-log-level: int value or associated keyword indicating how much to log.-log-destination: where to put the log file.but on a side note I've been thinking about making a pull request to add a server configuration file - so maybe I should take care of that first? Here are some config options we'll probably need: I am aware that this would add quite a lot of additional configuration options. Syncplay version and build type: Syncplay 1.6.9 / 1.7.0 Media player and version: MPV. To keep things simple, STDERR logging should probably be default, with the option to override the target file object to an actual physical file of the user's choice. To accommodate that, printing out dates on each log entry should be an optional feature. I would just put it all in one file - the location of which can be decided by the end user, but I imagine a lot of people (including myself) would just do something like -log-destination - to log to STDERR, and then just run the thing under systemd. I've already updated Microsoft Visual C 2015 Redistributable both x86 and 圆4 and still nothing seems to work. All the log shows is this: PySide2/init.py: Unable to import shiboken2 from C:\Program Files (x86)\Syncplay\lib/libsync.zip. Hopefully server administrators should know their way around grep and the like, so logging different events to different files seems like overkill to me. Errors in 'Syncplay.exe' - See the log file 'C:\Program Files (x86)\Syncplay\Syncplay.log for details. Furthermore, log_level would just be a simple int with associated codes ranging from DEBUG (show everything) to NONE. Which would simultaneously go on to log the event and optimally trigger external code for custom user scripts. Report_event("Sentence to help contextualize what's going on in log entry", contextual_data, log_level) Since there are a lot of events to log, and I also wanted to create some sort of event hook system, I think an optimal way to handle these two things would be with a single function - something like: Fix 1: Changing cache values Incorrect cache values can cause stuttering/lagging issues in VLC media player. Since we live in the SystemD era, logging to STDERR is going to be fine for a lot of people, including me - but some people might like logging to files to be a feature. Logging should be minimal by default, to protect privacy - perhaps with logging connections and disconnections like IRC servers typically do by default, but not logging everything that was said. Ideally, all these use cases should be satisfied by the same code, so as to minimize the need for future refactoring. In the case of someone running a private server that was subject to a hacking attempt, they might like to have a log that they can point fail2ban at so that they can automatically blacklist the IP. In the case of someone running a large server and troubleshooting a problem, they might want a log to tell them in detail what handshake happened and what JSON strings changed hands during the error. Logging should range from sparse to verbose, optimally.įor instance, in my use case, I'd just want a plaintext log of who said what. Enabling/disabling should be done in server config as with any other feature.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |