华为云NLP服务动手实践之分词

得先获取token，其实测试一些云服务的api用postman最简单，不过要是动手实践写代码的话，还是实践一些代码比较实在：

先获取token，

def get_token(username, password, area):
    """
    username, 华为云用户名
    password, 华为云密码，注意，如果带有特殊字符，如“@”神马的，需要转义
    area，哪个区，如北京四，可以写“cn-north-4”
    """
    auth_endpoint = "https://iam." + area + ".myhuaweicloud.com/v3/auth/tokens"
    header = {'content-type':'application/json'}
    data = { 
        "auth": { 
            "identity": { 
                "methods": [ 
                    "password" 
                ], 
                "password": { 
                    "user": { 
                        "name": username, 
                       "password": password, 
                        "domain": { 
                            "name": username 
                        } 
                    } 
                } 
            }, 
            "scope": { 
                "project": { 
                    "name": area 
                } 
            } 
        } 
    }
    #print(auth_endpoint)
    
    try:
        r = requests.post(url = auth_endpoint, headers = header, json = data, verify = False)
    except:
        #其实这样不规范
        print("get token error")
    
    token = r.headers['x-subject-token']
    return token

有了token，加上project_id就可以分词了，

def segment(token, area, project_id, text, pos_switch, criterion, lang):
    #project_id可以在我的凭证里面找到，不同的project对应不同的id
    url = "https://nlp-ext." + area + ".myhuaweicloud.com/v1/" + project_id + "/nlp-fundamental/segment"
    
    header = {'content-type':'application/json', 'X-Auth-Token':token}
    data = {
        "pos_swtich":pos_switch,
        "text":text,
        "lang":lang,
        "criterion":criterion
        }
    try:
        r = requests.post(url, json = data, headers = header, verify = False)
        #如果有代理，也可以加进来, "proojeces = proxy"
        print(r.text)#方便调试
        if(int(r.status_code) >= 600):
            logger.info('status_code:%s; url:%s'%(r.status_code, url))
            logger.error('Error: code:%s; msg:%s'%(r.status_code, r.text))
        seg = r.json()['words']
        return seg
    except:
        logger.error("Error: segment...")
        return ''