将 LaTeX 表达式转换为 PNG 文件

Question

目前，我和我的搭档正在编译一个函数，将 LaTeX 数学表达式转换为 PNG 文件，使用 Python 作为编程语言，使用 XeLaTeX 作为 LaTeX 编译器。这是我的代码。有一些字母是用我的母语写的，所以看起来很奇怪）：

\`from sympy import \*
import random
import os
import subprocess
#import pdf2image
from pdf2image import convert_from_path
import cv2
from concurrent.futures import ProcessPoolExecutor, as_completed
import pdfplumber
import pymupdf
import wand
from wand.image import Image
#PRACTICE MODE:

#1.TOPIC: TRIGONOMETRY
#Question type 1: Theory
#LaTeX-\>self-compiled function has corresponding PDF

def compile_latex(file_path):
os.chdir(os.path.dirname(file_path))
try:
subprocess.run(\['xelatex', file_path\], check=True)
except subprocess.CalledProcessError as e:
print("Error compiling LaTeX file", e)
exit(1)
pdf_file_path = file_path.replace('.tex', '.pdf')
if not os.path.exists(pdf_file_path):
print("PDF file was not created")
exit(1)
return pdf_file_path

def convert_pdf_to_png(pdf_file_path):
png_file_path = pdf_file_path.replace('.pdf', '.png')
try:
with Image(filename=pdf_file_path, resolution=240) as img:
img.compression_quality = 80
img.save(filename=png_file_path)
print("PDF successfully converted into PNG:", png_file_path)
except Exception as e:
print("Error converting PDF to PNG: ", e)
return png_file_path

def convert_multiple_pdf_to_png(pdf_files):
results=\[\]
with ProcessPoolExecutor() as executor:
futures={executor.submit(convert_pdf_to_png,pdf_file): pdf_file for pdf_file in pdf_files}
for future in as_completed(futures):
result=future.result()
results.append(result)
return results

#Note!: This is the function I wrote myself to compile individual image files. Please read the operating principle.

def MakeFile(phuongan,modauchofilelatex,endtext,count):
#Set variable b, which plays a key role in identifying which is the question file and which are the 4 corresponding answer files. Thanks to it, we can compile successfully and as desired
#READ the conditions CAREFULLY to help the machine know when to print the question file and when to print 4 corresponding answer files.
b=0

    folder="C:/Users/HP/Downloads/001_QO_folder"
    #pdf_files=[os.path.join(folder,file) for file in os.listdir(folder) if file.endswith('.pdf')]
    for i in range(len(phuongan)):
        if(b==0):
    
        #Create a ques file, set the count variable to type the question status, because after each print of 5 files containing questions and 4 answers, I will increase count+1.
         file_path= "C:/Users/HP/Downloads/001_QO_folder/001_Q0_%s_QUES.tex"%str(count)
         #pdf_file_path=os.path.abspath(file_path.replace('.tex','.pdf'))
         #png_file_path=os.path.abspath(file_path.replace('.tex','.png'))
         file=open("C:/Users/HP/Downloads/001_QO_folder/001_Q0_%s_QUES.tex"%str(count),'w',encoding='utf-8')
         file.write(modauchofilelatex+phuongan[0]+endtext)
         file.close()
         pdf_file_path=os.path.abspath(file_path.replace('.tex','.pdf'))
         #dvi_file_path=file_path.replace('.tex','.dvi')
         #xdv_file_path=file_path.replace('.tex','.xdv')
         compile_latex(file_path)
        #Convert pdf to image
         #image=convert_from_path(pdf_file_path)
        #Save as png format
         #image[0].save(png_file_path,'PNG')
        #Delete redundant files through their paths
         convert_pdf_to_png(pdf_file_path)
         #os.remove(file_path)
         #os.remove(pdf_file_path)
         #Each time the question is printed, the b's variable number will increase from 0 to 1 -> the score is not satisfied -> from there it can be identified when printing out 4 answers.
         b=b+1
        else:
         file_ans=open("C:/Users/HP/Downloads/001_QO_folder/001_Q0_%s_ANS%s.tex"%(str(count),str(i)),'w', encoding='utf-8')
         file_ans.write(modauchofilelatex+phuongan[i]+endtext)
         file_ans.close()
         file_ans_path = "C:/Users/HP/Downloads/001_QO_folder/001_Q0_%s_ANS%s.tex"%(str(count),str(i))
         pdf_file_ans_path=file_ans_path.replace('.tex','.pdf')
         png_file_ans_path=file_ans_path.replace('.tex','.png')
         #dvi_file_ans_path=file_ans_path.replace('.tex','.dvi')
         #xdv_file_ans_path=file_ans_path.replace('.tex','.xdv')
         compile_latex(file_ans_path)
         #imageans=convert_from_path(pdf_file_ans_path)
         #imageans[0].save(png_file_ans_path,'PNG')
         convert_from_path(pdf_file_ans_path)
         #os.remove(file_ans_path)
         #os.remove(pdf_file_ans_path)
        
        #results= convert_multiple_pdf_to_png(pdf_files)
        #for result in results:
            #print(result)

我尝试运行它几次。代码确实按照我的预期将 PDF 转换为 PNG，因为这是我能想到的唯一方法。但它打印出来的是整页。这是我所期望的[在此处输入图像描述](https://i.sstatic.net/BMr9Qvzu.png)，但我无法做到这一点。总而言之，有什么办法可以解决这个问题吗？

Answer 1

也许看看 pdfCropMargins ：“pdfCropMargins 程序是一个命令行应用程序，用于自动裁剪 PDF 文件的边距。”但它也可以在 python 中工作，如下面的文档示例所示。不过，您可能需要调整函数参数。

from pdfCropMargins import crop
crop(["-p", "20", "-u", "-s", "paper1.pdf"])
crop(["-p", "0", "-gui", "paper2.pdf"])

他们的文档示例：