update mupdf

mupdf/Makefile

@@ -42,6 +42,9 @@ VERSION_PATCH = $(shell grep "define FZ_VERSION_PATCH" include/mupdf/fitz/versio
 
 ifeq ($(LINUX_OR_OPENBSD),yes)
   SO_VERSION = .$(VERSION_MINOR).$(VERSION_PATCH)
+  ifeq ($(OS),Linux)
+    SO_VERSION_LINUX := yes
+  endif
 endif
 
 # --- Commands ---
@@ -93,9 +96,11 @@ $(OUT)/%.exe: %.c
 	$(LINK_CMD)
 
 $(OUT)/%.$(SO)$(SO_VERSION):
-	$(LINK_CMD) $(LIB_LDFLAGS) $(THIRD_LIBS) $(LIBCRYPTO_LIBS)
-ifneq ($(SO_VERSION),)
+ifeq ($(SO_VERSION_LINUX),yes)
+	$(LINK_CMD) -Wl,-soname,$(notdir $@) $(LIB_LDFLAGS) $(THIRD_LIBS) $(LIBCRYPTO_LIBS)
 	ln -sf $(notdir $@) $(patsubst %$(SO_VERSION), %, $@)
+else
+	$(LINK_CMD) $(LIB_LDFLAGS) $(THIRD_LIBS) $(LIBCRYPTO_LIBS)
 endif
 
 $(OUT)/%.def: $(OUT)/%.$(SO)$(SO_VERSION)
@@ -631,14 +636,14 @@ install-shared-c: install-shared-check shared install-headers
 	install -d $(DESTDIR)$(libdir)
 	install -m 644 $(OUT)/libmupdf.$(SO)$(SO_VERSION) $(DESTDIR)$(libdir)/
 ifneq ($(OS),OpenBSD)
-	ln -s libmupdf.$(SO)$(SO_VERSION) $(DESTDIR)$(libdir)/libmupdf.$(SO)
+	ln -sf libmupdf.$(SO)$(SO_VERSION) $(DESTDIR)$(libdir)/libmupdf.$(SO)
 endif
 
 install-shared-c++: install-shared-c c++
 	install -m 644 platform/c++/include/mupdf/*.h $(DESTDIR)$(incdir)/mupdf
 	install -m 644 $(OUT)/libmupdfcpp.$(SO)$(SO_VERSION) $(DESTDIR)$(libdir)/
 ifneq ($(OS),OpenBSD)
-	ln -s libmupdfcpp.$(SO)$(SO_VERSION) $(DESTDIR)$(libdir)/libmupdfcpp.$(SO)
+	ln -sf libmupdfcpp.$(SO)$(SO_VERSION) $(DESTDIR)$(libdir)/libmupdfcpp.$(SO)
 endif
 
 install-shared-python: install-shared-c++ python

mupdf/docs/src/language-bindings.rst

@@ -1547,6 +1547,28 @@ functions and class methods.]
     */
     FZ_FUNCTION std::string fz_get_glyph_name2(fz_context *ctx, fz_font *font, int glyph);
 
+    /**
+    Extra struct containing fz_install_load_system_font_funcs()'s args,
+    which we wrap with virtual_fnptrs set to allow use from Python/C# via
+    Swig Directors.
+    */
+    typedef struct fz_install_load_system_font_funcs_args
+    {{
+        fz_load_system_font_fn *f;
+        fz_load_system_cjk_font_fn *f_cjk;
+        fz_load_system_fallback_font_fn *f_fallback;
+    }} fz_install_load_system_font_funcs_args;
+
+    /**
+    Alternative to fz_install_load_system_font_funcs() that takes args in a
+    struct, to allow use from Python/C# via Swig Directors.
+    */
+    void fz_install_load_system_font_funcs2(fz_context *ctx, fz_install_load_system_font_funcs_args* args);
+
+    /* Internal singleton state to allow Swig Director class to find
+    fz_install_load_system_font_funcs_args class wrapper instance. */
+    extern void* fz_install_load_system_font_funcs2_state;
+
 
 Python/C# bindings details
 ---------------------------------------------------------------
@@ -1640,6 +1662,22 @@ Non-standard API or implementation
 * `pdf_lookup_metadata(pdfdocument, key)`: Return key value or None if not found:
 * `pdf_set_annot_color()`: Takes single `color` arg which must be float or tuple of 1-4 floats.
 * `pdf_set_annot_interior_color()`: Takes single `color` arg which must be float or tuple of 1-4 floats.
+* `fz_install_load_system_font_funcs()`: Takes Python callbacks with no `ctx` arg,
+  which can return `None`, `fz_font*` or a `mupdf.FzFont`.
+
+  Example usage (from `scripts/mupdfwrap_test.py:test_install_load_system_font()`)::
+
+    def font_f(name, bold, italic, needs_exact_metrics):
+        print(f'font_f(): Looking for font: {name=} {bold=} {italic=} {needs_exact_metrics=}.')
+        return mupdf.fz_new_font_from_file(...)
+    def f_cjk(name, ordering, serif):
+        print(f'f_cjk(): Looking for font: {name=} {ordering=} {serif=}.')
+        return None
+    def f_fallback(script, language, serif, bold, italic):
+        print(f'f_fallback(): looking for font: {script=} {language=} {serif=} {bold=} {italic=}.')
+        return None
+    mupdf.fz_install_load_system_font_funcs(font_f, f_cjk, f_fallback)
+
 
 Making MuPDF function pointers call Python code
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

mupdf/include/mupdf/fitz/archive.h

@@ -210,13 +210,6 @@ int fz_is_tar_archive(fz_context *ctx, fz_stream *file);
 */
 int fz_is_libarchive_archive(fz_context *ctx, fz_stream *file);
 
-/**
-	Detect if stream object is a chm archive.
-
-	Assumes that the stream object is seekable.
-*/
-int fz_is_chm_archive(fz_context *ctx, fz_stream *file);
-
 /**
 	Detect if stream object is a cfb archive.
 
@@ -267,26 +260,6 @@ fz_archive *fz_open_libarchive_archive(fz_context *ctx, const char *filename);
 */
 fz_archive *fz_open_libarchive_archive_with_stream(fz_context *ctx, fz_stream *file);
 
-/**
-	Open a chm file as an archive.
-
-	An exception is thrown if the file is not recognised as a chm.
-
-	filename: a path to an archive file as it would be given to
-	open(2).
-*/
-fz_archive *fz_open_chm_archive(fz_context *ctx, const char *filename);
-
-/**
-	Open a chm file as an archive.
-
-	Open an archive using a seekable stream object rather than
-	opening a file or directory on disk.
-
-	An exception is thrown if the file is not recognised as a chm.
-*/
-fz_archive *fz_open_chm_archive_with_stream(fz_context *ctx, fz_stream *file);
-
 /**
 	Open a cfb file as an archive.
 
@@ -440,8 +413,6 @@ fz_archive_handler;
 
 FZ_DATA extern const fz_archive_handler fz_libarchive_archive_handler;
 
-FZ_DATA extern const fz_archive_handler fz_chm_archive_handler;
-
 void fz_register_archive_handler(fz_context *ctx, const fz_archive_handler *handler);
 
 /**

mupdf/include/mupdf/fitz/document.h

@@ -313,51 +313,23 @@ typedef fz_link *(fz_page_create_link_fn)(fz_context *ctx, fz_page *page, fz_rec
 */
 typedef void (fz_page_delete_link_fn)(fz_context *ctx, fz_page *page, fz_link *link);
 
-/**
-	Function type to open a document from a file.
-
-	filename: file to open
-
-	Pointer to opened document. Throws exception in case of error.
-*/
-typedef fz_document *(fz_document_open_fn)(fz_context *ctx, const char *filename);
-
 /**
 	Function type to open a
 	document from a file.
 
 	stream: fz_stream to read document data from. Must be
 	seekable for formats that require it.
 
-	Pointer to opened document. Throws exception in case of error.
-*/
-typedef fz_document *(fz_document_open_with_stream_fn)(fz_context *ctx, fz_stream *stream);
-
-/**
-	Function type to open a document from a
-	file, with accelerator data.
-
-	filename: file to open
-
-	accel: accelerator file
-
-	Pointer to opened document. Throws exception in case of error.
-*/
-typedef fz_document *(fz_document_open_accel_fn)(fz_context *ctx, const char *filename, const char *accel);
-
-/**
-	Function type to open a document from a file,
-	with accelerator data.
-
-	stream: fz_stream to read document data from. Must be
-	seekable for formats that require it.
+	accel: fz_stream to read accelerator data from. May be
+	NULL. May be ignored.
 
-	accel: fz_stream to read accelerator data from. Must be
-	seekable for formats that require it.
+	dir: 'Directory context' in which the document is loaded;
+	associated content from (like images for an html stream
+	will be loaded from this). Maybe NULL. May be ignored.
 
 	Pointer to opened document. Throws exception in case of error.
 */
-typedef fz_document *(fz_document_open_accel_with_stream_fn)(fz_context *ctx, fz_stream *stream, fz_stream *accel);
+typedef fz_document *(fz_document_open_fn)(fz_context *ctx, fz_stream *stream, fz_stream *accel, fz_archive *dir);
 
 /**
 	Recognize a document type from
@@ -375,13 +347,16 @@ typedef int (fz_document_recognize_fn)(fz_context *ctx, const char *magic);
 /**
 	Recognize a document type from stream contents.
 
-	stream: stream contents to recognise.
+	stream: stream contents to recognise (may be NULL if document is
+	a directory).
+
+	dir: directory context from which stream is loaded.
 
 	Returns a number between 0 (not recognized) and 100
 	(fully recognized) based on how certain the recognizer
 	is that this is of the required type.
 */
-typedef int (fz_document_recognize_content_fn)(fz_context *ctx, fz_stream *stream);
+typedef int (fz_document_recognize_content_fn)(fz_context *ctx, fz_stream *stream, fz_archive *dir);
 
 /**
 	Type for a function to be called when processing an already opened page.
@@ -425,13 +400,29 @@ const fz_document_handler *fz_recognize_document_content(fz_context *ctx, const
 	Given a magic find a document handler that can handle a
 	document of this type.
 
-	stream: the file stream to sample.
+	stream: the file stream to sample. May be NULL if the document is
+	a directory.
 
 	magic: Can be a filename extension (including initial period) or
 	a mimetype.
 */
 const fz_document_handler *fz_recognize_document_stream_content(fz_context *ctx, fz_stream *stream, const char *magic);
 
+/**
+	Given a magic find a document handler that can handle a
+	document of this type.
+
+	stream: the file stream to sample. May be NULL if the document is
+	a directory.
+
+	dir: an fz_archive representing the directory from which the
+	stream was opened (or NULL).
+
+	magic: Can be a filename extension (including initial period) or
+	a mimetype.
+*/
+const fz_document_handler *fz_recognize_document_stream_and_dir_content(fz_context *ctx, fz_stream *stream, fz_archive *dir, const char *magic);
+
 /**
 	Open a document file and read its basic structure so pages and
 	objects can be located. MuPDF will try to repair broken
@@ -462,9 +453,30 @@ fz_document *fz_open_accelerated_document(fz_context *ctx, const char *filename,
 
 	magic: a string used to detect document type; either a file name
 	or mime-type.
+
+	stream: a stream representing the contents of the document file.
+
+	NOTE: The caller retains ownership of 'stream' - the document will take its
+	own reference if required.
 */
 fz_document *fz_open_document_with_stream(fz_context *ctx, const char *magic, fz_stream *stream);
 
+/**
+	Open a document using the specified stream object rather than
+	opening a file on disk.
+
+	magic: a string used to detect document type; either a file name
+	or mime-type.
+
+	stream: a stream representing the contents of the document file.
+
+	dir: a 'directory context' for those filetypes that need it.
+
+	NOTE: The caller retains ownership of 'stream' and 'dir' - the document will
+	take its own references if required.
+*/
+fz_document *fz_open_document_with_stream_and_dir(fz_context *ctx, const char *magic, fz_stream *stream, fz_archive *dir);
+
 /**
 	Open a document using a buffer rather than opening a file on disk.
 */
@@ -476,9 +488,34 @@ fz_document *fz_open_document_with_buffer(fz_context *ctx, const char *magic, fz
 
 	magic: a string used to detect document type; either a file name
 	or mime-type.
+
+	stream: a stream of the document contents.
+
+	accel: NULL, or a stream of the 'accelerator' contents for this document.
+
+	NOTE: The caller retains ownership of 'stream' and 'accel' - the document will
+	take its own references if required.
 */
 fz_document *fz_open_accelerated_document_with_stream(fz_context *ctx, const char *magic, fz_stream *stream, fz_stream *accel);
 
+/**
+	Open a document using the specified stream object rather than
+	opening a file on disk.
+
+	magic: a string used to detect document type; either a file name
+	or mime-type.
+
+	stream: a stream of the document contents.
+
+	accel: NULL, or a stream of the 'accelerator' contents for this document.
+
+	dir: NULL, or the 'directory context' for the stream contents.
+
+	NOTE: The caller retains ownership of 'stream', 'accel' and 'dir' - the document will
+	take its own references if required.
+*/
+fz_document *fz_open_accelerated_document_with_stream_and_dir(fz_context *ctx, const char *magic, fz_stream *stream, fz_stream *accel, fz_archive *dir);
+
 /**
 	Query if the document supports the saving of accelerator data.
 */
@@ -994,11 +1031,8 @@ struct fz_document_handler
 {
 	fz_document_recognize_fn *recognize;
 	fz_document_open_fn *open;
-	fz_document_open_with_stream_fn *open_with_stream;
 	const char **extensions;
 	const char **mimetypes;
-	fz_document_open_accel_fn *open_accel;
-	fz_document_open_accel_with_stream_fn *open_accel_with_stream;
 	fz_document_recognize_content_fn *recognize_content;
 };
 

mupdf/include/mupdf/fitz/font.h

@@ -62,6 +62,44 @@ int fz_unicode_from_glyph_name_strict(const char *name);
 const char **fz_duplicate_glyph_names_from_unicode(int unicode);
 const char *fz_glyph_name_from_unicode_sc(int unicode);
 
+/**
+ * A text decoder (to read arbitrary encodings and convert to unicode).
+ */
+typedef struct fz_text_decoder fz_text_decoder;
+
+struct fz_text_decoder {
+	// get maximum size estimate of converted text (fast)
+	int (*decode_bound)(fz_text_decoder *dec, unsigned char *input, int n);
+
+	// get exact size of converted text (slow)
+	int (*decode_size)(fz_text_decoder *dec, unsigned char *input, int n);
+
+	// convert text into output buffer
+	void (*decode)(fz_text_decoder *dec, char *output, unsigned char *input, int n);
+
+	// for internal use only; do not touch!
+	void *table1;
+	void *table2;
+};
+
+/* Initialize a text decoder using an IANA encoding name.
+ * See source/fitz/text-decoder.c for the exact list of supported encodings.
+ * Will throw an exception if the requested encoding is not available.
+ *
+ * The following is a subset of the supported encodings (see source/fitz/text-decoder.c for the full list):
+ *   iso-8859-1
+ *   iso-8859-7
+ *   koi8-r
+ *   euc-jp
+ *   shift_jis
+ *   euc-kr
+ *   euc-cn
+ *   gb18030
+ *   euc-tw
+ *   big5
+ */
+void fz_init_text_decoder(fz_context *ctx, fz_text_decoder *dec, const char *encoding);
+
 /**
 	An abstract font handle.
 */