leptonica: don't open files by name; use memory buffers

Avoids encoding issues and makes error trap unnecessary in some cases.

src/ocrmypdf/leptonica.py

@@ -296,9 +296,11 @@ class Pix(LeptonicaObject):
         Leptonica can load TIFF, PNM (PBM, PGM, PPM), PNG, and JPEG.  If
         loading fails then the object will wrap a C null pointer.
         """
-        filename = fspath(path)
-        with _LeptonicaErrorTrap():
-            return cls(lept.pixRead(os.fsencode(filename)))
+        with open(path, 'rb') as py_file:
+            data = py_file.read()
+            buffer = ffi.from_buffer(data)
+            with _LeptonicaErrorTrap():
+                return cls(lept.pixReadMem(buffer, len(buffer)))
 
     def write_implied_format(self, path, jpeg_quality=0, jpeg_progressive=0):
         """Write pix to the filename, with the extension indicating format.
@@ -306,11 +308,19 @@ class Pix(LeptonicaObject):
         jpeg_quality -- quality (iff JPEG; 1 - 100, 0 for default)
         jpeg_progressive -- (iff JPEG; 0 for baseline seq., 1 for progressive)
         """
-        filename = fspath(path)
-        with _LeptonicaErrorTrap():
-            lept.pixWriteImpliedFormat(
-                os.fsencode(filename), self._cdata, jpeg_quality, jpeg_progressive
-            )
+        lept_format = lept.getImpliedFileFormat(os.fsencode(path))
+        with open(path, 'wb') as py_file:
+            data = ffi.new('l_uint8 **pdata')
+            size = ffi.new('size_t *psize')
+            with _LeptonicaErrorTrap():
+                if lept_format == lept.L_JPEG_ENCODE:
+                    lept.pixWriteMemJpeg(
+                        data, size, self._cdata, jpeg_quality, jpeg_progressive
+                    )
+                else:
+                    lept.pixWriteMem(data, size, self._cdata, lept_format)
+            buffer = ffi.buffer(data[0], size[0])
+            py_file.write(buffer)
 
     @classmethod
     def frompil(self, pillow_image):

src/ocrmypdf/lib/compile_leptonica.py

@@ -16,6 +16,8 @@
 # You should have received a copy of the GNU General Public License
 # along with OCRmyPDF.  If not, see <http://www.gnu.org/licenses/>.
 
+from pathlib import Path
+
 from cffi import FFI
 
 ffibuilder = FFI()
@@ -221,9 +223,15 @@ ffibuilder.cdef(
     """
 PIX * pixRead ( const char *filename );
 PIX * pixReadMem ( const l_uint8 *data, size_t size );
+PIX * pixReadStream ( FILE *fp, l_int32 hint );
 PIX * pixScale ( PIX *pixs, l_float32 scalex, l_float32 scaley );
 l_int32 pixFindSkew ( PIX *pixs, l_float32 *pangle, l_float32 *pconf );
 l_int32 pixWriteImpliedFormat ( const char *filename, PIX *pix, l_int32 quality, l_int32 progressive );
+l_int32 getImpliedFileFormat ( const char *filename );
+l_ok pixWriteStream ( FILE *fp, PIX *pix, l_int32 format );
+l_ok pixWriteStreamJpeg ( FILE *fp, PIX *pixs, l_int32 quality, l_int32 progressive );
+l_ok pixWriteMem ( l_uint8 **pdata, size_t *psize, PIX *pix, l_int32 format );
+l_ok pixWriteMemJpeg ( l_uint8 **pdata, size_t *psize, PIX *pix, l_int32 quality, l_int32 progressive );
 l_int32
 pixWriteMemPng(l_uint8  **pdata,
                size_t    *psize,

tests/test_lept.py

@@ -94,16 +94,6 @@ def test_leptonica_compile(tmp_path):
     ffibuilder.compile(tmpdir=fspath(tmp_path), target=fspath(tmp_path / 'lepttest.*'))
 
 
-def test_with_stderr(capsys):
-    # pytest redirects stderr too; we must disable this for the test to be valid
-    with capsys.disabled():
-        with pytest.raises(FileNotFoundError):
-            lept.Pix.open("does_not_exist1")
-
-
-def test_without_stderr(capsys):
-    # pytest redirects stderr too; we must disable this for the test to be valid
-    with capsys.disabled():
-        with patch('sys.stderr', new=None):
-            with pytest.raises(FileNotFoundError):
-                lept.Pix.open("does_not_exist2")
+def test_file_not_found():
+    with pytest.raises(FileNotFoundError):
+        lept.Pix.open("does_not_exist1")