camera_list.csv (was Re: G7X II - need help finding led details.)

I noticed g7x2 is in trunk. I believe an update to camera_list.csv is in order...

I believe this line is missing from camera_list.csv:

g7x2,101a,PREALPHA,,SKIP_AUTOBUILD

I believe this line is missing from camera_list.csv:

Quote

g7x2,101a,PREALPHA,,SKIP_AUTOBUILD

It's not really a port at this point, just some early test code that blinks an LED.

Now, I'm confused. Shouldn't camera_list.csv reflect the platform/ directory structure?

Shouldn't camera_list.csv reflect the platform/ directory structure?

Why would it?  Right now, early work on ports can be shared and archived in the svn repository with no requirement that they actually build or work?  The ports in camera_list.csv are the ones that the autobuild uses to build released versions ( alpha, beta, stable).

Edit :  I suppose an argument could be made for the use of SKIP_AUTOBUILD in the camera_list.csv file ...
 

Edit :  I suppose an argument could be made for the use of SKIP_AUTOBUILD in the camera_list.csv file ...

That would put it in the batch operations that use -noskip. The main one is batch stubs builds, which would be fine including the partial G7X II code.

Doesn't matter to me one way or the other.

Now, I'm confused. Shouldn't camera_list.csv reflect the platform/ directory structure?

It defines how the batch builds are done. Does this matter to you for some reason, or are you just pointing out a possible mistake?

It defines how the batch builds are done. Does this matter to you for some reason, or are you just pointing out a possible mistake?

Any inconsistency between the source tree and camera_list.csv causes the CHIMP camera metadata generator tool to fail.

Edit: I just noticed that the Wifi button is called "WiFi" in g7x2 (as well as in sx700hs, sx60hs, g5x), whereas it's "Wifi" in g7x (and in six others). In addition, s95 has a "Play" button (should be "Playback").
Edit 2: There are also "VIDEO" and "Disp"... starting to look like a lost battle.

Full list:

a1200	Disp	Display
a2200	Disp	Display
a3100	Disp	Display
a3200	Disp	Display
a3300	Disp	Display
d20	VIDEO	Video
g10	Disp	Display
g11	Disp	Display
g12	Disp	Display
g16	Shortcut	Shrtcut
g5x	WiFi	Wifi
g7x2	WiFi	Wifi
s95	Play	Playback
s95	Disp	Display
sx170is	Disp	Display
sx400is	Disp	Display
sx410is	Disp	Display
sx510hs	Disp	Display
sx60hs	WiFi	Wifi
sx700hs	WiFi	Wifi

@reyalp
I guess this needs to be moved to a separate thread.

Any inconsistency between the source tree and camera_list.csv causes the CHIMP camera metadata generator tool to fail.

Sounds like you need to fix your tool then. The CHDK batch (build) operations are working, and they are parsing the csv. It is even possible to make a custom csv (a small subset of the original) and the batch will continue to work just fine.

Quote from: dmitrys on 29 / August / 2019, 14:30:45

Any inconsistency between the source tree and camera_list.csv causes the CHIMP camera metadata generator tool to fail.

Sounds like you need to fix your tool then. The CHDK batch (build) operations are working, and they are parsing the csv.

The tool was specifically designed to fail in such a case, so there's nothing to *fix* there. I suppose I could change it to produce a warning and proceed, although IMHO adding a LED test to trunk wasn't such a great idea.