CPAPtalk.com

Thanks patch and icon works great.

Works awesome on my windows XP system

An OSX version of the software is now available at:
https://s3.amazonaws.com/thefreemanclan ... pyHead.dmg

Again, this should be mainly considered an interim solution for series 60 users until JediMark is back up on his feet and we can get the main version updated. Feedback welcome!

Hello just got switched to the 560 after my other machine died. Just wondering if anyone knows of a source to get the patch for Fedora Core Linux. I would love to get up and running, but can wait until Jedi releases a new version if necessary. Thanks to everyone involved in Sleepyhead fantastic program.

sydneyearly wrote:Hello just got switched to the 560 after my other machine died. Just wondering if anyone knows of a source to get the patch for Fedora Core Linux. I would love to get up and running, but can wait until Jedi releases a new version if necessary. Thanks to everyone involved in Sleepyhead fantastic program.

The source is at:
git@github.com:rich0/rich0-sleepyhead.git
use branch rich-changes (if doing a repo clone add a -b rich-changes to the command)

A Gentoo ebuild (what I use) is in the source tree. Build instructions are on the sleepyhead website and apply as written, other than the URL and branch. Should work fine 32-bit or 64-bit. I guess if there is demand I could build a static binary, or if somebody wants to package it in .rpm or .deb feel free.

Hi All,

Not doing programming much in the last few years, any assistance would be appreciated.

git clone -b rich-changes git@github.com:rich0/rich0-sleepyhead.git
Cloning into 'rich0-sleepyhead'...
Permission denied (publickey).
fatal: The remote end hung up unexpectedly

Thanks in advance,

Brian

BrianD wrote: git clone -b rich-changes git@github.com:rich0/rich0-sleepyhead.git
Cloning into 'rich0-sleepyhead'...
Permission denied (publickey).
fatal: The remote end hung up unexpectedly

Hmm, try this one:
git clone -b rich-test git://github.com/rich0/rich0-sleepyhead.git

I posted a URL that had my username in it, which isn't going to work for anonymous access, and I also renamed the branch to better reflect its long-term purpose.

Also, we're merging the series 60 code into the mainline repository - but series 60 hasn't been merged just yet. rich-test has been rebased so it reflects everything Mark merged into the master repository, plus my prs1 fixes including series 60.

It looks like the patched version doesn't work on some PRS1 60 series BiPAP machines. You get ridiculous AHI over 100, and no event data.

viewtopic.php?f=1&t=84359&p=765271#p765140

archangle wrote:It looks like the patched version doesn't work on some PRS1 60 series BiPAP machines. You get ridiculous AHI over 100, and no event data.

Yep, working on getting some BiPap files to Mark and Rich so that little bug can be squashed.

rich0 wrote:An OSX version of the software is now available at:
https://s3.amazonaws.com/thefreemanclan ... pyHead.dmg

Again, this should be mainly considered an interim solution for series 60 users until JediMark is back up on his feet and we can get the main version updated. Feedback welcome!

Nice fix!

THANK YOU!!!

Nate

Thanks Rich,


I was able to pull down the sources, I will now give building a try. I do have the new bipap machine so I am not sure from some of the other comments it will behave right once built, but I will give it a try anyway.


Brian

Hi Rich,

I can confirm that on Linux the patch doesn't work right for the 60 Series Bipap. It built fine, but most of the graphs were missing and the AHI was with crazy numbers. I saw AHI over 1000. Is there anything I can do to help get this resolved ?


Brian

BrianD wrote:Hi Rich,

I can confirm that on Linux the patch doesn't work right for the 60 Series Bipap. It built fine, but most of the graphs were missing and the AHI was with crazy numbers. I saw AHI over 1000. Is there anything I can do to help get this resolved ?
Brian

This was actually on my to-do list for today (looking up what I need and putting out a request). However, I actually ran into a bug with newer versions of x11/libxcb/mesa so I spent half the day trying to track that down instead.

I might post a request detailing what I need. If you want to PM me a tarball of your sd card that would certainly be sufficient to get started.

Rich

BrianD wrote:Hi Rich,

I can confirm that on Linux the patch doesn't work right for the 60 Series Bipap. It built fine, but most of the graphs were missing and the AHI was with crazy numbers. I saw AHI over 1000. Is there anything I can do to help get this resolved ?


Brian

When I was trying to decode the data formats, it looks like the 50 series APAP and CPAP machines used different software than the 50 series advanced PAP machines. I think the 560 APAP code was derived from the advanced machine software.

There is a table in the source code that looks like it was an early attempt to decode the other formats but now that code is still active and not the new patches. I haven't looked at the new patches but the PR parsing code may need a rewrite as some of the code doesn't agree with other parsing.

-tim wrote:When I was trying to decode the data formats, it looks like the 50 series APAP and CPAP machines used different software than the 50 series advanced PAP machines. I think the 560 APAP code was derived from the advanced machine software.

There is a table in the source code that looks like it was an early attempt to decode the other formats but now that code is still active and not the new patches. I haven't looked at the new patches but the PR parsing code may need a rewrite as some of the code doesn't agree with other parsing.

Yeah, the parser is basically a couple of different routines in one. The ASV and 560 APAP share a common branch and the author of the patch did a decent job minimizing duplication. Other machines have completely independent functions. Having an ASV myself I am more familiar with that branch anyway since I've done some work on it.

I've got a file now, and once i fix some segfaults with some X11/xcb/mesa upgrades I'll try to take a look at it.