tests: tighten the type-matrix guards

Add two assertions surfaced by review of the override path: control.php
must not survive its rename to control.html (a dual-write regression
would leave both), and gen.php?id=5 (a query/extension-less URL served
image/png) must keep its .png and not be mangled to .html. Both exercise
the "override still fires" direction that the suppression cases don't.

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

src/htsname.c

@@ -138,6 +138,30 @@ static void cleanEndingSpaceOrDot(char *s) {
   }
 }
 
+/* Should the wire Content-Type override the URL's own extension when naming the
+   saved file? True only when the type is patchable (may_unknown2) and doing so
+   would not clobber a URL extension that already maps to a specific, non-HTML
+   type. This is the #267 mangle guard: a .png served as text/html (or with no
+   type) stays named .png. */
+static int wire_patches_ext(httrackp *opt, const char *wiremime,
+                            const char *file) {
+  char urlmime[256];
+
+  if (may_unknown2(opt, wiremime, file))
+    return 0; /* type kept verbatim (keep-list / bogus-multiple) */
+  urlmime[0] = '\0';
+  /* type implied by the URL extension, only when confidently known (flag 0) */
+  if (!get_httptype_sized(opt, urlmime, sizeof(urlmime), file, 0))
+    return 1; /* URL ext implies no known type: trust the wire type */
+  if (strfield2(wiremime, urlmime))
+    return 0; /* wire agrees with the ext: keep it (no .htm->.html churn) */
+  /* wire disagrees: keep a specific non-HTML ext against an html/empty claim */
+  if (!is_hypertext_mime(opt, urlmime, file) &&
+      (is_html_mime_type(wiremime) || !strnotempty(wiremime)))
+    return 0;
+  return 1;
+}
+
 // forme le nom du fichier à sauver (save) à partir de fil et adr
 // système intelligent, qui renomme en cas de besoin (exemple: deux INDEX.HTML et index.html)
 int url_savename(lien_adrfilsave *const afs,
@@ -325,7 +349,10 @@ int url_savename(lien_adrfilsave *const afs,
   }
 
   /* replace shtml to html.. */
-  if (opt->savename_delayed == HTS_SAVENAME_DELAYED_HARD)
+  /* HARD delays every type, except one the user pinned with --assume: honor it
+     immediately (ishtml() consults the user type), no delayed name (#56) */
+  if (opt->savename_delayed == HTS_SAVENAME_DELAYED_HARD &&
+      !is_userknowntype(opt, fil))
     is_html = -1;               /* ALWAYS delay type */
   else
     is_html = ishtml(opt, fil);
@@ -380,7 +407,7 @@ int url_savename(lien_adrfilsave *const afs,
               if (strnotempty(r.cdispo)) {        /* filename given */
                 ext_chg = 2;      /* change filename */
                 strcpybuff(ext, r.cdispo);
-              } else if (!may_unknown2(opt, r.contenttype, fil)) {        // on peut patcher à priori?
+              } else if (wire_patches_ext(opt, r.contenttype, fil)) {
                 if (give_mimext(s, sizeof(s),
                                 r.contenttype)) { // recognized extension
                   ext_chg = 1;
@@ -425,7 +452,8 @@ int url_savename(lien_adrfilsave *const afs,
               if (strnotempty(headers->r.cdispo)) {        /* filename given */
                 ext_chg = 2;      /* change filename */
                 strcpybuff(ext, headers->r.cdispo);
-              } else if (!may_unknown2(opt, headers->r.contenttype, headers->url_fil)) {    // on peut patcher à priori? (pas interdit ou pas de type)
+              } else if (wire_patches_ext(opt, headers->r.contenttype,
+                                          headers->url_fil)) {
                 char s[16];
                 if (give_mimext(
                         s, sizeof(s),
@@ -653,7 +681,8 @@ int url_savename(lien_adrfilsave *const afs,
                     if (strnotempty(back[b].r.cdispo)) {        /* filename given */
                       ext_chg = 2;      /* change filename */
                       strcpybuff(ext, back[b].r.cdispo);
-                    } else if (!may_unknown2(opt, back[b].r.contenttype, back[b].url_fil)) {    // on peut patcher à priori? (pas interdit ou pas de type)
+                    } else if (wire_patches_ext(opt, back[b].r.contenttype,
+                                                back[b].url_fil)) {
                       if (give_mimext(
                               s, sizeof(s),
                               back[b].r.contenttype)) { // recognized extension

tests/15_local-types.test

@@ -0,0 +1,20 @@
+#!/bin/bash
+#
+# Content-Type vs URL-extension naming (issue #267 family). Under the default
+# delayed type check (-%N2), a bogus/missing html-ish wire type must not clobber
+# a URL extension that maps to a specific non-HTML type. The .html "mangle" names
+# are asserted absent so a regression that re-introduces it fails here.
+
+: "${top_srcdir:=..}"
+
+bash "$top_srcdir/tests/local-crawl.sh" --errors 0 \
+    --found 'types/notype.png' --not-found 'types/notype.html' \
+    --found 'types/lie.png' --not-found 'types/lie.html' \
+    --found 'types/page.htm' --not-found 'types/page.html' \
+    --found 'types/photo.png' \
+    --found 'types/script.js' \
+    --found 'types/style.css' \
+    --found 'types/data.json' \
+    --found 'types/control.html' --not-found 'types/control.php' \
+    --found 'types/gend61c.png' --not-found 'types/gend61c.html' \
+    httrack 'BASEURL/types/index.html'

tests/16_local-assume.test

@@ -0,0 +1,11 @@
+#!/bin/bash
+#
+# --assume under the default delayed type check (-%N2), issue #56. A user type
+# pinned with --assume must be honored immediately, not lost to the delayed
+# name: photo.png served as image/png but assumed text/html is saved as .html.
+
+: "${top_srcdir:=..}"
+
+bash "$top_srcdir/tests/local-crawl.sh" --errors 0 \
+    --found 'types/photo.html' --not-found 'types/photo.png' \
+    httrack 'BASEURL/types/photo.png' --assume png=text/html

tests/Makefile.am

@@ -51,6 +51,8 @@ TESTS = \
 	12_crawl_https.test \
 	13_crawl_proxy_https.test \
 	13_local-cookies.test \
-	14_local-https.test
+	14_local-https.test \
+	15_local-types.test \
+	16_local-assume.test
 
 CLEANFILES = check-network_sh.cache

tests/local-server.py

@@ -118,11 +118,69 @@ class Handler(SimpleHTTPRequestHandler):
         if self.command != "HEAD":
             self.wfile.write(body)
 
+    # --- type/extension matrix (issue #267 family) -------------------------
+
+    def send_raw(self, body, content_type):
+        """Send a raw body with an explicit Content-Type, or none at all when
+        content_type is None (to observe httrack's typeless-file naming)."""
+        self.send_response(200)
+        if content_type is not None:
+            self.send_header("Content-Type", content_type)
+        self.send_header("Content-Length", str(len(body)))
+        self.end_headers()
+        if self.command != "HEAD":
+            self.wfile.write(body)
+
+    # A fake-binary PNG-ish blob for the image/typeless cases.
+    FAKE_PNG = b"\x89PNG\r\n\x1a\n" + b"\x00" * 64
+
+    # path -> (body, content_type); content_type None means no header at all.
+    TYPE_MATRIX = {
+        "/types/control.php": (b"<html><body>control</body></html>", "text/html"),
+        "/types/photo.png": (FAKE_PNG, "image/png"),
+        "/types/notype.png": (FAKE_PNG, None),
+        "/types/lie.png": (FAKE_PNG, "text/html"),
+        "/types/page.htm": (b"<html><body>htm page</body></html>", "text/html"),
+        "/types/script.js": (b"var x = 1;\n", "application/javascript"),
+        "/types/style.css": (b"body { color: red; }\n", "text/css"),
+        "/types/data.json": (b'{"k": "v"}\n', "application/json"),
+        "/types/gen.php": (FAKE_PNG, "image/png"),
+    }
+
+    def route_types_index(self):
+        body = (
+            '\t<a href="control.php">control</a>\n'
+            '\t<img src="photo.png" />\n'
+            '\t<img src="notype.png" />\n'
+            '\t<img src="lie.png" />\n'
+            '\t<a href="page.htm">htm</a>\n'
+            '\t<script src="script.js"></script>\n'
+            '\t<link rel="stylesheet" href="style.css" />\n'
+            '\t<a href="data.json">json</a>\n'
+            '\t<img src="gen.php?id=5" />\n'
+        )
+        self.send_html(body)
+
+    def route_types(self):
+        path = urlsplit(self.path).path
+        body, ctype = self.TYPE_MATRIX[path]
+        self.send_raw(body, ctype)
+
     ROUTES = {
         "/cookies/entrance.php": route_entrance,
         "/cookies/second.php": route_second,
         "/cookies/third.php": route_third,
         "/robots.txt": route_robots,
+        "/types/index.html": route_types_index,
+        "/types/control.php": route_types,
+        "/types/photo.png": route_types,
+        "/types/notype.png": route_types,
+        "/types/lie.png": route_types,
+        "/types/page.htm": route_types,
+        "/types/script.js": route_types,
+        "/types/style.css": route_types,
+        "/types/data.json": route_types,
+        "/types/gen.php": route_types,
     }
 
     # --- dispatch ----------------------------------------------------------