feat(file): name html as a file type, without a renderer for it

CHANGELOG.md

@@ -22,6 +22,10 @@ The release run heads these entries with the version and opens a fresh
 - The zip backend is `miniz/3.1.1`, up from `3.0.2`. Rendered output is
   unchanged.
 
+- New `FileType::hypertext_markup_language` for `html`, `htm`, `xhtml`,
+  `text/html` and `application/xhtml+xml`. Classification only: no `open`, no
+  `translate_html`, and never detected from its bytes.
+
 ## v6.11.0 - 2026-08-29
 
 - No view declares `<base target="_blank">` any more. A link back into what

README.md

@@ -42,6 +42,7 @@ opening one throws:
 - wpd (WordPerfect)
 - xlsb (Excel binary workbook — an OOXML package whose workbook parts are
   binary rather than spreadsheetml)
+- html / htm / xhtml (route to a web view; not detected from its bytes)
 
 ## Asking what is supported
 

apple/include/OdrCoreObjC/ODRFile.h

@@ -87,6 +87,8 @@ typedef NS_ENUM(NSInteger, ODRFileType) {
   ODRFileTypeIworkPages,
   ODRFileTypeIworkNumbers,
   ODRFileTypeIworkKeynote,
+
+  ODRFileTypeHypertextMarkupLanguage,
 } NS_SWIFT_NAME(FileType);
 
 typedef NS_ENUM(NSInteger, ODRFileCategory) {

apple/src/ODRFile.mm

@@ -95,6 +95,8 @@
 ODR_SAME_ENUM(ODRFileTypeIworkPages, odr::FileType::iwork_pages);
 ODR_SAME_ENUM(ODRFileTypeIworkNumbers, odr::FileType::iwork_numbers);
 ODR_SAME_ENUM(ODRFileTypeIworkKeynote, odr::FileType::iwork_keynote);
+ODR_SAME_ENUM(ODRFileTypeHypertextMarkupLanguage,
+              odr::FileType::hypertext_markup_language);
 
 ODR_SAME_ENUM(ODRFileCategoryUnknown, odr::FileCategory::unknown);
 ODR_SAME_ENUM(ODRFileCategoryText, odr::FileCategory::text);

jni/java/app/opendocument/core/FileType.java

@@ -17,7 +17,7 @@ public enum FileType {
   QUICKTIME_VIDEO, THIRD_GENERATION_PARTNERSHIP_VIDEO, MATROSKA_VIDEO,
   AUDIO_VIDEO_INTERLEAVE, SCALABLE_VECTOR_GRAPHICS, WINDOWS_ICON, JPEG_XL,
   JPEG_2000, PHOTOSHOP_DOCUMENT, WINDOWS_METAFILE, ENHANCED_METAFILE, XML,
-  IWORK_PAGES, IWORK_NUMBERS, IWORK_KEYNOTE;
+  IWORK_PAGES, IWORK_NUMBERS, IWORK_KEYNOTE, HYPERTEXT_MARKUP_LANGUAGE;
 
   static FileType fromNative(int code) {
     return code < 0 ? null : values()[code];

python/src/bind_file.cpp

@@ -86,7 +86,9 @@ void odr_python::bind_file(py::module_ &m) {
       .value("xml", odr::FileType::xml)
       .value("iwork_pages", odr::FileType::iwork_pages)
       .value("iwork_numbers", odr::FileType::iwork_numbers)
-      .value("iwork_keynote", odr::FileType::iwork_keynote);
+      .value("iwork_keynote", odr::FileType::iwork_keynote)
+      .value("hypertext_markup_language",
+             odr::FileType::hypertext_markup_language);
 
   py::enum_<odr::FileCategory>(m, "FileCategory")
       .value("unknown", odr::FileCategory::unknown)

src/odr/file.hpp

@@ -162,6 +162,10 @@ enum class FileType {
   // and no fixture pins those two, so nothing detects or decodes them yet.
   iwork_numbers,
   iwork_keynote,
+
+  // Classification only - never detected from its bytes.
+  // https://en.wikipedia.org/wiki/HTML
+  hypertext_markup_language,
 };
 
 /// @brief Collection of file categories.

src/odr/internal/file_type_table.cpp

@@ -262,6 +262,9 @@ constexpr std::array keynote_mimetypes{
     "application/x-iwork-keynote-sffkey"sv,
 };
 
+constexpr std::array html_extensions{"html"sv, "htm"sv, "xhtml"sv};
+constexpr std::array html_mimetypes{"text/html"sv, "application/xhtml+xml"sv};
+
 // The single source of truth behind every public format lookup; `odr_test`
 // asserts one row per `FileType` and capabilities that match the engines.
 //
@@ -798,6 +801,16 @@ constexpr std::array table{
          .open = true,
          .translate_html = true,
          .color_scheme = true}},
+
+    // Named but not decoded. Not `detect_by_content` - html has no dependable
+    // signature, so the caller routes on the file name.
+    Row{FileType::hypertext_markup_language,
+        "html"sv,
+        html_extensions,
+        html_mimetypes,
+        FileCategory::text,
+        DocumentType::unknown,
+        {}},
 };
 
 /// Finds the row whose list, selected by @p list, contains @p needle.

test/src/odr_test.cpp

@@ -27,7 +27,8 @@ namespace {
 std::vector<FileType> every_file_type() {
   std::vector<FileType> result;
   for (auto i = static_cast<std::size_t>(FileType::unknown);
-       i <= static_cast<std::size_t>(FileType::iwork_keynote); ++i) {
+       i <= static_cast<std::size_t>(FileType::hypertext_markup_language);
+       ++i) {
     result.push_back(static_cast<FileType>(i));
   }
   return result;
@@ -129,6 +130,28 @@ TEST(FileTypeTable, canonical_alias_is_the_first_one) {
   }
 }
 
+/// Without the row an `.html` decodes as xml and shows its own source.
+TEST(FileTypeTable, html_is_named_but_not_decoded) {
+  const FileType html = FileType::hypertext_markup_language;
+
+  EXPECT_EQ(file_type_by_file_extension("html"), html);
+  EXPECT_EQ(file_type_by_file_extension("htm"), html);
+  EXPECT_EQ(file_type_by_file_extension("xhtml"), html);
+  EXPECT_EQ(file_type_by_mimetype("text/html"), html);
+  EXPECT_EQ(mimetype_by_file_type(html), "text/html");
+  EXPECT_EQ(file_category_by_file_type(html), FileCategory::text);
+
+  const FileTypeCapabilities capabilities = capabilities_by_file_type(html);
+  EXPECT_FALSE(capabilities.detect_by_content);
+  EXPECT_FALSE(capabilities.open);
+  EXPECT_FALSE(capabilities.translate_html);
+
+  const std::string page = "<!DOCTYPE html><html><body><p>hi</p></body></html>";
+  EXPECT_THROW(std::ignore =
+                   open(File::from_memory(page), html, Logger::null()),
+               UnknownFileType);
+}
+
 /// `FileType::unknown` is the only type we refuse to name a MIME type for.
 TEST(FileTypeTable, only_unknown_has_no_mimetype) {
   for (const FileType type : every_file_type()) {