Refactor dataclasses

Makefile

@@ -83,7 +83,7 @@ smoke-test: tests/resources/2files/kant_aufklaerung_1784_0020.tif
 	eynollah -m $(CURDIR) layout -di $(<D) -o $(TMPDIR) 
 	test -s $(TMPDIR)/euler_rechenkunst01_1738_0025.xml
 	# mbreorder, directory mode (overwrite):
-	eynollah -m $(CURDIR) machine-based-reading-order -di $(<D) -o $(TMPDIR) 
+	eynollah -m $(CURDIR) reorder -mb -di $(<D) -o $(TMPDIR) 
 	fgrep -q http://schema.primaresearch.org/PAGE/gts/pagecontent/2019-07-15 $(TMPDIR)/$(basename $(<F)).xml
 	fgrep -c -e RegionRefIndexed $(TMPDIR)/$(basename $(<F)).xml
 	# binarize:

src/eynollah/cli/__init__.py

@@ -1,6 +1,7 @@
 from .cli import main
 from .cli_binarize import binarize_cli
 from .cli_enhance import enhance_cli
+from .cli_extract_page import extract_page_cli
 from .cli_extract_images import extract_images_cli
 from .cli_layout import layout_cli
 from .cli_models import models_cli
@@ -10,7 +11,8 @@
 main.add_command(binarize_cli, 'binarization')
 main.add_command(enhance_cli, 'enhancement')
 main.add_command(layout_cli, 'layout')
-main.add_command(readingorder_cli, 'machine-based-reading-order')
+main.add_command(readingorder_cli, 'reorder')
 main.add_command(models_cli, 'models')
 main.add_command(ocr_cli, 'ocr')
+main.add_command(extract_page_cli, 'extract-page')
 main.add_command(extract_images_cli, 'extract-images')

src/eynollah/cli/cli_extract_images.py

@@ -82,7 +82,7 @@ def extract_images_cli(
     ignore_page_extraction,
 ):
     """
-    Detect Layout (with optional image enhancement and reading order detection)
+    Detect image regions only
     """
     assert enable_plotting or not save_images, "Plotting with -si also requires -ep"
     assert not enable_plotting  or save_images, "Plotting with -ep also requires -si"

src/eynollah/cli/cli_extract_page.py

@@ -0,0 +1,80 @@
+import click
+
+@click.command(context_settings=dict(
+    help_option_names=['-h', '--help'],
+    show_default=True))
+@click.option(
+    "--image",
+    "-i",
+    help="input image filename",
+    type=click.Path(exists=True, dir_okay=False),
+)
+
+@click.option(
+    "--out",
+    "-o",
+    help="directory for output PAGE-XML files",
+    type=click.Path(exists=True, file_okay=False),
+    required=True,
+)
+@click.option(
+    "--overwrite",
+    "-O",
+    help="overwrite (instead of skipping) if output xml exists",
+    is_flag=True,
+)
+@click.option(
+    "--dir_in",
+    "-di",
+    help="directory of input images (instead of --image)",
+    type=click.Path(exists=True, file_okay=False),
+)
+@click.option(
+    "--input_binary",
+    "-ib",
+    is_flag=True,
+    help="In general, eynollah uses RGB as input, but if the input document is very dark, very bright or for any other reason you can turn on internal binarization here. When set, eynollah will binarize the RGB input document first.",
+)
+@click.option(
+    "--num_col_upper",
+    "-ncu",
+    default=0,
+    type=click.IntRange(min=0),
+    help="lower limit of columns in document image; 0 means autodetected from model",
+)
+@click.option(
+    "--num_col_lower",
+    "-ncl",
+    default=0,
+    type=click.IntRange(min=0),
+    help="upper limit of columns in document image; 0 means autodetected from model",
+)
+@click.pass_context
+def extract_page_cli(
+    ctx,
+    image,
+    out,
+    overwrite,
+    dir_in,
+    input_binary,
+    num_col_upper,
+    num_col_lower,
+):
+    """
+    Detect image regions only
+    """
+    assert bool(image) != bool(dir_in), "Either -i (single input) or -di (directory) must be provided, but not both."
+
+    from ..extract_page import EynollahPageExtractor
+    extractor = EynollahPageExtractor(
+        model_zoo=ctx.obj.model_zoo,
+        input_binary=input_binary,
+        num_col_upper=num_col_upper,
+        num_col_lower=num_col_lower,
+    )
+    extractor.run(overwrite=overwrite,
+                 image_filename=image,
+                 dir_in=dir_in,
+                 dir_out=out,
+    )
+

src/eynollah/cli/cli_readingorder.py

@@ -3,6 +3,12 @@
 @click.command(context_settings=dict(
     help_option_names=['-h', '--help'],
     show_default=True))
+@click.option(
+    "--model_based",
+    "-mb",
+    help="use machine-learning model instead of heuristic rules",
+    is_flag=True,
+)
 @click.option(
     "--input",
     "-i",
@@ -15,24 +21,40 @@
     help="directory of PAGE-XML input files (instead of --input)",
     type=click.Path(exists=True, file_okay=False),
 )
+@click.option(
+    "--dir_imgs",
+    "-dim",
+    help="directory of image input files (in addition to --dir_in or --input; filename stems must match the XML files, with image file format suffixes). Not needed for --model_based.",
+    type=click.Path(exists=True, file_okay=False),
+)
 @click.option(
     "--out",
     "-o",
     help="directory for output images",
     type=click.Path(exists=True, file_okay=False),
     required=True,
 )
+@click.option(
+    "--overwrite",
+    "-O",
+    help="overwrite (instead of skipping) if output xml exists",
+    is_flag=True,
+)
 @click.pass_context
-def readingorder_cli(ctx, input, dir_in, out):
+def readingorder_cli(ctx, model_based, input, dir_in, dir_imgs, out, overwrite):
     """
-    Generate ReadingOrder with a ML model
+    Generate ReadingOrder for existing segmentation from ML model or from heuristic rules
     """
-    from ..mb_ro_on_layout import Reorder
+    from ..reorder import Reorder
     assert bool(input) != bool(dir_in), "Either -i (single input) or -di (directory) must be provided, but not both."
+    assert bool(model_based) or bool(dir_imgs), "For heuristic reading order, -dim must be provided, too."
     orderer = Reorder(model_zoo=ctx.obj.model_zoo,
-                      device=ctx.obj.device)
-    orderer.run(xml_filename=input,
+                      device=ctx.obj.device,
+                      model_based=model_based)
+    orderer.run(overwrite=overwrite,
+                xml_filename=input,
                 dir_in=dir_in,
+                dir_imgs=dir_imgs,
                 dir_out=out,
     )