update ui, fix a bug that can not save cropped pdf to target directory

.gitignore

@@ -4,6 +4,8 @@ __pycache__/
 *.py[cod]
 *$py.class
 
+config.json
+
 # C extensions
 *.so
 

cropper.py

@@ -16,6 +16,12 @@ def clip(value, min_value, max_value):
             value = max_value
     return value
 
+def parse_workdir(filepath):
+    workdir = '/'.join('/'.join(filepath.split('\\')).split('/')[:-1])
+    if workdir == '':
+        workdir = '.'
+    return workdir
+
 def crop(args):
     infile = args.input
     # process filepath
@@ -31,9 +37,8 @@ def crop(args):
     thresh = args.thresh
 
     # parse working dir.
-    workdir = '/'.join('/'.join(infile.split('\\')).split('/')[:-1])
-    if workdir == '':
-        workdir = '.'
+    inworkdir = parse_workdir(infile)
+    outworkdir = parse_workdir(outfile)
 
     if infile.split('.')[-1].lower() in ['ppt', 'pptx']:
         # lets process pptx.
@@ -46,7 +51,7 @@ def crop(args):
         print('convert {0} to {1}'.format(infile, inbase + '.pdf'))
         slides.SaveAs(inbase + '.pdf', 32)
         slides.close()
-        powerpoint.Quit()
+        # powerpoint.Quit()
         # read comments.
         ppt = Presentation(infile)
         slides = ppt.slides
@@ -119,7 +124,7 @@ def crop(args):
             names = ['{0}_{1}.pdf'.format(outbase, i+1) for i in range(len(pdf))]
         if args.names is not None:
             names = args.names
-            namesfile = os.path.join(workdir, names)
+            namesfile = os.path.join(inworkdir, names)
             if os.path.isfile(namesfile) and names.split('.')[-1] == 'txt':
                 with open(namesfile, 'r', encoding='utf-8') as f:
                     names = f.read()
@@ -131,7 +136,7 @@ def crop(args):
             names = [n + '.pdf' if n[:-4] != '.pdf' else n for n in names]
         if len(names) < len(pdf):
             names += ['{0}_{1}.pdf'.format(outbase, i+1) for i in range(len(names), len(pdf))]
-        names = [os.path.join(workdir, name) for name in names]
+        names = [os.path.join(outworkdir, name) for name in names]
         for i, (name, page) in enumerate(zip(names, pdf)):
             page_pdf = fitz.Document()
             page_pdf.insertPDF(pdf, from_page=i, to_page=i)

main.py

@@ -86,10 +86,34 @@ def parse_and_crop(conf):
 # parser.add_argument('--mute', '-m', default=False, action='store_true', help='do not display output file path.')
 # args = parser.parse_args()
 
+
+import os
+import json
 from ui import Loader
+defaults = {
+    "input": "",
+    "output": "",
+    "background_color": "255,255,255",
+    "border": "0.0",
+    "zoom": "1.0",
+    "thresh": "1.0",
+    "split": "true",
+    "names": "",
+    "visual": "false",
+    "mute": "false"
+}
+config_path = 'config.json'
+if os.path.isfile(config_path):
+    with open(config_path, 'r', encoding='utf-8') as f:
+        updater = json.load(f)
+        defaults.update(updater)
+else:
+    with open(config_path, 'w+', encoding='utf-8') as f:
+        json.dump(defaults, f, ensure_ascii=False, indent=4)
+
 conf = {
         "input": {"name": "源文件", "type": "readfile", "extension": ("PDF & PPT", ".pdf .pptx")},
-        "output": {"name": "保存路径", "type": "savefile"},
+        "output": {"name": "保存路径", "type": "savefile", "initial": "output.pdf"},
         "background_color": {"name": "背景颜色", "type": "color", "default": "255,255,255"},
         "border": {"name": "留白", "type": "str", "default": "0.0"},
         "zoom": {"name": "缩放等级", "type": "str", "default": "1.0"},
@@ -99,6 +123,28 @@ conf = {
         "visual": {"name": "显示裁切框", "type": "str", "default": "false"},
         "mute": {"name": "显示保存文件", "type": "str", "default": "false"},
     }
+
+introduction = [
+"==== 使用说明 ====",
+"- 源文件：输入的文件，可以为PDF或者PPT，如果是PPT，会首先自动调用PowerPoint将给定的PPT转变为PDF然后进行裁剪（PPT裁剪第一次需要调用COM对象因此速度稍慢，建议耐心等待，裁切一次后速度会恢复到正常水平",
+"- 保存路径：保存的文件名，仅适用于非拆分模式。如果留空，则保存路径为源文件所在目录，保存的文件名为\"源文件名_crop.pdf\"",
+"- 背景颜色：哪一种颜色会被认为是背景",
+"- 留白：不紧贴有效内容裁切，预留一个给定大小的白边。",
+"- 缩放等级：裁切是基于视觉裁切的，因此裁切过程中会首先渲染PDF，采用默认值即可，更高的缩放会将PDF渲染为更高分辨率的图片从而提高裁剪精度，但计算时间也会相应增加。",
+"- 阈值：像素值差异多少会被认为是前景。",
+"- 拆分：是否将源文件自动拆分为每页一个的单个文件。默认文件名采取下划线+数字命名。如果源文件是PPT且PPT备注不为空，则首先采用PPT的备注作为当前页保存的文件名。",
+"- 页名称：手动指定拆分模式下每一页的名称，具有最高优先级，用逗号分隔。留空则使用默认的文件名或者PPT中备注的文件名。",
+"- 显示裁切框：在右侧的日志区域输出裁切框坐标",
+"- 显示保存文件：显示保存的文件名。",
+]
+help_msg = '\n'.join(introduction)
+
+# write defaults.
+for key in defaults:
+    conf[key]['default'] = defaults[key]
 root = tk.Tk()
-Loader(master=root, conf=conf, execution=parse_and_crop, title="PDF/PPT自动裁边")
+Loader(master=root, conf=conf, execution=parse_and_crop, title="PDF/PPT自动裁边", help_msg=help_msg)
+
+
+print(help_msg)
 root.mainloop()

ui.py

@@ -29,7 +29,7 @@ def text_on_image(img, text, pos, size=16, color='#000000'):
     return img
 
 class Loader(tk.Frame):
-    def __init__(self, master=None, conf={}, execution=None, title=None):
+    def __init__(self, master=None, conf={}, execution=None, title=None, help_msg=None):
         super().__init__(master)
         self.master = master
         self.conf = conf
@@ -44,7 +44,7 @@ class Loader(tk.Frame):
         # self.font = tk.Font(family='', size=40,weight='',slant='',underline='',overstrike='')
         if self.title is not None:
             self.master.title(self.title)
-    
+        self.help_msg = help_msg
     def add_resource(self, obj, name):
         if name in self.resources:
             print('Name [{0}] Repeated.'.format(name))
@@ -137,6 +137,12 @@ class Loader(tk.Frame):
                 conf[name] = self.resources[key].get()
         self.execution(conf)
     
+    def clear(self):
+        self.resources['textbox_log'].delete("1.0", 'end')
+    
+    def showhelp(self):
+        print(self.help_msg)
+    
     def create_widgets(self):
         self.rowcount = 0
         for key, value in self.conf.items():
@@ -153,7 +159,7 @@ class Loader(tk.Frame):
             destination = 'entry_' + key
             if dtype in ['readfile', 'savefile', 'directory']:
                 extension = value.get('extension', None)
-                initialfile = default
+                initialfile = value.get('initial', None)
                 args = {
                     "dst": destination,
                     "choose_type": dtype,
@@ -161,7 +167,7 @@ class Loader(tk.Frame):
                     "initial_file": initialfile
                 }
                 self.add_button(
-                    text="选择文件",
+                    text="<选择文件",
                     command=lambda args=args:self.choose_file(**args),
                     size=(16, 16),
                     layout={"row":self.rowcount, "column": 2, "padx": 5},
@@ -171,7 +177,7 @@ class Loader(tk.Frame):
                 self.conf[key]['type'] = 'str'
             elif dtype == 'color':
                 self.add_button(
-                    text="选择颜色",
+                    text="<选择颜色",
                     command=lambda dst=destination:self.choose_color(dst),
                     size=(16, 16),
                     layout={"row":self.rowcount, "column": 2, "padx": 5},
@@ -181,15 +187,29 @@ class Loader(tk.Frame):
 
             self.rowcount += 1
         self.add_button(
-            text="         裁剪           ",
+            text="清空输出>",
+            command=self.clear,
+            size=(16, 16),
+            layout={"row":self.rowcount-2, "column": 2, "pady": 5, "sticky": "e"},
+            name='clear_output'
+        )
+        self.add_button(
+            text="显示帮助>",
+            command=self.showhelp,
+            size=(16, 16),
+            layout={"row":self.rowcount-1, "column": 2, "pady": 5, "sticky": "e"},
+            name='show_help'
+        )
+        self.add_button(
+            text="        开始裁剪        ",
             command=self.execute,
             size=(16, 16),
-            layout={"row":self.rowcount, "column": 0, "columnspan": 2, "pady": 5},
+            layout={"row":self.rowcount, "column": 0, "columnspan": "3", "pady": 5},
             name='execute_final'
         )
         self.add_textbox(
             width=40, 
-            layout={"row":0, "column": 3, "rowspan": self.rowcount, "padx": 5, "pady": 0},
+            layout={"row":0, "column": 3, "rowspan": self.rowcount + 1, "padx": 5, "pady": 5, "sticky": "ns"},
             name="textbox_log"
         )
         sys.stdout = StdSimulator(self.resources['textbox_log'])