ARCHIVE [OSCAR feature request]

[srlevine1]

Is there a safe way to archive a date range of data so they remain accessible by date if reimported? 

Thanks,

Steve

[sawinglogz]

I'm not quite sure what you're asking. What do you mean by reimported?

On a guess, I think you're probably best creating a new profile and saving off your old profile somewhere. I think you could then delete much of the old profile folder in your OSCAR_Data folder, and then just put it back when you want to review it. But I haven't tested that and don't know what kind of hiccups you might encounter.

[srlevine1]

I'm not quite sure what you're asking. What do you mean by reimported?

On a guess, I think you're probably best creating a new profile and saving off your old profile somewhere. I think you could then delete much of the old profile folder in your OSCAR_Data folder, and then just put it back when you want to review it. But I haven't tested that and don't know what kind of hiccups you might encounter.

I would not like to lose the data and would like to preserve it for further use -- hence archive the data by date range, store it on a DVD or USB drive and then  "re-import" the data back into the system if it is needed for historical reports and progress checks.

Perhaps I should request additional functionality to be added to Oscar ...

Archive to [path][foldername]
Date Range [00/00/0000] to [00/00/0000]
GO

Import from [path][foldername]
Date Range [00/00/0000] to [00/00/0000]
CONVERT TO LATEST VERSION [Y/N]
GO

The format could be proprietary for convenience. -- Perhaps with a record header that includes the version which would allow for more functionality.

[pholynyk]

Unidee has asked for something similar, and I said then there were some difficulties with the STR.edf file. It starts on some date, and is added to each day without regard for calendar boundaries.

Basiscally, we would have to write a new STR file, or rather a set of STR files, since any gaps require a new start date. I have been thinking about reading the STR file one day at a time during the import; that could be extended to write something one day at a time.

So I would say it's possible, but it would probably come as part of a complete re-design and re-write of the ResMed loader. I'm not sure how the other loaders might do it; right now the oximeter data is not backed up for re-import, for example.

[unidee]

Yes. I have asked backuping old data with range of days. For example last year to backup, last month to backup etc. And then user could be able to remove that data from the data folder. And when he or she wants look again the bacuped data, he or she could restore. Now the term is not backup than archive. The operation is similair.

I got answer that Oscar data is difficult to separate to such ranges.

[unidee]

right now the oximeter data is not backed up for re-import, for example.

Oximetry data should be backed up. Now I backup data with copying whole OSCAR_data folder to usb-stick. Or tar & gzip it to tarball and copy it to some media like usb memory-stick. I think that oximetry data goes among other data to backup with this method.

And the cpap data importin can be reorganized. Then it is possible to split data to time ranges easily. I think that one month is a small enough piece of data to backup or archive or simply remove. And I have some data from year  2015. I need not them anymore, but it would be nice to look them in future. So I can remember how hard was it in first year to sleep with cpap machine. 

[sawinglogz]

There are LOTS of improvements that would make a huge difference. The problem is that there are so many.

Once 1.1.1 is out I plan to tackle oximetry import. Backing up imports (especially for direct downloads) is one of the items on that list.

[Gideon]

There is so much to do that one of the problems is deciding when to call a new version. There are always things that are left undone.
I expect v1.2.0 to contain, once again, major rewrites to allow future changes.

[unidee]

Do an archiving or backuping feature carefully and with time. We users can make backups manually now. When you carefully plan those functions, there are fewer bugs to fix. I think so.

And right now, I think Oscar is working really well, at least for me. Thanks for that.