update mupdf

mupdf/docs/src/language-bindings.rst

@@ -181,7 +181,7 @@ Extensions beyond the basic C API
 * There are various custom class methods and constructors.
 
 * There are extra functions for generating a text representation of 'POD'
-  structs and their C++ wrapper classes.
+  (plain old data) structs and their C++ wrapper classes.
 
   For example for `fz_rect` we provide these functions:
 
@@ -890,6 +890,12 @@ Building and installing the Python bindings using `pip`
   available from pypi.org so pip will fail to install prerequisites from
   `pypackage.toml`.
 
+  Instead one can run `setup.py` directly:
+
+  .. code-block:: shell
+
+      cd mupdf && setup.py install
+
 
 Building the Python bindings
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

mupdf/include/mupdf/fitz/archive.h

@@ -88,6 +88,9 @@ fz_archive *fz_open_directory(fz_context *ctx, const char *path);
 
 /**
 	Determine if a given path is a directory.
+
+	In the case of the path not existing, or having no access
+	we will return 0.
 */
 int fz_is_directory(fz_context *ctx, const char *path);
 
@@ -214,6 +217,13 @@ int fz_is_libarchive_archive(fz_context *ctx, fz_stream *file);
 */
 int fz_is_chm_archive(fz_context *ctx, fz_stream *file);
 
+/**
+	Detect if stream object is a cfb archive.
+
+	Assumes that the stream object is seekable.
+*/
+int fz_is_cfb_archive(fz_context *ctx, fz_stream *file);
+
 /**
 	Open a tar archive file.
 
@@ -277,6 +287,26 @@ fz_archive *fz_open_chm_archive(fz_context *ctx, const char *filename);
 */
 fz_archive *fz_open_chm_archive_with_stream(fz_context *ctx, fz_stream *file);
 
+/**
+	Open a cfb file as an archive.
+
+	An exception is thrown if the file is not recognised as a cfb.
+
+	filename: a path to an archive file as it would be given to
+	open(2).
+*/
+fz_archive *fz_open_cfb_archive(fz_context *ctx, const char *filename);
+
+/**
+	Open a cfb 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_cfb_archive_with_stream(fz_context *ctx, fz_stream *file);
+
 /**
 	fz_archive: zip implementation
 */

mupdf/include/mupdf/html.h

@@ -0,0 +1,59 @@
+// Copyright (C) 2023 Artifex Software, Inc.
+//
+// This file is part of MuPDF.
+//
+// MuPDF is free software: you can redistribute it and/or modify it under the
+// terms of the GNU Affero General Public License as published by the Free
+// Software Foundation, either version 3 of the License, or (at your option)
+// any later version.
+//
+// MuPDF is distributed in the hope that it will be useful, but WITHOUT ANY
+// WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+// FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more
+// details.
+//
+// You should have received a copy of the GNU Affero General Public License
+// along with MuPDF. If not, see <https://www.gnu.org/licenses/agpl-3.0.en.html>
+//
+// Alternative licensing terms are available from the licensor.
+// For commercial licensing, see <https://www.artifex.com/> or contact
+// Artifex Software, Inc., 39 Mesa Street, Suite 108A, San Francisco,
+// CA 94129, USA, for further information.
+
+// This header allows people to easily build HTML-based document handlers.
+
+#ifndef MUPDF_HTML_HTML_H
+#define MUPDF_HTML_HTML_H
+
+#include "mupdf/fitz/system.h"
+#include "mupdf/fitz/context.h"
+#include "mupdf/fitz/document.h"
+
+/*
+	HTML types required
+*/
+typedef struct fz_html_s fz_html;
+typedef struct fz_html_font_set_s fz_html_font_set;
+
+typedef struct
+{
+	const char *format_name;
+	fz_buffer *(*convert_to_html)(fz_context *ctx, fz_html_font_set *set, fz_buffer *buf, const char *user_css);
+	int try_xml;
+	int try_html5;
+	int patch_mobi;
+} fz_htdoc_format_t;
+
+fz_document *fz_htdoc_open_document_with_buffer(fz_context *ctx, fz_archive *zip, fz_buffer *buf, const fz_htdoc_format_t *format);
+
+fz_document *fz_htdoc_open_document_with_file_and_dir(fz_context *ctx, const char *dirname, const char *filename, const fz_htdoc_format_t *format);
+
+fz_document *fz_htdoc_open_document_with_file(fz_context *ctx, const char *filename, const fz_htdoc_format_t *format);
+
+fz_document *fz_htdoc_open_document_with_stream_and_dir(fz_context *ctx, const char *dirname, fz_stream *stm, const fz_htdoc_format_t *format);
+
+fz_document *fz_htdoc_open_document_with_stream(fz_context *ctx, fz_stream *file, const fz_htdoc_format_t *format);
+
+
+
+#endif /* MUPDF_HTML_HTML_H */

mupdf/platform/wasm/examples/mutool.js

@@ -17,8 +17,9 @@
 //
 // Alternative licensing terms are available from the licensor.
 // For commercial licensing, see <https://www.artifex.com/> or contact
-// Artifex Software, Inc., 1305 Grant Avenue - Suite 200, Novato,
-// CA 94945, U.S.A., +1(415)492-9861, for further information.
+// Artifex Software, Inc., 39 Mesa Street, Suite 108A, San Francisco,
+// CA 94129, USA, for further information.
+
 
 "use strict"
 

mupdf/platform/wasm/lib/mupdf.c

@@ -17,8 +17,8 @@
 //
 // Alternative licensing terms are available from the licensor.
 // For commercial licensing, see <https://www.artifex.com/> or contact
-// Artifex Software, Inc., 1305 Grant Avenue - Suite 200, Novato,
-// CA 94945, U.S.A., +1(415)492-9861, for further information.
+// Artifex Software, Inc., 39 Mesa Street, Suite 108A, San Francisco,
+// CA 94129, USA, for further information.
 
 // TODO: Story
 // TODO: DOM

mupdf/platform/wasm/lib/mupdf.js

@@ -17,8 +17,8 @@
 //
 // Alternative licensing terms are available from the licensor.
 // For commercial licensing, see <https://www.artifex.com/> or contact
-// Artifex Software, Inc., 1305 Grant Avenue - Suite 200, Novato,
-// CA 94945, U.S.A., +1(415)492-9861, for further information.
+// Artifex Software, Inc., 39 Mesa Street, Suite 108A, San Francisco,
+// CA 94129, USA, for further information.
 
 "use strict"
 

mupdf/platform/wasm/viewer/mupdf-view-page.js

@@ -17,8 +17,8 @@
 //
 // Alternative licensing terms are available from the licensor.
 // For commercial licensing, see <https://www.artifex.com/> or contact
-// Artifex Software, Inc., 1305 Grant Avenue - Suite 200, Novato,
-// CA 94945, U.S.A., +1(415)492-9861, for further information.
+// Artifex Software, Inc., 39 Mesa Street, Suite 108A, San Francisco,
+// CA 94129, USA, for further information.
 
 /* eslint-disable no-unused-vars */
 

mupdf/platform/wasm/viewer/mupdf-view-worker.js

@@ -17,8 +17,8 @@
 //
 // Alternative licensing terms are available from the licensor.
 // For commercial licensing, see <https://www.artifex.com/> or contact
-// Artifex Software, Inc., 1305 Grant Avenue - Suite 200, Novato,
-// CA 94945, U.S.A., +1(415)492-9861, for further information.
+// Artifex Software, Inc., 39 Mesa Street, Suite 108A, San Francisco,
+// CA 94129, USA, for further information.
 
 /* global mupdf */
 

mupdf/platform/wasm/viewer/mupdf-view.js

@@ -17,8 +17,8 @@
 //
 // Alternative licensing terms are available from the licensor.
 // For commercial licensing, see <https://www.artifex.com/> or contact
-// Artifex Software, Inc., 1305 Grant Avenue - Suite 200, Novato,
-// CA 94945, U.S.A., +1(415)492-9861, for further information.
+// Artifex Software, Inc., 39 Mesa Street, Suite 108A, San Francisco,
+// CA 94129, USA, for further information.
 
 "use strict"
 

mupdf/platform/win32/libmupdf.vcxproj

@@ -580,6 +580,7 @@
     <ClCompile Include="..\..\source\fitz\pool.c" />
     <ClCompile Include="..\..\source\fitz\printf.c" />
     <ClCompile Include="..\..\source\fitz\random.c" />
+    <ClCompile Include="..\..\source\fitz\uncfb.c" />
     <ClCompile Include="..\..\source\fitz\unchm.c" />
     <ClCompile Include="..\..\source\fitz\unlibarchive.c" />
     <ClCompile Include="..\..\source\fitz\warp.c" />
@@ -620,6 +621,7 @@
     <ClCompile Include="..\..\source\html\mobi.c" />
     <ClCompile Include="..\..\source\html\office.c" />
     <ClCompile Include="..\..\source\html\story-writer.c" />
+    <ClCompile Include="..\..\source\html\txt.c" />
     <ClCompile Include="..\..\source\html\xml-dom.c" />
     <ClCompile Include="..\..\source\pdf\pdf-annot.c" />
     <ClCompile Include="..\..\source\pdf\pdf-appearance.c" />
@@ -740,6 +742,7 @@
     <ClInclude Include="..\..\include\mupdf\fitz\write-pixmap.h" />
     <ClInclude Include="..\..\include\mupdf\fitz\writer.h" />
     <ClInclude Include="..\..\include\mupdf\fitz\xml.h" />
+    <ClInclude Include="..\..\include\mupdf\html.h" />
     <ClInclude Include="..\..\include\mupdf\memento.h" />
     <ClInclude Include="..\..\include\mupdf\pdf.h" />
     <ClInclude Include="..\..\include\mupdf\pdf\annot.h" />
@@ -809,4 +812,4 @@
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
   <ImportGroup Label="ExtensionTargets">
   </ImportGroup>
-</Project>
+</Project>

mupdf/platform/win32/libmupdf.vcxproj.filters

@@ -657,6 +657,12 @@
     <ClCompile Include="..\..\source\fitz\heap.c">
       <Filter>fitz</Filter>
     </ClCompile>
+    <ClCompile Include="..\..\source\html\txt.c">
+      <Filter>html</Filter>
+    </ClCompile>
+    <ClCompile Include="..\..\source\fitz\uncfb.c">
+      <Filter>fitz</Filter>
+    </ClCompile>
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="..\..\source\fitz\draw-imp.h">
@@ -929,5 +935,8 @@
     <ClInclude Include="..\..\include\mupdf\fitz\heap-imp.h">
       <Filter>!include\fitz</Filter>
     </ClInclude>
+    <ClInclude Include="..\..\include\mupdf\html.h">
+      <Filter>!include</Filter>
+    </ClInclude>
   </ItemGroup>
 </Project>

mupdf/scripts/wdev.py

@@ -12,7 +12,7 @@
 
 
 class WindowsVS:
-    '''
+    r'''
     Windows only. Finds locations of Visual Studio command-line tools. Assumes
     VS2019-style paths.
 
@@ -101,7 +101,7 @@ def default(value, name):
 
         # Find vcvars.bat.
         #
-        vcvars = f'{directory}\\VC\Auxiliary\\Build\\vcvars{cpu.bits}.bat'
+        vcvars = f'{directory}\\VC\\Auxiliary\\Build\\vcvars{cpu.bits}.bat'
         assert os.path.isfile( vcvars), f'No match for: {vcvars}'
 
         # Find cl.exe.

mupdf/scripts/wrap/__main__.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 
-'''
+r'''
 Support for generating C++ and python wrappers for the mupdf API.
 
 Overview:
@@ -428,7 +428,7 @@ class instance and return them as a tuple in both Python and
 
         Windows:
             > py -m venv pylocal
-            > pylocal\\Scripts\\activate
+            > pylocal\Scripts\activate
             (pylocal) > pip install libclang pyqt5
             (pylocal) > cd ...\mupdf
             (pylocal) > python setup.py install
@@ -898,10 +898,10 @@ class instance and return them as a tuple in both Python and
             Compare generated class methods with functions called by platform/gl
             code.
 
-        python3 -m cProfile -s cumulative ./scripts/mupdfwrap.py -b 0
+        python3 -m cProfile -s cumulative ./scripts/mupdfwrap.py --venv -b 0
             Profile generation of C++ source code.
 
-        ./scripts/mupdfwrap.py --venv --swig-windows-auto -b all -t
+        ./scripts/mupdfwrap.py --venv -b all -t
             Build and test on Windows.
 
 
@@ -1137,7 +1137,7 @@ def test( dir_so, expected_command):
         assert command == expected_command, f'\nExpected: {expected_command}\nBut:      {command}'
 
     mupdf_root = os.path.abspath( f'{__file__}/../../../')
-    infix = 'CXX=clang++ ' if state.state_.openbsd else ''
+    infix = 'CXX=c++ ' if state.state_.openbsd else ''
 
     test(
             'shared-release',
@@ -1192,9 +1192,9 @@ def _get_m_command( build_dirs, j=None, make=None, m_target=None, m_vars=None):
         if state.state_.openbsd:
             # Need to run gmake, not make. Also for some
             # reason gmake on OpenBSD sets CC to clang, but
-            # CXX to g++, so need to force CXX=clang++ too.
+            # CXX to g++, so need to force CXX=c++ too.
             #
-            make = 'CXX=clang++ gmake'
+            make = 'CXX=c++ gmake'
     if not make:
         make = 'make'
 
@@ -1999,7 +1999,7 @@ def fn(name):
                     elif build_csharp:
                         cpp_path = f'{build_dirs.dir_mupdf}/platform/csharp/mupdfcpp_swig.cpp'
                         out_so = f'{build_dirs.dir_so}/mupdfcsharp.so'  # todo: append {so_version} ?
-
+                    cpp_path = os.path.relpath(cpp_path)    # So we don't expose build machine details in __FILE__.
                     if state.state_.openbsd:
                         # clang needs around 2G on OpenBSD.
                         #
@@ -2972,17 +2972,21 @@ def system(command):
                     jlib.system(f'"{sys.executable}" -m venv --system-site-packages {venv}', out='log', verbose=1)
                 else:
                     jlib.system(f'"{sys.executable}" -m venv {venv}', out='log', verbose=1)
+
                 if state.state_.windows:
-                    command = f'{venv}\\Scripts\\activate.bat'
+                    command_venv_enter = f'{venv}\\Scripts\\activate.bat'
                 else:
-                    command = f'. {venv}/bin/activate'
-                command += f' && python -m pip install --upgrade pip'
+                    command_venv_enter = f'. {venv}/bin/activate'
+
+                command = f'{command_venv_enter} && python -m pip install --upgrade pip'
                 if state.state_.openbsd:
                     jlib.log( 'Not installing libclang on openbsd; we assume py3-llvm is installed.')
                     command += f' && python -m pip install --upgrade swig setuptools'
                 else:
                     command += f' && python -m pip install{force_reinstall} --upgrade libclang swig setuptools'
-                command += f' && python {shlex.quote(sys.argv[0])}'
+                jlib.system(command, out='log', verbose=1)
+
+                command = f'{command_venv_enter} && python {shlex.quote(sys.argv[0])}'
                 while 1:
                     try:
                         command += f' {shlex.quote(args.next())}'