Flag framework configures loguru logstore

Project scenario ：

loguru The logstore is a very good third-party logstore . It's plug and play , There are many ways to scroll the log 、 Automatically compress log files 、 Regular deletion and other functions . besides , Multithreading security 、 Highlight log 、 Log alarm and other functions are not a problem .
At present, the project adopts flask frame , In the past, the system was native logging, I want to see if I can loguru Integrate to flask in .

---

Problem description

Access to information through the Internet , Found some github Share the configuration method above , But in the configuration rotation After the parameter （ Log by time or size , Meet certain conditions to cut logs ）, There is a problem with the process of log writing .

PermissionError: [WinError 32]  Another program is using this file , Process unreachable .

---

Cause analysis ：

Preliminary analysis of the causes ,rotation Parameters , The function is once the log file size exceeds the specified size , A new log file will be generated . When modifying the name of the original log file ,flask The app is writing to the current log file , Make it impossible to operate .

---

Solution ：

Give the configuration parameters of the initialized log LOGFILE, add to {time}

LOGFILE = "./files/logs/log_{time}.log"

Then you can smoothly in flask Use in loguru, And the log will take effect according to the corresponding parameters .

---

Let's share with you about rotation Guidelines for using parameters ：

Scroll by time

For example, scroll by time , All we need to do is logger.add Add one of the parameters rotation Parameters ：

from loguru import logger
logger.add("file_2.log", rotation="12:00") #  Every day 12:00 A new file will be created 
logger.debug("That's it, beautiful and simple logging!")

such , If the current time passes the set time , It will generate a new log file . If not, use the original log file ：

As shown in the figure , After the set time , Then the original file_2.log rename , And add a new file_2.log file .

Scroll by size

In addition to scrolling the log by time ,Loguru You can also scroll by log size ：

from loguru import logger
logger.add("file_1.log", rotation="1 MB") #  Scroll large log files 
logger.debug("That's it, beautiful and simple logging!")

such , Once the log file size exceeds 1 MB A new log file will be generated .

Compress logs

If you don't want to delete the original log file ,Loguru It also supports direct compression of logs ：

from loguru import logger
logger.add("file_Y.log", compression="zip") #  Compress logs