Merge pull request #226 from JorjMcKie/master

Uploading first patch

README.md

@@ -1,4 +1,4 @@
-# PyMuPDF 1.14.0
+# PyMuPDF 1.14.1
 
 ![logo](https://github.com/rk700/PyMuPDF/blob/master/demo/pymupdf.jpg)
 
@@ -14,7 +14,7 @@ On **[PyPI](https://pypi.org/project/PyMuPDF)** since August 2016: [![](https://
 
 # Introduction
 
-This is **version 1.14.0 of PyMuPDF (formerly python-fitz)**, a Python binding with support for [MuPDF 1.14.0](http://mupdf.com/) - "a lightweight PDF, XPS, and E-book viewer".
+This is **version 1.14.1 of PyMuPDF (formerly python-fitz)**, a Python binding with support for [MuPDF 1.14.x](http://mupdf.com/) - "a lightweight PDF, XPS, and E-book viewer".
 
 MuPDF can access files in PDF, XPS, OpenXPS, CBZ, EPUB and FB2 (e-books) formats, and it is known for its top performance and high rendering quality.
 
@@ -30,14 +30,15 @@ The platform tag for Mac OSX is `macosx_10_6_intel`.
 
 The platform tag for Linux is `manylinux1_x86_64`, which makes these wheels usable on Debian, Ubuntu and most other variations.
 
-On other operating systems you need to generate PyMuPDF yourself. And of course you can choose to do so for a wheel-supported platform, too. Before you can do this, you must download and generate MuPDF. This process depends very much on your system. For most platforms, the MuPDF source contains prepared procedures for achieving this.
+On other operating systems you need to generate PyMuPDF yourself. And of course you can choose to do so for a wheel-supported platform, too. Before you can do this, you must download and generate MuPDF. This process depends very much on your system. For most platforms, the MuPDF source contains prepared procedures for achieving this. Please observe the following general steps:
 
-* Be sure to download the official MuPDF source release from [here](https://mupdf.com/downloads). Do **not use** MuPDF's [GitHub repo](https://github.com/ArtifexSoftware/mupdf). It contains their current **development source**, which is **not compatible** with this PyMuPDF version most of the time.
+* Be sure to download the official MuPDF source release from [here](https://mupdf.com/downloads/archive). Do **not use** MuPDF's [GitHub repo](https://github.com/ArtifexSoftware/mupdf). It contains their current **development source**, which is **not compatible** with this PyMuPDF version most of the time.
 
-* The repo's `fitz` folder contains a few files whose names start with an underscore. These files contain configuration data and hotfixes and each must be copy-renamed to its correct target location of the downloaded MuPDF source **before you generate MuPDF**. Currently, these files are:
-  - fitz configuration file `_mupdf_config.h` copy-replace to `mupdf/include/fitz/config.h`.
-  - fitz error module `_error.c`, copy-replace to `mupdf/source/fitz/error.c`.
-  - PDF device module `_pdf-device.c` copy-replace `mupdf/source/pdf/pdf-device.c`.
+* The repo's `fitz` folder contains a few files whose names start with an underscore `"_"`. These files contain configuration data and hotfixes. Each one must be copy-renamed to its correct target location of the MuPDF source that you have downloaded, **before you generate MuPDF**. Currently, these files are:
+  - fitz configuration file `_mupdf_config.h` copy-replace to: `mupdf/include/fitz/config.h`. It contains configuration data like e.g. which fonts to support.
+  - fitz error module `_error.c`, copy-replace to: `mupdf/source/fitz/error.c`. It redirects MuPDF warnings and errors so they can be intercepted by PyMuPDF.
+  - PDF device module `_pdf-device.c` copy-replace to: `mupdf/source/pdf/pdf-device.c`. It fixes a bug which caused method `Document.convertToPDF()` to bring down the interpeter.
+  - Now MuPDF can be generated.
 
 Once this is done, adjust directories in ``setup.py`` and run ``python setup.py install``.
 

fitz/__init__.py

@@ -13,6 +13,9 @@
 TOOLS         = fitz.Tools()
 fitz.TOOLS    = TOOLS
 
+if fitz.VersionFitz != fitz.TOOLS.mupdf_version():
+    raise ValueError("MuPDF library mismatch %s <> %s" % (fitz.VersionFitz, fitz.TOOLS.mupdf_version()))
+
 import fitz.utils
 
 # copy functions to their respective fitz classes

fitz/fitz.i

@@ -145,10 +145,10 @@ fz_set_stderr(gctx, JM_fitz_stderr);
 if (JM_fitz_stderr && JM_fitz_stdout)
     {;}
 else
-    PySys_WriteStdout("error redirecting stdout/stderr!\n");
+    PySys_WriteStderr("error redirecting stdout/stderr!\n");
 
-JM_error_log  = PyList_New(0);
-JM_output_log = PyList_New(0);
+JM_error_log  = PyByteArray_FromStringAndSize("", 0);
+JM_output_log = PyByteArray_FromStringAndSize("", 0);
 
 //-----------------------------------------------------------------------------
 // STOP redirect stdout/stderr
@@ -172,6 +172,7 @@ struct DeviceWrapper {
 // include version information and several other helpers
 //-----------------------------------------------------------------------------
 %pythoncode %{
+import os
 import weakref
 from binascii import hexlify
 import math
@@ -211,18 +212,27 @@ struct fz_document_s
         FITZEXCEPTION(fz_document_s, !result)
 
         %pythonprepend fz_document_s %{
-            if not filename or type(filename) == str:
+            if not filename or type(filename) is str:
                 pass
-            elif type(filename) == unicode:
-                filename = filename.encode('utf8')
             else:
-                raise TypeError("filename must be string or None")
-            self.name = filename if filename else ""
+                if str is bytes:                 # Python 2
+                    if type(filename) is unicode:
+                        filename = filename.encode("utf8")
+                else:
+                    filename = str(filename)     # should take care of pathlib
+
             self.streamlen = len(stream) if stream else 0
-            if stream and not (filename or filetype):
-                raise ValueError("filetype missing with stream specified")
-            if stream and type(stream) not in (bytes, bytearray):
-                raise ValueError("stream must be bytes or bytearray")
+
+            self.name = ""
+            if filename and self.streamlen == 0:
+                self.name = filename
+
+            if self.streamlen > 0:
+                if not (filename or filetype):
+                    raise ValueError("filetype missing with stream specified")
+                if type(stream) not in (bytes, bytearray):
+                    raise ValueError("stream must be bytes or bytearray")
+
             self.isClosed    = False
             self.isEncrypted = 0
             self.metadata    = None
@@ -2102,18 +2112,18 @@ struct fz_page_s {
                 inklist = pdf_new_array(gctx, annot->page->doc, n0);
                 for (j = 0; j < n0; j++)
                 {
-                    sublist = PySequence_GetItem(list, j);
+                    sublist = PySequence_ITEM(list, j);
                     n1 = PySequence_Size(sublist);
                     stroke = pdf_new_array(gctx, annot->page->doc, 2 * n1);
                     for (i = 0; i < n1; i++)
                     {
-                        p = PySequence_GetItem(sublist, i);
+                        p = PySequence_ITEM(sublist, i);
                         if (!PySequence_Check(p) || PySequence_Size(p) != 2)
                             THROWMSG("3rd level entries must be pairs of floats");
-                        x = PyFloat_AsDouble(PySequence_GetItem(p, 0));
+                        x = PyFloat_AsDouble(PySequence_ITEM(p, 0));
                         if (PyErr_Occurred())
                             THROWMSG("invalid point coordinate");
-                        y = PyFloat_AsDouble(PySequence_GetItem(p, 1));
+                        y = PyFloat_AsDouble(PySequence_ITEM(p, 1));
                         if (PyErr_Occurred())
                             THROWMSG("invalid point coordinate");
                         Py_CLEAR(p);
@@ -4559,7 +4569,7 @@ struct fz_annot_s
             if (n>0)
             {
                 for (i=0; i<n; i++)
-                    col[i] = (float) PyFloat_AsDouble(PySequence_GetItem(ccol, i));
+                    col[i] = (float) PyFloat_AsDouble(PySequence_ITEM(ccol, i));
                 fz_try(gctx)
                     pdf_set_annot_color(gctx, annot, n, col);
                 fz_catch(gctx)
@@ -4577,7 +4587,7 @@ struct fz_annot_s
                     return;
                 }
                 for (i=0; i<n; i++)
-                    col[i] = (float) PyFloat_AsDouble(PySequence_GetItem(icol, i));
+                    col[i] = (float) PyFloat_AsDouble(PySequence_ITEM(icol, i));
                 fz_try(gctx)
                     pdf_set_annot_interior_color(gctx, annot, n, col);
                 fz_catch(gctx)
@@ -5998,16 +6008,34 @@ struct Tools
         }
 
         %pythoncode%{@property%}
-        PyObject *fitz_stderr()
+        char *fitz_stdout()
         {
-            return PyUnicode_Join(Py_BuildValue("s", ""), JM_error_log);
+            return PyByteArray_AS_STRING(JM_output_log);
+        }
+
+        %feature("autodoc","Empty fitz output log.") empty_error_log;
+        void fitz_stdout_reset()
+        {
+            Py_CLEAR(JM_output_log);
+            JM_output_log = PyByteArray_FromStringAndSize("", 0);
+        }
+
+        %pythoncode%{@property%}
+        char *fitz_stderr()
+        {
+            return PyByteArray_AS_STRING(JM_error_log);
         }
 
         %feature("autodoc","Empty fitz error log.") empty_error_log;
         void fitz_stderr_reset()
         {
             Py_CLEAR(JM_error_log);
-            JM_error_log  = PyList_New(0);
+            JM_error_log  = PyByteArray_FromStringAndSize("", 0);
+        }
+
+        char *mupdf_version()
+        {
+            return FZ_VERSION;
         }
 
         PyObject *transform_rect(PyObject *rect, PyObject *matrix)

fitz/fitz.py

@@ -96,6 +96,7 @@ class _object:
     _newclass = 0
 
 
+import os
 import weakref
 from binascii import hexlify
 import math
@@ -105,9 +106,9 @@ class _object:
 
 
 VersionFitz = "1.14.0"
-VersionBind = "1.14.0"
-VersionDate = "2018-11-16 05:14:22"
-version = (VersionBind, VersionFitz, "20181116051422")
+VersionBind = "1.14.1"
+VersionDate = "2018-11-18 17:07:00"
+version = (VersionBind, VersionFitz, "20181118170700")
 
 
 class Matrix():
@@ -1572,18 +1573,27 @@ class Document(_object):
     def __init__(self, filename=None, stream=None, filetype=None, rect=None, width=0, height=0, fontsize=11):
         """__init__(self, filename=None, stream=None, filetype=None, rect=None, width=0, height=0, fontsize=11) -> Document"""
 
-        if not filename or type(filename) == str:
+        if not filename or type(filename) is str:
             pass
-        elif type(filename) == unicode:
-            filename = filename.encode('utf8')
         else:
-            raise TypeError("filename must be string or None")
-        self.name = filename if filename else ""
+            if str is bytes:                 # Python 2
+                if type(filename) is unicode:
+                    filename = filename.encode("utf8")
+            else:
+                filename = str(filename)     # should take care of pathlib
+
         self.streamlen = len(stream) if stream else 0
-        if stream and not (filename or filetype):
-            raise ValueError("filetype missing with stream specified")
-        if stream and type(stream) not in (bytes, bytearray):
-            raise ValueError("stream must be bytes or bytearray")
+
+        self.name = ""
+        if filename and self.streamlen == 0:
+            self.name = filename
+
+        if self.streamlen > 0:
+            if not (filename or filetype):
+                raise ValueError("filetype missing with stream specified")
+            if type(stream) not in (bytes, bytearray):
+                raise ValueError("stream must be bytes or bytearray")
+
         self.isClosed    = False
         self.isEncrypted = 0
         self.metadata    = None
@@ -4095,8 +4105,19 @@ def glyph_cache_empty(self):
 
     @property
 
+    def fitz_stdout(self):
+        """fitz_stdout(self) -> char *"""
+        return _fitz.Tools_fitz_stdout(self)
+
+
+    def fitz_stdout_reset(self):
+        """fitz_stdout_reset(self)"""
+        return _fitz.Tools_fitz_stdout_reset(self)
+
+    @property
+
     def fitz_stderr(self):
-        """fitz_stderr(self) -> PyObject *"""
+        """fitz_stderr(self) -> char *"""
         return _fitz.Tools_fitz_stderr(self)
 
 
@@ -4105,6 +4126,11 @@ def fitz_stderr_reset(self):
         return _fitz.Tools_fitz_stderr_reset(self)
 
 
+    def mupdf_version(self):
+        """mupdf_version(self) -> char *"""
+        return _fitz.Tools_mupdf_version(self)
+
+
     def transform_rect(self, rect, matrix):
         """transform_rect(self, rect, matrix) -> PyObject *"""
         return _fitz.Tools_transform_rect(self, rect, matrix)