Fixing issue #463

pymupdf · Mar 12, 2020 · c876582 · c876582
1 parent 5eaeddc
commit c876582
Show file tree

Hide file tree

Showing 7 changed files with 98 additions and 55 deletions.
diff --git a/PKG-INFO b/PKG-INFO
@@ -9,7 +9,7 @@ Home-page: https://github.com/pymupdf/PyMuPDF
 Download-url: https://github.com/pymupdf/PyMuPDF
 Summary: PyMuPDF is a Python binding for the PDF rendering library MuPDF
 Description:
-        Release date: March 05, 2020
+        Release date: March 10, 2020
 
         Authors
         =======

diff --git a/docs/changes.rst b/docs/changes.rst
@@ -6,10 +6,11 @@ Changes in Version 1.16.12
 * **Fixed** issue `#447 <https://github.com/pymupdf/PyMuPDF/issues/447>`_
 * **Fixed** issue `#461 <https://github.com/pymupdf/PyMuPDF/issues/461>`_.
 * **Fixed** issue `#397 <https://github.com/pymupdf/PyMuPDF/issues/397>`_.
-* **Added** JavaScript support to PDF form fields, thereby addressing `#454 <https://github.com/pymupdf/PyMuPDF/issues/454>`_.
-* **Added** a new annotation method :meth:`Annot.delete_responses`, which removes 'Popup' and response annotations referring to the current one.
+* **Fixed** issue `#463 <https://github.com/pymupdf/PyMuPDF/issues/463>`_.
+* **Added** JavaScript support to PDF form fields, thereby fixing `#454 <https://github.com/pymupdf/PyMuPDF/issues/454>`_.
+* **Added** a new annotation method :meth:`Annot.delete_responses`, which removes 'Popup' and response annotations referring to the current one. Mainly serves data protection purposes.
 * **Added** a new form field method :meth:`Widget.reset`, which resets the field value to its default.
-* **Changed** and extended handling of redactions: images and XObjects are removed if *contained* in a redaction rectangle. Any partial overlaps will just be covered by the redaction background color. Now a text can be specified to be inserted in the rectangle area to **replace the deleted original** text. This resolves `#434 <https://github.com/pymupdf/PyMuPDF/issues/434>`_.
+* **Changed** and extended handling of redactions: images and XObjects are removed if *contained* in a redaction rectangle. Any partial only overlaps will just be covered by the redaction background color. Now an *overlay* text can be specified to be inserted in the rectangle area to **take the place the deleted original** text. This resolves `#434 <https://github.com/pymupdf/PyMuPDF/issues/434>`_.
 
 Changes in Version 1.16.11
 ---------------------------

diff --git a/docs/installation.rst b/docs/installation.rst
@@ -55,27 +55,7 @@ Windows binaries are provided for Python **32-bit and 64-bit** versions.
 
 **Overview of wheel names (PyMuPDF version is x.xx.xx):**
 
-* ``PyMuPDF-x.xx.xx-cp27-cp27m-macosx_10_6_intel.whl``
-* ``PyMuPDF-x.xx.xx-cp27-cp27m-manylinux2010_x86_64.whl``
-* ``PyMuPDF-x.xx.xx-cp27-cp27m-win32.whl``
-* ``PyMuPDF-x.xx.xx-cp27-cp27m-win_amd64.whl``
-* ``PyMuPDF-x.xx.xx-cp27-cp27mu-manylinux2010_x86_64.whl``
-* ``PyMuPDF-x.xx.xx-cp35-cp35m-macosx_10_6_intel.whl``
-* ``PyMuPDF-x.xx.xx-cp35-cp35m-manylinux2010_x86_64.whl``
-* ``PyMuPDF-x.xx.xx-cp35-cp35m-win32.whl``
-* ``PyMuPDF-x.xx.xx-cp35-cp35m-win_amd64.whl``
-* ``PyMuPDF-x.xx.xx-cp36-cp36m-macosx_10_6_intel.whl``
-* ``PyMuPDF-x.xx.xx-cp36-cp36m-manylinux2010_x86_64.whl``
-* ``PyMuPDF-x.xx.xx-cp36-cp36m-win32.whl``
-* ``PyMuPDF-x.xx.xx-cp36-cp36m-win_amd64.whl``
-* ``PyMuPDF-x.xx.xx-cp37-cp37m-macosx_10_6_intel.whl``
-* ``PyMuPDF-x.xx.xx-cp37-cp37m-manylinux2010_x86_64.whl``
-* ``PyMuPDF-x.xx.xx-cp37-cp37m-win32.whl``
-* ``PyMuPDF-x.xx.xx-cp37-cp37m-win_amd64.whl``
-* ``PyMuPDF-x.xx.xx-cp38-cp38-macosx_10_9_x86_64.whl``
-* ``PyMuPDF-x.xx.xx-cp38-cp38-manylinux2010_x86_64.whl``
-* ``PyMuPDF-x.xx.xx-cp38-cp38-win32.whl``
-* ``PyMuPDF-x.xx.xx-cp38-cp38-win_amd64.whl``
+.. literalinclude:: wheelnames.txt
 
 
 Older versions can be found in the releases directory of our home page https://github.com/pymupdf/PyMuPDF/releases.

diff --git a/docs/page.rst b/docs/page.rst
@@ -984,13 +984,13 @@ This is available for PDF documents only. There are basically two groups of meth
 
    .. attribute:: CropBoxPosition
 
-      Contains the displacement of the page's */CropBox* for a PDF, otherwise the top-left coordinates of :attr:`Page.rect`.
+      Contains the top-left point of the page's */CropBox* for a PDF, otherwise *Point(0, 0)*.
 
       :type: :ref:`Point`
 
    .. attribute:: CropBox
 
-      The page's */CropBox* for a PDF, else :attr:`Page.rect`.
+      The page's */CropBox* for a PDF. Always the **unrotated** page rectangle is returned. For a non-PDF this will always equal the page rectangle.
 
       :type: :ref:`Rect`
 
@@ -1006,7 +1006,7 @@ This is available for PDF documents only. There are basically two groups of meth
 
       :type: :ref:`Rect`
 
-      .. note:: For most PDF documents and for all other types, *page.rect == page.CropBox == page.MediaBox* is true. However, for some PDFs the visible page is a true subset of */MediaBox*. In this case the above attributes help to correctly locate page elements.
+      .. note:: For most PDF documents and for all other types, *page.rect == page.CropBox == page.MediaBox* is true. However, for some PDFs the visible page is a true subset of */MediaBox*. Also, if the page is rotated, its ``Page.rect`` may not equal ``Page.CropBox``. In these cases the above attributes help to correctly locate page elements.
 
    .. attribute:: firstLink
 

diff --git a/docs/wheelnames.txt b/docs/wheelnames.txt
@@ -0,0 +1,21 @@
+PyMuPDF-x.xx.xx-cp27-cp27m-macosx_10_9_x86_64.whl
+PyMuPDF-x.xx.xx-cp27-cp27m-manylinux2010_x86_64.whl
+PyMuPDF-x.xx.xx-cp27-cp27m-win32.whl
+PyMuPDF-x.xx.xx-cp27-cp27m-win_amd64.whl
+PyMuPDF-x.xx.xx-cp27-cp27mu-manylinux2010_x86_64.whl
+PyMuPDF-x.xx.xx-cp35-cp35m-macosx_10_9_x86_64.whl
+PyMuPDF-x.xx.xx-cp35-cp35m-manylinux2010_x86_64.whl
+PyMuPDF-x.xx.xx-cp35-cp35m-win32.whl
+PyMuPDF-x.xx.xx-cp35-cp35m-win_amd64.whl
+PyMuPDF-x.xx.xx-cp36-cp36m-macosx_10_9_x86_64.whl
+PyMuPDF-x.xx.xx-cp36-cp36m-manylinux2010_x86_64.whl
+PyMuPDF-x.xx.xx-cp36-cp36m-win32.whl
+PyMuPDF-x.xx.xx-cp36-cp36m-win_amd64.whl
+PyMuPDF-x.xx.xx-cp37-cp37m-macosx_10_9_x86_64.whl
+PyMuPDF-x.xx.xx-cp37-cp37m-manylinux2010_x86_64.whl
+PyMuPDF-x.xx.xx-cp37-cp37m-win32.whl
+PyMuPDF-x.xx.xx-cp37-cp37m-win_amd64.whl
+PyMuPDF-x.xx.xx-cp38-cp38-macosx_10_9_x86_64.whl
+PyMuPDF-x.xx.xx-cp38-cp38-manylinux2010_x86_64.whl
+PyMuPDF-x.xx.xx-cp38-cp38-win32.whl
+PyMuPDF-x.xx.xx-cp38-cp38-win_amd64.whl
diff --git a/fitz/fitz.i b/fitz/fitz.i
@@ -2592,8 +2592,6 @@ if len(pyliste) == 0 or min(pyliste) not in range(len(self)) or max(pyliste) not
 
 
             xrefObject = get_pdf_object
-            xref_object = get_pdf_object
-            xref_stream = xrefStream
 
 
             def __repr__(self):
@@ -3673,28 +3671,56 @@ annot._erase()
         }
 
         //---------------------------------------------------------------------
-        // MediaBox size: width, height of /MediaBox (PDF only)
+        // MediaBox: get the /MediaBox (PDF only)
         //---------------------------------------------------------------------
-        PARENTCHECK(MediaBoxSize)
+        PARENTCHECK(MediaBox)
         %pythoncode %{@property%}
-        %feature("autodoc","Retrieve width, height of /MediaBox.") MediaBoxSize;
-        %pythonappend MediaBoxSize %{
-        val = Point(val)
-        if not bool(val):
-            r = self.rect
-            val = Point(r.width, r.height)
+        %feature("autodoc","Retrieve the /MediaBox.") MediaBox;
+        %pythonappend MediaBox %{
+        val = Rect(val)
         %}
-        PyObject *MediaBoxSize()
+        PyObject *MediaBox()
         {
-            PyObject *p = JM_py_from_point(fz_make_point(0, 0));
             pdf_page *page = pdf_page_from_fz_page(gctx, $self);
-            if (!page) return p;
-            fz_rect r = fz_empty_rect;
-            pdf_obj *o = pdf_dict_get_inheritable(gctx, page->obj, PDF_NAME(MediaBox));
-            if (!o) return p;
+            if (!page) return JM_py_from_rect(fz_bound_page(gctx, $self));
+
+            fz_rect mediabox, cropbox, page_mediabox;
+            PyObject *rect = NULL;
+            pdf_obj *obj;
+            float userunit = 1;
+
+            obj = pdf_dict_get(gctx, page->obj, PDF_NAME(UserUnit));
+            if (pdf_is_real(gctx, obj))
+            {
+                userunit = pdf_to_real(gctx, obj);
+            }
+
+            mediabox = pdf_to_rect(gctx, pdf_dict_get_inheritable(gctx, page->obj, PDF_NAME(MediaBox)));
+            if (fz_is_empty_rect(mediabox))
+            {
+                mediabox.x0 = 0;
+                mediabox.y0 = 0;
+                mediabox.x1 = 612;
+                mediabox.y1 = 792;
+            }
+
+            cropbox = pdf_to_rect(gctx,
+                pdf_dict_get_inheritable(gctx, page->obj, PDF_NAME(CropBox)));
+            if (!fz_is_empty_rect(cropbox))
+            {
+                mediabox = fz_intersect_rect(mediabox, cropbox);
+            }
+            page_mediabox.x0 = fz_min(mediabox.x0, mediabox.x1);
+            page_mediabox.y0 = fz_min(mediabox.y0, mediabox.y1);
+            page_mediabox.x1 = fz_max(mediabox.x0, mediabox.x1);
+            page_mediabox.y1 = fz_max(mediabox.y0, mediabox.y1);
+
+            if (page_mediabox.x1 - page_mediabox.x0 < 1 ||
+                page_mediabox.y1 - page_mediabox.y0 < 1)
+                page_mediabox = fz_unit_rect;
+
+            return JM_py_from_rect(page_mediabox);
 
-            r = pdf_to_rect(gctx, o);
-            return JM_py_from_point(fz_make_point(r.x1 - r.x0, r.y1 - r.y0));
         }
 
         //---------------------------------------------------------------------
@@ -3712,6 +3738,7 @@ annot._erase()
             pdf_obj *o = pdf_dict_get_inheritable(gctx, page->obj, PDF_NAME(CropBox));
             if (!o) return p;                    // no CropBox specified
             fz_rect cbox = pdf_to_rect(gctx, o);
+            Py_DECREF(p);
             return JM_py_from_point(fz_make_point(cbox.x0, cbox.y0));;
         }
 
@@ -4253,7 +4280,12 @@ def insertFont(self, fontname="helv", fontfile=None, fontbuffer=None,
         //---------------------------------------------------------------------
         PARENTCHECK(_getTransformation)
         %feature("autodoc","Return page transformation matrix.") _getTransformation;
-        %pythonappend _getTransformation %{val = Matrix(val)%}
+        %pythonappend _getTransformation %{
+        if self.rotation % 360 == 0:
+            val = Matrix(val)
+        else:
+            val = Matrix(1, 0, 0, -1, 0, self.CropBox.height)
+        %}
         PyObject *_getTransformation()
         {
             fz_matrix ctm = fz_identity;
@@ -4454,18 +4486,25 @@ def insertFont(self, fontname="helv", fontfile=None, fontbuffer=None,
 
         @property
         def CropBox(self):
+            """Rectangle /CropBox IGNORING any page rotation."""
+            rotation = self.rotation  # page rotation
+            width = self.rect.width  # page width
+            height = self.rect.height  # page height
+            if rotation % 180 != 0:  # rotation by odd number of 90
+                width, height = height, width  # so revert width and height
             x0 = self.CropBoxPosition.x
-            y0 = self.MediaBoxSize.y - self.CropBoxPosition.y - self.rect.height
-            x1 = x0 + self.rect.width
-            y1 = y0 + self.rect.height
+            y0 = self.MediaBox.height - self.CropBoxPosition.y - height
+            x1 = x0 + width
+            y1 = y0 + height
             return Rect(x0, y0, x1, y1)
 
         @property
-        def MediaBox(self):
-            return Rect(0, 0, self.MediaBoxSize)
-        
-        clean_contents = _cleanContents
-        get_contents = _getContents
+        def MediaBoxSize(self):
+            return Point(self.MediaBox.width, self.MediaBox.height)
+
+        cleanContents = _cleanContents
+        getContents = _getContents
+        getTransformation = _getTransformation
 
         %}
     }

diff --git a/fitz/utils.py b/fitz/utils.py
@@ -2419,8 +2419,10 @@ def __init__(self, page):
         self.width = page.MediaBoxSize.x
         self.x = page.CropBoxPosition.x
         self.y = page.CropBoxPosition.y
+
         self.pctm = page._getTransformation()  # page transf. matrix
         self.ipctm = ~self.pctm  # inverted transf. matrix
+
         self.draw_cont = ""
         self.text_cont = ""
         self.totalcont = ""