feat:pdftojson py 修改

30.PdfToJson/PocTxt_PY/pocTXTHandler.py

@@ -158,8 +158,10 @@ class poc:
         self.file_name = re.sub('.html', '', file)
         self.data = {
 # =============================================================================
+# =============================================================================
 #             '文件名': re.sub('.html', '', file),
 # =============================================================================
+# =============================================================================
             '发行人': self.issuer(),
 # =============================================================================
 #             '发行人': file.split('20')[0],
@@ -400,23 +402,27 @@ class poc:
         else:
             ret = ''
         ret = re.sub('[：,，:“”【】 为]', '', ret)
-        if re.search('[0-9]', ret):
+        if re.search('[0-9]+', ret):
             if '亿' in ret:
                 ret = re.sub(u'[\u4e00-\u9fff]', '', ret)
                 try:
                     ret = str(float(ret) * 100000000)
                 except:
                     ret = ''
+            else:
+                ret = re.sub(u'[\u4e00-\u9fff]', '', ret)
+                if len(ret) > 12:
+                    ret = ''
         else:
             ret = re.findall(r'[一壹二贰三叁四肆五伍六陆七柒八捌九玖十拾百佰千仟万萬亿億]+', ret)
             if ret:
                 ret = ret[0]
             else:
                 ret = ''
-        if not re.search('[0-9]', ret):
+        if not re.search('[0-9]+', ret):
             try:
                 ret = str(chinese_to_arabic(ret))
-            except Exception as err:
+            except:
                 pass
         if re.match(r'\d+',ret):   
             return str(round(float(ret),2))
@@ -500,36 +506,39 @@ class poc:
                     start, end = s.span()
                     ret = t[(start+1):end]
                     try:
-                        if (t[end] == '-') or (t[end] == '至') or (t[end] == '、') or (t[end] == '—') or (t[end] == '及'):
+                        if end < len(t) and ((t[end] == '-') or (t[end] == '至') or (t[end] == '、') or (t[end] == '—') or (t[end] == '及')):
                             a = re.findall('[^\-至、—]*?日', t[(end+1):])
                             if not a:
                                 ret = re.sub('：|【|】|期|指|〔|〕| |为|\[|\]|\(|\)|（|）|发行|变更', '', ret)
+                                print ('aa')
                                 return ret
                             ret = ret+'至'+a[0]
-# =============================================================================
-#                         if re.match([0-9],ret) is False:
-#                             return ''
-# =============================================================================
+    # =============================================================================
+    #                         if re.match([0-9],ret) is False:
+    #                             return ''
+    # =============================================================================
                         ret = re.sub('：|【|】|期|指|〔|〕| |为|\[|\]|\(|\)|（|）|发行|变更', '', ret)
                         tmp_file_name = self.file_name
                         tmp_file_name = re.sub(r'\(.{2}\)','',tmp_file_name)
                         tmp_file_name = re.sub(r'（.{2}）','',tmp_file_name)
                         tmp_file_name = re.sub(r'\.txt','',tmp_file_name)
                         tmp_file_name = re.sub(r'期','',tmp_file_name)
-                        ret = re.sub(tmp_file_name,'',ret)
+                        if tmp_file_name in ret:
+                            print('########')
+                        ret = ret.replace(tmp_file_name,'')
                         ret = re.sub('起息日','',ret)
-                        ret=  ret.split(':')[-1]
-                        if ret[:0] =='日':
-                            ret =ret[1:]
-                        if re.findall('[0-9]+',ret):
-                            return ret[1:]
+                        ret = ret.split(':')[-1]
+                        if re.search('[0-9]+',ret):
+                            if ret[0] == '日':
+                                return ret[1:]
+                            else:
+                                return ret
                         else:
                             return ''
-                    except Exception as err:
-                        print(err)
+                    except:
+                        pass
                 else:
                     print('warning:', t)
-
                 return ret
 
     def value_date(self):
@@ -584,8 +593,10 @@ if __name__ == '__main__':
 
     # single instance:
 
+# =============================================================================
+#     folder= './full'
+# =============================================================================
     folder= 'D:\\qqFILE\\result\\full'
-    folder2= 'D:\\data\\full'
     filelist = os.listdir(folder)
     filelist = [file for file in filelist if re.search('.txt', file)]
     # file = filelist[0]