graph for each column

[Peal]

Hello Gustaf,
don't worry about your timing, this is still a hobby project

Surprisingly, the delimiter is a tab:

Delimiter.png (109.41 KiB) Viewed 26592 times

Maybe without a title row, it's first searching for the delimiter (comma being the first valid one) and only after that trying to identify the decimal separator?

Here's some raw data for testing:

Code: Select all

#4326							
#Kraft 	#Geschwindigkeit 	#Pneumatik Druck 	#Hydraulik Druck 	#Weg 	#INI 	#INI 	#Signal
-2,52	0,00	0,36	0,16	103,73	0,00	0,82	1,00
-2,53	0,00	0,36	0,21	103,76	0,00	0,82	1,00
-2,50	0,00	0,36	0,26	103,76	0,00	0,82	1,00
-2,53	0,00	0,36	0,16	103,76	0,00	0,82	1,00
-2,52	0,00	0,36	0,20	103,76	0,00	0,82	1,00
-2,52	0,00	0,36	0,30	103,76	0,00	0,82	1,00
-2,53	0,00	0,36	0,44	103,73	0,00	0,82	1,00
-2,50	0,00	0,36	0,43	103,73	0,00	0,82	1,00
-2,53	0,00	0,36	0,33	103,73	0,00	0,82	1,00
-2,52	0,00	0,36	0,25	103,73	0,00	0,82	1,00
-2,52	0,00	0,36	0,26	103,73	0,00	0,82	1,00
-2,53	0,00	0,36	0,26	103,73	0,00	0,82	1,00
-2,51	0,00	0,36	0,13	103,73	0,00	0,82	1,00
-2,53	0,00	0,36	0,11	103,73	0,00	0,82	1,00
-2,52	0,00	0,36	0,14	103,73	0,00	0,82	1,00
-2,51	0,00	0,36	0,18	103,73	0,00	0,82	1,00
-2,53	0,00	0,36	0,24	103,76	0,00	0,82	1,00
-2,51	0,00	0,36	0,23	103,73	0,00	0,82	1,00
-2,52	0,00	0,36	0,34	103,76	0,00	0,82	1,00
-2,52	0,00	0,36	0,34	103,73	0,00	0,82	1,00
-2,51	0,00	0,36	0,31	103,73	0,00	0,82	1,00
-2,53	0,00	0,36	0,33	103,76	0,00	0,82	1,00
-2,51	0,00	0,36	0,24	103,73	0,00	0,82	1,00
-2,52	0,00	0,36	0,24	103,73	0,00	0,82	1,00
-2,52	0,00	0,36	0,31	103,73	0,00	0,82	1,00

What i forgot to tell you: without the first "#" in the title row, everything gets detected correctly.
So that's my current workaround.

Have a nice week!

[Waveworks]

Hi Peal, thank you for your feedback and the example file, that's very helpful. It turns out you're better at debugging my application than I am, because you're absolutely right: The value separator detection simply starts from the beginning and looks for either a comma, a semicolon or a tab, and when the first two rows are commented out, it will hit a comma before a tab and will take that comma as a value separator. So this is a bug - it should of course ignore commas when you have explicitly said that comma is your decimal separator. Sorry about that! I will fix it for the next version.

I'm glad you found a workaround. Another workaround is to simply set the Value separator in the Preferences -> CSV Import to Tab.

Best regards,
Gustaf

[Peal]

Always happy to help

I understand that this is not a trivial problem.
Just changing the order of detection would only shift the potential for bugs to auto detection of decimal separators with explicitly defined delimiters.

You somehow need to prioritise hard defined symbols over the automatic detection algorithms.

Poking around a bit more, i found out that it's possible to set the comma character as delimiter and decimal point at the same time
One more cultural difference that introduces programming problems

Although i'm looking forward to the new version, don't pressure yourself because of this problem. The current workaround is easy and fast.