如何通过 Python 中的 API 将块中的文件上传到 SharePoint （Office 365）（使用 startupload/continueupload/finishupload）

这是我的工作解决方案。我认为我的错误是urllib.parse.quote（）URL;但我不确定。

有以下几种方法：

1. GetFileByServerRelativePath（decodedurl='{URL}'） 例如，url = f“/sites/{site_name}/_api/web/GetFileByServerRelativePath（decodedurl='/sites/{site_name}/{folder_path}/{file_name}'）/startupload（uploadId=guid'{GUID}'）”

2. GetFileById（'{unique_file_id}'） 例如，file_url = f“/sites/{site_name}/_api/web/getfilebyid（'{unique_file_id}'）/startupload（uploadId=guid'{GUID}'）”

常规：

• 如果服务器上没有现有文件，则必须创建一个文件，否则这两种方法都不起作用。
• 将urllib.parse.quote（）与GetFileByServerRelativePath（decodedurl='{URL}'）一起使用不适用于我的解决方案

方法：

1. 使用“.../GetFileByServerRelativePath（decodedurl='/sites/{site_name}/{folder_path}/{file_name}'）”检查文件的 UniqueId
2. 如果服务器上不存在该文件，请创建新的（空的内存中）文件，通过“.../files/add（overwrite=true，url='{file_name}'）”上传它
3. 生成 GUID -> GUID = uuid.uuid4（）
4. 读取块中的文件 -> big_file_chunk = file.read（CHUNK_SIZE）
5. 以“.../startupload（uploadId=guid'{GUID}'）”开头，并保护响应中的文件偏移量
6. 继续“.../continueupload（uploadId=guid'{GUID}'，fileOffset={file_offset}）”，只要您不在最后一个chunk_count;此外，还可以保护 FileOffset 与响应
7. 使用“.../finishupload（uploadId=guid'{GUID}'，fileOffset={file_offset}）”完成最后一个块

扩展 URL：

host = "COMPANY.sharepoint.com"
site_name = "DUMMYSUPPLIER"
foler_path = "Shared%20Documents/TestFolder"
file_name = "FILE.EXT"
UniqueId = '3e6be666-8b5b-40fa-89ab-7cf9092a603d'
guid = '57dd389e-5325-4cc4-95cd-55af2131ae67'

url = "/sites/DUMMYSUPPLIER/_api/web/getfilebyid('3e6be666-8b5b-40fa-89ab-7cf9092a603d')/startupload(uploadId=guid'57dd389e-5325-4cc4-95cd-55af2131ae67')"
url = f"/sites/DUMMYSUPPLIER/_api/web/GetFileByServerRelativePath(decodedurl='/sites/DUMMYSUPPLIER/Shared%20Documents/TestFolder/FILE.EXT')/startupload(uploadId=guid'57dd389e-5325-4cc4-95cd-55af2131ae67')"

使用 GetFileById（'{unique_file_id}'） 的解决方案：

def _uploadLargeFile(host: str, bearer_token: str, site_name: str, folder_path: str, file_name: str) -> http.client.HTTPResponse:    
    """
    Uploads a file in chunks to the server and overwrites exisitng files. (Recommended for large files; mandatory for file sizes >= 250MB.)
    """
    headers = {
            'Authorization': f'Bearer {bearer_token}',
            'Accept': 'application/json;odata=verbose',
            'Content-Type': 'application/octet-stream;odata=verbose'
            }                              

    unique_file_id = get_file_unique_id(host=host, bearer_token=bearer_token, site_name=site_name, folder_path=folder_path, file_name=file_name)
    if not unique_file_id:
        unique_file_id, resp = create_and_upload_empty_file(host=host, bearer_token=bearer_token, site_name=site_name, folder_path=folder_path, file_name=file_name)

    GUID = uuid.uuid4()
    file_size = os.stat(file_name).st_size
    with open(file_name, "rb") as file:
        continue_upload = True
        file_offset = 0 # While uploading in chunks the fileoffset can be calculated as follows: fileoffset = chunk_count * CHUNK_SIZE
        while continue_upload:
            big_file_chunk = file.read(CHUNK_SIZE)

            conn = http.client.HTTPSConnection(host=host)
            
            if not big_file_chunk or _is_last_chunk(file_size, file_offset, CHUNK_SIZE): # Finish Upload
                file_url = f"/sites/{site_name}/_api/web/getfilebyid('{unique_file_id}')/finishupload(uploadId=guid'{GUID}',fileOffset={file_offset})"
                url = urllib.parse.quote(file_url)
                conn.request("POST", url, big_file_chunk, headers, encode_chunked=True)
                resp = conn.getresponse()                
                continue_upload = False
            elif file_offset == 0: # Start Upload
                file_url = f"/sites/{site_name}/_api/web/getfilebyid('{unique_file_id}')/startupload(uploadId=guid'{GUID}')"
                url = urllib.parse.quote(file_url)
                conn.request("POST", url, big_file_chunk, headers, encode_chunked=True)
                resp = conn.getresponse()
                resp_json = json.load(resp)
                file_offset = int(resp_json['d']['StartUpload'])
            else: # continue Upload
                file_url = f"/sites/{site_name}/_api/web/getfilebyid('{unique_file_id}')/continueupload(uploadId=guid'{GUID}',fileOffset={file_offset})"
                url = urllib.parse.quote(file_url)
                conn.request("POST", url, big_file_chunk, headers, encode_chunked=True)
                resp = conn.getresponse()
                resp_json = json.load(resp)
                file_offset = int(resp_json['d']['ContinueUpload'])

            if continue_upload:
                 prect = (file_offset / file_size) * 100
                 print(f"Uploading file '{file_name}': {prect:.2f}% uploaded ({file_offset}/{file_size})")
            else:
                 print(f"Uploading file '{file_name}': {100.0:.1f}% uploaded ({file_size}/{file_size})")
        return resp
        
        
def get_file_unique_id(host: str, bearer_token: str, site_name: str, folder_path: str, file_name: str) -> str | None:
    """
    Returns the files UniqueId if it exists otherwise None.
    """

    headers = {
    'Authorization': f'Bearer {bearer_token}',
    'Accept': 'application/json;odata=verbose',
    'Content-Type': 'application/octet-stream;odata=verbose'
    }

    url = f"/sites/{site_name}/_api/Web/GetFileByServerRelativePath(decodedurl='/sites/{site_name}/{folder_path}/{file_name}')"

    conn = http.client.HTTPSConnection(host)
    conn.request(method="POST", url=url, body=None, headers=headers)
    resp = conn.getresponse()
    resp_json = json.load(resp)
    
    if 'd' in resp_json:       
        return resp_json['d']['UniqueId']
    else:
        return None

        
def create_and_upload_empty_file(host: str, bearer_token: str, site_name: str, folder_path: str, file_name: str) -> (str, http.client.HTTPResponse):
    """
    Creates an empty file (in-memory) and uploads it the server. Returns the UniqueId of the uploaded file.
    """
    emptyfile = io.BytesIO(b"") # In-Memory file content

    headers = {
    'Authorization': f'Bearer {bearer_token}',
    'Accept': 'application/json;odata=verbose',
    'Content-Type': 'application/octet-stream;odata=verbose'
    }

    file_url = f"/sites/{site_name}/_api/web/getfolderbyserverrelativeurl('{folder_path}')/files/add(overwrite=true,url='{file_name}')"
    url = urllib.parse.quote(file_url)
    
    conn = http.client.HTTPSConnection(host)
    conn.request("POST", url, emptyfile, headers)
    resp = conn.getresponse()
    resp_json = json.load(resp)    
    if 'd' in resp_json:       
        return (resp_json['d']['UniqueId'], resp)
    else:
        return (None, resp)


def _is_last_chunk(file_size, file_offset, chunk_size) -> bool:
    return file_size - file_offset <= chunk_size