Cumulus Support

Just a quick note for those people using Cumulus Toolbox -- if you are also testing version 1.9.1 Beta of Cumulus please do not use the "File Editor" in 'Toolbox as it is not aware of the additional fields added to the log files in Cumulus 1.9.1!

I hope to have a new version out in the next few days, with several other new features too

Is there a way that we could avoid the problem of extra fields breaking the editor? I assume the grid could just be made to load whatever was there, but the headings would be a problem? I don't want to add a header line to the files, but what if there were a separate file which listed the fields? Or does that make it all to difficult?

At the moment I am building the grid to load whatever it reads, and crucially also save in that same way it found it. If there are extra columns in the data they will appear as 'unknown' in the header.

A separate file for headings could be an idea. I guess we could store a core config centrally -- if Toolbox finds extra columns it would refer to this file for updated config.

daj wrote:At the moment I am building the grid to load whatever it reads, and crucially also save in that same way it found it. If there are extra columns in the data they will appear as 'unknown' in the header.

Ah, I see, that's good. So it does work, but just has these 'unknown' headings for the new columns?

steve wrote:...So it does work....

'Will work' that way in the next update; which will include the new headings, but do 'unknown' if you add extras in the future. However, I like the idea of a header file to refer to for future expansion

By saying 'building the grid' I really meant 'building' as is coding it for the next release. I can see how it was confusing

OK, got it.

So how about if the Cumulus installer creates two files, one called dayfileheader.txt and the other monthlyfileheader.txt; each one has a single line with a comma-separated list of headings for each file? Or should we avoid the dreaded variable comma and use a different separator?

Perfect.

Let's use a better separator -- what about the | ? I think it is relatively safe

Your choice

Yes, that vertical bar thing whatever it's called.

Slightly off topic for this section, butI was just thinking, why can't the headers be the first row of the data files themselves, with the same separators as the data. The big advantage would be that programs such as Excel would be able to import data including the header rows.

• But I suppose it depends on two things:
• Cumulus being able to create the data files with header rows.
• Cumulus being able to ignore the first line of the file as being a header and not data.

Alternatively, I would prefer the same separator rather than the "pipe character". This way it could be slotted into the monthly file after the month is complete for easy data manipulation. Then again, I am being lazy, as I could create my own header files.

Bob

tjaliwalpa wrote: [*]Cumulus being able to create the data files with header rows.
[*]Cumulus being able to ignore the first line of the file as being a header and not data.

Yes; but I'd prefer to keep things simple rather than introduce a possible way for things to go wrong.

Alternatively, I would prefer the same separator rather than the "pipe character". This way it could be slotted into the monthly file after the month is complete for easy data manipulation.

Cumulus would have to change as in your second point if that were to work.

The problem with using the comma for the separator is that in some countries, it's not a comma. If I knew then what I know now, I wouldn't have used the dratted thing.

just to complete this post, a new build is available

https://cumulus.hosiene.co.uk/viewtopic.php?f=18&t=3549