Segment not recognized

[mute mutnig]

Dear all, 

some time ago I created a segment in GC, usually GC recognized the segment during import of the Garmin fit file.

This time not.

Looking at the map, it looks quiet the same as other rides.

How can I get GC to recognize the segment?

(found a ink to the source code for changing parameters - but I do not want (to be honest: not able to) to compile my own GC for WIN/LINUX/MacOS)

Thanks 

Helmut

[Ale Martinez]

Hi, the source code includes debug statements which can be enabled only at compile time, what I could do is a change to allow enabling them a runtime so you can see the output at the console when starting GoldenCheetah with the command line option: --debug-rules gc.routes.debug=true, let me know if you are interested.

Cheers, Ale.

[mute mutnig]

Hi Ale, 

yes please, if feasible please provide a linux appimage.

I would like to try.

BR,

Helmut

[Ale Martinez]

Done in https://github.com/GoldenCheetah/GoldenCheetah/commit/7cdb14ca7aab8c9748551bd0370c63a6160609d2, when ci finishes the AppImage will be in https://ci.appveyor.com/project/Joern-R/goldencheetah-knhd8/builds/53173210 > Ubuntu2204 > Artifacts, downloads have a quota so you may have to wait and retry.

[Ale Martinez]

El martes, 2 de diciembre de 2025 a la(s) 12:38:48 p.m. UTC-3, Ale Martinez escribió:

Done in https://github.com/GoldenCheetah/GoldenCheetah/commit/7cdb14ca7aab8c9748551bd0370c63a6160609d2, when ci finishes the AppImage will be in https://ci.appveyor.com/project/Joern-R/goldencheetah-knhd8/builds/53173210 > Ubuntu2204 > Artifacts, downloads have a quota so you may have to wait and retry.

 https://ci.appveyor.com/project/Joern-R/goldencheetah-knhd8/builds/53176386/job/bobbhdgyt5h2eg5p/artifacts

[mute mutnig]

Dear Ale, 

just installed an imported the fit were the segment is not detected: no debug output during import - segment not recognized:

 ./GoldenCheetah_v3.7_x64_debug.AppImage --debug-rules gc.routes.debug=true
QApplication: invalid style override 'gtk' passed, ignoring it.
Available styles: Windows, Fusion
Python loaded [3.7.17 (default, Jun  6 2023, 20:10:09)  [GCC 11.3.0]]
GoldenCheetah: redirecting log messages (stderr) to file  "/home/mute/GoldenCheetah/goldencheetah.log"

I deleted the log file, and restarted the import no messages were written to the log file.

I also get no message if I import a fit where GC recognizes the segment.

Am I doing something wrong?

Helmut

[Ale Martinez]

Likely you need to add quotation marks around debug rules, try  ./GoldenCheetah_v3.7_x64_debug.AppImage --debug-rules "gc.routes.debug=true"

See https://github.com/GoldenCheetah/GoldenCheetah/wiki/Command-line-options for the full details

[mute mutnig]

Yes, this was the fault, no I get debug info in the log file:

Debug: Opening ride:  "2025_12_01_11_29_29.json"  for  "Golden_C_Sprint Unterpremstätten"
Debug:     Start point identified...
Debug:     Start time  2552 M
Debug:     WARNING route diverge at  2555 ( 2555 ) after  3 secs for  0.1126 km  46.9688 - 15.4054 / 46.9685 - 15.404 M
Debug:     WARNING route diverge at  2556 ( 2556 ) after  4 secs for  0.120537 km  46.9688 - 15.4054 / 46.9685 - 15.4039 M
Debug:     WARNING route diverge at  2566 ( 2566 ) after  14 secs for  0.124231 km  46.9688 - 15.4054 / 46.9684 - 15.4039 M
Debug:     STOP route diverge at  2566 ( 2566 ) after  14 secs for  0.124231 km  46.9688 - 15.4054 /

My interpretation: GC see I passed the starting point but then i went some were but not on the segment.

This picture shows the segment: it starts right after the western roundabout:

At my recent ride I came from the eastern roundabout went around the western into the segment, GC found the starting point on the way to the western roundabout, and stopped because of the wrong direction, GC did not find the starting point on my way on the segment again.....

Same on today's ride were I went again from east to west and then on the segment form west to east...

Any chance to let GC even in situations like this recognize the segment?

(for sure I can provide all the files - just instruct me how to export correct)

Cheers,

Helmut

[Ale Martinez]

El miércoles, 3 de diciembre de 2025 a la(s) 1:19:04 p.m. UTC-3, mut...@gmail.com escribió:

Yes, this was the fault, no I get debug info in the log file:

Debug: Opening ride:  "2025_12_01_11_29_29.json"  for  "Golden_C_Sprint Unterpremstätten"
Debug:     Start point identified...
Debug:     Start time  2552 M
Debug:     WARNING route diverge at  2555 ( 2555 ) after  3 secs for  0.1126 km  46.9688 - 15.4054 / 46.9685 - 15.404 M
Debug:     WARNING route diverge at  2556 ( 2556 ) after  4 secs for  0.120537 km  46.9688 - 15.4054 / 46.9685 - 15.4039 M
Debug:     WARNING route diverge at  2566 ( 2566 ) after  14 secs for  0.124231 km  46.9688 - 15.4054 / 46.9684 - 15.4039 M
Debug:     STOP route diverge at  2566 ( 2566 ) after  14 secs for  0.124231 km  46.9688 - 15.4054 /

My interpretation: GC see I passed the starting point but then i went some were but not on the segment.

This picture shows the segment: it starts right after the western roundabout:

At my recent ride I came from the eastern roundabout went around the western into the segment, GC found the starting point on the way to the western roundabout, and stopped because of the wrong direction, GC did not find the starting point on my way on the segment again.....

Same on today's ride were I went again from east to west and then on the segment form west to east...

I think this happens because you passed the second time for the starting point while the algorithm was still trying to match the segment starting from the first time, once it gives up it should go to an state looking for the starting point again.

As a user some workarounds are to take a longer loop after passing the first time when you ride or, afterwards just remove the samples when you pass the first time creating a small gap.

As a developer the algorithm could consider the initial point as a restart when inside a segment, but I didn't write that code nor I know it in detail.

Any chance to let GC even in situations like this recognize the segment?

May be, all we need is a developer interested enough in the issue to analyze the code and propose a change, since this is an isolated piece of code it may be a project for someone willing to start coding in GoldenCheetah.

 

(for sure I can provide all the files - just instruct me how to export correct)

config/routes.xml and the activity file (original or json) would be enough, but honestly, I am not that interested, may be others are.