Flawed handling of line breaks (and tabs)

[CountBigBang]

The (single) log file is read, but the file content (in this simple example just two lines) is shown as a single entry, i.e. ignoring line breaks (see screenshot). The log file is UTF-8 encoded, uses the common \n end-of-line character and is displayed correctly in text editors on Linux and Windows.

[CountBigBang]

I did some testing and found the reason for the problem. The log viewer expects a log format where each entry starts with [{loglevel}]. I used a format without the brackets, so no end-of-lines where detected. This should at least be mentioned in the README to prevent weary debug sessions. Of course, a more flexible approach would allow for individual formats since I am not aware of any convention specifying the log file format in Python or Django. The examples in the Django documentation do not make use of brackets.

[agusmakmun]

@CountBigBang the character [ and ] are required for this plugin to identify the next line. you can checkout at this lines.

# pattern = "; |\[INFO\] |\[DEBUG\] |\[WARNING\] |\[ERROR\] |\[CRITICAL\] "
patterns = [']OFNI[', ']GUBED[', ']GNINRAW[', ']RORRE[', ']LACITIRC[']

[ghost]

@CountBigBang I changed utils to match besed on regular expressions, not gone into too much detail on the patterns but works for my logs. patterns = ['\].*OFNI.*\[', '\].*GUBED.*\[', '\].*GNINRAW.*\[', '\].*RORRE.*\[', '\].*LACITIRC.*\['] if any([re.search(p,line) for p in patterns]):

[ghost]

wanted to follow up my previous comment for anyone else who finds this - I'm on AWS EC2 and found that the below date pattern worked the best; patterns = [r'\]*(Z|.*)[0-9]{3}(.|,)([0-9]{1}[0-5]{1}:){2}[0-9]{1}[0-2]{1}(T|\s)[0-9]{1}[0-3]{1}-[0-9]{1}[0-1]{1}-[0-9]{4}\[*$']

[agusmakmun]

@seanpconkie now you can use the custom pattern at settings.py for newest version v1.0.7

pip install django-log-viewer --upgrade

LOG_VIEWER_PATTERNS = [']OFNI[', ']GUBED[', ']GNINRAW[', ']RORRE[', ']LACITIRC[']

to your custom pattern;

LOG_VIEWER_PATTERNS = [r'\]*(Z|.*)[0-9]{3}(.|,)([0-9]{1}[0-5]{1}:){2}[0-9]{1}[0-2]{1}(T|\s)[0-9]{1}[0-3]{1}-[0-9]{1}[0-1]{1}-[0-9]{4}\[*$']

[sgvolpe]

I think there's also an issue with the exclude argument. by default is '', which in my case makes the script read just the first entry.