无法让pyinstaller使用--onefile和--noConsole

Question

我有一个gui程序，我用它来处理扫描的pdf，并寻找带有二维码的页面，这告诉程序在哪里可以将pdf分成多个文件。我的程序本质上是一个函数列表，当每个步骤完成时，这些函数会一个接一个地相互调用。

我遇到麻烦的地方是当我试图用PyInstaller将它编译成exe时。我的程序作为python文件运行良好，并且在指定--noConsole和--onefile选项时也能正常工作。我不能工作的是当我两者都用的时候。出于某种原因，试图同时使用--onefile和--noConsole获得一个工作程序不可避免地会失败。它的靴子很好，但总是在同一部分失败。以下是失败部分的代码：

import PySimpleGUI as sg
import shutil
import pathlib
from pyzbar.pyzbar import decode
from pyzbar.pyzbar import ZBarSymbol
from json import loads
from pdf2image import convert_from_path
import fitz
import win32com.client
import win32timezone
import os
import os.path
from os import path
from datetime import datetime, timedelta
from sys import exit
import tempfile

def scan_reader(filepath):
    '''takes a filepath pointing to a pdf file and returns the number of pages, as well as any QR data that could be pulled from the pages'''

    # Converts the pdf into a list of images, one for each page
    images = convert_from_path(filepath, size = (500, None), dpi = 300, poppler_path = 'bin', grayscale=True)
    # For each page image: searches page for QR, and if qr is found it adds it to the page_qr_data dictionary in the format page number:qr data
    page_qr_data = {}
    for idx, image in enumerate(images):
        # Decode the QR image
        detectedBarcodes = decode(image, symbols=[ZBarSymbol.QRCODE])       
        # If not detected then continue to the next page
        if not detectedBarcodes:
            continue
        else:
            # If QR detected, adds the page number to page_qr_data as a key and makes a list of the qr data for the value
            page_qr_data[idx] = []
            for barcode in detectedBarcodes:
                page_qr_data[idx].append(barcode.data)

    return len(images), page_qr_data

def scan_reader_window():
    '''Presents the user with a list of files that have been pulled and asks them to select which ones they'd like to import'''
    files = os.listdir(temp_pdf_dir)
    receivers_dict = {}
    page_lengths = {}
    i = 0
    for file in files:
        if sg.one_line_progress_meter('Processing files...', i+1, len(files), f'(Step 2 of 2) Processing files...\n{file}', key = 'file_processor', orientation = 'horizontal'):
            filepath = f'{temp_pdf_dir}/{file}'
            page_lengths[file], raw_scan_data = scan_reader(filepath)
            for key, value in raw_scan_data.copy().items():
                if len(value) > 1:
                    del raw_scan_data[key]
                raw_scan_data[key] = value[0]
    
            for key, value in raw_scan_data.copy().items():
                try:
                    raw_scan_data[key] = loads(value)
                except:
                    del raw_scan_data[key]
    
            for key, value in raw_scan_data.copy().items():
                if 'DB_FILENAME: ' not in value:
                    del raw_scan_data[key]
                else:
                    raw_scan_data[key] = value[20:]
            receivers_dict[file] = raw_scan_data
            i += 1
        elif i == len(files)-1:
            filepath = f'{temp_pdf_dir}/{file}'
            page_lengths[file], raw_scan_data = scan_reader(filepath)
            for key, value in raw_scan_data.copy().items():
                if len(value) > 1:
                    del raw_scan_data[key]
                raw_scan_data[key] = value[0]
    
            for key, value in raw_scan_data.copy().items():
                try:
                    raw_scan_data[key] = loads(value)
                except:
                    del raw_scan_data[key]
    
            for key, value in raw_scan_data.copy().items():
                if 'DB_FILENAME: ' not in value:
                    del raw_scan_data[key]
                else:
                    raw_scan_data[key] = value[20:]
            receivers_dict[file] = raw_scan_data
            i += 1
        else:
            clutter = os.listdir(temp_pdf_dir)
            for file in clutter:
                os.remove(f'{temp_pdf_dir}/{file}')
            exit()

    
    layout = []

    for value in receivers_dict.values():
        for filename in value.values():
            layout.append([sg.Checkbox(filename, default = True, background_color = 'white')])

    window = sg.Window('Scans', resizable = True, icon = parker_icon).Layout([    [sg.Text('Check all of the scans you wish to import.')],
                                                                                    [sg.Column(layout, scrollable=True, background_color = 'white', size = (350,350), expand_x = True, expand_y = True, justification = 'center', key = 'scans_column')],
                                                                                    [sg.Text("Save location:"), sg.Input(default_text = 'F:/Receiving/Receivers', key = 'selected_files'), sg.FolderBrowse(initial_folder = 'F:/Receiving/Receivers')],
                                                                                    [sg.Button('Continue'), sg.Button('Cancel')]    ])
    while True:
        event, values = window.read()
        if event == sg.WIN_CLOSED or event=='Cancel':
            clutter = os.listdir(temp_pdf_dir)
            for file in clutter:
                os.remove(f'{temp_pdf_dir}/{file}')
            exit()
        elif event == "Continue":
            save_location = values.pop('selected_files')
            del values['Browse']

            page_download_dict = {}
            for filename in receivers_dict.keys():
                page_download_dict[filename] = {}

            for key, value in receivers_dict.items():
                for page, filename in value.items():
                    page_download_dict[key][filename] = [page]

            for key, value in page_download_dict.items():
                for idx, (po_key, page_list) in enumerate(reversed(value.items())):
                    if idx == 0:
                        page_list.append(page_lengths[key]-1)
                        start_page = page_list[0]
                    else:
                        page_list.append(start_page-1)
                        start_page = page_list[0]

            i = 0
            for value in page_download_dict.values():
                for v in value.values():
                    v.append(values[i])
                    i += 1

            window.close()
            pdf_split_page(page_download_dict, len(values), save_location)

同样，当指定了--noconsole或--onefile时，它可以作为python文件和exe完美地工作。只有当我尝试两者都做的时候，它才会失败，而我一辈子也想不出为什么。有人有什么想法吗？谢谢！

编辑：这似乎是一个poppler问题。我尝试使用--debug运行它，在它崩溃后，我得到弹出消息“无法获得页计数。poppler是否安装并在路径中？”我修改了代码，使其“poppler_path='bin'”，并将bin添加到我的规范文件中，如下所示:datas=[('d://coding/bin','bin')]，但仍然没有成功。另外，因为它是被请求的，这里是来自pyinstaller的日志（对不起，当我从控制台复制if时，它会删除换行符，不知道如何更好地格式化）：

Answer 1

我终于找到了解决办法。问题出在subprocess.popen（）上，我使用的pdf2image模块使用它。使用这里找到的方法，我可以修改文件pdf2image.py，现在我的程序可以同时使用--onefile和--noConsole。