Merge pull request #32 from breezedeus/dev

fix: results from `self.recognize_by_clf` have no `line_number`

Makefile

@@ -2,7 +2,7 @@ package:
 	rm -rf build
 	python setup.py sdist bdist_wheel
 
-VERSION = 0.2.3
+VERSION = 0.2.3.1
 upload:
 	python -m twine upload  dist/pix2text-$(VERSION)* --verbose
 

RELEASE.md

@@ -1,5 +1,10 @@
 # Release Notes
 
+## Update 2023.07.14：发布 **V0.2.3.1**
+
+主要变更：
+* 修复了 `self.recognize_by_clf` 返回结果中不包含 `line_number` 字段导致 `merge_line_texts` 报错的bug。
+
 ## Update 2023.07.03：发布 **V0.2.3**
 
 主要变更：

pix2text/__version__.py

@@ -1,4 +1,4 @@
 # coding: utf-8
 # Copyright (C) 2022-2023, [Breezedeus](https://www.breezedeus.com).
 
-__version__ = '0.2.3'
+__version__ = '0.2.3.1'

pix2text/utils.py

@@ -420,19 +420,20 @@ def merge_line_texts(
     line_margin_list = []  # 每行的最左边和左右边的x坐标
     isolated_included = []  # 每行是否包含了 `isolated` 类型的数学公式
     for o in out:
-        if len(out_texts) <= o['line_number']:
+        line_number = o.get('line_number', 0)
+        if len(out_texts) <= line_number:
             out_texts.append([])
             line_margin_list.append([0, 0])
             isolated_included.append(False)
-        out_texts[o['line_number']].append(o['text'])
-        line_margin_list[o['line_number']][1] = max(
-            line_margin_list[o['line_number']][1], float(o['position'][2, 0])
+        out_texts[line_number].append(o['text'])
+        line_margin_list[line_number][1] = max(
+            line_margin_list[line_number][1], float(o['position'][2, 0])
         )
-        line_margin_list[o['line_number']][0] = min(
-            line_margin_list[o['line_number']][0], float(o['position'][0, 0])
+        line_margin_list[line_number][0] = min(
+            line_margin_list[line_number][0], float(o['position'][0, 0])
         )
         if o['type'] == 'isolated':
-            isolated_included[o['line_number']] = True
+            isolated_included[line_number] = True
 
     line_text_list = [smart_join(o) for o in out_texts]