fix page to png

@@ -17,8 +17,10 @@ args = parser.parse_args()
 LOG_BASE = '[pdf to img]'
 LOG_BASE = '[pdf to img]'
 
 
 # 页面保存为png图片参数
 # 页面保存为png图片参数
-ZOOM_X = ZOOM_Y = 2.0
-trans = fitz.Matrix(ZOOM_X, ZOOM_X).preRotate(0)  # zoom factor 2 in each dimension
+ZOOM_X_1 = ZOOM_Y_1 = 1.0
+ZOOM_X_2 = ZOOM_Y_2 = 2.0
+trans_1 = fitz.Matrix(ZOOM_X_1, ZOOM_X_1).preRotate(0)  # zoom factor 1 in each dimension
+trans_2 = fitz.Matrix(ZOOM_X_2, ZOOM_X_2).preRotate(0)  # zoom factor 2 in each dimension
 
 
 # 特殊filter处理
 # 特殊filter处理
 ADOBE_FILTER_SET = {'FlateDecode', 'JPXDecode', 'JBIG2Decode'}
 ADOBE_FILTER_SET = {'FlateDecode', 'JPXDecode', 'JBIG2Decode'}
@@ -38,13 +40,17 @@ class PDFHandler:
         self.img_dir_path = os.path.join(target_path, os.path.splitext(os.path.basename(path))[0])
         self.img_dir_path = os.path.join(target_path, os.path.splitext(os.path.basename(path))[0])
         self.xref_set = set()
         self.xref_set = set()
 
 
-    def get_img_save_path(self, pno, img_index=0, ext='png'):
+    def get_img_save_path(self, pno, img_index=0, ext='jpeg'):
         return os.path.join(self.img_dir_path, 'page_{0}_img_{1}.{2}'.format(pno, img_index, ext))
         return os.path.join(self.img_dir_path, 'page_{0}_img_{1}.{2}'.format(pno, img_index, ext))
 
 
     def page_to_png(self, page):
     def page_to_png(self, page):
-        pm = page.getPixmap(matrix=trans, alpha=False)
+        if page.MediaBoxSize.x > 1500 or page.MediaBoxSize.y > 1500:
+            pm = page.getPixmap(matrix=trans_1, alpha=False)
+        else:
+            pm = page.getPixmap(matrix=trans_2, alpha=False)
         img_save_path = self.get_img_save_path(page.number)
         img_save_path = self.get_img_save_path(page.number)
-        pm.writePNG(img_save_path)
+        pm.writeImage(img_save_path)
+        # pm.writePNG(img_save_path)
 
 
     @staticmethod
     @staticmethod
     def getimage(pix):
     def getimage(pix):
@@ -247,7 +253,8 @@ def main():
         failed_list = []
         failed_list = []
         for parent, dirnames, filenames in os.walk(pdf_path):
         for parent, dirnames, filenames in os.walk(pdf_path):
             # 图片保存目录
             # 图片保存目录
-            target_path = os.path.realpath(args.output) if args.output else parent
+            # target_path = os.path.realpath(args.output) if args.output else parent
+            target_path = parent.replace(pdf_path, args.output)
             for pdf_file in filenames:
             for pdf_file in filenames:
                 if not pdf_file.endswith('pdf') and not pdf_file.endswith('PDF'):
                 if not pdf_file.endswith('pdf') and not pdf_file.endswith('PDF'):
                     continue
                     continue