Add PDF linearization

src/ocrmypdf/_pipeline.py

@@ -716,6 +716,13 @@ def convert_to_pdfa(input_pdf, input_ps_stub, context):
     return output_file
 
 
+def should_linearize(working_file, context):
+    filesize = os.stat(working_file).st_size
+    if filesize > (context.options.fast_web_view * 1_000_000):
+        return True
+    return False
+
+
 def metadata_fixup(working_file, context):
     output_file = context.get_path('metafix.pdf')
     options = context.options
@@ -749,11 +756,14 @@ def metadata_fixup(working_file, context):
                 context.log.info(
                     "The following metadata fields were not copied: %r", not_copied
                 )
-
     pdf.save(
         output_file,
         compress_streams=True,
+        preserve_pdfa=True,
         object_stream_mode=pikepdf.ObjectStreamMode.generate,
+        linearize=(  # Don't linearize if optimize() will be linearizing too
+            should_linearize(working_file, context) if options.optimize == 0 else False
+        ),
     )
     original.close()
     pdf.close()
@@ -762,7 +772,13 @@ def metadata_fixup(working_file, context):
 
 def optimize_pdf(input_file, context):
     output_file = context.get_path('optimize.pdf')
-    optimize(input_file, output_file, context)
+    save_settings = dict(
+        compress_streams=True,
+        preserve_pdfa=True,
+        object_stream_mode=pikepdf.ObjectStreamMode.generate,
+        linearize=should_linearize(input_file, context),
+    )
+    optimize(input_file, output_file, context, save_settings)
     return output_file
 
 

src/ocrmypdf/api.py

@@ -196,6 +196,7 @@ def ocr(  # pylint: disable=unused-argument
     pdfa_image_compression=None,
     user_words=None,
     user_patterns=None,
+    fast_web_view=None,
     keep_temporary_files=None,
     progress_bar=None,
     tesseract_env=None,

src/ocrmypdf/cli.py

@@ -466,6 +466,18 @@ advanced.add_argument(
     metavar='FILE',
     help="Specify the location of the Tesseract user patterns file.",
 )
+advanced.add_argument(
+    '--fast-web-view',
+    type=numeric(float, 0),
+    default=1.0,
+    metavar="MEGABYTES",
+    help="If the size of file is more than this threshold (in MB), then "
+    "linearize the PDF for fast web viewing. This allows the PDF to be "
+    "displayed before it is fully downloaded in web browsers, but increases "
+    "the space required slightly. By default we skip this for small files "
+    "which do not benefit. If the threshold is 0 it will be apply to all files. "
+    "Set the threshold very high to disable.",
+)
 
 debugging = parser.add_argument_group(
     "Debugging", "Arguments to help with troubleshooting and debugging"

src/ocrmypdf/optimize.py

@@ -449,7 +449,7 @@ def transcode_pngs(pike, images, image_name_fn, root, log, options):
         im_obj.write(compdata.read(), filter=Name.FlateDecode, decode_parms=dparms)
 
 
-def optimize(input_file, output_file, context):
+def optimize(input_file, output_file, context, save_settings):
     log = context.log
     options = context.options
     if options.optimize == 0:
@@ -479,11 +479,7 @@ def optimize(input_file, output_file, context):
 
         target_file = Path(output_file).with_suffix('.opt.pdf')
         pike.remove_unreferenced_resources()
-        pike.save(
-            target_file,
-            preserve_pdfa=True,
-            object_stream_mode=pikepdf.ObjectStreamMode.generate,
-        )
+        pike.save(target_file, **save_settings)
 
     input_size = Path(input_file).stat().st_size
     output_size = Path(target_file).stat().st_size
@@ -497,8 +493,11 @@ def optimize(input_file, output_file, context):
     log.info(f"Optimize ratio: {ratio:.2f} savings: {(100 * savings):.1f}%")
 
     if savings < 0:
-        log.info("Optimize did not improve the file - discarded")
-        re_symlink(input_file, output_file)
+        log.info("Image optimization did not improve the file - discarded")
+        # We still need to save the file
+        with pikepdf.open(input_file) as pike:
+            pike.remove_unreferenced_resources()
+            pike.save(output_file, **save_settings)
     else:
         re_symlink(target_file, output_file)
 
@@ -535,7 +534,16 @@ def main(infile, outfile, level, jobs=1):
     with TemporaryDirectory() as td:
         context = PDFContext(options, td, infile, None)
         tmpout = Path(td) / 'out.pdf'
-        optimize(infile, tmpout, context)
+        optimize(
+            infile,
+            tmpout,
+            context,
+            dict(
+                compress_streams=True,
+                preserve_pdfa=True,
+                object_stream_mode=pikepdf.ObjectStreamMode.generate,
+            ),
+        )
         copy(fspath(tmpout), fspath(outfile))