Update TextProcessor.py

TextProcessor.py

@@ -29,6 +29,7 @@ def paragraphs(folderpath):
         baselines[i] = min(pg_x1s)
         rights[i] = max(pg_x2s)
 
+    is_inblock = False
     for (j, line_text) in enumerate(line_texts):
         if j == 0:
             para = []
@@ -38,7 +39,8 @@ def paragraphs(folderpath):
             prior_median = (baselines[pg_inds[j - 1]] + rights[pg_inds[j - 1]]) / 2
             current_median = (baselines[pg_inds[j]] + rights[pg_inds[j]]) / 2
 
-            prior_endswith_period = re.search('[:\.]([^A-z]{0,2})$',line_texts[j - 1].strip()) is not None  # Include colon?
+            prior_endswith_period = re.search('[:\.]([^A-z]{0,2})$',
+                                              line_texts[j - 1].strip()) is not None  # Include colon?
             prior_is_section_header = re.search('^([ABCDEIVX]+)$', line_texts[j - 1].strip()) is not None
             current_is_section_header = re.search('^([ABCDEIVX]+)$', line_texts[j].strip()) is not None
             prior_is_asterisk = re.search('^([\s\*]+)$', line_texts[j - 1].strip()) is not None
@@ -57,7 +59,12 @@ def paragraphs(folderpath):
             is_start_blockquote = (prior_endswith_period and current_supertabbed and prior_more_left)
             is_after_blockquote = (prior_endswith_period and not current_supertabbed and prior_supertabbed)
             is_after_disposition = (prior_right_margin and current_tabbed)
-            is_after_oneline_paragraph = (prior_tabbed and current_tabbed and not prior_supertabbed and not current_supertabbed)
+            is_after_oneline_paragraph = (prior_tabbed and current_tabbed and not prior_supertabbed and not current_supertabbed and not is_inblock)
+
+            if is_start_blockquote:
+                is_inblock = True
+            if is_after_blockquote:
+                is_inblock = False
 
             if is_section_header or is_the_classic or is_after_oneline_paragraph or is_start_blockquote or is_after_blockquote or is_after_disposition:
                 paras.append(para)
@@ -88,6 +95,4 @@ def process_file(folderpath):
         x1, y1, x2, y2, para_first_line, pg_ind = indent
         image = cv2.imread(folderpath + '/' + str(pg_ind) + '-processed.png')
         cv2.circle(image, (x1 - 15, int(0.5 * (y1 + y2))), radius=1, color=(240, 32, 160), thickness=2)
-        cv2.imwrite(folderpath + '/' + str(pg_ind) + '-processed.png', image)
-
-process_file('PDF Cases/462_122')
+        cv2.imwrite(folderpath + '/' + str(pg_ind) + '-processed.png', image)