Add option to explicitly add interword spaces to HOCR pdf-renderer

This commit includes an optional work around for limitations of the
PDF.js viewer described in
https://github.com/jbarlow83/OCRmyPDF/issues/133. Here is explicitly
add an addition space to text elements before drawing them on the PDF
canvas when using the HOCR renderer. This option does not apply to
other pdf renderers in OCRmyPDF and is turned off by default.

ocrmypdf/__main__.py

@@ -304,6 +304,9 @@ advanced.add_argument(
 advanced.add_argument(
     '--user-patterns', metavar='FILE',
     help="Specify the location of the Tesseract user patterns file.")
+advanced.add_argument(
+    '--interword-spaces', action='store_true',
+    help="Add spaces between words with HOCR transformation.")
 
 debugging = parser.add_argument_group(
     "Debugging",
@@ -463,7 +466,11 @@ def check_options_advanced(options, log):
             "--pdfa-image-compression argument has no effect when "
             "--output-type is not 'pdfa', 'pdfa-1', or 'pdfa-2'"
         )
-
+    if options.interword_spaces and options.pdf_renderer != 'hocr':
+        log.warning(
+            "--interword-spaces argument has no effect when "
+            "--pdf-renderer is not 'hocr'"
+        )
 
 def check_options_metadata(options, log):
     import unicodedata

ocrmypdf/hocrtransform.py

@@ -137,7 +137,7 @@ class HocrTransform():
         return s
 
     def to_pdf(self, outFileName, imageFileName=None, showBoundingboxes=False,
-               fontname="Helvetica", invisibleText=False):
+               fontname="Helvetica", invisibleText=False, interwordSpaces=False):
         """
         Creates a PDF file with an image superimposed on top of the text.
         Text is positioned according to the bounding box of the lines in
@@ -180,7 +180,7 @@ class HocrTransform():
                 ".//%sspan[@class='ocrx_word']" % (self.xmlns)) is not None:
             elemclass = "ocrx_word"
 
-        # itterate all text elements
+        # iterate all text elements
         # light green for bounding box of word/line
         pdf.setStrokeColorRGB(1, 0, 0)
         pdf.setLineWidth(0.5)		# bounding box line width
@@ -196,6 +196,12 @@ class HocrTransform():
             if len(elemtxt) == 0:
                 continue
 
+            # if the advanced option `--interword-spaces` is true, append a space
+            # to the end of each text element to allow simpler PDF viewers such
+            # as PDF.js to better recognize words in search and copy and paste
+            if interwordSpaces:
+                elemtxt += ' '
+
             pxl_coords = self.element_coordinates(elem)
             pt = self.pt_from_pixel(pxl_coords)
 
@@ -242,10 +248,12 @@ if __name__ == "__main__":
                         help='Resolution of the image that was OCRed')
     parser.add_argument('-i', '--image', default=None,
                         help='Path to the image to be placed above the text')
+    parser.add_argument('--interword-spaces', action='store_true',
+                         default=False, help='Add spaces between words')
     parser.add_argument('hocrfile', help='Path to the hocr file to be parsed')
     parser.add_argument(
         'outputfile', help='Path to the PDF file to be generated')
     args = parser.parse_args()
 
     hocr = HocrTransform(args.hocrfile, args.resolution)
-    hocr.to_pdf(args.outputfile, args.image, args.boundingboxes)
+    hocr.to_pdf(args.outputfile, args.image, args.boundingboxes, interwordSpaces=args.interword_spaces)

ocrmypdf/pipeline.py

@@ -639,8 +639,8 @@ def render_hocr_page(
 
     hocrtransform = HocrTransform(hocr, dpi)
     hocrtransform.to_pdf(output_file, imageFileName=None,
-                         showBoundingboxes=False, invisibleText=True)
-
+                         showBoundingboxes=False, invisibleText=True,
+                         interwordSpaces=options.interword_spaces)
 
 def flatten_groups(groups):
     for obj in groups:
@@ -664,8 +664,8 @@ def render_hocr_debug_page(
 
     hocrtransform = HocrTransform(hocr, dpi)
     hocrtransform.to_pdf(output_file, imageFileName=None,
-                         showBoundingboxes=True, invisibleText=False)
-
+                         showBoundingboxes=True, invisibleText=False,
+                         interwordSpaces=options.interword_spaces)
 
 def combine_layers(
         infiles,