Initial parsing script

parse_scandata.py

@@ -0,0 +1,98 @@
+#!/usr/bin/env python3.6
+
+import math
+import sys
+import os
+import xmltodict
+
+#from pprint import pprint
+
+from os.path import join, exists
+
+from PIL import Image, ImageDraw
+
+# find -iname \*.jp2 -exec opj_decompress -i {} -o {}.ppm \;
+
+BASE_PATH = sys.argv[1]
+ITEM_ID = sys.argv[2]
+FILTER_PAGE_NR = None
+
+if len(sys.argv) > 3:
+    # ex: 0020, or ANY
+    FILTER_PAGE_NR = sys.argv[3]
+
+if len(sys.argv) > 4:
+    OUTPATH = sys.argv[4]
+
+r = xmltodict.parse(open(join(BASE_PATH, ITEM_ID, ITEM_ID + '_scandata.xml'), 'rb'))
+
+relevant = r['book']['pageData']['page']
+
+#pprint(relevant)
+#PIL.ImageDraw.Draw.polygon(xy, fill=None, outline=None)
+
+for r in relevant:
+
+    # Only want 'Normal' pagetype, for now
+    pagetype = r['pageType']
+    if pagetype != 'Normal':
+        continue
+
+    orig_file_name = r['origFileName']
+    rotation = int(r['rotateDegree'])
+    skew = float(r['skewAngle'])
+    apply_skew = r['skewActive']
+    crop = r['cropBox']
+    crop_x, crop_y, crop_w, crop_h = int(crop['x']), int(crop['y']), int(crop['w']), int(crop['h'])
+
+    if FILTER_PAGE_NR and FILTER_PAGE_NR != 'ANY' and orig_file_name[0:4] != FILTER_PAGE_NR:
+        continue
+
+    #print(pagetype, orig_file_name, rotation, skew, apply_skew, crop_x, crop_y, crop_w, crop_h)
+
+    skew_r = math.radians(-skew)
+
+    if apply_skew:
+        top_l = (crop_x, crop_y)
+        top_r = (crop_x + crop_w * math.cos(skew_r), crop_y + crop_w * math.sin(skew_r))
+        bottom_l = (crop_x + crop_h * math.sin(skew_r), crop_y + crop_h * math.cos(skew_r))
+        bottom_r = (top_r[0], bottom_l[1] + crop_w * math.sin(skew_r))
+    else:
+        top_l = (crop_x, crop_y)
+        top_r = (crop_x + crop_w, crop_y)
+        bottom_l = (crop_x, crop_y + crop_h)
+        bottom_r = (top_r[0], bottom_l[1])
+
+    coords = [str(int(x)) for x in top_l + top_r + bottom_r + bottom_l]
+
+    data = [ITEM_ID + '_orig_' + orig_file_name.replace('.jpg', '.jp2.ppm')]
+    data.extend(coords)
+    print(','.join(data))
+
+    if OUTPATH:
+        outbase = join(OUTPATH, ITEM_ID)
+        if not exists(outbase):
+            os.makedirs(outbase)
+        i = Image.open(join(BASE_PATH + ITEM_ID, ITEM_ID + '_orig_jp2', ITEM_ID + '_orig_' + orig_file_name.replace('.jpg', '.jp2')))
+        #i = Image.open(join(BASE_PATH + ITEM_ID, ITEM_ID + '_orig_jp2', ITEM_ID + '_orig_' + orig_file_name.replace('.jpg', '.jp2.ppm')))
+        i = i.rotate(-rotation, expand=True)
+        iw, ih = i.size
+
+        p = join(outbase, orig_file_name).replace('.jpg', '.orig.jpg')
+        print(p)
+        isave=i.copy()
+        isave.thumbnail((iw/4, ih/4), Image.ANTIALIAS)
+        isave.save(p)
+        #i.save(p)
+
+        idraw = ImageDraw.Draw(i)
+        idraw.rectangle([0, 0, iw, ih], outline=0, fill=0)
+        idraw.polygon([top_l, top_r, bottom_r, bottom_l], outline=255, fill=255)
+
+        p = join(outbase, orig_file_name).replace('.jpg', '.coloured.jpg')
+        print(p)
+
+        isave=i.copy()
+        isave.thumbnail((iw/4, ih/4), Image.ANTIALIAS)
+        isave.save(p)
+        #i.save(p)