Matulog is a GUI implemented in Matlab for the analysis of ulog files. It is not finished yet but I hope people still find it useful.
To try it out just clone or download the repo from here. Matlab and pyulog is needed for it to run. It has only been tested on Linux so far so please create an issue or write here if it does not work.
image.png1600873 77.2 KB
when you said "The inconvenience about this board is that there is not enough room for one" in response to a comment posted regarding programming the ulog, did you mean that there isn't enough space to reprogram. I am still a little confused on how to reprogram the logger to record information from 3 accelerometers (such as the ADXL345 triaxial board). Could you provide more information?
Download Ulog
Download Zip 🔥 https://geags.com/2y3BdD 🔥
ulogOBJ = ulogreader(filePath) reads the ULOG file from the specified path and returns an object containing information about the file. The information in filePath is used to set the FileName property.
openulog() initializes the library and returns afile descriptor corresponding to the error channel or -1 onerror. The ident and facility parameters arethe same as in the syslog(3) function openlog. Theoptions parameter is a bit field which indicateslogging options. This includes all the options used in thesyslog(3) function openlog, as well as two added bythis package:
Use localtime(3) as the time base. The default is to usegmtime(3), so the default is for the time stamps to be inUTC. Note that this default is different than the syslog(3)package which uses localtime, and that the ctime(3) styleformating (dictated by the protocol) includes no timezonefield. So, mixing ulog(3) messages and syslog(3) messages inthe same file can be confusing as you first look at the timestamps.
ulog() is analogous to the syslog() call, thearguments are the same. The %m format directive is notsupported, but see serror() below. None of our programs useulog directly. Instead, we use the convenience wrappersserror(), uerror(), unotice(), uinfo() and udebug()described below.
vulog() is functionally identical to the ulog()function but for a single va_list argument. See stdarg(3)for more information. On success, the number of charactersactually written is returned. This number will be zero ifthe priority of the message was insufficient to have itprinted. On failure, the value -1 is returned.
setulogmask() is analogous to the syslog(3)setlogmask() call, the argument is the same. A majordifference is that this call actually sets the log priorityas documented. (In the syslog() implementations we haveseen, setlogmask() was missing or a NOOP).
serror() is equivalent to ulog(LOG_ERR, ...) also.However, if the global errno is not 0, indicating a systemcall error, a string describing the error is appended to theerror message in the fashion of perror(3). This function isused for reporting error conditions which are the directresult of a system call error.
rollulogpri() cycles through the followingpriority mask states: neither LOG_INFO nor LOG_DEBUG; justLOG_INFO; LOG_INFO and LOG_DEBUG. Each call moves to thenext state in the order above. If just LOG_DEBUG was setinitially, the cycle is joined at "LOG_INFO andLOG_DEBUG" by a call to rollulogpri(). Used forsetting verbosity.
This package will work with old (4.2 BSD) syslogd asshipped on Ultrix, but the functionality is tailored to the4.3 BSD syslog. You may wish to use the LOG_NOTIME option toopenulog() to avoid having two dates per line in yourlogfile on such systems.
ULOG has been deprecated, and if you don't have module ipt_ULOG you should move on to the newer NFLOG target. ulogd handles both of these, even though it is still called "ulog". Check out man iptables-extensions.
Addition of ulog_message_info_multiple_header_s and ulog_message_flag_bits_s messages and the ability to append data to a log.This is used to add crash data to an existing log. If data is appended to a log that is cut in the middle of a message, it cannot be parsed with version 1 parsers. Other than that forward and backward compatibility is given if parsers ignore unknown messages.
anylogger is a logging facade thatallows code to use logging without getting coupled to a specific loggingsystem. You can use that code with any logging system out there. ulog hasanylogger support built in. For other loggers, adapters are available.
In ulog, logging is always sent to exactly one channel. By default,two channels exist: output, for messages of loggers that are indebug mode, or that are at an enabled level, and drain, for thosemessages that are filtered away.
The default outputs are not very special, but the entire machinery is in place foryou to easily add any custom outputs you could ever need. You can define additionaloutputs by making ulog use a mod with an outputs key:
Formatting is another feature that separates ulog from debug andloglevel. debug has formatting, but it is hardcoded and messes up thecallstack and there is not much you can do about it. loglevel does not messup the callstack, but it also has no formatting at all out of the box.
ulog uses the new optionsin kurly v2 to make tag open/close markers optional and enable nicer looking formatstrings. This means that lvl name perf and {lvl} {name} {perf} areequivalent. When using no open/close markers, any non-identifier symbolsfollowing the tag name are considered nested text. For example for the formatstring name:22, the name format will receive ':22' as ctx.text. Thisallows for parameterized formats, as ulog has done to supportpadding options.
Browsers have great debug consoles these days. They even include stacktraceinfo for higher-level messages. But they did mess one thing up imho; themessages at these higher levels are indented a bit more than the othermessages, making the logging harder to read. This can be clearly seen in thescreenshot from ulog v2.0.0-beta-11, which did not yet have alignment:
ulog features a simple, yet powerful and flexible configuration mechanism. OnNode JS, we can configure ulog via program arguments, environment variablesor a configuration file. On browsers, we use querystring arguments orlocalStorage. On both platforms, the configuration is monitored and changes toit are picked up by ulog at runtime without the need to restart theapplication.
ulog extends this configuration mechanism. With debug, you can only turn loggers on and off, but ulog allows for a much more varied range of options. This is achieved by extending the configuration syntax so it also accepts the value applicable for the loggers matching the expression. Also we allow a semicolon-separated list of such expression=value pairs. For example, to set the logger test to debug and my:* loggers except for my:lib to info, we could write:
ulog is completely modular, consisting of a microscopically small core and awhole bunch of mods. The mods are small pieces of functionality, designed tobe usable stand-alone or in combination with each other. All functionality thatulog adds to anylogger, it adds in the form of mods. All, except for theability to add mods itself, in the form of method ulog.use, that is in core.
ulog being built with mods makes it easy to use lazy loading to reduce theminimum footprint. If you use ulog from a script tag(not recommended), you can get it for free. Otherwise, you have to configureyour bundler to code-split.
This works because the anylogger API is backwards compatible with that of debug. And because ulog has native anylogger support, once you import ulog in your entry point, all those libraries that were using debug start to use ulog automagically!
ulog is intended for use on embedded Linux applications that aremultithreaded and require a single thread to provide hard real-timeguarantees. Logs made using ulog are sent through a synchronouschannelbefore being processed, if logging cannot be completed in constant time,the message is dropped and an error flag is set.
To avoid this, ulog can be used to offloaded heavy I/O into a separatethread, if enough execution threads are available and scheduling is setupcorrectly, this will guarantee that logging never takes longer than astring format and a few small allocations.
The origin of ulog: RT-Thread has always lacked a small, useful log component, and the birth of ulog complements this short board. It will be open sourced as a basic component of RT-Thread, allowing our developers to use a simple and easy-to-use logging system to improve development efficiency.
Static and dynamic levelsClassify according to whether the log can be modified during the run phase. The dynamic level that can be modified during the run phase can only be called static level in the compilation phase. Logs that are lower than the static level (here specifically the logs using the LOG_X API) will not be compiled into the ROM and will not be output or displayed. The dynamic level can control logs that their level are higher than or equal to the static level. Logs that are lower than the dynamic level are filtered out when ulog is running.
Global level and module levelClassification by scope. Each file (module) can also be set to a separate log level in ulog. The global level scope is larger than the module level, that is, the module level can only control module logs higher than or equal to the global level.
The tag attribute of each log can also be output and displayed. At the same time, ulog can also set the output level of each tag (module) corresponding to the log. The log of the current unimportant module can be selectively closed, which not only reduces ROM resources, but also helps developers filter irrelevant logs.
The macro ulog_x(LOG_TAG, __VA_ARGS__): x corresponds to a different level of shorthand. The parameter LOG_TAG is the log label, the parameter ... is the log content, and the format is the same as printf. This API is useful when you use different tag output logs in one file. 2351a5e196