add arxiv.py

Author: yzy1996

Python+arXiv/README.md

@@ -0,0 +1,19 @@
+# Usage
+
+1. run
+
+```shell
+python arxiv.py
+```
+
+2. input `arxiv number` and get the result, for example:
+
+![image-20220419004604986](https://raw.githubusercontent.com/yzy1996/Image-Hosting/master/image-20220419004604986.png)
+
+3. copy the text and (shift) paste to your markdown file.
+
+[Lifting 2D StyleGAN for 3D-Aware Face Generation](https://arxiv.org/abs/2011.13126)  
+*Yichun Shi, Divyansh Aggarwal, Anil K. Jain*  
+**[`CVPR 2021`] (``)** 
+
+4. modify the information as you wish.

Python+arXiv/arxiv.py

@@ -0,0 +1,162 @@
+from PyPDF2 import PdfFileReader
+from urllib import request
+import re
+from soupsieve import match
+
+from tqdm import trange
+
+from pathlib import Path
+
+'''
+pipeline: 输入文章标题任意格式, 得到标准模板样式
+'''
+
+class Information():
+    def __init__(self, query_id=None, query_title=None) -> None:
+        
+        if query_id != None:
+            self.query_url = f'http://export.arxiv.org/api/query?id_list={query_id}'
+        elif query_title != None:
+            query_title = query_title.replace(' ', '+')
+            self.query_url = f'https://export.arxiv.org/api/query?search_query=all:{query_title}&max_results=1'
+
+        self.strInf = request.urlopen(self.query_url).read().decode('utf-8')
+
+        self._re_process()
+
+    # 正则表达式解析
+    def _re_process(self):
+
+        Id = r'<id>http://arxiv.org/abs/(.*)</id>'
+        Title = r'<title>([\s\S]*)</title>' # 有时候名字太长了，会换行
+        Authors = r'<author>\s*<name>(.*)</name>\s*</author>'
+        Year = r'<published>(\d{4}).*</published>'
+
+        id_version = re.findall(Id, self.strInf)[0]
+        id = id_version[0:-2]
+        
+        title = re.findall(Title, self.strInf)[0]
+        title = re.sub(r'\n\s', '', title) # 去掉换行
+        title_sub = re.sub(r'[^\w\s-]', '', title) # 去掉标点  
+
+        authors = re.findall(Authors, self.strInf)
+
+        year = re.findall(Year, self.strInf)[0]
+
+        self.id_version = id_version
+        self.id = id
+        self.title = title
+        self.title_sub = title_sub
+        self.authors = authors
+        self.year = year
+        self.publish = ''
+        self.affiliation = ''
+        
+        self.abs_url = f'https://arxiv.org/abs/{self.id}'
+        self.pdf_url = f'https://arxiv.org/pdf/{self.id}'
+
+    def _get_publish(self):
+
+        # 读取 txt 预定义会议名称
+
+        with open(r'conf_list.txt') as f:
+            lines = [line.strip() for line in f]
+        reg = '|'.join(lines)
+
+        # obtain form arxiv comments
+        Publish = f'<arxiv:comment xmlns:arxiv="http://arxiv.org/schemas/atom">[\s\S]*(({reg}).*?\d{{4}})[\s\S]*</arxiv:comment>'
+        publish = re.findall(Publish, self.strInf)
+
+        if publish != []:
+            self.publish = publish[0][0]
+
+            # todo 处理例如 CVPR2020 -> CVPR 2020
+            # re.sub(r"(?<=\w)(?=(?:\w\w)+$)", " ", text)
+
+        else:
+            # 未来对接整个互联网搜索
+            self.publish = 'arXiv ' + self.year
+
+    def _get_affiliation(self):
+
+        # obtain from pdf file
+        # 判断这个文件是否存在
+        pdf_file = Path(f'{self.year}_{self.title_sub}.pdf')
+
+        if pdf_file.exists():
+            with pdf_file.open('rb') as f:
+                pdf = PdfFileReader(f)
+
+                first_page = pdf.getPage(0).extractText()
+                first_page = first_page.split()
+        
+            authors1 = self.authors[0].replace(' ', '')
+            self.affiliation = first_page[first_page.index(authors1) + 1]
+
+
+    def write_notes(self):
+
+        self._get_publish()
+        self._get_affiliation()
+
+        # 组合处理
+        title_url = f'[{self.title}]({self.abs_url})  '
+        authors = ', '.join(self.authors)
+        authors = f'*{authors}*  '
+
+        publish = f'**[`{self.publish}`] (`{self.affiliation}`)** '
+
+        print(title_url)
+        print(authors)
+        print(publish)
+
+
+    # download pdf from the web
+    def download(self):
+  
+        request.urlretrieve(self.pdf_url, f'{self.year}_{self.title_sub}.pdf')
+
+
+def verify_local_version(filename='11.pdf'):
+
+    with open(filename, 'rb') as f:
+        pdf = PdfFileReader(f)
+        first_page = pdf.getPage(0).extractText()
+        first_page = first_page.split()
+
+    # 查到本地文件的版本 v-x
+    id_version_local = first_page[-5][6:]
+    id = id_version_local[:-2]
+
+    information = Information(query_id=id)
+
+    if information.id_version != id_version_local:
+
+        print('>>>Downloading the latest version!!!')
+        information.download()
+        
+
+if __name__ == "__main__":
+
+    # query with title
+    # input_title = 'Image-to-Image Translation with Conditional Adversarial Networks'
+    # information = Information(query_title=input_title)
+    # information.write_notes()
+
+    # # query with id
+    # id = '2103.13413'
+    # information = Information(query_id=id)
+    # information.write_notes()
+
+    # # 
+    # verify_local_version()
+    while True:
+        id = input("type id: ")
+
+        # 先判断 id 是否有效，形如：2103.13413
+        if re.match(id, r'\t'):
+            pass
+
+
+        information = Information(query_id=id)
+        information.write_notes()

Python+arXiv/conf_list.txt

@@ -0,0 +1,9 @@
+ICML
+NeurIPS
+ICLR
+CVPR
+ICCV
+ECCV
+AAAI
+IJCAI
+3DV