC;
 /* Permission bits for created folders and files (mkdir and chmod).
    PROTECT_FOLDER is owner-only. With HTS_ACCESS set (the default) the ACCESS_
    modes also grant group/other read; otherwise they stay owner-only. */
-#define HTS_PROTECT_FOLDER (S_IRUSR|S_IWUSR|S_IXUSR)
+#define HTS_PROTECT_FOLDER (S_IRUSR | S_IWUSR | S_IXUSR)
 
 #if HTS_ACCESS
-#define HTS_ACCESS_FILE (S_IRUSR|S_IWUSR|S_IRGRP|S_IROTH)
+#define HTS_ACCESS_FILE (S_IRUSR | S_IWUSR | S_IRGRP | S_IROTH)
 
-#define HTS_ACCESS_FOLDER (S_IRUSR|S_IWUSR|S_IXUSR|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH)
+#define HTS_ACCESS_FOLDER                                                      \
+  (S_IRUSR | S_IWUSR | S_IXUSR | S_IRGRP | S_IXGRP | S_IROTH | S_IXOTH)
 #else
-#define HTS_ACCESS_FILE (S_IRUSR|S_IWUSR)
+#define HTS_ACCESS_FILE (S_IRUSR | S_IWUSR)
 
-#define HTS_ACCESS_FOLDER (S_IRUSR|S_IWUSR|S_IXUSR)
+#define HTS_ACCESS_FOLDER (S_IRUSR | S_IWUSR | S_IXUSR)
 #endif
 
 /* Sanity-check that the required preprocessor switches are defined */
@@ -427,7 +434,11 @@ typedef int T_SOC;
 #endif
 
 /* fflush sur stdout */
-#define io_flush { fflush(stdout); fflush(stdin); }
+#define io_flush                                                               \
+  {                                                                            \
+    fflush(stdout);                                                            \
+    fflush(stdin);                                                             \
+  }
 
 /* HTSLib */
 
@@ -447,7 +458,7 @@ typedef int T_SOC;
 
 #ifdef _DEBUG
 // trace mallocs
-//#define HTS_TRACE_MALLOC
+// #define HTS_TRACE_MALLOC
 #ifdef HTS_TRACE_MALLOC
 typedef unsigned long int t_htsboundary;
 
@@ -524,7 +535,13 @@ static const t_htsboundary htsboundary = 0xDEADBEEF;
 #if _HTS_WIDE
 extern FILE *DEBUG_fp;
 
-#define DEBUG_W(A)  { if (DEBUG_fp==NULL) DEBUG_fp=fopen("bug.out","wb"); fprintf(DEBUG_fp,":>"A); fflush(DEBUG_fp); }
+#define DEBUG_W(A)                                                             \
+  {                                                                            \
+    if (DEBUG_fp == NULL)                                                      \
+      DEBUG_fp = fopen("bug.out", "wb");                                       \
+    fprintf(DEBUG_fp, ":>" A);                                                 \
+    fflush(DEBUG_fp);                                                          \
+  }
 #undef _
 #define _ ,
 #endif

src/htslib.c

@@ -2605,6 +2605,8 @@ int ident_url_absolute(const char *url, lien_adrfil *adrfil) {
     for(i = 0; adrfil->fil[i] != '\0'; i++)
       if (adrfil->fil[i] == '\\')
         adrfil->fil[i] = '/';
+    // collapse ../ like the http branch above (path-traversal safety)
+    fil_simplifie(adrfil->fil);
   }
 
   // no hostname

src/htsmodules.h

@@ -69,41 +69,41 @@ typedef struct hash_struct hash_struct;
 #define HTS_DEF_FWSTRUCT_htsmoduleStruct
 typedef struct htsmoduleStruct htsmoduleStruct;
 #endif
-typedef int (*t_htsAddLink) (htsmoduleStruct * str, char *link);
+typedef int (*t_htsAddLink)(htsmoduleStruct *str, char *link);
 
 /** Per-object context passed to a parser module for one downloaded file.
     Field access classes are noted; engine owns all pointers unless stated. */
 struct htsmoduleStruct {
   /* Read-only elements */
-  const char *filename;         /* filename (C:\My Web Sites\...) */
-  int size;                     /* size of filename (should be > 0) */
-  const char *mime;             /* MIME type of the object */
-  const char *url_host;         /* incoming hostname (www.foo.com) */
-  const char *url_file;         /* incoming filename (/bar/bar.gny) */
+  const char *filename; /* filename (C:\My Web Sites\...) */
+  int size;             /* size of filename (should be > 0) */
+  const char *mime;     /* MIME type of the object */
+  const char *url_host; /* incoming hostname (www.foo.com) */
+  const char *url_file; /* incoming filename (/bar/bar.gny) */
 
   /* Write-only */
-  const char *wrapper_name;     /* name of wrapper (static string) */
-  char *err_msg;                /* if an error occurred, the error message (max. 1KB) */
+  const char *wrapper_name; /* name of wrapper (static string) */
+  char *err_msg; /* if an error occurred, the error message (max. 1KB) */
 
   /* Read/Write */
-  int relativeToHtmlLink;       /* set this to 1 if all urls you pass to addLink
-                                   are in fact relative to the html file where your
-                                   module was originally */
+  int relativeToHtmlLink; /* set this to 1 if all urls you pass to addLink
+                             are in fact relative to the html file where your
+                             module was originally */
 
   /* Callbacks */
-  t_htsAddLink addLink;         /* call this function when links are 
-                                   being detected. it if not your responsability to decide
-                                   if the engine will keep them, or not. */
+  t_htsAddLink addLink; /* call this function when links are
+                           being detected. it if not your responsability to
+                           decide if the engine will keep them, or not. */
 
   /* Optional */
-  char *localLink;              /* if non null, the engine will write there the local
-                                   relative filename of the link added by addLink(), or
-                                   the absolute path if the link was refused by the wizard */
-  int localLinkSize;            /* size of the optionnal buffer */
+  char *localLink;   /* if non null, the engine will write there the local
+                        relative filename of the link added by addLink(), or
+                        the absolute path if the link was refused by the wizard */
+  int localLinkSize; /* size of the optionnal buffer */
 
   /* User-defined */
-  void *userdef;                /* can be used by callback routines
-                                 */
+  void *userdef; /* can be used by callback routines
+                  */
 
   /* The parser httrackp structure (may be used) */
   httrackp *opt;
@@ -117,7 +117,6 @@ struct htsmoduleStruct {
   int *ptr_;
   const char *page_charset_;
   /* Internal use - please don't touch */
-
 };
 
 #ifdef __cplusplus
@@ -126,11 +125,11 @@ extern "C" {
 
 /** Module lifecycle hooks. Init/PlugInit return 1 on success, 0 on failure;
     Exit returns its own status (ignored by the engine). */
-typedef int (*t_htsWrapperInit) (char *fn, char *args);
+typedef int (*t_htsWrapperInit)(char *fn, char *args);
 
-typedef int (*t_htsWrapperExit) (void);
+typedef int (*t_htsWrapperExit)(void);
 
-typedef int (*t_htsWrapperPlugInit) (char *args);
+typedef int (*t_htsWrapperPlugInit)(char *args);
 
 /* Library internal definictions */
 #ifdef HTS_INTERNAL_BYTECODE
@@ -138,7 +137,7 @@ typedef int (*t_htsWrapperPlugInit) (char *args);
 /** Capabilities string ("-noV6", "-nossl", ...) followed by "+name" for each
     loaded module. Returned pointer aliases opt->state.HTbuff; do not free, and
     it is overwritten by the next call. */
-HTSEXT_API const char *hts_get_version_info(httrackp * opt);
+HTSEXT_API const char *hts_get_version_info(httrackp *opt);
 
 /** Static capabilities string set by htspe_init(); valid for the process
     lifetime, do not free. */
@@ -154,7 +153,7 @@ extern void htspe_uninit(void);
 /** Run the external-parser callbacks for the object described by str.
     Returns the parse callback result (>=0) on a handled object, or -1 if no
     module claimed it or its wrapper_name is blacklisted. */
-extern int hts_parse_externals(htsmoduleStruct * str);
+extern int hts_parse_externals(htsmoduleStruct *str);
 
 /** Nonzero if IPv6 support was compiled in (== HTS_INET6). */
 extern int V6_is_available;

src/htsnet.h

@@ -112,10 +112,10 @@ struct SOCaddr {
 
 /** Pointer to the port field (network byte order) for the active family.
     Asserts on NULL or an unset/unknown family. */
-static HTS_INLINE HTS_UNUSED in_port_t* SOCaddr_sinport_(SOCaddr *const addr,
-                                                         const char *file, const int line) {
+static HTS_INLINE HTS_UNUSED in_port_t *
+SOCaddr_sinport_(SOCaddr *const addr, const char *file, const int line) {
   assertf_(addr != NULL, file, line);
-  switch(addr->m_addr.sa.sa_family) {
+  switch (addr->m_addr.sa.sa_family) {
   case AF_INET:
     return &addr->m_addr.in.sin_port;
     break;
@@ -125,7 +125,7 @@ static HTS_INLINE HTS_UNUSED in_port_t* SOCaddr_sinport_(SOCaddr *const addr,
     break;
 #endif
   default:
-    assertf_(! "invalid structure", file, line);
+    assertf_(!"invalid structure", file, line);
     return 0;
     break;
   }
@@ -133,10 +133,11 @@ static HTS_INLINE HTS_UNUSED in_port_t* SOCaddr_sinport_(SOCaddr *const addr,
 
 /** Length of the active sockaddr (sockaddr_in or sockaddr_in6), or 0 if the
     family is unset/unknown. The 0 case doubles as the "not valid" test. */
-static HTS_INLINE HTS_UNUSED socklen_t SOCaddr_size_(const SOCaddr*const addr,
-                                                     const char *file, const int line) {
+static HTS_INLINE HTS_UNUSED socklen_t SOCaddr_size_(const SOCaddr *const addr,
+                                                     const char *file,
+                                                     const int line) {
   assertf_(addr != NULL, file, line);
-  switch(addr->m_addr.sa.sa_family) {
+  switch (addr->m_addr.sa.sa_family) {
   case AF_INET:
     return sizeof(addr->m_addr.in);
     break;
@@ -152,8 +153,8 @@ static HTS_INLINE HTS_UNUSED socklen_t SOCaddr_size_(const SOCaddr*const addr,
 }
 
 /** Reset to the unset state (family AF_UNSPEC), making the address invalid. */
-static HTS_INLINE HTS_UNUSED void SOCaddr_clear_(SOCaddr*const addr,
-                                                 const char *file, const int line) {
+static HTS_INLINE HTS_UNUSED void
+SOCaddr_clear_(SOCaddr *const addr, const char *file, const int line) {
   assertf_(addr != NULL, file, line);
   addr->m_addr.sa.sa_family = AF_UNSPEC;
 }
@@ -191,14 +192,16 @@ static HTS_INLINE HTS_UNUSED void SOCaddr_clear_(SOCaddr*const addr,
 
 /** Set the port (host-order argument, stored network-order) on the active
  * family. */
-#define SOCaddr_initport(server, port) do { \
-  SOCaddr_sinport(server) = htons((in_port_t) (port)); \
-} while(0)
+#define SOCaddr_initport(server, port)                                         \
+  do {                                                                         \
+    SOCaddr_sinport(server) = htons((in_port_t) (port));                       \
+  } while (0)
 
 /** Initialize as an all-zero IPv4 wildcard (INADDR_ANY) address; returns its
     sockaddr length. */
-static HTS_INLINE HTS_UNUSED socklen_t SOCaddr_initany_(SOCaddr*const addr,
-                                                        const char *file, const int line) {
+static HTS_INLINE HTS_UNUSED socklen_t SOCaddr_initany_(SOCaddr *const addr,
+                                                        const char *file,
+                                                        const int line) {
   assertf_(addr != NULL, file, line);
   memset(&addr->m_addr.in, 0, sizeof(addr->m_addr.in));
   addr->m_addr.in.sin_family = AF_INET;
@@ -206,17 +209,20 @@ static HTS_INLINE HTS_UNUSED socklen_t SOCaddr_initany_(SOCaddr*const addr,
 }
 
 /** Initialize server as an IPv4 wildcard (INADDR_ANY) address. */
-#define SOCaddr_initany(server) do { \
-  SOCaddr_initany_(&(server), __FILE__, __LINE__); \
-} while(0)
+#define SOCaddr_initany(server)                                                \
+  do {                                                                         \
+    SOCaddr_initany_(&(server), __FILE__, __LINE__);                           \
+  } while (0)
 
 /** Populate server from data. data_size selects the source form: a full
     sockaddr_in / sockaddr_in6, or a raw 4-byte (IPv4) / 16-byte (IPv6) address
     with port zeroed. Any other size leaves an AF_INET shell. Returns the
     resulting sockaddr length. */
-static HTS_UNUSED socklen_t SOCaddr_copyaddr_(SOCaddr*const server, 
-                                              const void *data, const size_t data_size,
-                                              const char *file, const int line) {
+static HTS_UNUSED socklen_t SOCaddr_copyaddr_(SOCaddr *const server,
+                                              const void *data,
+                                              const size_t data_size,
+                                              const char *file,
+                                              const int line) {
   assertf_(server != NULL, file, line);
   assertf_(data != NULL, file, line);
 
@@ -248,32 +254,35 @@ static HTS_UNUSED socklen_t SOCaddr_copyaddr_(SOCaddr*const server,
 
 /** Copy hpaddr (length hpsize) into server, writing the result length into the
     lvalue server_len (int). See SOCaddr_copyaddr_ for accepted forms. */
-#define SOCaddr_copyaddr(server, server_len, hpaddr, hpsize) do { \
-  server_len = (int) SOCaddr_copyaddr_(&(server), hpaddr, hpsize, __FILE__, __LINE__); \
-} while(0)
+#define SOCaddr_copyaddr(server, server_len, hpaddr, hpsize)                   \
+  do {                                                                         \
+    server_len = (int) SOCaddr_copyaddr_(&(server), hpaddr, hpsize, __FILE__,  \
+                                         __LINE__);                            \
+  } while (0)
 
 /** Like SOCaddr_copyaddr but discards the result length. */
-#define SOCaddr_copyaddr2(server, hpaddr, hpsize) do { \
-  (void) SOCaddr_copyaddr_(&(server), hpaddr, hpsize, __FILE__, __LINE__); \
-} while(0)
+#define SOCaddr_copyaddr2(server, hpaddr, hpsize)                              \
+  do {                                                                         \
+    (void) SOCaddr_copyaddr_(&(server), hpaddr, hpsize, __FILE__, __LINE__);   \
+  } while (0)
 
 /** Copy one SOCaddr (src) into another (dest), preserving family and port. */
-#define SOCaddr_copy_SOCaddr(dest, src) do { \
-  SOCaddr_copyaddr_(&(dest), &(src).m_addr.sa, SOCaddr_size(src), __FILE__, __LINE__); \
-} while(0)
+#define SOCaddr_copy_SOCaddr(dest, src)                                        \
+  do {                                                                         \
+    SOCaddr_copyaddr_(&(dest), &(src).m_addr.sa, SOCaddr_size(src), __FILE__,  \
+                      __LINE__);                                               \
+  } while (0)
 
 /** Write the numeric (dotted/colon) host of ss into namebuf (capacity
     namebuflen), scope id stripped. On failure namebuf becomes "". */
-static HTS_UNUSED void SOCaddr_inetntoa_(char *namebuf, size_t namebuflen, 
-                                         SOCaddr *const ss,
-                                         const char *file, const int line) {
+static HTS_UNUSED void SOCaddr_inetntoa_(char *namebuf, size_t namebuflen,
+                                         SOCaddr *const ss, const char *file,
+                                         const int line) {
   assertf_(namebuf != NULL, file, line);
   assertf_(ss != NULL, file, line);
 
-  if (getnameinfo(&ss->m_addr.sa, sizeof(ss->m_addr),
-                  namebuf, namebuflen, 
-                  NULL, 0, 
-                  NI_NUMERICHOST) == 0) {
+  if (getnameinfo(&ss->m_addr.sa, sizeof(ss->m_addr), namebuf, namebuflen, NULL,
+                  0, NI_NUMERICHOST) == 0) {
     /* remove scope id(s) */
     char *const pos = strchr(namebuf, '%');
     if (pos != NULL) {
@@ -285,11 +294,12 @@ static HTS_UNUSED void SOCaddr_inetntoa_(char *namebuf, size_t namebuflen,
 }
 
 /** Numeric host of ss into namebuf (capacity namebuflen); "" on failure. */
-#define SOCaddr_inetntoa(namebuf, namebuflen, ss) \
+#define SOCaddr_inetntoa(namebuf, namebuflen, ss)                              \
   SOCaddr_inetntoa_(namebuf, namebuflen, &(ss), __FILE__, __LINE__)
 
 /** Single-char family tag: '1' for IPv4, '2' otherwise (used in the cache). */
-#define SOCaddr_getproto(ss) ( SOCaddr_size(ss) == sizeof(struct sockaddr_in) ? '1' : '2')
+#define SOCaddr_getproto(ss)                                                   \
+  (SOCaddr_size(ss) == sizeof(struct sockaddr_in) ? '1' : '2')
 
 /** Length type for socket APIs (getsockname, accept, ...). */
 typedef socklen_t SOClen;

src/htsopt.h

@@ -72,6 +72,7 @@ typedef struct String String;
 #endif
 #ifndef HTS_DEF_STRUCT_String
 #define HTS_DEF_STRUCT_String
+
 struct String {
   char *buffer_;
   size_t length_;
@@ -80,7 +81,7 @@ struct String {
 #endif
 
 /* Defines */
-#define CATBUFF_SIZE (STRING_SIZE*2*2)
+#define CATBUFF_SIZE (STRING_SIZE * 2 * 2)
 
 #define STRING_SIZE 2048
 
@@ -108,7 +109,7 @@ struct htsfilters {
 };
 
 /* User callbacks chain */
-typedef int (*htscallbacksfncptr) (void);
+typedef int (*htscallbacksfncptr)(void);
 
 typedef struct htscallbacks htscallbacks;
 
@@ -179,6 +180,7 @@ typedef struct lien_url lien_url;
 
 #ifndef HTS_DEF_DEFSTRUCT_hts_log_type
 #define HTS_DEF_DEFSTRUCT_hts_log_type
+
 typedef enum hts_log_type {
   LOG_PANIC,
   LOG_ERROR,
@@ -278,16 +280,17 @@ struct htslibhandles {
 
 /* Javascript parser flags */
 typedef enum htsparsejava_flags {
-  HTSPARSE_NONE = 0,            // don't parse
-  HTSPARSE_DEFAULT = 1,         // parse default (all)
-  HTSPARSE_NO_CLASS = 2,        // don't parse .java
-  HTSPARSE_NO_JAVASCRIPT = 4,   // don't parse .js
-  HTSPARSE_NO_AGGRESSIVE = 8    // don't aggressively parse .js or .java
+  HTSPARSE_NONE = 0,          // don't parse
+  HTSPARSE_DEFAULT = 1,       // parse default (all)
+  HTSPARSE_NO_CLASS = 2,      // don't parse .java
+  HTSPARSE_NO_JAVASCRIPT = 4, // don't parse .js
+  HTSPARSE_NO_AGGRESSIVE = 8  // don't aggressively parse .js or .java
 } htsparsejava_flags;
 
 /* Link-rewriting style for saved pages (opt->urlmode). */
 #ifndef HTS_DEF_DEFSTRUCT_hts_urlmode
 #define HTS_DEF_DEFSTRUCT_hts_urlmode
+
 typedef enum hts_urlmode {
   HTS_URLMODE_ABSOLUTE = 0, /**< absolute URL (http://host/path) everywhere */
   HTS_URLMODE_ABSOLUTE_FILE = 1, /**< legacy file: form, unused */
@@ -301,6 +304,7 @@ typedef enum hts_urlmode {
 /* Cache policy for updates and retries (opt->cache). */
 #ifndef HTS_DEF_DEFSTRUCT_hts_cachemode
 #define HTS_DEF_DEFSTRUCT_hts_cachemode
+
 typedef enum hts_cachemode {
   HTS_CACHE_NONE = 0,       /**< no cache */
   HTS_CACHE_PRIORITY = 1,   /**< cache takes priority over the network */
@@ -311,6 +315,7 @@ typedef enum hts_cachemode {
 /* Interactive wizard level (opt->wizard). */
 #ifndef HTS_DEF_DEFSTRUCT_hts_wizard
 #define HTS_DEF_DEFSTRUCT_hts_wizard
+
 typedef enum hts_wizard {
   HTS_WIZARD_NONE = 0, /**< no wizard */
   HTS_WIZARD_ASK = 1,  /**< wizard asks questions */
@@ -321,6 +326,7 @@ typedef enum hts_wizard {
 /* robots.txt / meta-robots obedience level (opt->robots). */
 #ifndef HTS_DEF_DEFSTRUCT_hts_robots
 #define HTS_DEF_DEFSTRUCT_hts_robots
+
 typedef enum hts_robots {
   HTS_ROBOTS_NEVER = 0,        /**< ignore robots rules */
   HTS_ROBOTS_SOMETIMES = 1,    /**< partial obedience (default) */
@@ -406,34 +412,34 @@ struct httrackp {
   /* */
   hts_wizard wizard; /**< interactive wizard level (none/ask/auto) */
   hts_boolean flush; /**< fflush() log files after each write */
-  int travel;   /**< link-following scope (same domain, etc.) */
-  int seeker;   /**< allowed direction: go up and/or down the tree */
-  int depth;    /**< maximum recursion depth (-rN) */
-  int extdepth; /**< maximum recursion depth outside the start domain */
+  int travel;        /**< link-following scope (same domain, etc.) */
+  int seeker;        /**< allowed direction: go up and/or down the tree */
+  int depth;         /**< maximum recursion depth (-rN) */
+  int extdepth;      /**< maximum recursion depth outside the start domain */
   hts_urlmode
       urlmode; /**< saved-link rewriting style (relative, absolute, etc.) */
-  hts_boolean no_type_change;   // do not change file type according to MIME
-  hts_log_type debug;           /**< debug logging level */
-  int getmode;           /**< what to fetch (HTML, images, ...) bitmask */
-  FILE *log;             /**< informational log stream; NULL mutes it */
-  FILE *errlog;          /**< error log stream; NULL mutes it */
-  LLint maxsite;         /**< max total bytes for the whole mirror */
-  LLint maxfile_nonhtml; /**< max bytes per non-HTML file */
-  LLint maxfile_html;    /**< max bytes per HTML file */
-  int maxsoc;            /**< max simultaneous sockets (-cN) */
-  LLint fragment;        /**< split site after this many bytes */
+  hts_boolean no_type_change; // do not change file type according to MIME
+  hts_log_type debug;         /**< debug logging level */
+  int getmode;                /**< what to fetch (HTML, images, ...) bitmask */
+  FILE *log;                  /**< informational log stream; NULL mutes it */
+  FILE *errlog;               /**< error log stream; NULL mutes it */
+  LLint maxsite;              /**< max total bytes for the whole mirror */
+  LLint maxfile_nonhtml;      /**< max bytes per non-HTML file */
+  LLint maxfile_html;         /**< max bytes per HTML file */
+  int maxsoc;                 /**< max simultaneous sockets (-cN) */
+  LLint fragment;             /**< split site after this many bytes */
   hts_boolean
       nearlink; /**< also fetch images/data adjacent to a page but off-site */
   hts_boolean makeindex;  /**< build a top-level index.html */
   hts_boolean kindex;     /**< build a keyword index */
   hts_boolean delete_old; /**< delete locally obsolete files after update */
-  int timeout;    /**< connection timeout in seconds */
-  int rateout;    /**< minimum transfer rate (bytes/s) before abort */
-  int maxtime;    /**< max total mirror duration in seconds */
-  int maxrate;    /**< max transfer rate cap (bytes/s) */
-  float maxconn;  /**< max connections per second */
-  int waittime;   /**< scheduled start time (wall-clock seconds) */
-  hts_cachemode cache; /**< cache generation mode */
+  int timeout;            /**< connection timeout in seconds */
+  int rateout;            /**< minimum transfer rate (bytes/s) before abort */
+  int maxtime;            /**< max total mirror duration in seconds */
+  int maxrate;            /**< max transfer rate cap (bytes/s) */
+  float maxconn;          /**< max connections per second */
+  int waittime;           /**< scheduled start time (wall-clock seconds) */
+  hts_cachemode cache;    /**< cache generation mode */
   // int aff_progress;     // progress bar
   hts_boolean shell; /**< driven by a shell over stdin/stdout pipes */
   t_proxy proxy;     /**< proxy configuration */
@@ -446,12 +452,12 @@ struct httrackp {
   hts_boolean
       delayed_cached;   // delayed type check can be cached to speedup updates
   hts_boolean mimehtml; /**< produce a single MIME/MHTML archive */
-  hts_boolean user_agent_send;  /**< send a User-Agent header */
-  String user_agent;            /**< User-Agent value (e.g. httrack/1.0) */
-  String referer;               /**< Referer value to send */
-  String from;                  /**< From value to send */
-  String path_log;              /**< directory for cache and logs */
-  String path_html;             /**< output directory for the mirror */
+  hts_boolean user_agent_send; /**< send a User-Agent header */
+  String user_agent;           /**< User-Agent value (e.g. httrack/1.0) */
+  String referer;              /**< Referer value to send */
+  String from;                 /**< From value to send */
+  String path_log;             /**< directory for cache and logs */
+  String path_html;            /**< output directory for the mirror */
   String path_html_utf8; /**< output directory for the mirror, UTF-8 form */
   String path_bin;       /**< directory for HTML templates */
   int retry;             /**< extra retries on a failed transfer */
@@ -459,49 +465,49 @@ struct httrackp {
   hts_boolean maketrack; /**< maintain an operations-statistics log */
   int parsejava;         /**< Java/JS parsing mode; see htsparsejava_flags */
   int hostcontrol; /**< ban slow/timing-out hosts; see hts_hostcontrol bits */
-  hts_boolean errpage;   /**< generate an error page on 404 and similar */
+  hts_boolean errpage; /**< generate an error page on 404 and similar */
   hts_boolean
       check_type; /**< probe unknown-type links (cgi/asp/dir) and follow moves
                    */
-  hts_boolean all_in_cache; /**< keep all retrieved data in the cache */
-  hts_robots robots; /**< robots.txt handling level */
+  hts_boolean all_in_cache;      /**< keep all retrieved data in the cache */
+  hts_robots robots;             /**< robots.txt handling level */
   hts_boolean external;          /**< render external links as error pages */
   hts_boolean passprivacy;       /**< strip passwords from external links */
   hts_boolean includequery;      /**< include the query string in saved names */
   hts_boolean mirror_first_page; /**< only mirror the links of the first page */
-  String sys_com;               /**< system command to run */
-  hts_boolean sys_com_exec;     /**< actually execute sys_com */
-  hts_boolean accept_cookie;    /**< accept and send cookies */
-  t_cookie *cookie;             /**< cookie store */
-  hts_boolean http10;           /**< force HTTP/1.0 */
-  hts_boolean nokeepalive;      /**< disable keep-alive */
-  hts_boolean nocompression;    /**< disable content compression */
-  hts_boolean sizehack;         /**< treat same-size response as "updated" */
-  hts_boolean urlhack;          // force "url normalization" to avoid loops
-  hts_boolean tolerant;         /**< accept an incorrect Content-Length */
+  String sys_com;                /**< system command to run */
+  hts_boolean sys_com_exec;      /**< actually execute sys_com */
+  hts_boolean accept_cookie;     /**< accept and send cookies */
+  t_cookie *cookie;              /**< cookie store */
+  hts_boolean http10;            /**< force HTTP/1.0 */
+  hts_boolean nokeepalive;       /**< disable keep-alive */
+  hts_boolean nocompression;     /**< disable content compression */
+  hts_boolean sizehack;          /**< treat same-size response as "updated" */
+  hts_boolean urlhack;           // force "url normalization" to avoid loops
+  hts_boolean tolerant;          /**< accept an incorrect Content-Length */
   hts_boolean
       parseall; /**< parse aggressively, including unknown tags with links */
   hts_boolean parsedebug; /**< parser debug mode */
   hts_boolean norecatch;  /**< do not re-fetch files the user deleted locally */
   hts_verbosedisplay verbosedisplay; /**< animated text progress display */
-  String footer;      /**< footer/info line injected into pages */
-  int maxcache;       /**< in-memory cache backing limit (bytes) */
+  String footer; /**< footer/info line injected into pages */
+  int maxcache;  /**< in-memory cache backing limit (bytes) */
   // int maxcache_anticipate; // maximum links to anticipate (upper bound)
-  hts_boolean ftp_proxy;        /**< use the HTTP proxy for FTP too */
-  String filelist;              /**< file listing URLs to include */
-  String urllist;               /**< file listing filters to include */
-  htsfilters filters;           /**< filter pointers (+/-pattern rules) */
-  hash_struct *hash;            // hash structure
-  lien_url **liens;             // links
-  int lien_tot;                 // top index of "links" heap (always out-of-range)
-  lien_buffers *liensbuf;       // links buffers
-  robots_wizard *robotsptr;     // robots ptr
-  String lang_iso;              /**< Accept-Language value (en, fr, ...) */
-  String accept;                // Accept:
-  String headers;               // Additional headers
-  String mimedefs;              // ext1=mimetype1\next2=mimetype2..
-  String mod_blacklist;         /**< blacklisted modules */
-  hts_boolean convert_utf8;     // filenames UTF-8 conversion (3.46)
+  hts_boolean ftp_proxy;    /**< use the HTTP proxy for FTP too */
+  String filelist;          /**< file listing URLs to include */
+  String urllist;           /**< file listing filters to include */
+  htsfilters filters;       /**< filter pointers (+/-pattern rules) */
+  hash_struct *hash;        // hash structure
+  lien_url **liens;         // links
+  int lien_tot;             // top index of "links" heap (always out-of-range)
+  lien_buffers *liensbuf;   // links buffers
+  robots_wizard *robotsptr; // robots ptr
+  String lang_iso;          /**< Accept-Language value (en, fr, ...) */
+  String accept;            // Accept:
+  String headers;           // Additional headers
+  String mimedefs;          // ext1=mimetype1\next2=mimetype2..
+  String mod_blacklist;     /**< blacklisted modules */
+  hts_boolean convert_utf8; // filenames UTF-8 conversion (3.46)
   //
   int maxlink;   /**< max number of links */
   int maxfilter; /**< max number of filters */
@@ -587,17 +593,17 @@ typedef struct htsrequest htsrequest;
 struct htsrequest {
   short int user_agent_send; /**< send a User-Agent header */
   short int http11; /**< sign the request as HTTP/1.1 rather than HTTP/1.0 */
-  short int nokeepalive;        /**< disable keep-alive */
-  short int range_used;         /**< a Range header is in use */
-  short int nocompression;      /**< disable compression */
-  short int flush_garbage;      // recycled
-  const char *user_agent;       /**< User-Agent value */
-  const char *referer;          /**< Referer value */
-  const char *from;             /**< From value */
-  const char *lang_iso;         /**< Accept-Language value */
-  const char *accept;           /**< Accept value */
-  const char *headers;          /**< extra request headers */
-  htsrequest_proxy proxy;       /**< proxy for this request */
+  short int nokeepalive;   /**< disable keep-alive */
+  short int range_used;    /**< a Range header is in use */
+  short int nocompression; /**< disable compression */
+  short int flush_garbage; // recycled
+  const char *user_agent;  /**< User-Agent value */
+  const char *referer;     /**< Referer value */
+  const char *from;        /**< From value */
+  const char *lang_iso;    /**< Accept-Language value */
+  const char *accept;      /**< Accept value */
+  const char *headers;     /**< extra request headers */
+  htsrequest_proxy proxy;  /**< proxy for this request */
 };
 
 /* Result of a connection / header fetch. */
@@ -629,8 +635,8 @@ struct htsblk {
   short int is_file; /**< 1 if a file descriptor rather than a socket */
   T_SOC soc;         /**< socket id */
   SOCaddr address;   /**< peer IP address */
-  int address_size;             // IP address structure length (unused internally)
-  FILE *fp;                     /**< file handle for file:// */
+  int address_size;  // IP address structure length (unused internally)
+  FILE *fp;          /**< file handle for file:// */
 #if HTS_USEOPENSSL
   short int ssl; /**< nonzero if this is an SSL connection (https) */
   // BIO* ssl_soc;          // SSL structure
@@ -712,7 +718,7 @@ struct lien_back {
   LLint chunk_blocksize; /**< data size declared by the chunk */
   LLint compressed_size; /**< compressed size (stats only) */
   //
-  //int links_index;        // to access liens[links_index]
+  // int links_index;        // to access liens[links_index]
   //
   char info[256]; /**< status text, e.g. for FTP */
   int stop_ftp;   /**< stop flag for FTP */

src/htsparse.c

@@ -317,6 +317,27 @@ static int is_http_method(const char *s, size_t len) {
   return 0;
 }
 
+/* Percent-encode '(' and ')' in a link emitted into an unquoted url(...) (CSS
+   or JS): a literal ')' closes the token early and the UA mis-parses the value
+   (#163). The UA decodes %28/%29 back to the saved-on-disk name. */
+static void escape_url_parens(char *const s, const size_t size) {
+  char BIGSTK buff[HTS_URLMAXSIZE * 2];
+  size_t i, j;
+
+  for (i = 0, j = 0; s[i] != '\0' && j + 3 < size && j + 3 < sizeof(buff);
+       i++) {
+    if (s[i] == '(' || s[i] == ')') {
+      buff[j++] = '%';
+      buff[j++] = '2';
+      buff[j++] = s[i] == '(' ? '8' : '9';
+    } else {
+      buff[j++] = s[i];
+    }
+  }
+  buff[j] = '\0';
+  strlcpybuff(s, buff, size);
+}
+
 /* Main parser */
 int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
   char catbuff[CATBUFF_SIZE];
@@ -3027,6 +3048,10 @@ int htsparse(htsmoduleStruct * str, htsmoduleStructExtended * stre) {
                           /* Never escape high-chars (we don't know the encoding!!) */
                           inplace_escape_uri_utf(tempo, sizeof(tempo));
 
+                          // unquoted url() (CSS/JS): keep parens escaped
+                          if (ending_p == ')')
+                            escape_url_parens(tempo, sizeof(tempo));
+
                           //if (!no_esc_utf)
                           //  escape_uri(tempo);     // escape with %xx
                           //else {

src/htssafe.h

@@ -48,7 +48,7 @@ Please visit our Website: http://www.httrack.com
 /** Assert error callback. **/
 #ifndef HTS_DEF_FWSTRUCT_htsErrorCallback
 #define HTS_DEF_FWSTRUCT_htsErrorCallback
-typedef void (*htsErrorCallback) (const char *msg, const char *file, int line);
+typedef void (*htsErrorCallback)(const char *msg, const char *file, int line);
 #ifdef __cplusplus
 extern "C" {
 #endif
@@ -58,12 +58,13 @@ HTSEXT_API htsErrorCallback hts_get_error_callback(void);
 #endif
 #endif
 
-#define HTSSAFE_ABORT_FUNCTION(A,B,C) do { \
-  htsErrorCallback callback = hts_get_error_callback(); \
-  if (callback != NULL) { \
-    callback(A,B,C); \
-  } \
-} while(0)
+#define HTSSAFE_ABORT_FUNCTION(A, B, C)                                        \
+  do {                                                                         \
+    htsErrorCallback callback = hts_get_error_callback();                      \
+    if (callback != NULL) {                                                    \
+      callback(A, B, C);                                                       \
+    }                                                                          \
+  } while (0)
 
 #endif
 
@@ -75,7 +76,8 @@ HTSEXT_API htsErrorCallback hts_get_error_callback(void);
 /**
  * Fatal assertion check.
  */
-#define assertf__(exp, sexp, file, line) (void) ( (exp) || (abortf_(sexp, file, line), 0) )
+#define assertf__(exp, sexp, file, line)                                       \
+  (void) ((exp) || (abortf_(sexp, file, line), 0))
 
 /**
  * Fatal assertion check.
@@ -106,12 +108,13 @@ static HTS_UNUSED void abortf_(const char *exp, const char *file, int line) {
 #if (defined(__GNUC__) && !defined(__cplusplus))
 
 /* Note: char[] and const char[] are compatible */
-#define HTS_IS_CHAR_BUFFER(VAR) ( __builtin_types_compatible_p ( typeof (VAR), char[] ) )
+#define HTS_IS_CHAR_BUFFER(VAR)                                                \
+  (__builtin_types_compatible_p(typeof(VAR), char[]))
 #else
 /* Note: a bit lame as char[8] won't be seen. */
-#define HTS_IS_CHAR_BUFFER(VAR) ( sizeof(VAR) != sizeof(char*) )
+#define HTS_IS_CHAR_BUFFER(VAR) (sizeof(VAR) != sizeof(char *))
 #endif
-#define HTS_IS_NOT_CHAR_BUFFER(VAR) ( ! HTS_IS_CHAR_BUFFER(VAR) )
+#define HTS_IS_NOT_CHAR_BUFFER(VAR) (!HTS_IS_CHAR_BUFFER(VAR))
 
 /* Compile-time checks. */
 static HTS_UNUSED void htssafe_compile_time_check_(void) {
@@ -201,60 +204,74 @@ static char *strncatbuff_ptr_(char *dest, const char *src, size_t n) {
  */
 #if (defined(__GNUC__) && !defined(__cplusplus))
 
-#define strncatbuff(A, B, N) __builtin_choose_expr( HTS_IS_CHAR_BUFFER(A), \
-  strncat_safe_(A, sizeof(A), B, \
-  HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), N, \
-  "overflow while appending '" #B "' to '"#A"'", __FILE__, __LINE__), \
-  strncatbuff_ptr_((A), (B), (N)) )
+#define strncatbuff(A, B, N)                                                   \
+  __builtin_choose_expr(                                                       \
+      HTS_IS_CHAR_BUFFER(A),                                                   \
+      strncat_safe_(A, sizeof(A), B,                                           \
+                    HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), N,    \
+                    "overflow while appending '" #B "' to '" #A "'", __FILE__, \
+                    __LINE__),                                                 \
+      strncatbuff_ptr_((A), (B), (N)))
 #else
-#define strncatbuff(A, B, N) \
-  ( HTS_IS_NOT_CHAR_BUFFER(A) \
-  ? strncat(A, B, N) \
-  : strncat_safe_(A, sizeof(A), B, \
-  HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), N, \
-  "overflow while appending '" #B "' to '"#A"'", __FILE__, __LINE__) )
+#define strncatbuff(A, B, N)                                                   \
+  (HTS_IS_NOT_CHAR_BUFFER(A)                                                   \
+       ? strncat(A, B, N)                                                      \
+       : strncat_safe_(A, sizeof(A), B,                                        \
+                       HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), N, \
+                       "overflow while appending '" #B "' to '" #A "'",        \
+                       __FILE__, __LINE__))
 #endif
 
 /**
  * Append characters of "B" to "A".
- * If "A" is a char[] variable whose size is not sizeof(char*), then the size 
+ * If "A" is a char[] variable whose size is not sizeof(char*), then the size
  * is assumed to be the capacity of this array.
  */
 #if (defined(__GNUC__) && !defined(__cplusplus))
 
-#define strcatbuff(A, B) __builtin_choose_expr( HTS_IS_CHAR_BUFFER(A), \
-  strncat_safe_(A, sizeof(A), B, \
-  HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), (size_t) -1, \
-  "overflow while appending '" #B "' to '"#A"'", __FILE__, __LINE__), \
-  strcatbuff_ptr_((A), (B)) )
+#define strcatbuff(A, B)                                                       \
+  __builtin_choose_expr(                                                       \
+      HTS_IS_CHAR_BUFFER(A),                                                   \
+      strncat_safe_(A, sizeof(A), B,                                           \
+                    HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B),       \
+                    (size_t) -1,                                               \
+                    "overflow while appending '" #B "' to '" #A "'", __FILE__, \
+                    __LINE__),                                                 \
+      strcatbuff_ptr_((A), (B)))
 #else
-#define strcatbuff(A, B) \
-  ( HTS_IS_NOT_CHAR_BUFFER(A) \
-  ? strcat(A, B) \
-  : strncat_safe_(A, sizeof(A), B, \
-  HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), (size_t) -1, \
-  "overflow while appending '" #B "' to '"#A"'", __FILE__, __LINE__) )
+#define strcatbuff(A, B)                                                       \
+  (HTS_IS_NOT_CHAR_BUFFER(A)                                                   \
+       ? strcat(A, B)                                                          \
+       : strncat_safe_(A, sizeof(A), B,                                        \
+                       HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B),    \
+                       (size_t) -1,                                            \
+                       "overflow while appending '" #B "' to '" #A "'",        \
+                       __FILE__, __LINE__))
 #endif
 
 /**
  * Copy characters from "B" to "A".
- * If "A" is a char[] variable whose size is not sizeof(char*), then the size 
+ * If "A" is a char[] variable whose size is not sizeof(char*), then the size
  * is assumed to be the capacity of this array.
  */
 #if (defined(__GNUC__) && !defined(__cplusplus))
 
-#define strcpybuff(A, B) __builtin_choose_expr( HTS_IS_CHAR_BUFFER(A), \
-  strcpy_safe_(A, sizeof(A), B, \
-  HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), \
-  "overflow while copying '" #B "' to '"#A"'", __FILE__, __LINE__), \
-  strcpybuff_ptr_((A), (B)) )
+#define strcpybuff(A, B)                                                       \
+  __builtin_choose_expr(                                                       \
+      HTS_IS_CHAR_BUFFER(A),                                                   \
+      strcpy_safe_(A, sizeof(A), B,                                            \
+                   HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B),        \
+                   "overflow while copying '" #B "' to '" #A "'", __FILE__,    \
+                   __LINE__),                                                  \
+      strcpybuff_ptr_((A), (B)))
 #else
-#define strcpybuff(A, B) \
-  ( HTS_IS_NOT_CHAR_BUFFER(A) \
-  ? strcpy(A, B) \
-  : strcpy_safe_(A, sizeof(A), B, \
-  HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), \
-  "overflow while copying '" #B "' to '"#A"'", __FILE__, __LINE__) )
+#define strcpybuff(A, B)                                                       \
+  (HTS_IS_NOT_CHAR_BUFFER(A)                                                   \
+       ? strcpy(A, B)                                                          \
+       : strcpy_safe_(A, sizeof(A), B,                                         \
+                      HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B),     \
+                      "overflow while copying '" #B "' to '" #A "'", __FILE__, \
+                      __LINE__))
 #endif
 
 /*
@@ -268,10 +285,10 @@ static char *strncatbuff_ptr_(char *dest, const char *src, size_t n) {
 /**
  * Append characters of "B" to "A", "A" having a maximum capacity of "S".
  */
-#define strlcatbuff(A, B, S) \
-  strncat_safe_(A, S, B, \
-  HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), (size_t) -1, \
-  "overflow while appending '" #B "' to '"#A"'", __FILE__, __LINE__)
+#define strlcatbuff(A, B, S)                                                   \
+  strncat_safe_(A, S, B, HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B),  \
+                (size_t) -1, "overflow while appending '" #B "' to '" #A "'",  \
+                __FILE__, __LINE__)
 
 /**
  * Append at most "N" characters of "B" to "A", "A" having a maximum capacity
@@ -285,17 +302,18 @@ static char *strncatbuff_ptr_(char *dest, const char *src, size_t n) {
 /**
  * Copy characters of "B" to "A", "A" having a maximum capacity of "S".
  */
-#define strlcpybuff(A, B, S) \
-  strcpy_safe_(A, S, B, \
-  HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B), \
-  "overflow while copying '" #B "' to '"#A"'", __FILE__, __LINE__)
+#define strlcpybuff(A, B, S)                                                   \
+  strcpy_safe_(A, S, B, HTS_IS_NOT_CHAR_BUFFER(B) ? (size_t) -1 : sizeof(B),   \
+               "overflow while copying '" #B "' to '" #A "'", __FILE__,        \
+               __LINE__)
 
 /** strnlen replacement (autotools). **/
-#if ( ! defined(_WIN32) && ! defined(HAVE_STRNLEN) )
+#if (!defined(_WIN32) && !defined(HAVE_STRNLEN))
 
 static HTS_UNUSED size_t strnlen(const char *s, size_t maxlen) {
   size_t i;
-  for(i = 0 ; i < maxlen && s[i] != '\0' ; i++) ;
+  for (i = 0; i < maxlen && s[i] != '\0'; i++)
+    ;
   return i;
 }
 #endif
@@ -304,13 +322,14 @@ static HTS_UNUSED size_t strnlen(const char *s, size_t maxlen) {
    Aborts if source is NULL or has no NUL within that capacity. The sentinel
    sizeof_source == (size_t)-1 means "capacity unknown", and falls back to the
    unbounded strlen (used when the source is a pointer rather than an array). */
-static HTS_INLINE HTS_UNUSED size_t strlen_safe_(const char *source, const size_t sizeof_source, 
+static HTS_INLINE HTS_UNUSED size_t strlen_safe_(const char *source,
+                                                 const size_t sizeof_source,
                                                  const char *file, int line) {
   size_t size;
-  assertf_( source != NULL, file, line );
-  size = sizeof_source != (size_t) -1 
-    ? strnlen(source, sizeof_source) : strlen(source);
-  assertf_( size < sizeof_source, file, line );
+  assertf_(source != NULL, file, line);
+  size = sizeof_source != (size_t) -1 ? strnlen(source, sizeof_source)
+                                      : strlen(source);
+  assertf_(size < sizeof_source, file, line);
   return size;
 }
 
@@ -319,10 +338,10 @@ static HTS_INLINE HTS_UNUSED size_t strlen_safe_(const char *source, const size_
    source's capacity or (size_t)-1 if unknown. Aborts if the result (existing
    dest length + appended bytes + NUL) would not fit sizeof_dest: this NEVER
    truncates. Always NUL-terminates on success. */
-static HTS_INLINE HTS_UNUSED char* strncat_safe_(char *const dest, const size_t sizeof_dest,
-                                                 const char *const source, const size_t sizeof_source, 
-                                                 const size_t n,
-                                                 const char *exp, const char *file, int line) {
+static HTS_INLINE HTS_UNUSED char *
+strncat_safe_(char *const dest, const size_t sizeof_dest,
+              const char *const source, const size_t sizeof_source,
+              const size_t n, const char *exp, const char *file, int line) {
   const size_t source_len = strlen_safe_(source, sizeof_source, file, line);
   const size_t dest_len = strlen_safe_(dest, sizeof_dest, file, line);
   /* note: "size_t is an unsigned integral type" ((size_t) -1 is positive) */
@@ -337,12 +356,14 @@ static HTS_INLINE HTS_UNUSED char* strncat_safe_(char *const dest, const size_t
 /* Core bounded copy: empties dest then appends all of source via
    strncat_safe_. sizeof_dest is dest's total capacity (NUL included). Aborts
    (no truncation) if source plus its NUL would not fit. */
-static HTS_INLINE HTS_UNUSED char* strcpy_safe_(char *const dest, const size_t sizeof_dest,
-                                                const char *const source, const size_t sizeof_source, 
-                                                const char *exp, const char *file, int line) {
+static HTS_INLINE HTS_UNUSED char *
+strcpy_safe_(char *const dest, const size_t sizeof_dest,
+             const char *const source, const size_t sizeof_source,
+             const char *exp, const char *file, int line) {
   assertf_(sizeof_dest != 0, file, line);
   dest[0] = '\0';
-  return strncat_safe_(dest, sizeof_dest, source, sizeof_source, (size_t) -1, exp, file, line);
+  return strncat_safe_(dest, sizeof_dest, source, sizeof_source, (size_t) -1,
+                       exp, file, line);
 }
 
 /**
@@ -360,9 +381,9 @@ static HTS_INLINE HTS_UNUSED char* strcpy_safe_(char *const dest, const size_t s
  * htsbuff_ptr(). The buffer is kept NUL-terminated; htsbuff_str() returns it.
  */
 typedef struct {
-  char *buf;        /* backing buffer (kept NUL-terminated) */
-  size_t cap;       /* total capacity of buf, including the NUL */
-  size_t len;       /* current length, excluding the NUL */
+  char *buf;  /* backing buffer (kept NUL-terminated) */
+  size_t cap; /* total capacity of buf, including the NUL */
+  size_t len; /* current length, excluding the NUL */
 } htsbuff;
 
 static HTS_INLINE HTS_UNUSED htsbuff htsbuff_ptr_(char *buf, size_t cap) {
@@ -384,23 +405,29 @@ static HTS_INLINE HTS_UNUSED htsbuff htsbuff_ptr_(char *buf, size_t cap) {
 #if (defined(__GNUC__) && !defined(__cplusplus))
 
 /* 0 for an array, a -1 array-size compile error for a pointer. */
-#define htsbuff_must_be_array_(A) \
-  (sizeof(char[1 - 2 * !!__builtin_types_compatible_p(typeof(A), typeof(&(A)[0]))]) - 1)
+#define htsbuff_must_be_array_(A)                                              \
+  (sizeof(char[1 - 2 * !!__builtin_types_compatible_p(typeof(A),               \
+                                                      typeof(&(A)[0]))]) -     \
+   1)
 
-#define htsbuff_array(ARR) htsbuff_ptr_((ARR), sizeof(ARR) + htsbuff_must_be_array_(ARR))
+#define htsbuff_array(ARR)                                                     \
+  htsbuff_ptr_((ARR), sizeof(ARR) + htsbuff_must_be_array_(ARR))
 #else
 #define htsbuff_array(ARR) htsbuff_ptr_((ARR), sizeof(ARR))
 #endif
 /** Builder over pointer P of known capacity N (N includes the NUL). */
-#define htsbuff_ptr(P, N)  htsbuff_ptr_((P), (N))
+#define htsbuff_ptr(P, N) htsbuff_ptr_((P), (N))
 
-/** Append at most n characters of s (stopping at its NUL). Aborts on overflow. */
-static HTS_INLINE HTS_UNUSED void htsbuff_catn(htsbuff *b, const char *s, size_t n) {
+/** Append at most n characters of s (stopping at its NUL). Aborts on overflow.
+ */
+static HTS_INLINE HTS_UNUSED void htsbuff_catn(htsbuff *b, const char *s,
+                                               size_t n) {
   const size_t add = strnlen(s, n);
   /* Overflow-safe: keep the (potentially huge) 'add' alone on one side. The
      maintained invariant len < cap makes 'cap - len' >= 1 (no underflow), so
      'add < cap - len' cannot wrap the way 'len + add < cap' could. */
-  assertf__(add < b->cap - b->len, "htsbuff append overflow", __FILE__, __LINE__);
+  assertf__(add < b->cap - b->len, "htsbuff append overflow", __FILE__,
+            __LINE__);
   memcpy(b->buf + b->len, s, add);
   b->len += add;
   b->buf[b->len] = '\0';
@@ -433,15 +460,21 @@ static HTS_INLINE HTS_UNUSED const char *htsbuff_str(const htsbuff *b) {
    added bounds checking. freet() also NULLs the freed pointer and tolerates
    NULL. memcpybuff() despite the name is a raw memcpy: the caller owns the
    bounds. */
-#define malloct(A)          malloc(A)
+#define malloct(A) malloc(A)
 
-#define calloct(A,B)        calloc((A), (B))
+#define calloct(A, B) calloc((A), (B))
 
-#define freet(A)            do { if ((A) != NULL) { free(A); (A) = NULL; } } while(0)
+#define freet(A)                                                               \
+  do {                                                                         \
+    if ((A) != NULL) {                                                         \
+      free(A);                                                                 \
+      (A) = NULL;                                                              \
+    }                                                                          \
+  } while (0)
 
-#define strdupt(A)          strdup(A)
+#define strdupt(A) strdup(A)
 
-#define realloct(A,B)       realloc(A, B)
+#define realloct(A, B) realloc(A, B)
 
 #define memcpybuff(A, B, N) memcpy((A), (B), (N))
 

src/htsstrings.h

@@ -41,11 +41,11 @@ Please visit our Website: http://www.httrack.com
 /* GCC extension */
 #ifndef HTS_UNUSED
 #ifdef __GNUC__
-#define HTS_UNUSED __attribute__ ((unused))
+#define HTS_UNUSED __attribute__((unused))
 
-#define HTS_STATIC static __attribute__ ((unused))
+#define HTS_STATIC static __attribute__((unused))
 
-#define HTS_PRINTF_FUN(fmt, arg) __attribute__ ((format (printf, fmt, arg)))
+#define HTS_PRINTF_FUN(fmt, arg) __attribute__((format(printf, fmt, arg)))
 #else
 #define HTS_UNUSED
 #define HTS_STATIC static
@@ -60,6 +60,7 @@ typedef struct String String;
 #endif
 #ifndef HTS_DEF_STRUCT_String
 #define HTS_DEF_STRUCT_String
+
 /**
  * Growable owned string.
  *
@@ -86,7 +87,7 @@ struct String {
 
 /** Allocator **/
 #ifndef STRING_REALLOC
-#define STRING_REALLOC(BUFF, SIZE) ( (char*) realloc(BUFF, SIZE) )
+#define STRING_REALLOC(BUFF, SIZE) ((char *) realloc(BUFF, SIZE))
 
 #define STRING_FREE(BUFF) free(BUFF)
 #endif
@@ -96,11 +97,11 @@ struct String {
 #endif
 
 /** Initializer for an empty String (NULL buffer). Use to declare or reset. **/
-#define STRING_EMPTY { (char*) NULL, 0, 0 }
+#define STRING_EMPTY {(char *) NULL, 0, 0}
 
 /** Read-only buffer pointer. NULL until the String has been written to.
     Invalidated by any subsequent growing operation. **/
-#define StringBuff(BLK) ( (const char*) ((BLK).buffer_) )
+#define StringBuff(BLK) ((const char *) ((BLK).buffer_))
 
 /** Read/write buffer pointer. Same NULL/invalidation rules as StringBuff. **/
 #define StringBuffRW(BLK) ((BLK).buffer_)
@@ -109,56 +110,60 @@ struct String {
 #define StringLength(BLK) ((BLK).length_)
 
 /** Non-zero if the String holds at least one byte. **/
-#define StringNotEmpty(BLK) ( StringLength(BLK) > 0 )
+#define StringNotEmpty(BLK) (StringLength(BLK) > 0)
 
 /** Allocated capacity in bytes, including room for the terminating NUL. **/
 #define StringCapacity(BLK) ((BLK).capacity_)
 
 /** Byte at POS (read). No bounds check; POS must be < StringLength. **/
-#define StringSub(BLK, POS) ( StringBuff(BLK)[POS] )
+#define StringSub(BLK, POS) (StringBuff(BLK)[POS])
 
 /** Byte at POS (read/write). No bounds check; POS must be < StringLength. **/
-#define StringSubRW(BLK, POS) ( StringBuffRW(BLK)[POS] )
+#define StringSubRW(BLK, POS) (StringBuffRW(BLK)[POS])
 
 /** Subcharacter (read/write) **/
-#define StringSubRW(BLK, POS) ( StringBuffRW(BLK)[POS] )
+#define StringSubRW(BLK, POS) (StringBuffRW(BLK)[POS])
 
 /** Byte POS positions from the end (read). POS==1 is the last byte. **/
-#define StringRight(BLK, POS) ( StringBuff(BLK)[StringLength(BLK) - POS] )
+#define StringRight(BLK, POS) (StringBuff(BLK)[StringLength(BLK) - POS])
 
 /** Byte POS positions from the end (read/write). POS==1 is the last byte. **/
-#define StringRightRW(BLK, POS) ( StringBuffRW(BLK)[StringLength(BLK) - POS] )
+#define StringRightRW(BLK, POS) (StringBuffRW(BLK)[StringLength(BLK) - POS])
 
 /** Drop the last byte and re-terminate. Undefined if the String is empty
     (no length check; would underflow). **/
-#define StringPopRight(BLK) do { \
-  StringBuffRW(BLK)[--StringLength(BLK)] = '\0'; \
-} while(0)
+#define StringPopRight(BLK)                                                    \
+  do {                                                                         \
+    StringBuffRW(BLK)[--StringLength(BLK)] = '\0';                             \
+  } while (0)
 
 /** Grow so capacity_ >= CAPACITY (total bytes, including the NUL). May realloc
     (invalidating prior buffer pointers); aborts via STRING_ASSERT on OOM.
     Never shrinks. **/
-#define StringRoomTotal(BLK, CAPACITY) do { \
-  const size_t capacity_ = (size_t) (CAPACITY); \
-  while ((BLK).capacity_ < capacity_) { \
-    if ((BLK).capacity_ < 16) { \
-      (BLK).capacity_ = 16; \
-    } else { \
-      (BLK).capacity_ *= 2; \
-    } \
-    (BLK).buffer_ = STRING_REALLOC((BLK).buffer_, (BLK).capacity_); \
-    STRING_ASSERT((BLK).buffer_ != NULL); \
-  } \
-} while(0)
+#define StringRoomTotal(BLK, CAPACITY)                                         \
+  do {                                                                         \
+    const size_t capacity_ = (size_t) (CAPACITY);                              \
+    while ((BLK).capacity_ < capacity_) {                                      \
+      if ((BLK).capacity_ < 16) {                                              \
+        (BLK).capacity_ = 16;                                                  \
+      } else {                                                                 \
+        (BLK).capacity_ *= 2;                                                  \
+      }                                                                        \
+      (BLK).buffer_ = STRING_REALLOC((BLK).buffer_, (BLK).capacity_);          \
+      STRING_ASSERT((BLK).buffer_ != NULL);                                    \
+    }                                                                          \
+  } while (0)
 
 /** Reserve room for SIZE more bytes beyond the current length (plus the NUL).
     May realloc, invalidating prior buffer pointers. **/
-#define StringRoom(BLK, SIZE) StringRoomTotal(BLK, StringLength(BLK) + (SIZE) + 1)
+#define StringRoom(BLK, SIZE)                                                  \
+  StringRoomTotal(BLK, StringLength(BLK) + (SIZE) + 1)
 
 /** Reserve room for SIZE more bytes and return the (post-realloc) RW buffer,
     for appending in place. Does not update length_; the caller must. **/
 #define StringBuffN(BLK, SIZE) StringBuffN_(&(BLK), SIZE)
-HTS_STATIC char *StringBuffN_(String * blk, int size) {
+
+HTS_STATIC char *StringBuffN_(String *blk, int size) {
   StringRoom(*blk, size);
   return StringBuffRW(*blk);
 }
@@ -166,40 +171,44 @@ HTS_STATIC char *StringBuffN_(String * blk, int size) {
 /** Zero the fields (NULL buffer, no allocation). Use on an uninitialized
     String only; does NOT free an existing buffer (use StringFree to reset
     an owned one), so calling it on a live String leaks. **/
-#define StringInit(BLK) do { \
-  (BLK).buffer_ = NULL; \
-  (BLK).capacity_ = 0; \
-  (BLK).length_ = 0; \
-} while(0)
+#define StringInit(BLK)                                                        \
+  do {                                                                         \
+    (BLK).buffer_ = NULL;                                                      \
+    (BLK).capacity_ = 0;                                                       \
+    (BLK).length_ = 0;                                                         \
+  } while (0)
 
 /** Truncate to length 0, keeping the allocation. Forces a non-NULL buffer
     (allocates if empty) and writes the leading NUL, so StringBuff is "". **/
-#define StringClear(BLK) do { \
-  (BLK).length_ = 0; \
-  StringRoom(BLK, 0); \
-  (BLK).buffer_[0] = '\0'; \
-} while(0)
+#define StringClear(BLK)                                                       \
+  do {                                                                         \
+    (BLK).length_ = 0;                                                         \
+    StringRoom(BLK, 0);                                                        \
+    (BLK).buffer_[0] = '\0';                                                   \
+  } while (0)
 
 /** Set length_ to SIZE, or to strlen(buffer_) if SIZE is negative. Caller
     asserts SIZE fits the existing content; does not (re)allocate. **/
-#define StringSetLength(BLK, SIZE) do { \
-  if (SIZE >= 0) { \
-    (BLK).length_ = SIZE; \
-  } else { \
-    (BLK).length_ = strlen((BLK).buffer_); \
-  } \
-} while(0)
+#define StringSetLength(BLK, SIZE)                                             \
+  do {                                                                         \
+    if (SIZE >= 0) {                                                           \
+      (BLK).length_ = SIZE;                                                    \
+    } else {                                                                   \
+      (BLK).length_ = strlen((BLK).buffer_);                                   \
+    }                                                                          \
+  } while (0)
 
 /** Release the owned buffer and reset to the empty state (NULL buffer).
     Idempotent; safe on an already-empty String. **/
-#define StringFree(BLK) do { \
-  if ((BLK).buffer_ != NULL) { \
-    STRING_FREE((BLK).buffer_); \
-    (BLK).buffer_ = NULL; \
-  } \
-  (BLK).capacity_ = 0; \
-  (BLK).length_ = 0; \
-} while(0)
+#define StringFree(BLK)                                                        \
+  do {                                                                         \
+    if ((BLK).buffer_ != NULL) {                                               \
+      STRING_FREE((BLK).buffer_);                                              \
+      (BLK).buffer_ = NULL;                                                    \
+    }                                                                          \
+    (BLK).capacity_ = 0;                                                       \
+    (BLK).length_ = 0;                                                         \
+  } while (0)
 
 /** Take ownership of a NUL-terminated heap string STR (the String will free
     it). Frees any current buffer first. STR MUST have been allocated by an
@@ -207,48 +216,52 @@ HTS_STATIC char *StringBuffN_(String * blk, int size) {
     freed or used by the caller afterwards. length_/capacity_ are set to
     strlen(STR) (capacity_ here excludes the NUL, so the next append reallocs).
    **/
-#define StringSetBuffer(BLK, STR) do { \
-  size_t len__ = strlen( STR ); \
-  StringFree(BLK); \
-  (BLK).buffer_ = ( STR ); \
-  (BLK).capacity_ = len__; \
-  (BLK).length_ = len__; \
-} while(0)
+#define StringSetBuffer(BLK, STR)                                              \
+  do {                                                                         \
+    size_t len__ = strlen(STR);                                                \
+    StringFree(BLK);                                                           \
+    (BLK).buffer_ = (STR);                                                     \
+    (BLK).capacity_ = len__;                                                   \
+    (BLK).length_ = len__;                                                     \
+  } while (0)
 
 /** Append SIZE raw bytes from STR (NULs allowed as data). Grows as needed and
     re-terminates with a NUL after the appended bytes. STR must not alias
     BLK's buffer (a realloc would invalidate it). **/
-#define StringMemcat(BLK, STR, SIZE) do { \
-  const char* str_mc_ = (STR); \
-  const size_t size_mc_ = (size_t) (SIZE); \
-  StringRoom(BLK, size_mc_); \
-  if (size_mc_ > 0) { \
-    memcpy((BLK).buffer_ + (BLK).length_, str_mc_, size_mc_); \
-    (BLK).length_ += size_mc_; \
-  } \
-  *((BLK).buffer_ + (BLK).length_) = '\0'; \
-} while(0)
+#define StringMemcat(BLK, STR, SIZE)                                           \
+  do {                                                                         \
+    const char *str_mc_ = (STR);                                               \
+    const size_t size_mc_ = (size_t) (SIZE);                                   \
+    StringRoom(BLK, size_mc_);                                                 \
+    if (size_mc_ > 0) {                                                        \
+      memcpy((BLK).buffer_ + (BLK).length_, str_mc_, size_mc_);                \
+      (BLK).length_ += size_mc_;                                               \
+    }                                                                          \
+    *((BLK).buffer_ + (BLK).length_) = '\0';                                   \
+  } while (0)
 
 /** Replace content with SIZE raw bytes from STR (NULs allowed as data).
     Same non-aliasing requirement as StringMemcat. **/
-#define StringMemcpy(BLK, STR, SIZE) do { \
-  (BLK).length_ = 0; \
-  StringMemcat(BLK, STR, SIZE); \
-} while(0)
+#define StringMemcpy(BLK, STR, SIZE)                                           \
+  do {                                                                         \
+    (BLK).length_ = 0;                                                         \
+    StringMemcat(BLK, STR, SIZE);                                              \
+  } while (0)
 
 /** Append one byte and re-terminate. Grows as needed. **/
-#define StringAddchar(BLK, c) do { \
-  String * const s__ = &(BLK); \
-  char c__ = (c); \
-  StringRoom(*s__, 1); \
-  StringBuffRW(*s__)[StringLength(*s__)++] = c__; \
-  StringBuffRW(*s__)[StringLength(*s__)  ] = 0; \
-} while(0)
+#define StringAddchar(BLK, c)                                                  \
+  do {                                                                         \
+    String *const s__ = &(BLK);                                                \
+    char c__ = (c);                                                            \
+    StringRoom(*s__, 1);                                                       \
+    StringBuffRW(*s__)[StringLength(*s__)++] = c__;                            \
+    StringBuffRW(*s__)[StringLength(*s__)] = 0;                                \
+  } while (0)
 
 /** Hand the buffer to the caller and reset the String to empty (NULL buffer).
     The returned pointer is now owned by the caller, who must STRING_FREE() it.
     Returns NULL if the String was empty. **/
-HTS_STATIC char *StringAcquire(String * blk) {
+HTS_STATIC char *StringAcquire(String *blk) {
   char *buff = StringBuffRW(*blk);
 
   StringBuffRW(*blk) = NULL;
@@ -259,7 +272,7 @@ HTS_STATIC char *StringAcquire(String * blk) {
 
 /** Return an independent deep copy of *src (its own allocation). The caller
     owns the result and must StringFree it. **/
-HTS_STATIC String StringDup(const String * src) {
+HTS_STATIC String StringDup(const String *src) {
   String s = STRING_EMPTY;
 
   StringMemcat(s, StringBuff(*src), StringLength(*src));
@@ -270,7 +283,7 @@ HTS_STATIC String StringDup(const String * src) {
     ownership transfers and the caller keeps no dangling alias. Frees any
     current buffer first. *str MUST be allocator-compatible (see
     StringSetBuffer). No-op if str or *str is NULL. **/
-HTS_STATIC void StringAttach(String * blk, char **str) {
+HTS_STATIC void StringAttach(String *blk, char **str) {
   StringFree(*blk);
   if (str != NULL && *str != NULL) {
     StringBuffRW(*blk) = *str;
@@ -281,43 +294,46 @@ HTS_STATIC void StringAttach(String * blk, char **str) {
 
 /** Append the C string STR (up to its NUL). No-op if STR is NULL. STR must not
     alias BLK's buffer. **/
-#define StringCat(BLK, STR) do { \
-  const char *const str__ = ( STR ); \
-  if (str__ != NULL) { \
-    const size_t size__ = strlen(str__); \
-    StringMemcat(BLK, str__, size__); \
-  } \
-} while(0)
+#define StringCat(BLK, STR)                                                    \
+  do {                                                                         \
+    const char *const str__ = (STR);                                           \
+    if (str__ != NULL) {                                                       \
+      const size_t size__ = strlen(str__);                                     \
+      StringMemcat(BLK, str__, size__);                                        \
+    }                                                                          \
+  } while (0)
 
 /** Append at most SIZE leading bytes of the C string STR. No-op if STR is
     NULL. STR must not alias BLK's buffer. **/
-#define StringCatN(BLK, STR, SIZE) do { \
-  const char *str__ = ( STR ); \
-  if (str__ != NULL) { \
-    size_t size__ = strlen(str__); \
-    if (size__ > (SIZE)) { \
-      size__ = (SIZE); \
-    } \
-    StringMemcat(BLK, str__, size__); \
-  } \
-} while(0)
+#define StringCatN(BLK, STR, SIZE)                                             \
+  do {                                                                         \
+    const char *str__ = (STR);                                                 \
+    if (str__ != NULL) {                                                       \
+      size_t size__ = strlen(str__);                                           \
+      if (size__ > (SIZE)) {                                                   \
+        size__ = (SIZE);                                                       \
+      }                                                                        \
+      StringMemcat(BLK, str__, size__);                                        \
+    }                                                                          \
+  } while (0)
 
 /** Replace content with at most SIZE leading bytes of the C string STR.
     If STR is NULL, clears to "". STR must not alias BLK's buffer. **/
-#define StringCopyN(BLK, STR, SIZE) do { \
-  const char *str__ = ( STR ); \
-  const size_t usize__ = (SIZE); \
-  (BLK).length_ = 0; \
-  if (str__ != NULL) { \
-    size_t size__ = strlen(str__); \
-    if (size__ > usize__ ) { \
-      size__ = usize__; \
-    } \
-    StringMemcat(BLK, str__, size__); \
-  } else { \
-    StringClear(BLK); \
-  } \
-} while(0)
+#define StringCopyN(BLK, STR, SIZE)                                            \
+  do {                                                                         \
+    const char *str__ = (STR);                                                 \
+    const size_t usize__ = (SIZE);                                             \
+    (BLK).length_ = 0;                                                         \
+    if (str__ != NULL) {                                                       \
+      size_t size__ = strlen(str__);                                           \
+      if (size__ > usize__) {                                                  \
+        size__ = usize__;                                                      \
+      }                                                                        \
+      StringMemcat(BLK, str__, size__);                                        \
+    } else {                                                                   \
+      StringClear(BLK);                                                        \
+    }                                                                          \
+  } while (0)
 
 /** Replace blk's content with a copy of String blk2. blk and blk2 must be
     distinct Strings (use StringCopyOverlapped if they may be the same). **/
@@ -326,23 +342,25 @@ HTS_STATIC void StringAttach(String * blk, char **str) {
 /** Replace content with a copy of the C string STR. If STR is NULL, clears to
     "". STR must not alias BLK's buffer (use StringCopyOverlapped if it might).
    **/
-#define StringCopy(BLK, STR) do { \
-  const char *str__ = ( STR ); \
-  if (str__ != NULL) { \
-    size_t size__ = strlen(str__); \
-    StringMemcpy(BLK, str__, size__); \
-  } else { \
-    StringClear(BLK); \
-  } \
-} while(0)
+#define StringCopy(BLK, STR)                                                   \
+  do {                                                                         \
+    const char *str__ = (STR);                                                 \
+    if (str__ != NULL) {                                                       \
+      size_t size__ = strlen(str__);                                           \
+      StringMemcpy(BLK, str__, size__);                                        \
+    } else {                                                                   \
+      StringClear(BLK);                                                        \
+    }                                                                          \
+  } while (0)
 
 /** Like StringCopy but safe when STR aliases BLK's own buffer: copies via a
     temporary, so a self-copy or overlap is well-defined. **/
-#define StringCopyOverlapped(BLK, STR) do { \
-  String s__ = STRING_EMPTY; \
-  StringCopy(s__, STR); \
-  StringCopyS(BLK, s__); \
-  StringFree(s__); \
-} while(0)
+#define StringCopyOverlapped(BLK, STR)                                         \
+  do {                                                                         \
+    String s__ = STRING_EMPTY;                                                 \
+    StringCopy(s__, STR);                                                      \
+    StringCopyS(BLK, s__);                                                     \
+    StringFree(s__);                                                           \
+  } while (0)
 
 #endif

src/htswrap.h

@@ -57,17 +57,17 @@ extern "C" {
 #endif
 
 /** Legacy no-op retained for ABI compatibility; always returns 1. */
-HTSEXT_API int htswrap_init(void);      // LEGACY
+HTSEXT_API int htswrap_init(void); // LEGACY
 
 /** Legacy no-op retained for ABI compatibility; always returns 1. */
-HTSEXT_API int htswrap_free(void);      // LEGACY
+HTSEXT_API int htswrap_free(void); // LEGACY
 
 #ifdef __cplusplus
 }
 #endif
 
-//HTSEXT_API int htswrap_add(httrackp * opt, const char *name, void *fct);
-//HTSEXT_API uintptr_t htswrap_read(httrackp * opt, const char *name);
+// HTSEXT_API int htswrap_add(httrackp * opt, const char *name, void *fct);
+// HTSEXT_API uintptr_t htswrap_read(httrackp * opt, const char *name);
 
 #endif
 

src/httrack-library.h

@@ -73,6 +73,7 @@ typedef struct strc_int2bytes2 strc_int2bytes2;
 #endif
 #ifndef HTS_DEF_DEFSTRUCT_hts_log_type
 #define HTS_DEF_DEFSTRUCT_hts_log_type
+
 /** Log severity levels, most to least severe. A message is emitted only if its
     level is <= opt->debug. LOG_ERRNO is a flag OR'd into the level to append
     ": <strerror(errno)>" to the message. */
@@ -97,7 +98,7 @@ typedef struct hts_stat_struct hts_stat_struct;
     retain them. */
 #ifndef HTS_DEF_FWSTRUCT_htsErrorCallback
 #define HTS_DEF_FWSTRUCT_htsErrorCallback
-typedef void (*htsErrorCallback) (const char *msg, const char *file, int line);
+typedef void (*htsErrorCallback)(const char *msg, const char *file, int line);
 #endif
 
 /* Helpers for plugging callbacks
@@ -111,29 +112,35 @@ requires: htsdefines.h */
  * CALLBACKARG_USERDEF(). Allocates a t_hts_callbackarg with hts_malloc (not
  * checked for OOM); it is freed by hts_free_opt().
  */
-#define CHAIN_FUNCTION(OPT, MEMBER, FUNCTION, ARGUMENT) do { \
-  t_hts_callbackarg *carg = (t_hts_callbackarg*) hts_malloc(sizeof(t_hts_callbackarg)); \
-  carg->userdef = ( ARGUMENT ); \
-  carg->prev.fun = (void*) ( OPT )->callbacks_fun-> MEMBER .fun; \
-  carg->prev.carg = ( OPT )->callbacks_fun-> MEMBER .carg; \
-  ( OPT )->callbacks_fun-> MEMBER .fun = ( FUNCTION ); \
-  ( OPT )->callbacks_fun-> MEMBER .carg = carg; \
-} while(0)
+#define CHAIN_FUNCTION(OPT, MEMBER, FUNCTION, ARGUMENT)                        \
+  do {                                                                         \
+    t_hts_callbackarg *carg =                                                  \
+        (t_hts_callbackarg *) hts_malloc(sizeof(t_hts_callbackarg));           \
+    carg->userdef = (ARGUMENT);                                                \
+    carg->prev.fun = (void *) (OPT)->callbacks_fun->MEMBER.fun;                \
+    carg->prev.carg = (OPT)->callbacks_fun->MEMBER.carg;                       \
+    (OPT)->callbacks_fun->MEMBER.fun = (FUNCTION);                             \
+    (OPT)->callbacks_fun->MEMBER.carg = carg;                                  \
+  } while (0)
 
-/* The following helpers are useful only if you know that an existing callback migh be existing before before the call to CHAIN_FUNCTION()
-If your functions were added just after hts_create_opt(), no need to make the previous function check */
+/* The following helpers are useful only if you know that an existing callback
+migh be existing before before the call to CHAIN_FUNCTION() If your functions
+were added just after hts_create_opt(), no need to make the previous function
+check */
 
 /** Inside a chained callback, return the ARGUMENT pointer originally passed to
     CHAIN_FUNCTION(), or NULL when CARG is NULL. */
-#define CALLBACKARG_USERDEF(CARG) ( ( (CARG) != NULL ) ? (CARG)->userdef : NULL )
+#define CALLBACKARG_USERDEF(CARG) (((CARG) != NULL) ? (CARG)->userdef : NULL)
 
 /** Return the callback of type NAME that this one chained over, cast to its
     function-pointer type, or NULL. Call it to forward to the prior handler. */
-#define CALLBACKARG_PREV_FUN(CARG, NAME) ( (t_hts_htmlcheck_ ##NAME) ( ( (CARG) != NULL ) ? (CARG)->prev.fun : NULL ) )
+#define CALLBACKARG_PREV_FUN(CARG, NAME)                                       \
+  ((t_hts_htmlcheck_##NAME)(((CARG) != NULL) ? (CARG)->prev.fun : NULL))
 
 /** Return the carg of the callback this one chained over (pass it when
    forwarding to the CALLBACKARG_PREV_FUN result), or NULL. */
-#define CALLBACKARG_PREV_CARG(CARG) ( ( (CARG) != NULL ) ? (CARG)->prev.carg : NULL )
+#define CALLBACKARG_PREV_CARG(CARG)                                            \
+  (((CARG) != NULL) ? (CARG)->prev.carg : NULL)
 
 /* Functions */
 
@@ -162,7 +169,7 @@ HTSEXT_API int hts_main(int argc, char **argv);
     hts_main() to set options or plug callbacks on opt first. Blocks until the
     mirror ends and returns the engine exit code. The caller keeps ownership of
     opt and must release it with hts_free_opt(). */
-HTSEXT_API int hts_main2(int argc, char **argv, httrackp * opt);
+HTSEXT_API int hts_main2(int argc, char **argv, httrackp *opt);
 
 /* Options handling */
 /** Allocate and default-initialize an option set, preloading the bundled parser
@@ -174,7 +181,7 @@ HTSEXT_API httrackp *hts_create_opt(void);
     modules, DNS cache, owned strings, and the structure). NULL is accepted. The
     pointer is invalid afterward. Do not call while a mirror is running on that
     opt; wait until hts_has_stopped() is true. */
-HTSEXT_API void hts_free_opt(httrackp * opt);
+HTSEXT_API void hts_free_opt(httrackp *opt);
 
 /** Return sizeof(httrackp) as the library sees it, for caller-vs-library struct
     ABI mismatch checks. */
@@ -184,16 +191,16 @@ HTSEXT_API size_t hts_sizeof_opt(void);
     Returns NULL if opt is NULL. The result aliases a single process-global
     static: it is not thread-safe and is overwritten by the next call, so copy
     out the fields you need. */
-HTSEXT_API const hts_stat_struct* hts_get_stats(httrackp * opt);
+HTSEXT_API const hts_stat_struct *hts_get_stats(httrackp *opt);
 
 /** Legacy no-op retained for API compatibility. */
-HTSEXT_API void set_wrappers(httrackp * opt);   /* LEGACY */
+HTSEXT_API void set_wrappers(httrackp *opt); /* LEGACY */
 
 /** Load a plugin shared library and run its hts_plug(opt, argv) entry point. On
     success the handle is recorded in opt and unloaded by hts_free_opt().
     @return 1 if loaded and hts_plug succeeded; 0 if loaded but hts_plug was
     missing or refused; -1 if the library could not be loaded. */
-HTSEXT_API int plug_wrapper(httrackp * opt, const char *moduleName,
+HTSEXT_API int plug_wrapper(httrackp *opt, const char *moduleName,
                             const char *argv);
 
 /** Install the process-global assertion/error callback (NULL clears it). Not
@@ -212,12 +219,12 @@ HTSEXT_API hts_boolean hts_log(httrackp *opt, const char *prefix,
 /** printf-style log at level @p type (an hts_log_type, optionally |LOG_ERRNO).
     Forwards to the registered log callback, and when the level is <= opt->debug
     also to opt->log. @p format must be non-NULL. */
-HTSEXT_API void hts_log_print(httrackp * opt, int type, const char *format,
-                              ...) HTS_PRINTF_FUN(3, 4);
+HTSEXT_API void hts_log_print(httrackp *opt, int type, const char *format, ...)
+    HTS_PRINTF_FUN(3, 4);
 
 /** va_list form of hts_log_print(). @p opt may be NULL (only the callback
    runs). Preserves errno. @p format must be non-NULL. */
-HTSEXT_API void hts_log_vprint(httrackp * opt, int type, const char *format,
+HTSEXT_API void hts_log_vprint(httrackp *opt, int type, const char *format,
                                va_list args);
 
 /** Install the process-global log callback invoked by hts_log_vprint() for
@@ -231,7 +238,7 @@ hts_set_log_vprint_callback(void (*callback)(httrackp *opt, int type,
     result is written into and aliases a 2048-byte scratch buffer inside opt: it
     is valid until that buffer is next used, and must not be freed. opt must be
     non-NULL. */
-HTSEXT_API const char *hts_get_version_info(httrackp * opt);
+HTSEXT_API const char *hts_get_version_info(httrackp *opt);
 
 /** Static build-features string (TLS, zlib, ipv6, and so on). Process-global
     storage; do not free or modify. */
@@ -241,21 +248,22 @@ HTSEXT_API const char *hts_is_available(void);
 HTSEXT_API const char *hts_version(void);
 
 /* Wrapper functions */
-HTSEXT_API int htswrap_init(void);      // DEPRECATED - DUMMY FUNCTION
+HTSEXT_API int htswrap_init(void); // DEPRECATED - DUMMY FUNCTION
 
-HTSEXT_API int htswrap_free(void);      // DEPRECATED - DUMMY FUNCTION
+HTSEXT_API int htswrap_free(void); // DEPRECATED - DUMMY FUNCTION
 
 /** Register callback @p fct under @p name in opt's callback table (for example
     "start", "check-html", "linkdetected"). Returns 1 on success, 0 if @p name
    is not a known slot. Prefer CHAIN_FUNCTION(), which preserves any prior
    callback. */
-HTSEXT_API int htswrap_add(httrackp * opt, const char *name, void *fct);
+HTSEXT_API int htswrap_add(httrackp *opt, const char *name, void *fct);
 
 /** Return the function pointer registered under @p name in opt as a uintptr_t,
    or 0 if none or unknown. */
-HTSEXT_API uintptr_t htswrap_read(httrackp * opt, const char *name);
+HTSEXT_API uintptr_t htswrap_read(httrackp *opt, const char *name);
 
-/* Internal library allocators, if a different libc is being used by the client */
+/* Internal library allocators, if a different libc is being used by the client
+ */
 /** strdup() through the library allocator. Returns a heap copy freed with
     hts_free(), or NULL on failure. */
 HTSEXT_API char *hts_strdup(const char *string);
@@ -272,13 +280,13 @@ HTSEXT_API void *hts_realloc(void *const data, const size_t size);
 HTSEXT_API void hts_free(void *data);
 
 /* Other functions */
-HTSEXT_API int hts_resetvar(void);      // DEPRECATED - DUMMY FUNCTION
+HTSEXT_API int hts_resetvar(void); // DEPRECATED - DUMMY FUNCTION
 
 /** (Re)build the top-level index.html aggregating every mirror project found
     under @p path. @p binpath is the data root used to locate the
     templates/topindex-*.html files, falling back to built-in templates. Writes
     <path>/index.html. @return 1 on success, 0 on failure. */
-HTSEXT_API int hts_buildtopindex(httrackp * opt, const char *path,
+HTSEXT_API int hts_buildtopindex(httrackp *opt, const char *path,
                                  const char *binpath);
 
 /** Scan every mirror project under @p path and return a CRLF-separated list:
@@ -321,14 +329,14 @@ HTSEXT_API hts_boolean catch_url(T_SOC soc, char *url, char *method,
 /** Whether the engine is parsing HTML. Returns 0 if not, otherwise the percent
     done (at least 1). @p flag >= 0 also requests a progress refresh; pass a
     negative value to query without side effects. */
-HTSEXT_API int hts_is_parsing(httrackp * opt, int flag);
+HTSEXT_API int hts_is_parsing(httrackp *opt, int flag);
 
 /** Current background phase: 0 none, 1 testing links, 2 purge, 3, 4 scheduling,
     5 waiting for a slot. */
-HTSEXT_API int hts_is_testing(httrackp * opt);
+HTSEXT_API int hts_is_testing(httrackp *opt);
 
 /** Nonzero once the engine has begun its exit sequence. */
-HTSEXT_API int hts_is_exiting(httrackp * opt);
+HTSEXT_API int hts_is_exiting(httrackp *opt);
 
 /*HTSEXT_API int hts_setopt(httrackp* opt); DEPRECATED ; see copy_htsopt() */
 
@@ -344,15 +352,15 @@ HTSEXT_API hts_boolean hts_resetaddurl(httrackp *opt);
 /** Apply the runtime-tunable options from @p from onto @p to, to adjust a live
     mirror. Only fields set to a non-sentinel value are copied; the rest of @p
    to is left untouched. The user-agent string is deep-copied. @return 0. */
-HTSEXT_API int copy_htsopt(const httrackp * from, httrackp * to);
+HTSEXT_API int copy_htsopt(const httrackp *from, httrackp *to);
 
 /** Return the engine's last error message, or NULL. The string is owned by
     @p opt; do not free it, and use it only while @p opt lives. */
-HTSEXT_API char *hts_errmsg(httrackp * opt);
+HTSEXT_API char *hts_errmsg(httrackp *opt);
 
 /** Get or set the transfer-pause flag. @p p >= 0 sets it (nonzero means
    paused); a negative value queries. @return the current pause flag. */
-HTSEXT_API int hts_setpause(httrackp * opt, int);
+HTSEXT_API int hts_setpause(httrackp *opt, int);
 
 /** Ask the running mirror to terminate (sets the stop flag under the state
    lock, so it is safe to call from another thread). @p force is currently
@@ -363,15 +371,15 @@ HTSEXT_API int hts_request_stop(httrackp *opt, hts_boolean force);
 /** Queue a single in-progress file, by URL, to be cancelled by the engine.
     @p url is copied internally. Takes the state lock, so it is thread-safe.
     @return the underlying push result. */
-HTSEXT_API int hts_cancel_file_push(httrackp * opt, const char *url);
+HTSEXT_API int hts_cancel_file_push(httrackp *opt, const char *url);
 
 /** Cancel the in-progress link-testing phase. Effective only while a test runs.
  */
-HTSEXT_API void hts_cancel_test(httrackp * opt);
+HTSEXT_API void hts_cancel_test(httrackp *opt);
 
 /** Cancel the in-progress HTML parsing. Effective only while parsing is active.
  */
-HTSEXT_API void hts_cancel_parsing(httrackp * opt);
+HTSEXT_API void hts_cancel_parsing(httrackp *opt);
 
 /** Nonzero once the mirror has fully ended. Read under the engine state lock,
    so safe to poll from another thread. Wait for this before hts_free_opt(). */
@@ -416,19 +424,19 @@ HTSEXT_API void qsec2str(char *st, TStamp t);
    is reused, and a given strc is not reentrant. Use one strc per
    concurrently-live result. */
 /** Format @p n as a decimal string into @p strc and return it. */
-HTSEXT_API char *int2char(strc_int2bytes2 * strc, int n);
+HTSEXT_API char *int2char(strc_int2bytes2 *strc, int n);
 
 /** Format byte count @p n as "<num><unit>" (B/KiB/MiB/GiB and so on) into
     @p strc and return it. */
-HTSEXT_API char *int2bytes(strc_int2bytes2 * strc, LLint n);
+HTSEXT_API char *int2bytes(strc_int2bytes2 *strc, LLint n);
 
 /** Format a transfer rate @p n as "<num><unit>/s" into @p strc and return it.
  */
-HTSEXT_API char *int2bytessec(strc_int2bytes2 * strc, long int n);
+HTSEXT_API char *int2bytessec(strc_int2bytes2 *strc, long int n);
 
 /** Split byte count @p n into number and unit, returning a 2-element array
     {number, unit} stored inside @p strc. */
-HTSEXT_API char **int2bytes2(strc_int2bytes2 * strc, LLint n);
+HTSEXT_API char **int2bytes2(strc_int2bytes2 *strc, LLint n);
 
 /** Skip any "user[:pass]@" identification prefix in a URL, returning a pointer
     into the argument past it (or past the protocol if none). The result aliases
@@ -490,40 +498,50 @@ HTSEXT_API void unescape_amp(char *s);
 
 /** Percent-escape only spaces (' ' becomes "%20"); copy everything else
  * verbatim. */
-HTSEXT_API size_t escape_spc_url(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t escape_spc_url(const char *const src, char *const dest,
+                                 const size_t size);
 
 /** Aggressively percent-escape @p src for use as a single URL path segment
     (reserved, delimiter, unwise, special, avoid and mark characters). */
-HTSEXT_API size_t escape_in_url(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t escape_in_url(const char *const src, char *const dest,
+                                const size_t size);
 
 /** Percent-escape @p src as a URI, escaping only what is necessary and keeping
     '/' and other reserved characters. */
-HTSEXT_API size_t escape_uri(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t escape_uri(const char *const src, char *const dest,
+                             const size_t size);
 
 /** Like escape_uri() for a UTF-8 URI: also escapes reserved characters other
     than '/'. */
-HTSEXT_API size_t escape_uri_utf(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t escape_uri_utf(const char *const src, char *const dest,
+                                 const size_t size);
 
 /** Minimal "make safe" escape: percent-escapes only '"', ' ' and control
     characters, leaving an already-formed URL otherwise intact. */
-HTSEXT_API size_t escape_check_url(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t escape_check_url(const char *const src, char *const dest,
+                                   const size_t size);
 
 /** Append-variant of escape_spc_url(): escapes @p src after the existing
     NUL-terminated content of @p dest. Returns the bytes appended (excluding the
     NUL). */
-HTSEXT_API size_t append_escape_spc_url(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t append_escape_spc_url(const char *const src, char *const dest,
+                                        const size_t size);
 
 /** Append-variant of escape_in_url(). See append_escape_spc_url(). */
-HTSEXT_API size_t append_escape_in_url(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t append_escape_in_url(const char *const src, char *const dest,
+                                       const size_t size);
 
 /** Append-variant of escape_uri(). See append_escape_spc_url(). */
-HTSEXT_API size_t append_escape_uri(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t append_escape_uri(const char *const src, char *const dest,
+                                    const size_t size);
 
 /** Append-variant of escape_uri_utf(). See append_escape_spc_url(). */
-HTSEXT_API size_t append_escape_uri_utf(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t append_escape_uri_utf(const char *const src, char *const dest,
+                                        const size_t size);
 
 /** Append-variant of escape_check_url(). See append_escape_spc_url(). */
-HTSEXT_API size_t append_escape_check_url(const char *const src, char *const dest, const size_t size);
+HTSEXT_API size_t append_escape_check_url(const char *const src,
+                                          char *const dest, const size_t size);
 
 /** In-place variant of escape_spc_url(): escapes the NUL-terminated string in
     @p dest back into @p dest. */
@@ -543,32 +561,39 @@ HTSEXT_API size_t inplace_escape_check_url(char *const dest, const size_t size);
 
 /** Same escaping as escape_check_url() but returns @p dest instead of the byte
     count. */
-HTSEXT_API char *escape_check_url_addr(const char *const src, char *const dest, const size_t size);
+HTSEXT_API char *escape_check_url_addr(const char *const src, char *const dest,
+                                       const size_t size);
 
 /** Build a MIME/MHTML content-id token in @p dest from @p adr and @p fil:
     escape_in_url() both, then replace every '%' with 'X' so the result is one
     opaque token. */
-HTSEXT_API size_t make_content_id(const char *const adr, const char *const fil, char *const dest, const size_t size);
+HTSEXT_API size_t make_content_id(const char *const adr, const char *const fil,
+                                  char *const dest, const size_t size);
 
 /** Low-level percent-escaper backing the escape_* family. @p mode selects the
     character class to escape: 0 check_url, 1 in_url, 2 spc_url, 3 uri,
     30 uri_utf. @p max_size is the dest capacity including the NUL. */
-HTSEXT_API size_t x_escape_http(const char *const s, char *const dest, const size_t max_size, const int mode);
+HTSEXT_API size_t x_escape_http(const char *const s, char *const dest,
+                                const size_t max_size, const int mode);
 
 /** Strip all control characters (byte value < 32) from @p s in place. */
 HTSEXT_API void escape_remove_control(char *const s);
 
 /** HTML-escape for text output: rewrite '&' to "&amp;" and pass every other
    byte through unchanged. */
-HTSEXT_API size_t escape_for_html_print(const char *const s, char *const dest, const size_t size);
+HTSEXT_API size_t escape_for_html_print(const char *const s, char *const dest,
+                                        const size_t size);
 
 /** Like escape_for_html_print() but also convert every high byte (>= 128) to a
     numeric entity "&#xNN;". */
-HTSEXT_API size_t escape_for_html_print_full(const char *const s, char *const dest, const size_t size);
+HTSEXT_API size_t escape_for_html_print_full(const char *const s,
+                                             char *const dest,
+                                             const size_t size);
 
 /** Percent-decode @p s into @p catbuff (capacity @p size) and return @p
    catbuff. Decodes every "%xx" hex escape. */
-HTSEXT_API char *unescape_http(char *const catbuff, const size_t size, const char *const s);
+HTSEXT_API char *unescape_http(char *const catbuff, const size_t size,
+                               const char *const s);
 
 /** Percent-decode @p s into @p catbuff, but only the escapes that are safe to
     decode while keeping a valid URI (reserved, delimiter, unwise, control and
@@ -589,17 +614,16 @@ HTSEXT_API hts_boolean get_httptype_sized(httrackp *opt, char *s, size_t ssize,
     HTS_MIMETYPE_SIZE capacity. */
 HTS_DEPRECATED("use get_httptype_sized(opt, s, ssize, fil, flag)")
 
-HTSEXT_API void get_httptype(httrackp * opt, char *s, const char *fil,
-                             int flag);
+HTSEXT_API void get_httptype(httrackp *opt, char *s, const char *fil, int flag);
 
 /** Classify @p fil by its extension: 0 unknown, 1 known non-HTML, 2 known HTML.
     Consults the built-in table then user --assume rules. 0 for a NULL @p fil.
  */
-HTSEXT_API int is_knowntype(httrackp * opt, const char *fil);
+HTSEXT_API int is_knowntype(httrackp *opt, const char *fil);
 
 /** Like is_knowntype() but consults only the user --assume rules: 0 no rule,
     1 non-HTML, 2 HTML. */
-HTSEXT_API int is_userknowntype(httrackp * opt, const char *fil);
+HTSEXT_API int is_userknowntype(httrackp *opt, const char *fil);
 
 /** 1 if @p fil, an extension such as "asp" or "php" (not a full filename), is a
     known dynamic-page type, else 0. */
@@ -624,7 +648,7 @@ HTSEXT_API hts_boolean guess_httptype_sized(httrackp *opt, char *s,
     HTS_MIMETYPE_SIZE capacity. */
 HTS_DEPRECATED("use guess_httptype_sized(opt, s, ssize, fil)")
 
-HTSEXT_API void guess_httptype(httrackp * opt, char *s, const char *fil);
+HTSEXT_API void guess_httptype(httrackp *opt, char *s, const char *fil);
 
 /* Ugly string tools */
 /* These take a caller scratch buffer catbuff of capacity size and return it. On
@@ -633,11 +657,13 @@ HTSEXT_API void guess_httptype(httrackp * opt, char *s, const char *fil);
    time), not a pointer. */
 /** Concatenate @p a and @p b into @p catbuff (NULL or empty operands are
  * skipped). */
-HTSEXT_API char *concat(char *catbuff, size_t size, const char *a, const char *b);
+HTSEXT_API char *concat(char *catbuff, size_t size, const char *a,
+                        const char *b);
 
 /** Like concat(a, b) but convert '/' to the platform path separator (Windows).
  */
-HTSEXT_API char *fconcat(char *catbuff, size_t size, const char *a, const char *b);
+HTSEXT_API char *fconcat(char *catbuff, size_t size, const char *a,
+                         const char *b);
 
 /** Copy @p a into @p catbuff, converting '/' to the platform path separator
     (Windows). */
@@ -719,7 +745,7 @@ HTSEXT_API FILE *hts_fopen_utf8(const char *path, const char *mode);
 #define STAT hts_stat_utf8
 typedef struct _stat STRUCT_STAT;
 
-HTSEXT_API int hts_stat_utf8(const char *path, STRUCT_STAT * buf);
+HTSEXT_API int hts_stat_utf8(const char *path, STRUCT_STAT *buf);
 
 #define UNLINK hts_unlink_utf8
 HTSEXT_API int hts_unlink_utf8(const char *pathname);
@@ -731,12 +757,12 @@ HTSEXT_API int hts_rename_utf8(const char *oldpath, const char *newpath);
 
 HTSEXT_API int hts_mkdir_utf8(const char *pathname);
 
-#define UTIME(A,B) hts_utime_utf8(A,B)
+#define UTIME(A, B) hts_utime_utf8(A, B)
 
 typedef struct _utimbuf STRUCT_UTIMBUF;
 
 HTSEXT_API int hts_utime_utf8(const char *filename,
-                              const STRUCT_UTIMBUF * times);
+                              const STRUCT_UTIMBUF *times);
 #else
 #define FOPEN fopen
 #define STAT stat
@@ -748,7 +774,7 @@ typedef struct stat STRUCT_STAT;
 
 typedef struct utimbuf STRUCT_UTIMBUF;
 
-#define UTIME(A,B) utime(A,B)
+#define UTIME(A, B) utime(A, B)
 #endif
 #define HTS_DEF_FILEAPI
 #endif
@@ -756,20 +782,21 @@ typedef struct utimbuf STRUCT_UTIMBUF;
 /** Macro aimed to break at build-time if a size is not a sizeof() strictly
  *  greater than sizeof(char*). **/
 #undef COMPILE_TIME_CHECK_SIZE
-#define COMPILE_TIME_CHECK_SIZE(A) (void) ((void (*)(char[A - sizeof(char*) - 1])) NULL)
+#define COMPILE_TIME_CHECK_SIZE(A)                                             \
+  (void) ((void (*)(char[A - sizeof(char *) - 1])) NULL)
 
 /** Macro aimed to break at compile-time if a size is not a sizeof() strictly
  *  greater than sizeof(char*). **/
 #undef RUNTIME_TIME_CHECK_SIZE
-#define RUNTIME_TIME_CHECK_SIZE(A) assertf((A) != sizeof(void*))
+#define RUNTIME_TIME_CHECK_SIZE(A) assertf((A) != sizeof(void *))
 
-#define fconv(A,B,C) (COMPILE_TIME_CHECK_SIZE(B), fconv(A,B,C))
+#define fconv(A, B, C) (COMPILE_TIME_CHECK_SIZE(B), fconv(A, B, C))
 
-#define concat(A,B,C,D) (COMPILE_TIME_CHECK_SIZE(B), concat(A,B,C,D))
+#define concat(A, B, C, D) (COMPILE_TIME_CHECK_SIZE(B), concat(A, B, C, D))
 
-#define fconcat(A,B,C,D) (COMPILE_TIME_CHECK_SIZE(B), fconcat(A,B,C,D))
+#define fconcat(A, B, C, D) (COMPILE_TIME_CHECK_SIZE(B), fconcat(A, B, C, D))
 
-#define fslash(A,B,C) (COMPILE_TIME_CHECK_SIZE(B), fslash(A,B,C))
+#define fslash(A, B, C) (COMPILE_TIME_CHECK_SIZE(B), fslash(A, B, C))
 
 #ifdef __cplusplus
 }

src/webhttrack

@@ -4,131 +4,140 @@
 # Initializes the htsserver GUI frontend and launch the default browser
 
 BROWSEREXE=
-SRCHBROWSEREXE="x-www-browser www-browser iceape mozilla firefox-developer-edition firefox icecat iceweasel abrowser firebird galeon konqueror midori opera google-chrome chrome chromium chromium-browser netscape firefox-developer-edition"
+SRCHBROWSEREXE=(x-www-browser www-browser iceape mozilla firefox-developer-edition firefox icecat iceweasel abrowser firebird galeon konqueror midori opera google-chrome chrome chromium chromium-browser netscape firefox-developer-edition)
+# shellcheck disable=SC2153 # BROWSER is the standard freedesktop env var, not a typo
 if test -n "${BROWSER}"; then
-# sensible-browser will f up if BROWSER is not set
-SRCHBROWSEREXE="xdg-open sensible-browser ${SRCHBROWSEREXE}"
+    # sensible-browser will f up if BROWSER is not set
+    SRCHBROWSEREXE=(xdg-open sensible-browser "${SRCHBROWSEREXE[@]}")
 fi
 # Patch for Darwin/Mac by Ross Williams
-if test "`uname -s`" == "Darwin"; then
-# Darwin/Mac OS X uses a system 'open' command to find
-# the default browser. The -W flag causes it to wait for
-# the browser to exit
-BROWSEREXE="/usr/bin/open -W"
+if test "$(uname -s)" == "Darwin"; then
+    # Darwin/Mac OS X uses a system 'open' command to find
+    # the default browser. The -W flag causes it to wait for
+    # the browser to exit
+    BROWSEREXE="/usr/bin/open -W"
 fi
-BINWD=`dirname "$0"`
-SRCHPATH="$BINWD /usr/local/bin /usr/share/bin /usr/bin /usr/lib/httrack /usr/local/lib/httrack /usr/local/share/httrack /opt/local/bin /sw/bin ${HOME}/usr/bin ${HOME}/bin"
-SRCHPATH="$SRCHPATH "`echo $PATH | tr ":" " "`
-SRCHDISTPATH="$BINWD/../share $BINWD/.. /usr/share /usr/local /usr /local /usr/local/share ${HOME}/usr ${HOME}/usr/share /opt/local/share /sw ${HOME}/usr/local ${HOME}/usr/share"
+BINWD=$(dirname "$0")
+SRCHPATH=("$BINWD" /usr/local/bin /usr/share/bin /usr/bin /usr/lib/httrack /usr/local/lib/httrack /usr/local/share/httrack /opt/local/bin /sw/bin "${HOME}/usr/bin" "${HOME}/bin")
+IFS=':' read -ra pathdirs <<<"$PATH"
+for d in "${pathdirs[@]}"; do
+    # drop empty PATH fields, matching the old echo|tr word-split
+    test -n "$d" && SRCHPATH+=("$d")
+done
+SRCHDISTPATH=("$BINWD/../share" "$BINWD/.." /usr/share /usr/local /usr /local /usr/local/share "${HOME}/usr" "${HOME}/usr/share" /opt/local/share /sw "${HOME}/usr/local" "${HOME}/usr/share")
 
 ###
 # And now some famous cuisine
 
 function log {
-echo "$0($$): $@" >&2
-return 0
+    echo "$0($$): $*" >&2
+    return 0
 }
 
 function launch_browser {
-log "Launching $1"
-browser=$1
-url=$2
-log "Spawning browser.."
-${browser} "${url}"
-# note: browser can hiddenly use the -remote feature of
-# mozilla and therefore return immediately
-log "Browser (or helper) exited"
+    log "Launching $1"
+    browser=$1
+    url=$2
+    log "Spawning browser.."
+    ${browser} "${url}"
+    # note: browser can hiddenly use the -remote feature of
+    # mozilla and therefore return immediately
+    log "Browser (or helper) exited"
 }
 
 # First ensure that we can launch the server
 BINPATH=
-for i in ${SRCHPATH}; do
-	! test -n "${BINPATH}" && test -x ${i}/htsserver && BINPATH=${i}
+for i in "${SRCHPATH[@]}"; do
+    ! test -n "${BINPATH}" && test -x "${i}/htsserver" && BINPATH="${i}"
 done
-for i in ${SRCHDISTPATH}; do
-	! test -n "${DISTPATH}" && test -f "${i}/httrack/lang.def" && DISTPATH="${i}/httrack"
+for i in "${SRCHDISTPATH[@]}"; do
+    ! test -n "${DISTPATH}" && test -f "${i}/httrack/lang.def" && DISTPATH="${i}/httrack"
 done
 test -n "${BINPATH}" || ! log "Could not find htsserver" || exit 1
 test -n "${DISTPATH}" || ! log "Could not find httrack directory" || exit 1
-test -f ${DISTPATH}/lang.def || ! log "Could not find ${DISTPATH}/lang.def" || exit 1
-test -f ${DISTPATH}/lang.indexes || ! log "Could not find ${DISTPATH}/lang.indexes" || exit 1
-test -d ${DISTPATH}/lang || ! log "Could not find ${DISTPATH}/lang" || exit 1
-test -d ${DISTPATH}/html || ! log "Could not find ${DISTPATH}/html" || exit 1
+test -f "${DISTPATH}/lang.def" || ! log "Could not find ${DISTPATH}/lang.def" || exit 1
+test -f "${DISTPATH}/lang.indexes" || ! log "Could not find ${DISTPATH}/lang.indexes" || exit 1
+test -d "${DISTPATH}/lang" || ! log "Could not find ${DISTPATH}/lang" || exit 1
+test -d "${DISTPATH}/html" || ! log "Could not find ${DISTPATH}/html" || exit 1
 
 # Locale
 HTSLANG="${LC_MESSAGES}"
 ! test -n "${HTSLANG}" && HTSLANG="${LC_ALL}"
 ! test -n "${HTSLANG}" && HTSLANG="${LANG}"
-HTSLANG="`echo $LANG | cut -f1 -d'.' | cut -f1 -d'_'`"
-LANGN=`grep -E "^${HTSLANG}:" ${DISTPATH}/lang.indexes | cut -f2 -d':'`
+HTSLANG="$(echo "$LANG" | cut -f1 -d'.' | cut -f1 -d'_')"
+LANGN=$(grep -E "^${HTSLANG}:" "${DISTPATH}/lang.indexes" | cut -f2 -d':')
 ! test -n "${LANGN}" && LANGN=1
 
 # Find the browser
 # note: not all systems have sensible-browser or www-browser alternative
 # thefeore, we have to find a bit more if sensible-browser could not be found
 
-for i in ${SRCHBROWSEREXE}; do
-for j in ${SRCHPATH}; do
-if test -x ${j}/${i}; then
-BROWSEREXE=${j}/${i}
-fi
-test -n "$BROWSEREXE" && break
-done
-test -n "$BROWSEREXE" && break
+for i in "${SRCHBROWSEREXE[@]}"; do
+    for j in "${SRCHPATH[@]}"; do
+        if test -x "${j}/${i}"; then
+            BROWSEREXE="${j}/${i}"
+        fi
+        test -n "$BROWSEREXE" && break
+    done
+    test -n "$BROWSEREXE" && break
 done
 test -n "$BROWSEREXE" || ! log "Could not find any suitable browser" || exit 1
 
 # "browse" command
 if test "$1" = "browse"; then
-if test -f "${HOME}/.httrack.ini"; then
-INDEXF=`cat ${HOME}/.httrack.ini | tr '\r' '\n' | grep -E "^path=" | cut -f2- -d'='`
-if test -n "${INDEXF}" -a -d "${INDEXF}" -a -f "${INDEXF}/index.html"; then
-INDEXF="${INDEXF}/index.html"
-else
-INDEXF=""
-fi
-fi
-if ! test -n "$INDEXF"; then 
-INDEXF="${HOME}/websites/index.html"
-fi
-launch_browser "${BROWSEREXE}" "file://${INDEXF}"
-exit $?
+    if test -f "${HOME}/.httrack.ini"; then
+        INDEXF=$(tr '\r' '\n' <"${HOME}/.httrack.ini" | grep -E "^path=" | cut -f2- -d'=')
+        if test -n "${INDEXF}" -a -d "${INDEXF}" -a -f "${INDEXF}/index.html"; then
+            INDEXF="${INDEXF}/index.html"
+        else
+            INDEXF=""
+        fi
+    fi
+    if ! test -n "$INDEXF"; then
+        INDEXF="${HOME}/websites/index.html"
+    fi
+    launch_browser "${BROWSEREXE}" "file://${INDEXF}"
+    exit $?
 fi
 
 # Create a temporary filename
-TMPSRVFILE="$(mktemp ${TMPDIR:-/tmp}/.webhttrack.XXXXXXXX)" || ! log "Could not create the temporary file ${TMPSRVFILE}" || exit 1
+TMPSRVFILE="$(mktemp "${TMPDIR:-/tmp}/.webhttrack.XXXXXXXX")" || ! log "Could not create the temporary file ${TMPSRVFILE}" || exit 1
 # Launch htsserver binary and setup the server
-(${BINPATH}/htsserver "${DISTPATH}/" --ppid "$$" path "${HOME}/websites" lang "${LANGN}" $@; echo SRVURL=error) > ${TMPSRVFILE}&
+(
+    "${BINPATH}/htsserver" "${DISTPATH}/" --ppid "$$" path "${HOME}/websites" lang "${LANGN}" "$@"
+    echo SRVURL=error
+) >"${TMPSRVFILE}" &
 # Find the generated SRVURL
 SRVURL=
 MAXCOUNT=60
 while ! test -n "$SRVURL"; do
-MAXCOUNT=$[$MAXCOUNT - 1]
-test $MAXCOUNT -gt 0 || exit 1
-test $MAXCOUNT -lt 50 && echo "waiting for server to reply.."
-SRVURL=`grep -E URL= ${TMPSRVFILE} | cut -f2- -d=`
-test ! "$SRVURL" = "error" || ! log "Could not spawn htsserver" || exit 1
-test -n "$SRVURL" || sleep 1
+    MAXCOUNT=$((MAXCOUNT - 1))
+    test $MAXCOUNT -gt 0 || exit 1
+    test $MAXCOUNT -lt 50 && echo "waiting for server to reply.."
+    SRVURL=$(grep -E URL= "${TMPSRVFILE}" | cut -f2- -d=)
+    test ! "$SRVURL" = "error" || ! log "Could not spawn htsserver" || exit 1
+    test -n "$SRVURL" || sleep 1
 done
 
 # Cleanup function
+# shellcheck disable=SC2120 # $1 is an optional "signal caught" marker; bare calls are intentional
 function cleanup {
-test -n "$1" && log "Nasty signal caught, cleaning up.."
-# Do not kill if browser exited (chrome bug issue) ; server will die itself
-test -n "$1" && test -f ${TMPSRVFILE} && SRVPID=`grep -E PID= ${TMPSRVFILE} | cut -f2- -d=`
-test -n "${SRVPID}" && kill -9 ${SRVPID}
-test -f ${TMPSRVFILE} && rm ${TMPSRVFILE}
-test -n "$1" && log "..Done"
-return 0
+    test -n "$1" && log "Nasty signal caught, cleaning up.."
+    # Do not kill if browser exited (chrome bug issue) ; server will die itself
+    test -n "$1" && test -f "${TMPSRVFILE}" && SRVPID=$(grep -E PID= "${TMPSRVFILE}" | cut -f2- -d=)
+    test -n "${SRVPID}" && kill -9 "${SRVPID}"
+    test -f "${TMPSRVFILE}" && rm "${TMPSRVFILE}"
+    test -n "$1" && log "..Done"
+    return 0
 }
 
 # Cleanup in case of emergency
-trap "cleanup now; exit" 1 2 3 4 5 6 7 8 9 11 13 14 15 16 19 24 25
+trap "cleanup now; exit" HUP INT QUIT ILL TRAP ABRT BUS FPE SEGV PIPE ALRM TERM STKFLT XCPU XFSZ
 
 # Got SRVURL, launch browser
 launch_browser "${BROWSEREXE}" "${SRVURL}"
 
 # That's all, folks!
-trap "" 1 2 3 4 5 6 7 8 9 11 13 14 15 16 19 24 25
+trap "" HUP INT QUIT ILL TRAP ABRT BUS FPE SEGV PIPE ALRM TERM STKFLT XCPU XFSZ
 cleanup
 exit 0

tests/01_engine-charset.test

@@ -6,11 +6,11 @@ set -euo pipefail
 # charset -> UTF-8 conversion (hts_convertStringToUTF8).
 # -#3 <charset> <string> prints the string re-decoded from <charset> as UTF-8.
 conv() {
-	test "$(httrack -O /dev/null -#3 "$1" "$2")" == "$3" || exit 1
+    test "$(httrack -O /dev/null -#3 "$1" "$2")" == "$3" || exit 1
 }
 # crash probe: malformed input must exit cleanly, not abort.
 runs() {
-	httrack -O /dev/null -#3 "$1" "$2" >/dev/null 2>&1 || exit 1
+    httrack -O /dev/null -#3 "$1" "$2" >/dev/null 2>&1 || exit 1
 }
 
 # the source bytes below are UTF-8 (this file is UTF-8); "café" is 0x63 61 66 C3 A9.

tests/01_engine-entities.test

@@ -6,11 +6,11 @@ set -euo pipefail
 # HTML entity unescaping (hts_unescapeEntitiesWithCharset).
 # -#6 <string> prints the string with entities decoded (UTF-8 output).
 ent() {
-	test "$(httrack -O /dev/null -#6 "$1")" == "$2" || exit 1
+    test "$(httrack -O /dev/null -#6 "$1")" == "$2" || exit 1
 }
 # crash probe: malformed input must exit cleanly, not abort.
 runs() {
-	httrack -O /dev/null -#6 "$1" >/dev/null 2>&1 || exit 1
+    httrack -O /dev/null -#6 "$1" >/dev/null 2>&1 || exit 1
 }
 
 # named entities

tests/01_engine-filter.test

@@ -7,10 +7,10 @@ set -euo pipefail
 # -#0 <filter> <string> prints "<string> does match <filter>" or "... does NOT match ...".
 
 match() {
-	test "$(httrack -O /dev/null -#0 "$1" "$2")" == "$2 does match $1" || exit 1
+    test "$(httrack -O /dev/null -#0 "$1" "$2")" == "$2 does match $1" || exit 1
 }
 nomatch() {
-	test "$(httrack -O /dev/null -#0 "$1" "$2")" == "$2 does NOT match $1" || exit 1
+    test "$(httrack -O /dev/null -#0 "$1" "$2")" == "$2 does NOT match $1" || exit 1
 }
 
 # bare star matches everything
@@ -67,7 +67,7 @@ nomatch '*[\[]' 'a'
 # filter guide claims (GitHub #148); it parses as the class {'[','\'} followed
 # by a trailing literal ']'. These assertions document the current (buggy)
 # behavior so any future matcher fix is a deliberate, visible change.
-nomatch '*[\[\]]' '['   # not matched, despite the docs
-match '*[\[\]]' ']'     # only via the empty class-match + trailing ']'
-match '*[\[\]]' '[]'    # one of {'[','\'} then the trailing ']'
+nomatch '*[\[\]]' '[' # not matched, despite the docs
+match '*[\[\]]' ']'   # only via the empty class-match + trailing ']'
+match '*[\[\]]' '[]'  # one of {'[','\'} then the trailing ']'
 nomatch '*[\[\]]' '[]x'

tests/01_engine-mime.test

@@ -7,10 +7,10 @@ set -euo pipefail
 # -#2 <path> prints "<path> is '<mime>'" then "and its local type is '.<ext>'".
 
 mime() {
-	test "$(httrack -O /dev/null -#2 "$1" | head -1)" == "$1 is '$2'" || exit 1
+    test "$(httrack -O /dev/null -#2 "$1" | head -1)" == "$1 is '$2'" || exit 1
 }
 unknown() {
-	test "$(httrack -O /dev/null -#2 "$1" | head -1)" == "$1 is of an unknown MIME type" || exit 1
+    test "$(httrack -O /dev/null -#2 "$1" | head -1)" == "$1 is of an unknown MIME type" || exit 1
 }
 
 mime '/a/b.html' 'text/html'

tests/01_engine-parse.test

@@ -264,4 +264,63 @@ grep -Fq 'window.open("winopen.gif")' "$saved7" ||
 ! grep -Fq 'window.open("file://' "$saved7" ||
     ! echo "FAIL #218: window.open URL left absolute (not rewritten)" || exit 1
 
+# Parens in an unquoted url(...) (#163): the source %28/%29 decode to literal
+# '(' ')' in the saved name, but a literal ')' in the rewritten url() closes the
+# token early, so they must stay encoded. Negative control: without the fix the
+# %281%29 greps fail (parens are RFC2396 "mark" chars the escaper leaves alone).
+site8="$tmp/cssparens"
+mkdir -p "$site8"
+for f in 'img (1).gif' 'a(b)c(1).gif' 'q (4).gif'; do gif "$site8/$f"; done
+cat >"$site8/style.css" <<'EOF'
+.a { background: url(img%20%281%29.gif); }
+.b { background: url(a%28b%29c%281%29.gif); }
+.c { background: url("q%20%284%29.gif"); }
+EOF
+out8="$tmp/cssparens-out"
+crawl "$site8/style.css" "$out8"
+found "img (1).gif" "$out8"
+found "a(b)c(1).gif" "$out8"
+found "q (4).gif" "$out8"
+css8=$(find "$out8" -type f -path '*/file/*' -name style.css -print -quit)
+test -n "$css8" || ! echo "FAIL: saved style.css not found" || exit 1
+grep -Fq 'url(img%20%281%29.gif)' "$css8" ||
+    ! echo "FAIL #163: parens in unquoted url() not percent-encoded on rewrite" || exit 1
+grep -Fq 'url(a%28b%29c%281%29.gif)' "$css8" ||
+    ! echo "FAIL #163: not every paren in a url() was percent-encoded" || exit 1
+grep -Fq 'url("q%20%284%29.gif")' "$css8" ||
+    ! echo "FAIL #163: quoted url() altered or parens left literal on rewrite" || exit 1
+
+# The url() detector is not CSS-specific: <script> and inline style= get the
+# same encoding, but ordinary href/src (ending_p is the quote, not ')') keep
+# literal parens -- the attribute checks guard the gate against over-firing.
+site9="$tmp/urlparens"
+mkdir -p "$site9"
+for f in 'js (1).gif' 'inl (2).gif' 'asrc (3).gif' 'ahref (4).gif'; do gif "$site9/$f"; done
+cat >"$site9/index.html" <<EOF
+<html><body>
+<script>var bg = "url(js%20%281%29.gif)";</script>
+<div style="background-image:url(inl%20%282%29.gif)"></div>
+<img src="asrc%20%283%29.gif">
+<a href="ahref%20%284%29.gif">link</a>
+</body></html>
+EOF
+out9="$tmp/urlparens-out"
+crawl "$site9/index.html" "$out9"
+saved9=$(savedhtml "$out9")
+test -n "$saved9" || ! echo "FAIL: saved urlparens page not found" || exit 1
+# rewrite-only: the JS-string asset is not queued for download
+grep -Fq 'url(js%20%281%29.gif)' "$saved9" ||
+    ! echo "FAIL #163: parens in <script> url() not percent-encoded" || exit 1
+found "inl (2).gif" "$out9"
+grep -Fq 'url(inl%20%282%29.gif)' "$saved9" ||
+    ! echo "FAIL #163: parens in inline style url() not percent-encoded" || exit 1
+found "asrc (3).gif" "$out9"
+found "ahref (4).gif" "$out9"
+grep -Fq 'src="asrc%20(3).gif"' "$saved9" ||
+    ! echo "FAIL #163: parens in a plain src attribute were wrongly encoded" || exit 1
+grep -Fq 'href="ahref%20(4).gif"' "$saved9" ||
+    ! echo "FAIL #163: parens in a plain href attribute were wrongly encoded" || exit 1
+! grep -Eq '(src|href)="[^"]*%28' "$saved9" ||
+    ! echo "FAIL #163: gate over-fired onto a non-url() attribute link" || exit 1
+
 exit 0

tests/01_engine-relative.test

@@ -0,0 +1,68 @@
+#!/bin/bash
+#
+# lienrelatif (build relative path) + ident_url_relatif (resolve a link, collapse
+# ./ and ../). Regression net for #137/#162; expected values hand-computed.
+
+set -euo pipefail
+
+# relative path from <curr>'s directory to <link>
+rel() {
+    local got
+    got=$(httrack -O /dev/null -#l "$1" "$2")
+    test "$got" == "relative=$3" ||
+        {
+            echo "FAIL rel($1, $2): got '$got' want 'relative=$3'"
+            exit 1
+        }
+}
+
+# resolve <link> against origin <adr>/<fil> -> adr=.. fil=..
+ident() {
+    local got
+    got=$(httrack -O /dev/null -#i "$1" "$2" "$3")
+    test "$got" == "$4" ||
+        {
+            echo "FAIL ident($1, $2, $3): got '$got' want '$4'"
+            exit 1
+        }
+}
+
+### lienrelatif
+
+rel 'dir/page.html' 'dir/index.html' 'page.html'
+rel 'dir/page.html' 'dir/page.html' 'page.html' # self-link
+rel 'a.html' 'dir/index.html' '../a.html'
+rel 'x.html' 'a/b/c/index.html' '../../../x.html'
+rel 'h/a/x.jpg' 'h/a/sub/page.html' '../x.jpg'
+rel 'a/b/c/x.html' 'index.html' 'a/b/c/x.html'
+rel 'h/sub/x.jpg' 'h/page.html' 'sub/x.jpg'
+rel 'h/dir2/x.jpg' 'h/dir1/page.html' '../dir2/x.jpg' # sibling dir
+rel 'h/bc/x.jpg' 'h/b/page.html' '../bc/x.jpg'        # b/bc prefix trap
+rel 'h/b/x.jpg' 'h/bc/page.html' '../b/x.jpg'
+rel 'h2/img/x.jpg' 'h1/p/page.html' '../../h2/img/x.jpg' # cross-host
+rel 'img.cdn/photo.jpg' 'www.site/articles/2020/post.html' '../../../img.cdn/photo.jpg'
+rel 'h/a/' 'h/a/sub/page.html' '../' # link is ancestor dir
+rel 'x.html' 'page.html' 'x.html'
+rel 'dir/page.html?x=1' 'dir/index.html?y=2' 'page.html' # ? stripped
+
+### ident_url_relatif
+
+ident 'img.gif' 'www.foo.com' '/dir/page.html' 'adr=www.foo.com fil=/dir/img.gif'
+ident 'sub/img.gif' 'www.foo.com' '/dir/page.html' 'adr=www.foo.com fil=/dir/sub/img.gif'
+ident '/img.gif' 'www.foo.com' '/dir/page.html' 'adr=www.foo.com fil=/img.gif'
+# embedded ../ collapses (#137)
+ident '../img.gif' 'www.foo.com' '/dir/sub/page.html' 'adr=www.foo.com fil=/dir/img.gif'
+ident 'sub/../logo.png' 'www.foo.com' '/articles/2020/post.html' 'adr=www.foo.com fil=/articles/2020/logo.png'
+ident '../../pix/sub/../logo.png' 'www.foo.com' '/articles/2020/post.html' 'adr=www.foo.com fil=/pix/logo.png'
+ident '../../../../x.gif' 'www.foo.com' '/a/b/page.html' 'adr=www.foo.com fil=/x.gif' # above-root clamp
+ident '?page=2' 'www.foo.com' '/dir/index.html?old=1' 'adr=www.foo.com fil=/dir/index.html?page=2'
+ident 'http://other.com/a/b/../c/index.html' 'www.foo.com' '/p.html' 'adr=other.com fil=/a/c/index.html'
+# file:// collapses ../ like the other schemes; traversal contained, // authority kept
+ident 'file:///var/data/pix/sub/../logo.png' 'www.foo.com' '/p.html' 'adr=file:// fil=/var/data/pix/logo.png'
+ident 'file:///a/b/c/../../d/e.gif' 'www.foo.com' '/p.html' 'adr=file:// fil=/a/d/e.gif'
+ident 'file:///a/../../b' 'www.foo.com' '/p.html' 'adr=file:// fil=/b'
+ident 'file://srv/share/../x' 'www.foo.com' '/p.html' 'adr=file:// fil=//srv/x'
+ident 'mailto:foo@bar.com' 'www.foo.com' '/p.html' 'error=-1' # unsupported scheme
+ident 'javascript:void(0)' 'www.foo.com' '/p.html' 'error=-1'
+
+echo "OK"

tests/01_engine-simplify.test

@@ -5,7 +5,7 @@ set -euo pipefail
 
 # path simplify engine (fil_simplifie): collapses ./ and ../ segments.
 simp() {
-	test "$(httrack -O /dev/null -#1 "$1")" == "simplified=$2" || exit 1
+    test "$(httrack -O /dev/null -#1 "$1")" == "simplified=$2" || exit 1
 }
 
 simp './foo/bar/' 'foo/bar/'
@@ -26,3 +26,17 @@ simp './a/../../b' 'b'
 
 # empty segments ('//') are not dot-segments and are preserved, per RFC 3986
 simp 'a//b' 'a//b'
+simp 'a//b/../c' 'a//c'
+
+# absolute paths keep the leading '/'; above-root '..' is clamped to it
+simp '/a/../b' '/b'
+simp '/a/../../b' '/b'
+simp '/../x' '/x'
+
+# collapses to nothing -> './' (relative) or '/' (absolute)
+simp '..' './'
+simp 'a/..' './'
+simp '/' '/'
+
+simp 'a/b/..' 'a/'              # trailing bare '..'
+simp 'a/../b?x=../y' 'b?x=../y' # '?' freezes simplification

tests/01_engine-strsafe.test

@@ -21,9 +21,15 @@ test "$out" == "strsafe: OK" || exit 1
 # the bounded macro aborts (non-zero exit), so don't let set -e trip on it
 err=$(httrack -#8 overflow "this string is far too long for the buffer" 2>&1) || true
 case "$err" in
-	*"strsafe: NOT aborted"*) echo "over-capacity write was NOT caught" >&2; exit 1 ;;
-	*"overflow while copying"*) ;;
-	*) echo "expected htssafe overflow abort, got: $err" >&2; exit 1 ;;
+*"strsafe: NOT aborted"*)
+    echo "over-capacity write was NOT caught" >&2
+    exit 1
+    ;;
+*"overflow while copying"*) ;;
+*)
+    echo "expected htssafe overflow abort, got: $err" >&2
+    exit 1
+    ;;
 esac
 
 # Same guarantee for the htsbuff builder. The source is exactly the buffer
@@ -32,7 +38,13 @@ esac
 # aborted"). Match the specific htsbuff abort message, not just any assert.
 err=$(httrack -#8 overflow-buff "abcd" 2>&1) || true
 case "$err" in
-	*"strsafe: NOT aborted"*) echo "htsbuff over-capacity write was NOT caught" >&2; exit 1 ;;
-	*"htsbuff append overflow"*) ;;
-	*) echo "expected htsbuff overflow abort, got: $err" >&2; exit 1 ;;
+*"strsafe: NOT aborted"*)
+    echo "htsbuff over-capacity write was NOT caught" >&2
+    exit 1
+    ;;
+*"htsbuff append overflow"*) ;;
+*)
+    echo "expected htsbuff overflow abort, got: $err" >&2
+    exit 1
+    ;;
 esac

tests/10_crawl-simple.test

@@ -3,6 +3,6 @@
 
 set -euo pipefail
 
-bash check-network.sh ||  ! echo "skipping online unit tests" || exit 77
+bash check-network.sh || ! echo "skipping online unit tests" || exit 77
 
 bash crawl-test.sh --errors 0 --files 5 httrack http://ut.httrack.com/simple/basic.html

tests/11_crawl-cookies.test

@@ -3,10 +3,10 @@
 
 set -euo pipefail
 
-bash check-network.sh ||  ! echo "skipping online unit tests" || exit 77
+bash check-network.sh || ! echo "skipping online unit tests" || exit 77
 
 bash crawl-test.sh --errors 0 --files 3 \
-	--found ut.httrack.com/cookies/third.html \
-	--found ut.httrack.com/cookies/second.html \
-	--found ut.httrack.com/cookies/entrance.html \
-	httrack http://ut.httrack.com/cookies/entrance.php
+    --found ut.httrack.com/cookies/third.html \
+    --found ut.httrack.com/cookies/second.html \
+    --found ut.httrack.com/cookies/entrance.html \
+    httrack http://ut.httrack.com/cookies/entrance.php

tests/11_crawl-idna.test

@@ -3,21 +3,21 @@
 
 set -euo pipefail
 
-bash check-network.sh ||  ! echo "skipping online unit tests" || exit 77
+bash check-network.sh || ! echo "skipping online unit tests" || exit 77
 
 # unicode tests
 bash crawl-test.sh \
-	--errors 1 --files 5 \
-	--found 'café.ut.httrack.com/unicode-links/café3860.html' \
-	--found 'café.ut.httrack.com/unicode-links/café30f4.html' \
-	--found 'café.ut.httrack.com/unicode-links/café5e1f.html' \
-	--found 'café.ut.httrack.com/unicode-links/café7b30.html' \
-	httrack 'http://ut.httrack.com/unicode-links/idna.html' \
-	'+*.ut.httrack.com/*' --robots=0
+    --errors 1 --files 5 \
+    --found 'café.ut.httrack.com/unicode-links/café3860.html' \
+    --found 'café.ut.httrack.com/unicode-links/café30f4.html' \
+    --found 'café.ut.httrack.com/unicode-links/café5e1f.html' \
+    --found 'café.ut.httrack.com/unicode-links/café7b30.html' \
+    httrack 'http://ut.httrack.com/unicode-links/idna.html' \
+    '+*.ut.httrack.com/*' --robots=0
 
 # unicode tests (bogus links)
 bash crawl-test.sh \
-	--errors 0 --files 1 \
-	--found 'ut.httrack.com/unicode-links/idna_bogus.html' \
-	httrack 'http://ut.httrack.com/unicode-links/idna_bogus.html' \
-	'-*' --robots=0
+    --errors 0 --files 1 \
+    --found 'ut.httrack.com/unicode-links/idna_bogus.html' \
+    httrack 'http://ut.httrack.com/unicode-links/idna_bogus.html' \
+    '-*' --robots=0

tests/11_crawl-international.test

@@ -3,67 +3,67 @@
 
 set -euo pipefail
 
-bash check-network.sh ||  ! echo "skipping online unit tests" || exit 77
+bash check-network.sh || ! echo "skipping online unit tests" || exit 77
 
 # unicode tests
 bash crawl-test.sh \
-	--errors 1 --files 10 \
-	--found ut.httrack.com/unicode-links/caf%a91bce.html \
-	--found ut.httrack.com/unicode-links/café30f4.html \
-	--found ut.httrack.com/unicode-links/café3860.html \
-	--found ut.httrack.com/unicode-links/café463e.html \
-	--found ut.httrack.com/unicode-links/café5e1f.html \
-	--found ut.httrack.com/unicode-links/café7b30.html \
-	--found ut.httrack.com/unicode-links/café8007.html \
-	--found ut.httrack.com/unicode-links/café9fa8.html \
-	--found ut.httrack.com/unicode-links/caféae52.html \
-	--found ut.httrack.com/unicode-links/caféc009.html \
-	--found ut.httrack.com/unicode-links/utf8.html \
-	httrack http://ut.httrack.com/unicode-links/utf8.html
+    --errors 1 --files 10 \
+    --found ut.httrack.com/unicode-links/caf%a91bce.html \
+    --found ut.httrack.com/unicode-links/café30f4.html \
+    --found ut.httrack.com/unicode-links/café3860.html \
+    --found ut.httrack.com/unicode-links/café463e.html \
+    --found ut.httrack.com/unicode-links/café5e1f.html \
+    --found ut.httrack.com/unicode-links/café7b30.html \
+    --found ut.httrack.com/unicode-links/café8007.html \
+    --found ut.httrack.com/unicode-links/café9fa8.html \
+    --found ut.httrack.com/unicode-links/caféae52.html \
+    --found ut.httrack.com/unicode-links/caféc009.html \
+    --found ut.httrack.com/unicode-links/utf8.html \
+    httrack http://ut.httrack.com/unicode-links/utf8.html
 
 bash crawl-test.sh \
-	--errors 4 --files  7 \
-	--found ut.httrack.com/unicode-links/café3860.html \
-	--found ut.httrack.com/unicode-links/café9fa8.html \
-	--found ut.httrack.com/unicode-links/café30f4.html \
-	--found ut.httrack.com/unicode-links/café5e1f.html \
-	--found ut.httrack.com/unicode-links/café7b30.html \
-	--found ut.httrack.com/unicode-links/café8007.html \
-	--found ut.httrack.com/unicode-links/caf%e939bd.html \
-	--found ut.httrack.com/unicode-links/caf%e9ae52.html \
-	--found ut.httrack.com/unicode-links/caféaec2.html \
-	--found ut.httrack.com/unicode-links/caféfad6.html \
-	--found ut.httrack.com/unicode-links/default.html \
-	httrack http://ut.httrack.com/unicode-links/default.html
+    --errors 4 --files 7 \
+    --found ut.httrack.com/unicode-links/café3860.html \
+    --found ut.httrack.com/unicode-links/café9fa8.html \
+    --found ut.httrack.com/unicode-links/café30f4.html \
+    --found ut.httrack.com/unicode-links/café5e1f.html \
+    --found ut.httrack.com/unicode-links/café7b30.html \
+    --found ut.httrack.com/unicode-links/café8007.html \
+    --found ut.httrack.com/unicode-links/caf%e939bd.html \
+    --found ut.httrack.com/unicode-links/caf%e9ae52.html \
+    --found ut.httrack.com/unicode-links/caféaec2.html \
+    --found ut.httrack.com/unicode-links/caféfad6.html \
+    --found ut.httrack.com/unicode-links/default.html \
+    httrack http://ut.httrack.com/unicode-links/default.html
 
 bash crawl-test.sh \
-	--errors 2 --files  9 \
-	--found ut.httrack.com/unicode-links/caf%a9ae52.html \
-	--found ut.httrack.com/unicode-links/caf%a9bf59.html \
-	--found ut.httrack.com/unicode-links/café30f4.html \
-	--found ut.httrack.com/unicode-links/café3860.html \
-	--found ut.httrack.com/unicode-links/café5e1f.html \
-	--found ut.httrack.com/unicode-links/café647f.html \
-	--found ut.httrack.com/unicode-links/café7b30.html \
-	--found ut.httrack.com/unicode-links/café8007.html \
-	--found ut.httrack.com/unicode-links/caféaec2.html \
-	--found ut.httrack.com/unicode-links/caféfad6.html \
-	--found ut.httrack.com/unicode-links/iso88591.html \
-	httrack http://ut.httrack.com/unicode-links/iso88591.html
+    --errors 2 --files 9 \
+    --found ut.httrack.com/unicode-links/caf%a9ae52.html \
+    --found ut.httrack.com/unicode-links/caf%a9bf59.html \
+    --found ut.httrack.com/unicode-links/café30f4.html \
+    --found ut.httrack.com/unicode-links/café3860.html \
+    --found ut.httrack.com/unicode-links/café5e1f.html \
+    --found ut.httrack.com/unicode-links/café647f.html \
+    --found ut.httrack.com/unicode-links/café7b30.html \
+    --found ut.httrack.com/unicode-links/café8007.html \
+    --found ut.httrack.com/unicode-links/caféaec2.html \
+    --found ut.httrack.com/unicode-links/caféfad6.html \
+    --found ut.httrack.com/unicode-links/iso88591.html \
+    httrack http://ut.httrack.com/unicode-links/iso88591.html
 
 bash crawl-test.sh \
-	--errors 4 --files  9 \
-	--found ut.httrack.com/unicode-links/caf%a8%a6c72a.html \
-	--found ut.httrack.com/unicode-links/caf%a9bf59.html \
-	--found ut.httrack.com/unicode-links/café8007.html \
-	--found ut.httrack.com/unicode-links/cafébf43.html \
-	--found ut.httrack.com/unicode-links/cafédcd8.html \
-	--found ut.httrack.com/unicode-links/café2461.html \
-	--found ut.httrack.com/unicode-links/caf%a8%a61bce.html \
-	--found ut.httrack.com/unicode-links/caf%a9ae52.html \
-	--found ut.httrack.com/unicode-links/café7b30.html \
-	--found ut.httrack.com/unicode-links/café30f4.html \
-	--found ut.httrack.com/unicode-links/café5e1f.html \
-	--found ut.httrack.com/unicode-links/café3860.html \
-	--found ut.httrack.com/unicode-links/gb18030.html \
-	httrack http://ut.httrack.com/unicode-links/gb18030.html
+    --errors 4 --files 9 \
+    --found ut.httrack.com/unicode-links/caf%a8%a6c72a.html \
+    --found ut.httrack.com/unicode-links/caf%a9bf59.html \
+    --found ut.httrack.com/unicode-links/café8007.html \
+    --found ut.httrack.com/unicode-links/cafébf43.html \
+    --found ut.httrack.com/unicode-links/cafédcd8.html \
+    --found ut.httrack.com/unicode-links/café2461.html \
+    --found ut.httrack.com/unicode-links/caf%a8%a61bce.html \
+    --found ut.httrack.com/unicode-links/caf%a9ae52.html \
+    --found ut.httrack.com/unicode-links/café7b30.html \
+    --found ut.httrack.com/unicode-links/café30f4.html \
+    --found ut.httrack.com/unicode-links/café5e1f.html \
+    --found ut.httrack.com/unicode-links/café3860.html \
+    --found ut.httrack.com/unicode-links/gb18030.html \
+    httrack http://ut.httrack.com/unicode-links/gb18030.html

tests/11_crawl-longurl.test

@@ -3,10 +3,10 @@
 
 set -euo pipefail
 
-bash check-network.sh ||  ! echo "skipping online unit tests" || exit 77
+bash check-network.sh || ! echo "skipping online unit tests" || exit 77
 
 # http://code.google.com/p/httrack/issues/detail?id=42&can=1
 # we expect 2 errors only because other links are too longs (to be modified if suitable)
 bash crawl-test.sh --errors 2 --files 1 \
-	--found ut.httrack.com/overflow/longquerywithaccents.html \
-	httrack http://ut.httrack.com/overflow/longquerywithaccents.php
+    --found ut.httrack.com/overflow/longquerywithaccents.html \
+    httrack http://ut.httrack.com/overflow/longquerywithaccents.php

tests/11_crawl-parsing.test

@@ -3,45 +3,45 @@
 
 set -euo pipefail
 
-bash check-network.sh ||  ! echo "skipping online unit tests" || exit 77
+bash check-network.sh || ! echo "skipping online unit tests" || exit 77
 
 # http://code.google.com/p/httrack/issues/detail?id=4&can=1
 bash crawl-test.sh --errors 0 --files 4 \
-	--found ut.httrack.com/parsing/back5e1f.gif \
-	--found ut.httrack.com/parsing/events.html \
-	--found ut.httrack.com/parsing/fade230f4.gif \
-	--found ut.httrack.com/parsing/fade3860.gif \
-	httrack http://ut.httrack.com/parsing/events.html
+    --found ut.httrack.com/parsing/back5e1f.gif \
+    --found ut.httrack.com/parsing/events.html \
+    --found ut.httrack.com/parsing/fade230f4.gif \
+    --found ut.httrack.com/parsing/fade3860.gif \
+    httrack http://ut.httrack.com/parsing/events.html
 
 # http://code.google.com/p/httrack/issues/detail?id=2&can=1
 bash crawl-test.sh --errors 0 --files 3 \
-	--found ut.httrack.com/parsing/background-image.css \
-	--found ut.httrack.com/parsing/background-image.html \
-	--found ut.httrack.com/parsing/fade.gif \
-	httrack http://ut.httrack.com/parsing/background-image.html
+    --found ut.httrack.com/parsing/background-image.css \
+    --found ut.httrack.com/parsing/background-image.html \
+    --found ut.httrack.com/parsing/fade.gif \
+    httrack http://ut.httrack.com/parsing/background-image.html
 
 # javascript parsing
 bash crawl-test.sh --errors 0 --files 3 \
-	--found ut.httrack.com/parsing/back.gif \
-	--found ut.httrack.com/parsing/fade.gif \
-	--found ut.httrack.com/parsing/javascript.html \
-	httrack http://ut.httrack.com/parsing/javascript.html
+    --found ut.httrack.com/parsing/back.gif \
+    --found ut.httrack.com/parsing/fade.gif \
+    --found ut.httrack.com/parsing/javascript.html \
+    httrack http://ut.httrack.com/parsing/javascript.html
 
 # handling of + before query string
 bash crawl-test.sh --errors 0 --files 6 \
-	--found ut.httrack.com/parsing/escaping.html \
-	--found "ut.httrack.com/parsing/foo bar30f4.html" \
-	--found "ut.httrack.com/parsing/foo bar5e1f.html" \
-	--found "ut.httrack.com/parsing/foo+bar+plus3860.html" \
-	--found "ut.httrack.com/parsing/foo barae52.html" \
-	--found "ut.httrack.com/parsing/foo bar7b30.html" \
-	httrack http://ut.httrack.com/parsing/escaping.html
+    --found ut.httrack.com/parsing/escaping.html \
+    --found "ut.httrack.com/parsing/foo bar30f4.html" \
+    --found "ut.httrack.com/parsing/foo bar5e1f.html" \
+    --found "ut.httrack.com/parsing/foo+bar+plus3860.html" \
+    --found "ut.httrack.com/parsing/foo barae52.html" \
+    --found "ut.httrack.com/parsing/foo bar7b30.html" \
+    httrack http://ut.httrack.com/parsing/escaping.html
 
 # handling of # encoded in filename
 # see http://code.google.com/p/httrack/issues/detail?id=25
 bash crawl-test.sh --errors 2 --files 4 \
-	--found "ut.httrack.com/parsing/escaping2.html" \
-	--found "ut.httrack.com/parsing/++foo++bar++plus++.html" \
-	--found "ut.httrack.com/parsing/foo#bar#.html" \
-	--found "ut.httrack.com/parsing/foo bar.html" \
-	httrack http://ut.httrack.com/parsing/escaping2.html
+    --found "ut.httrack.com/parsing/escaping2.html" \
+    --found "ut.httrack.com/parsing/++foo++bar++plus++.html" \
+    --found "ut.httrack.com/parsing/foo#bar#.html" \
+    --found "ut.httrack.com/parsing/foo bar.html" \
+    httrack http://ut.httrack.com/parsing/escaping2.html

tests/12_crawl_https.test

@@ -3,11 +3,11 @@
 
 set -euo pipefail
 
-bash check-network.sh ||  ! echo "skipping online unit tests" || exit 77
+bash check-network.sh || ! echo "skipping online unit tests" || exit 77
 
 if test "${HTTPS_SUPPORT:-}" == "no"; then
-	echo "no https support compiled, skipping"
-	exit 77
+    echo "no https support compiled, skipping"
+    exit 77
 fi
 
 bash crawl-test.sh --errors 0 --files 5 httrack https://ut.httrack.com/simple/basic.html

tests/Makefile.am

@@ -35,6 +35,7 @@ TESTS = \
 	01_engine-mime.test \
 	01_engine-parse.test \
 	01_engine-rcfile.test \
+	01_engine-relative.test \
 	01_engine-simplify.test \
 	01_engine-strsafe.test \
 	02_manpage-regen.test \

tests/check-network.sh

@@ -6,39 +6,39 @@
 
 # do not enable online tests (./configure --disable-online-unit-tests)
 if test "$ONLINE_UNIT_TESTS" == "no"; then
-echo "online tests are disabled" >&2
-exit 1
+    echo "online tests are disabled" >&2
+    exit 1
 
 # enable online tests (--enable-online-unit-tests)
 elif test "$ONLINE_UNIT_TESTS" == "yes"; then
-exit 0
+    exit 0
 
 # check if online tests are reachable
 else
 
-# test url
-url=http://ut.httrack.com/enabled
+    # test url
+    url=http://ut.httrack.com/enabled
 
-# cache file name
-cache=check-network_sh.cache
+    # cache file name
+    cache=check-network_sh.cache
 
-# cached result ?
-if test -f $cache ; then
-	if grep -q "ok" $cache ; then
-		exit 0
-	else
-		echo "online tests are disabled (cached)" >&2
-		exit 1
-	fi
+    # cached result ?
+    if test -f $cache; then
+        if grep -q "ok" $cache; then
+            exit 0
+        else
+            echo "online tests are disabled (cached)" >&2
+            exit 1
+        fi
 
-# fetch single file
-elif bash crawl-test.sh --errors 0 --files 1 httrack --timeout=3 --max-time=3 "$url" 2>/dev/null >/dev/null ; then
-	echo "ok" > $cache
-	exit 0
-else
-	echo "error" > $cache
-	echo "online tests are disabled (auto)" >&2
-	exit 1
-fi
+    # fetch single file
+    elif bash crawl-test.sh --errors 0 --files 1 httrack --timeout=3 --max-time=3 "$url" 2>/dev/null >/dev/null; then
+        echo "ok" >$cache
+        exit 0
+    else
+        echo "error" >$cache
+        echo "online tests are disabled (auto)" >&2
+        exit 1
+    fi
 
 fi

tests/crawl-test.sh

@@ -2,185 +2,184 @@
 #
 
 function warning {
-  echo "** $*" >&2
-  return 0
+    echo "** $*" >&2
+    return 0
 }
 
 function die {
-  warning "$*"
-  exit 1
+    warning "$*"
+    exit 1
 }
 
 function debug {
-  if test -n "$verbose"; then
-    echo "$*" >&2
-  fi
+    if test -n "$verbose"; then
+        echo "$*" >&2
+    fi
 }
 
 function info {
-  printf "[$*] ..\t" >&2
+    printf '[%s] ..\t' "$*" >&2
 }
 
 function result {
-  echo "$*" >&2
+    echo "$*" >&2
 }
 
 function cleanup {
-  debug "cleaning function called"
-  if test -n "$tmpdir"; then
-    if test -d "$tmpdir"; then
-      if test -z "$nopurge"; then
-        debug "cleaning up $tmpdir"
-        rm -rf "$tmpdir"
-      fi
+    debug "cleaning function called"
+    if test -n "$tmpdir"; then
+        if test -d "$tmpdir"; then
+            if test -z "$nopurge"; then
+                debug "cleaning up $tmpdir"
+                rm -rf "$tmpdir"
+            fi
+        fi
+    fi
+    if test -n "$crawlpid"; then
+        debug "killing $crawlpid"
+        kill -9 "$crawlpid"
+        crawlpid=
     fi
-  fi
-  if test -n "$crawlpid"; then
-    debug "killing $crawlpid"
-    kill -9 "$crawlpid"
-    crawlpid=
-  fi
 }
 
 function usage {
-  cat << EOF
+    cat <<EOF
 usage: $0
 EOF
 }
 
 function assert_equals {
-  info "$1"
-  if test ! "$2" == "$3"; then
-    result "expected '$2', got '$3'"
-    exit 1
-  else
-    result "OK ($2)"
-  fi
+    info "$1"
+    if test ! "$2" == "$3"; then
+        result "expected '$2', got '$3'"
+        exit 1
+    else
+        result "OK ($2)"
+    fi
 }
 
 function start-crawl {
-  # parse args
-  pos=1
-  while test "$#" -ge "$pos" ; do
-    case "${!pos}" in
-    --debug)
-      verbose=1
-      ;;
-    --no-purge|--summary|--print-files)
-      ;;
-    --errors|--files|--found|--not-found|--directory)
-      pos=$[${pos}+1]
-      test "$#" -ge "$pos" || warning "missing argument" || return 1
-      ;;
-    httrack)
-      pos=$[${pos}+1]
-      break;
-      ;;
-    *)
-      warning "unrecognized option ${!pos}"
-      return 1
-      ;;
-    esac
-    pos=$[${pos}+1]
-  done
-  debug "remaining args: ${@:${pos}}"
+    # parse args
+    pos=1
+    while test "$#" -ge "$pos"; do
+        case "${!pos}" in
+        --debug)
+            verbose=1
+            ;;
+        --no-purge | --summary | --print-files) ;;
+        --errors | --files | --found | --not-found | --directory)
+            pos=$((pos + 1))
+            test "$#" -ge "$pos" || warning "missing argument" || return 1
+            ;;
+        httrack)
+            pos=$((pos + 1))
+            break
+            ;;
+        *)
+            warning "unrecognized option ${!pos}"
+            return 1
+            ;;
+        esac
+        pos=$((pos + 1))
+    done
+    debug "remaining args: ${*:pos}"
 
-  # ut/ won't exceed 2 minutes
-  moreargs="--quiet --max-time=120 --timeout=30 --connection-per-second=5"
+    # ut/ won't exceed 2 minutes
+    moreargs=(--quiet --max-time=120 --timeout=30 --connection-per-second=5)
 
-  # proxy environment ?
-  if test -n "$http_proxy"; then
-    moreargs="$moreargs --proxy $http_proxy"
-  fi
+    # proxy environment ?
+    if test -n "${http_proxy:-}"; then
+        moreargs+=(--proxy "$http_proxy")
+    fi
 
-  test -n "$tmpdir" || ! warning "no tmpdir" || return 1
-  tmp="${tmpdir}/crawl"
-  rm -rf "$tmp"
-  mkdir "$tmp" || ! warning "could not create $tmp" || return 1
-
-  which httrack >/dev/null || ! warning "could not find httrack" || return 1
-  ver=$(httrack -O /dev/null --version | sed -e 's/HTTrack version //')
-  test -n "$ver" || ! warning "could not run httrack" || return 1
-
-  # start crawl
-  log="${tmp}/log"
-  debug starting httrack -O "${tmp}" ${moreargs} ${@:${pos}}
-  info "running httrack ${@:${pos}}"
-  httrack -O "${tmp}" --user-agent="httrack $ver ut ($(uname -omrs))" ${moreargs} ${@:${pos}} >"${log}" 2>&1 &
-  crawlpid="$!"
-  debug "started cralwer on pid $crawlpid"
-  wait "$crawlpid"
-  result="$?"
-  crawlpid=
-  test "$result" -eq 0 || ! result "error code $result" || return 1
-  result "OK"
-  grep -iE "^[0-9\:]*[[:space:]]Error:" "${tmp}/hts-log.txt" >&2
-
-  # now audit
-  while test "$#" -gt 0; do
-    case "$1" in
-    --no-purge)
-      nopurge=1
-      ;;
-    --summary)
-      grep -E "^HTTrack Website Copier/[^ ]* mirror complete in " "${tmp}/hts-log.txt"
-      ;;
-    --print-files)
-      find "${tmp}" -mindepth 1 -type f
-      ;;
-    --errors)
-      shift
-      assert_equals "checking errors" "$1" "$(grep -iEc "^[0-9\:]*[[:space:]]Error:" "${tmp}/hts-log.txt")"
-      ;;
-    --found)
-      shift
-      info "checking for $1"
-      if test -f "${tmp}/$1" ; then
-        result "OK"
-      else
-        result "not found"
-        exit 1
-      fi
-      ;;
-    --not-found)
-      shift
-      info "checking for $1"
-      if test -f "${tmp}/$1" ; then
-        result "OK"
-      else
-        result "not found"
-        exit 1
-      fi
-      ;;
-    --directory)
-      shift
-      info "checking for $1"
-      if test -d "${tmp}/$1" ; then
-        result "OK"
-      else
-        result "not found"
-        exit 1
-      fi
-      ;;
-    --files)
-      shift
-      nFiles=$(grep -E "^HTTrack Website Copier/[^ ]* mirror complete in " "${tmp}/hts-log.txt" \
-        | sed -e 's/.*[[:space:]]\([^ ]*\)[[:space:]]files written.*/\1/g')
-      assert_equals "checking files" "$1" "$nFiles"
-      ;;
-    httrack)
-      break;
-      ;;
-    esac
-    shift
-  done
-
-  # cleanup
-  if test -z "$nopurge"; then
+    test -n "$tmpdir" || ! warning "no tmpdir" || return 1
+    tmp="${tmpdir}/crawl"
     rm -rf "$tmp"
-  else
-    tmpdir=
-  fi
+    mkdir "$tmp" || ! warning "could not create $tmp" || return 1
+
+    which httrack >/dev/null || ! warning "could not find httrack" || return 1
+    ver=$(httrack -O /dev/null --version | sed -e 's/HTTrack version //')
+    test -n "$ver" || ! warning "could not run httrack" || return 1
+
+    # start crawl
+    log="${tmp}/log"
+    debug starting httrack -O "${tmp}" "${moreargs[@]}" "${@:pos}"
+    info "running httrack ${*:pos}"
+    httrack -O "${tmp}" --user-agent="httrack $ver ut ($(uname -omrs))" "${moreargs[@]}" "${@:pos}" >"${log}" 2>&1 &
+    crawlpid="$!"
+    debug "started cralwer on pid $crawlpid"
+    wait "$crawlpid"
+    result="$?"
+    crawlpid=
+    test "$result" -eq 0 || ! result "error code $result" || return 1
+    result "OK"
+    grep -iE "^[0-9\:]*[[:space:]]Error:" "${tmp}/hts-log.txt" >&2
+
+    # now audit
+    while test "$#" -gt 0; do
+        case "$1" in
+        --no-purge)
+            nopurge=1
+            ;;
+        --summary)
+            grep -E "^HTTrack Website Copier/[^ ]* mirror complete in " "${tmp}/hts-log.txt"
+            ;;
+        --print-files)
+            find "${tmp}" -mindepth 1 -type f
+            ;;
+        --errors)
+            shift
+            assert_equals "checking errors" "$1" "$(grep -iEc "^[0-9\:]*[[:space:]]Error:" "${tmp}/hts-log.txt")"
+            ;;
+        --found)
+            shift
+            info "checking for $1"
+            if test -f "${tmp}/$1"; then
+                result "OK"
+            else
+                result "not found"
+                exit 1
+            fi
+            ;;
+        --not-found)
+            shift
+            info "checking for $1"
+            if test -f "${tmp}/$1"; then
+                result "OK"
+            else
+                result "not found"
+                exit 1
+            fi
+            ;;
+        --directory)
+            shift
+            info "checking for $1"
+            if test -d "${tmp}/$1"; then
+                result "OK"
+            else
+                result "not found"
+                exit 1
+            fi
+            ;;
+        --files)
+            shift
+            nFiles=$(grep -E "^HTTrack Website Copier/[^ ]* mirror complete in " "${tmp}/hts-log.txt" |
+                sed -e 's/.*[[:space:]]\([^ ]*\)[[:space:]]files written.*/\1/g')
+            assert_equals "checking files" "$1" "$nFiles"
+            ;;
+        httrack)
+            break
+            ;;
+        esac
+        shift
+    done
+
+    # cleanup
+    if test -z "$nopurge"; then
+        rm -rf "$tmp"
+    else
+        tmpdir=
+    fi
 }
 
 # check args
@@ -195,7 +194,7 @@ tmpdir=
 crawlpid=
 nopurge=
 verbose=
-trap "cleanup" 0 1 2 3 4 5 6 7 8 9 11 13 14 15 16 19 24 25
+trap cleanup EXIT HUP INT QUIT ILL TRAP ABRT BUS FPE SEGV PIPE ALRM TERM STKFLT XCPU XFSZ
 
 # working directory
 tmpdir="${tmptopdir}/httrack_ut.$$"

tests/run-all-tests.sh

@@ -2,19 +2,19 @@
 #
 
 error=0
-for i in *.test ; do
-	if bash $i ; then
-		echo "$i: passed" >&2
-	else
-		echo "$i: ERROR" >&2
-		error=$[${error}+1]
-	fi
+for i in *.test; do
+    if bash "$i"; then
+        echo "$i: passed" >&2
+    else
+        echo "$i: ERROR" >&2
+        error=$((error + 1))
+    fi
 done
 
 if test "$error" -eq 0; then
-	echo "all tests passed" >&2
+    echo "all tests passed" >&2
 else
-	echo "${error} test(s) failed" >&2
+    echo "${error} test(s) failed" >&2
 fi
 
 exit $error

tools/mk-sbuild-chroot.sh

@@ -0,0 +1,152 @@
+#!/usr/bin/env bash
+#
+# Bootstrap an sbuild chroot for the clean-room build gate (mkdeb.sh --sbuild).
+#
+# Uses the rootless unshare backend: no root, no schroot daemon. It builds a
+# minimal buildd chroot tarball into ~/.cache/sbuild/<dist>-<arch>.tar.zst, where
+# sbuild --dist=<dist> finds it automatically in unshare mode.
+#
+# Usage:
+#   tools/mk-sbuild-chroot.sh [options]
+#
+# Options:
+#   -d, --dist DIST       suite to bootstrap (default: unstable)
+#   -a, --arch ARCH       architecture (default: dpkg --print-architecture)
+#   -m, --mirror URL      apt mirror (default: http://deb.debian.org/debian)
+#       --components LIST  comma-separated components (default: main)
+#   -f, --force           rebuild even if the tarball already exists
+#       --write-sbuildrc  add "$chroot_mode = 'unshare';" to ~/.sbuildrc if absent
+#   -h, --help            show this help
+#
+# One-time setup; refresh later with sbuild-update or by rerunning with --force.
+# Requires mmdebstrap and the uidmap tools (newuidmap) for the unshare backend.
+
+set -euo pipefail
+
+readonly PROGNAME=${0##*/}
+
+die() {
+    printf '%s: error: %s\n' "$PROGNAME" "$*" >&2
+    exit 1
+}
+
+info() {
+    printf '==> %s\n' "$*" >&2
+}
+
+usage() {
+    sed -n '2,/^set -euo/{/^set -euo/!p}' "$0" | sed 's/^# \{0,1\}//'
+}
+
+need() {
+    local tool
+    for tool in "$@"; do
+        command -v "$tool" >/dev/null 2>&1 || die "required tool not found: $tool"
+    done
+}
+
+main() {
+    local dist=unstable
+    local arch=""
+    local mirror=http://deb.debian.org/debian
+    local components=main
+    local force=0
+    local write_sbuildrc=0
+
+    while [[ $# -gt 0 ]]; do
+        case $1 in
+        -d | --dist)
+            [[ $# -ge 2 ]] || die "missing argument for $1"
+            dist=$2
+            shift 2
+            ;;
+        -a | --arch)
+            [[ $# -ge 2 ]] || die "missing argument for $1"
+            arch=$2
+            shift 2
+            ;;
+        -m | --mirror)
+            [[ $# -ge 2 ]] || die "missing argument for $1"
+            mirror=$2
+            shift 2
+            ;;
+        --components)
+            [[ $# -ge 2 ]] || die "missing argument for $1"
+            components=$2
+            shift 2
+            ;;
+        -f | --force)
+            force=1
+            shift
+            ;;
+        --write-sbuildrc)
+            write_sbuildrc=1
+            shift
+            ;;
+        -h | --help)
+            usage
+            exit 0
+            ;;
+        *)
+            die "unknown option: $1 (try --help)"
+            ;;
+        esac
+    done
+
+    need mmdebstrap dpkg
+    # Unshare needs the setuid uid/gid mappers; mmdebstrap fails cryptically without.
+    command -v newuidmap >/dev/null 2>&1 ||
+        die "newuidmap not found; install the uidmap package for the unshare backend"
+
+    # Unshare maps a whole UID range, not just the caller's: the base install
+    # creates system users, and without an /etc/subuid+subgid range the install
+    # crashes (dpkg SIGSEGV) instead of erroring cleanly. Root uses mode=root and
+    # needs no range.
+    if [[ $(id -u) -ne 0 ]]; then
+        local me
+        me=$(id -un)
+        if ! grep -qs "^$me:" /etc/subuid || ! grep -qs "^$me:" /etc/subgid; then
+            # Suggest a range starting past every allocation in either file.
+            local start
+            start=$(awk -F: '{e = $2 + $3; if (e > m) m = e} END {print (m ? m : 100000)}' \
+                /etc/subuid /etc/subgid 2>/dev/null)
+            die "no /etc/subuid+subgid range for $me; the unshare backend needs one:
+  sudo usermod --add-subuids $start-$((start + 65535)) --add-subgids $start-$((start + 65535)) $me"
+        fi
+    fi
+
+    : "${arch:=$(dpkg --print-architecture)}"
+    local cache=$HOME/.cache/sbuild
+    local tarball=$cache/${dist}-${arch}.tar.zst
+
+    if [[ -e $tarball && $force -eq 0 ]]; then
+        info "chroot already exists: $tarball (use --force to rebuild)"
+    else
+        info "bootstrapping $dist/$arch chroot into $tarball"
+        mkdir -p "$cache"
+        mmdebstrap --variant=buildd --arch="$arch" --components="$components" \
+            "$dist" "$tarball" "$mirror"
+        info "chroot ready: $tarball"
+    fi
+
+    local rc=$HOME/.sbuildrc
+    local mode_line="\$chroot_mode = 'unshare';"
+    # shellcheck disable=SC2016  # $chroot_mode is literal regex text, not a shell var.
+    if grep -qsE '^[[:space:]]*\$chroot_mode[[:space:]]*=.*unshare' "$rc"; then
+        : # already configured (active, non-commented line)
+    elif [[ $write_sbuildrc -eq 1 ]]; then
+        info "enabling the unshare backend in $rc"
+        printf '%s\n' "$mode_line" >>"$rc"
+    else
+        cat >&2 <<EOF
+==> To use this chroot without passing --chroot-mode each time, add to $rc:
+      $mode_line
+    (or rerun with --write-sbuildrc). Then verify with:
+      sbuild --dist=$dist path/to/package.dsc
+    and build the release gate with:
+      tools/mkdeb.sh --source-only --sbuild
+EOF
+    fi
+}
+
+main "$@"

tools/mkdeb.sh

@@ -23,8 +23,17 @@
 #   -s, --source-only          build only the source package
 #   -u, --unsigned             do not sign anything (implies no release sigs)
 #       --no-release-artifacts skip the orig tarball .asc/.md5/.sha1
+#       --sbuild               additionally build the .dsc in a clean sbuild
+#                              chroot as a from-scratch verification gate
 #   -h, --help                 show this help
 #
+# --sbuild reproduces the buildd environment: it builds the source package in a
+# minimal chroot holding only the declared Build-Depends, so an FTBFS or a
+# missing dependency fails here instead of on the archive's buildds (which, with
+# a source-only upload, are otherwise the first clean build). It needs an sbuild
+# chroot for the changelog's distribution; create one once with the companion
+# tools/mk-sbuild-chroot.sh (rootless unshare backend).
+#
 # SOURCE_DATE_EPOCH is honored for reproducible output.
 
 set -euo pipefail
@@ -60,6 +69,7 @@ main() {
     local source_only=0
     local unsigned=0
     local release_artifacts=1
+    local sbuild=0
 
     while [[ $# -gt 0 ]]; do
         case $1 in
@@ -85,6 +95,10 @@ main() {
             release_artifacts=0
             shift
             ;;
+        --sbuild)
+            sbuild=1
+            shift
+            ;;
         -h | --help)
             usage
             exit 0
@@ -96,6 +110,7 @@ main() {
     done
 
     need git autoreconf debuild dcmd
+    [[ $sbuild -eq 1 ]] && need sbuild dpkg-parsechangelog
     if [[ $unsigned -eq 0 ]]; then
         need gpg
         [[ -n $key ]] || die "no signing key (pass --key or set DEBSIGN_KEYID, or use --unsigned)"
@@ -179,6 +194,30 @@ main() {
     [[ ${#changes[@]} -ge 1 ]] || die "debuild produced no .changes file"
     dcmd cp -- "${changes[@]}" "$outdir/"
 
+    # Clean-room build gate: rebuild the source package in a minimal chroot that
+    # holds only the declared Build-Depends, the same way the buildds will. An
+    # undeclared dependency or any FTBFS aborts the release here instead of
+    # surfacing after a source-only upload. Logs and clean-built debs land in
+    # $outdir/sbuild for inspection.
+    if [[ $sbuild -eq 1 ]]; then
+        local -a dscs
+        shopt -s nullglob
+        dscs=("$scratch"/*.dsc)
+        shopt -u nullglob
+        [[ ${#dscs[@]} -ge 1 ]] || die "no .dsc to sbuild"
+
+        local dist
+        dist=$(cd "$scratch/httrack-$ver" && dpkg-parsechangelog -S Distribution)
+        [[ $dist == UNRELEASED ]] && dist=unstable
+
+        info "clean-room build with sbuild (dist $dist)"
+        local sbdir=$outdir/sbuild
+        rm -rf -- "$sbdir"
+        mkdir -p "$sbdir"
+        (cd "$sbdir" && sbuild --dist="$dist" -- "${dscs[0]}")
+        info "sbuild clean-room build passed; logs in $sbdir"
+    fi
+
     # Release artifacts for the upstream tarball (detached sig + checksums).
     if [[ $release_artifacts -eq 1 && $unsigned -eq 0 ]]; then
         info "signing upstream tarball"