X-Star Log Files

[yrless]

Could be...I would do a fresh log file dump from the aircraft and use the python script that gets put into the "conv" directory...that's what I'm doing and I haven't seen any header errors..it sounds like the script you're running is looking for data columns that may have been moved or deleted.

 

[Jagerbomb52]

The python script file shows up inside a zip file called conv.zip.
Am I supposed to unzip it to a directory called conv ?

 

[Jagerbomb52]

 

[yrless]

View attachment 1292

Yep...just do the unzip right where it is and it'll produce the directory for you

 

[Jagerbomb52]

Still getting the invalid header. Going to try again later on my other computer. Starting to get frustrated so I need a break & a beer LOL

 

[Jagerbomb52]

Does it matter if the " marks are there?

python sdlog2_dump.py log001.bin -f "export.csv" -t "TIME" -d "," -n ""

 

[yrless]

Yes...keep everything verbatim...quotes and all.

 

[Jagerbomb52]

Well I have no idea what I am doing wrong then. two different computers and same error. I will upload a log file with the python script file from my last dump to my google drive and maybe you can give it a try in your spare time. I will post a link here when I get it uploaded.

 

[Jagerbomb52]

Have no idea where this log file is from but here ya go.

conv.rar

 

[yrless]

Ok.....I know what's going on......I made a stupid assumption......I pulled my data from another XSP which has 2.0.x FW loaded so I made the assumption that all the logs were produced within 2.0.x FW.....Bad assumption! The first few log files I retrieved were produced in 1.1.x FW and those were the ones I used with the Python script.....turns out the next 40+ log files are from 2.0.x FW........soooooo, bad news is I ran you around the tree a bit, good news is you know how to use this dang thing! I'm working on getting an updated script from the GitHub site

 

[Jagerbomb52]

Ok.....I know what's going on......I made a stupid assumption......I pulled my data from another XSP which has 2.0.x FW loaded so I made the assumption that all the logs were produced within 2.0.x FW.....Bad assumption! The first few log files I retrieved were produced in 1.1.x FW and those were the ones I used with the Python script.....turns out the next 40+ log files are from 2.0.x FW........soooooo, bad news is I ran you around the tree a bit, good news is you know how to use this dang thing! I'm working on getting an updated script from the GitHub site

This old dog only has a few more good runs around that dang tree

 

[yrless]

W

This old dog only has a few more good runs around that dang tree

Woof!

 

[yrless]

Here's the modified script from one of the gents on github...I ran it and it works...it will show an error upon completion in the cmd window but he said he's aware of one or two warts in this thing.....I'll do some more investigation when I get time but it appears to work very well.....one more lap around the tree

 

[Jagerbomb52]

Yes it works great.

What dataprofile are you using Dashware for the CSV?

 

[yrless]

I created a new profile called CSV skinny.....I prefer to work with a minimum amount of information with Dashware because it can be really buggy so I scale down what I need to just a few columns.....first screen shot shows the data I use and the second shows the resultant pro ject (although the actual data wasn't coupled to the widgets when I took the screen shot).

Edit....AGL in feet and Heading are computed values....they're not directly available from the datalog

 

[Jagerbomb52]

From my export.csv when I upload it to gpsvisualizer.com it only shows one waypoint. is there something else we need to do to the .csv file ?

 

[yrless]

If its a flight that's very limited in area then it'll look like one point....did you try to zoom in as much as possible to see if there are really multiple points there?


Edit...other possibility is that there are "0's" for Lat/Lon before acquisition so you may have to delete those rows in the csv.

 

[Jagerbomb52]

I need to find a good one to test.

 

[Jagerbomb52]

Is this the error you were talking about? So far I am still running around that tree LOL
And yes a lot of info seems to be missing near the top of the export.csv file.

C:\log\conv>python sdlog2_dump.py log001.bin -f "export.csv" -t "TIME" -d "," -n ""
Traceback (most recent call last):
File "sdlog2_dump.py", line 372, in <module>
_main()
File "sdlog2_dump.py", line 369, in _main
parser.process(fn)
File "sdlog2_dump.py", line 182, in process
self.__parseMsg(msg_descr)
File "sdlog2_dump.py", line 268, in __parseMsg
data = list(struct.unpack(msg_struct, self.__buffer[self.__ptr+self.MSG_HEADER_LEN:self.__ptr+msg_length]))
struct.error: unpack requires a buffer of 84 bytes

 

[yrless]

yep I see the same stuff........I also just noticed that a lot of data that I used previously to compute heading and AGL info is no longer in the log-data....so its either gone or quite possibly renamed which pisses me off because I have like 24 hours of development time into this stuff...arghh. Oh well.....I'm gonna finish this bag of air with a few chips in it and head out for some fixed wing RC flights