Port to ocrd core version 3.0.0

ocrd_cis/ocropy/clip.py

@@ -1,6 +1,7 @@
 from __future__ import absolute_import
+from logging import Logger
 
-import os.path
+from os.path import join
 import numpy as np
 from PIL import Image, ImageStat, ImageOps
 from shapely.geometry import Polygon
@@ -14,7 +15,6 @@
 from ocrd_utils import (
     getLogger,
     make_file_id,
-    assert_file_grp_cardinality,
     coordinates_of_segment,
     polygon_from_points,
     bbox_from_polygon,
@@ -24,22 +24,20 @@
     MIMETYPE_PAGE
 )
 
-from .. import get_ocrd_tool
 from .ocrolib import midrange, morph
 from .common import (
     # binarize,
-    pil2array, array2pil
-)
-
-TOOL = 'ocrd-cis-ocropy-clip'
+    array2pil, determine_zoom, pil2array)
 
 class OcropyClip(Processor):
+    logger: Logger
 
-    def __init__(self, *args, **kwargs):
-        self.ocrd_tool = get_ocrd_tool()
-        kwargs['ocrd_tool'] = self.ocrd_tool['tools'][TOOL]
-        kwargs['version'] = self.ocrd_tool['version']
-        super(OcropyClip, self).__init__(*args, **kwargs)
+    @property
+    def executable(self):
+        return 'ocrd-cis-ocropy-clip'
+
+    def setup(self):
+        self.logger = getLogger('processor.OcropyClip')
 
     def process(self):
         """Clip text regions / lines of the workspace at intersections with neighbours.
@@ -76,13 +74,10 @@ def process(self):
         # too. However, region-level clipping _must_ be run before region-level
         # deskewing, because that would make segments incomensurable with their
         # neighbours.
-        LOG = getLogger('processor.OcropyClip')
         level = self.parameter['level-of-operation']
-        assert_file_grp_cardinality(self.input_file_grp, 1)
-        assert_file_grp_cardinality(self.output_file_grp, 1)
 
         for (n, input_file) in enumerate(self.input_files):
-            LOG.info("INPUT FILE %i / %s", n, input_file.pageId or input_file.ID)
+            self.logger.info("INPUT FILE %i / %s", n, input_file.pageId or input_file.ID)
             file_id = make_file_id(input_file, self.output_file_grp)
 
             pcgts = page_from_file(self.workspace.download_file(input_file))
@@ -92,16 +87,8 @@ def process(self):
 
             page_image, page_coords, page_image_info = self.workspace.image_from_page(
                 page, page_id, feature_selector='binarized')
-            if self.parameter['dpi'] > 0:
-                zoom = 300.0/self.parameter['dpi']
-            elif page_image_info.resolution != 1:
-                dpi = page_image_info.resolution
-                if page_image_info.resolutionUnit == 'cm':
-                    dpi *= 2.54
-                LOG.info('Page "%s" uses %f DPI', page_id, dpi)
-                zoom = 300.0/dpi
-            else:
-                zoom = 1
+            zoom = determine_zoom(self.parameter['dpi'], page_image_info)
+            self.logger.info(f"Page '{page_id}' uses {self.parameter['dpi']} DPI")
 
             # FIXME: what about text regions inside table regions?
             regions = list(page.get_TextRegion())
@@ -120,7 +107,7 @@ def process(self):
                 page.get_TableRegion() +
                 page.get_UnknownRegion())
             if not num_texts:
-                LOG.warning('Page "%s" contains no text regions', page_id)
+                self.logger.warning('Page "%s" contains no text regions', page_id)
             background = ImageStat.Stat(page_image)
             # workaround for Pillow#4925
             if len(background.bands) > 1:
@@ -151,7 +138,7 @@ def process(self):
                 if level == 'region':
                     if region.get_AlternativeImage():
                         # FIXME: This should probably be an exception (bad workflow configuration).
-                        LOG.warning('Page "%s" region "%s" already contains image data: skipping',
+                        self.logger.warning('Page "%s" region "%s" already contains image data: skipping',
                                     page_id, region.id)
                         continue
                     shape = prep(shapes[i])
@@ -169,7 +156,7 @@ def process(self):
                 # level == 'line':
                 lines = region.get_TextLine()
                 if not lines:
-                    LOG.warning('Page "%s" region "%s" contains no text lines', page_id, region.id)
+                    self.logger.warning('Page "%s" region "%s" contains no text lines', page_id, region.id)
                     continue
                 region_image, region_coords = self.workspace.image_from_segment(
                     region, page_image, page_coords, feature_selector='binarized')
@@ -187,7 +174,7 @@ def process(self):
                 for j, line in enumerate(lines):
                     if line.get_AlternativeImage():
                         # FIXME: This should probably be an exception (bad workflow configuration).
-                        LOG.warning('Page "%s" region "%s" line "%s" already contains image data: skipping',
+                        self.logger.warning('Page "%s" region "%s" line "%s" already contains image data: skipping',
                                     page_id, region.id, line.id)
                         continue
                     shape = prep(shapes[j])
@@ -203,7 +190,7 @@ def process(self):
                                              input_file.pageId, file_id + '_' + region.id + '_' + line.id)
 
             # update METS (add the PAGE file):
-            file_path = os.path.join(self.output_file_grp, file_id + '.xml')
+            file_path = join(self.output_file_grp, file_id + '.xml')
             pcgts.set_pcGtsId(file_id)
             out = self.workspace.add_file(
                 ID=file_id,
@@ -212,13 +199,12 @@ def process(self):
                 local_filename=file_path,
                 mimetype=MIMETYPE_PAGE,
                 content=to_xml(pcgts))
-            LOG.info('created file ID: %s, file_grp: %s, path: %s',
+            self.logger.info('created file ID: %s, file_grp: %s, path: %s',
                      file_id, self.output_file_grp, out.local_filename)
 
     def process_segment(self, segment, segment_mask, segment_polygon, neighbours,
                         background_image, parent_image, parent_coords, parent_bin,
                         page_id, file_id):
-        LOG = getLogger('processor.OcropyClip')
         # initialize AlternativeImage@comments classes from parent, except
         # for those operations that can apply on multiple hierarchy levels:
         features = ','.join(
@@ -230,7 +216,7 @@ def process_segment(self, segment, segment_mask, segment_polygon, neighbours,
         segment_bbox = bbox_from_polygon(segment_polygon)
         for neighbour, neighbour_mask in neighbours:
             if not np.any(segment_mask > neighbour_mask):
-                LOG.info('Ignoring enclosing neighbour "%s" of segment "%s" on page "%s"',
+                self.logger.info('Ignoring enclosing neighbour "%s" of segment "%s" on page "%s"',
                          neighbour.id, segment.id, page_id)
                 continue
             # find connected components that (only) belong to the neighbour:
@@ -240,7 +226,7 @@ def process_segment(self, segment, segment_mask, segment_polygon, neighbours,
             num_foreground = np.count_nonzero(segment_mask * parent_bin)
             if not num_intruders:
                 continue
-            LOG.debug('segment "%s" vs neighbour "%s": suppressing %d of %d pixels on page "%s"',
+            self.logger.debug('segment "%s" vs neighbour "%s": suppressing %d of %d pixels on page "%s"',
                       segment.id, neighbour.id, num_intruders, num_foreground, page_id)
             # suppress in segment_mask so these intruders can stay in the neighbours
             # (are not removed from both sides)

ocrd_cis/ocropy/common.py

@@ -10,7 +10,7 @@
 from skimage.morphology import medial_axis
 import networkx as nx
 from PIL import Image
-
+from ocrd_models import OcrdExif
 from . import ocrolib
 from .ocrolib import morph, psegutils, sl
 # for decorators (type-checks etc):
@@ -2102,3 +2102,15 @@ def find_topological():
     #     rlabels[region_hull] = region
     # DSAVE('rlabels_closed', rlabels)
     return rlabels
+
+def determine_zoom(dpi: float, page_image_info: OcrdExif) -> float:
+    if dpi > 0:
+        zoom = 300.0/dpi
+    elif page_image_info.resolution != 1:
+        dpi = page_image_info.resolution
+        if page_image_info.resolutionUnit == 'cm':
+            dpi *= 2.54
+        zoom = 300.0/dpi
+    else:
+        zoom = 1
+    return zoom

ocrd_cis/ocropy/denoise.py

@@ -1,11 +1,10 @@
 from __future__ import absolute_import
-
-import os.path
+from logging import Logger
+from os.path import join
 
 from ocrd_utils import (
     getLogger,
     make_file_id,
-    assert_file_grp_cardinality,
     MIMETYPE_PAGE
 )
 from ocrd_modelfactory import page_from_file
@@ -14,20 +13,19 @@
 )
 from ocrd import Processor
 
-from .. import get_ocrd_tool
 from .common import (
     # binarize,
-    remove_noise)
-
-TOOL = 'ocrd-cis-ocropy-denoise'
+    determine_zoom, remove_noise)
 
 class OcropyDenoise(Processor):
+    logger: Logger
+
+    @property
+    def executable(self):
+        return 'ocrd-cis-ocropy-denoise'
 
-    def __init__(self, *args, **kwargs):
-        self.ocrd_tool = get_ocrd_tool()
-        kwargs['ocrd_tool'] = self.ocrd_tool['tools'][TOOL]
-        kwargs['version'] = self.ocrd_tool['version']
-        super(OcropyDenoise, self).__init__(*args, **kwargs)
+    def setup(self):
+        self.logger = getLogger('processor.OcropyDenoise')
 
     def process(self):
         """Despeckle the pages / regions / lines of the workspace.
@@ -50,13 +48,10 @@ def process(self):
 
         Produce a new output file by serialising the resulting hierarchy.
         """
-        LOG = getLogger('processor.OcropyDenoise')
         level = self.parameter['level-of-operation']
-        assert_file_grp_cardinality(self.input_file_grp, 1)
-        assert_file_grp_cardinality(self.output_file_grp, 1)
 
         for (n, input_file) in enumerate(self.input_files):
-            LOG.info("INPUT FILE %i / %s", n, input_file.pageId or input_file.ID)
+            self.logger.info("INPUT FILE %i / %s", n, input_file.pageId or input_file.ID)
             file_id = make_file_id(input_file, self.output_file_grp)
 
             pcgts = page_from_file(self.workspace.download_file(input_file))
@@ -67,24 +62,17 @@ def process(self):
             page_image, page_xywh, page_image_info = self.workspace.image_from_page(
                 page, page_id,
                 feature_selector='binarized' if level == 'page' else '')
-            if self.parameter['dpi'] > 0:
-                zoom = 300.0/self.parameter['dpi']
-            elif page_image_info.resolution != 1:
-                dpi = page_image_info.resolution
-                if page_image_info.resolutionUnit == 'cm':
-                    dpi *= 2.54
-                LOG.info('Page "%s" uses %f DPI', page_id, dpi)
-                zoom = 300.0/dpi
-            else:
-                zoom = 1
+
+            zoom = determine_zoom(self.parameter['dpi'], page_image_info)
+            self.logger.info(f"Page '{page_id}' uses {self.parameter['dpi']} DPI")
 
             if level == 'page':
                 self.process_segment(page, page_image, page_xywh, zoom,
                                      input_file.pageId, file_id)
             else:
                 regions = page.get_AllRegions(classes=['Text'], order='reading-order')
                 if not regions:
-                    LOG.warning('Page "%s" contains no text regions', page_id)
+                    self.logger.warning('Page "%s" contains no text regions', page_id)
                 for region in regions:
                     region_image, region_xywh = self.workspace.image_from_segment(
                         region, page_image, page_xywh,
@@ -95,7 +83,7 @@ def process(self):
                         continue
                     lines = region.get_TextLine()
                     if not lines:
-                        LOG.warning('Page "%s" region "%s" contains no text lines', page_id, region.id)
+                        self.logger.warning('Page "%s" region "%s" contains no text lines', page_id, region.id)
                     for line in lines:
                         line_image, line_xywh = self.workspace.image_from_segment(
                             line, region_image, region_xywh,
@@ -105,7 +93,7 @@ def process(self):
                                              file_id + '_' + region.id + '_' + line.id)
 
             # update METS (add the PAGE file):
-            file_path = os.path.join(self.output_file_grp, file_id + '.xml')
+            file_path = join(self.output_file_grp, file_id + '.xml')
             pcgts.set_pcGtsId(file_id)
             out = self.workspace.add_file(
                 ID=file_id,
@@ -114,15 +102,14 @@ def process(self):
                 local_filename=file_path,
                 mimetype=MIMETYPE_PAGE,
                 content=to_xml(pcgts))
-            LOG.info('created file ID: %s, file_grp: %s, path: %s',
+            self.logger.info('created file ID: %s, file_grp: %s, path: %s',
                      file_id, self.output_file_grp, out.local_filename)
 
     def process_segment(self, segment, segment_image, segment_xywh, zoom, page_id, file_id):
-        LOG = getLogger('processor.OcropyDenoise')
         if not segment_image.width or not segment_image.height:
-            LOG.warning("Skipping '%s' with zero size", file_id)
+            self.logger.warning("Skipping '%s' with zero size", file_id)
             return
-        LOG.info("About to despeckle '%s'", file_id)
+        self.logger.info("About to despeckle '%s'", file_id)
         bin_image = remove_noise(segment_image,
                                  maxsize=self.parameter['noise_maxsize']/zoom*300/72) # in pt
         # update METS (add the image file):