WX Parser fails on certain formats

[na7q]

One example.
KA7MYM>APRS,TCPIP*,qAC,AMBCWOP-2:@081706z4322.78N/12416.81W_187/000g002t...r000p028P028h...b10295L063AmbientCWOP.com"

[ab0oo]

I just got back in the country. Let me take a look at this and see what's going on. That's a pretty complex object, but should be at least mostly decoded.

[na7q]

It's complex and mostly there for sure. I'm attempting to get APRSdroid up to date with decoding abilities, and ran into this issue. Lots of work to do every which way!

[ab0oo]

I added this to src/test/resources/cwop.txt and then ran src/test/java/net/ab0oo/aprs/parser/TestFile against that entire CWOP dump, and it spit this back:
KA7MYM>APRS
Via Digis: TCPIP*,QAC,AMBCWOP-2
Raw Bytes: @081706z4322.78N/12416.81W_187/000g002t...r000p028P028h...b10295L063AmbientCWOP.com
Data Type Identifier: @
Create Timestamp: Wed Feb 12 08:41:54 PST 2025
Comment:
Class net.ab0oo.aprs.parser.PositionField
---POSITION---
Position Source Uncompressed
Is Compressed: false
Latitude: 43.37967
Longitude: -124.28017
Comment: null
Class net.ab0oo.aprs.parser.WeatherField
---WEATHER---
Wind Direction: 187
Wind Speed: 0
Wind Gust: 2
Temperature: null
Rain last hour: 0.0
Rain 24 hours: 0.28
Rain since 00:00: 0.28
Humidity: null
Pressure: 10295.0
Class net.ab0oo.aprs.parser.TimeField
---TIMESTAMP---
Reported Timestamp: 08 09:02

No errors, and the only thing the library is missing is the Luminosity > 1000 watts/meter^2, whcih looks like it should be 1063 in this case. This is off trunk, btw.
this line should be extracting that info, and it looks like it's not...
https://vscode.dev/github/ab0oo/javAPRSlib/blob/v3.1.1/src/main/java/net/ab0oo/aprs/parser/WeatherParser.java#L124

I'll bang on it a bit and see if I can figure out why it's failing to parse that. 25 years ago, the PCRE and I were best friends. I think those memory cells were overwritten by several years of having to sit through middle school girls basketball games in the mid 2000s.

Did you see anything else that was Not Right?

[na7q]

Something I would like to see fixed as well might be the Q construct. Making it all uppercase gives the appearance of it as a digipath and not a Q construct.

[ab0oo]

Ok, I just pushed a massive set of changes. Most of it was formatting and documentation (javadocs), but there are also some pretty important parsing fixes. I'm going to pull your working fork and see what I can do.