nParse - Yet another parser.

[magicfest2]

Ah yeah, thinking about it again and it's obvious that you're correct. The range that the game displays damage messages is super low for some reason. I had to be practically in melee range today trying to see how much damage my pet was doing. So, I guess there really isn't any way to have an accurate parser for this in EQ, unless you're a melee class that's always right next to the mobs.... And even then, you don't see damage for spells you didn't cast, right? Bleh. Guess my pipe dream is dead already.

[magicfest2]

Hey, following up on a different issue -- I was going to start working on a different feature, and figured I should probably base off the `dev` branch, since it looks like it is newer and contains some more modern stuff (like more valid pep8, and a requirements.txt) but it seems to be missing some stuff?

There are references to "widgets" which don't seem to exist, and no obvious standard package provides that. I am assuming during your refactoring you made a widgets directory and forgot to add it to the commit, so that code is only available locally for you?

I see it referenced here (and a few other places): https://github.com/nomns/nparse/blob...s/window.py#L6

[Mirox]

Originally Posted by magicfest2 [You must be logged in to view images. Log in or Register.]

Hey, following up on a different issue -- I was going to start working on a different feature, and figured I should probably base off the `dev` branch, since it looks like it is newer and contains some more modern stuff (like more valid pep8, and a requirements.txt) but it seems to be missing some stuff?

There are references to "widgets" which don't seem to exist, and no obvious standard package provides that. I am assuming during your refactoring you made a widgets directory and forgot to add it to the commit, so that code is only available locally for you?

I see it referenced here (and a few other places): https://github.com/nomns/nparse/blob...s/window.py#L6

I will take a look. I haven't had any time to work on nParse. I am positive I have local changes that haven't been synced since I was in the middle of working on features the last time I was able to work on it.

[magicfest2]

Cool, thanks! I've based my changes on master for now but should be easy to rebase them later.

One more question: what is your build process for releases? I'm *guessing* you are using PyInstaller, and I've been playing around with it, but I'm new to it and can't get it to work exactly the way I want. What is the exact command you use to generate a release?

[Mirox]

I use PyInstaller and use the spec file within the main directory -- "nparse_py.spec". So, I just run ".\pyinstaller nparse_py.spec". It's been so long since I've built it. I may use the --onefile flag or something to ensure the output is a single exe file. Then, everything that isn't imported into the executable needs to be copied into the dist directory where the executable is created. I believe I copy generic settings and most of the data directory.

[Mirox]

Originally Posted by Mirox [You must be logged in to view images. Log in or Register.]

I will take a look. I haven't had any time to work on nParse. I am positive I have local changes that haven't been synced since I was in the middle of working on features the last time I was able to work on it.

I've updated the dev branch -- widgets directory was missing as you surmised. It should be usable, but I can't ensure there are no bugs in it. There are major changes in the dev branch that are not complete, so there are some wonky things like a blank trigger window.

[Faxi]

Originally Posted by Mirox [You must be logged in to view images. Log in or Register.]

My bad, I had to do a quick update to v0.2.1. Go here for newest release.

https://github.com/nomns/nparse/releases

my Windows Defender freaked out over this... "Severe Trojan"

[magicfest2]

Originally Posted by Mirox [You must be logged in to view images. Log in or Register.]

I've updated the dev branch -- widgets directory was missing as you surmised. It should be usable, but I can't ensure there are no bugs in it. There are major changes in the dev branch that are not complete, so there are some wonky things like a blank trigger window.

Thanks! I'll continue polishing stuff on the master branch then, and port over to the dev branch once I'm done so I know any new bugs aren't something I've introduced. [You must be logged in to view images. Log in or Register.]

[magicfest2]

Originally Posted by Mirox [You must be logged in to view images. Log in or Register.]

I use PyInstaller and use the spec file within the main directory -- "nparse_py.spec". So, I just run ".\pyinstaller nparse_py.spec". It's been so long since I've built it. I may use the --onefile flag or something to ensure the output is a single exe file. Then, everything that isn't imported into the executable needs to be copied into the dist directory where the executable is created. I believe I copy generic settings and most of the data directory.

Ah yes, that specfile works great. I had never used PyInstall before, so I wasn't sure why it kept generating me a new specfile instead of using the existing one -- didn't realize I had to specify the specfile name in the build command instead of the python file. Thanks!

[Mirox]

Originally Posted by Faxi [You must be logged in to view images. Log in or Register.]

my Windows Defender freaked out over this... "Severe Trojan"

False positive. Some anti-virus do not like PyInstaller which is a program that creates the executable to run the python code.

Here is a report from VirusTotal: https://www.virustotal.com/gui/file/...d3c3/detection. You can see that the detections that do happen are all identifying different alerts, but most anti-virus see nothing wrong.

I would be more inclined to trust CrowdStrike for actual detection.