bwrc/embla-r

An R-package for reading physiologic signal data stored in the Embla Data Format (EBM).

embla-r's Issues

Time format issue in parse_ebm_record

Hi there,

thanks for the great embla-r funtions which I have been trying out. While doing so I noticed that there is a formatting error /typo for the extracted starting time in the parse_ebm_record, line 124
"%d%02d%02dT%02d%02d## %0d2.%02d" which results in considerable time shifts of the signal in particular compared to any extracted annotation.
With kind regards
Steph

Annotation

Does this package read annotations from *.ebm file? I assume both, the annotations and the signal, are stored in the same *.ebm file, please correct me if wrong.

cant install package

Error: Failed to install 'ebm' from GitHub:
(converted from warning) installation of package ‘C:/Users/Kai'/AppData/Local/Temp/RtmpekVnb0/filed5440a283f/ebm_0.9.0.tar.gz’ had non-zero exit status

i cant install the package, this error comes out

Issue with joining > 2 parts of recording

Hi again,

just wanted to let you know that there was another issue, when the recording signal has more than 2 starts, in read_ebm lines 278-282, where i and i+1 should be 2_(i-1) and 2_(i-1) + 1

        ## create filler time and data vector
        time_vector[[2*(i-1)]] <- seq.int(from = t_stop[[i - 1]], to = t_start[[i]], by = (1 / header$EBM_R_SAMPLING_RATE))
        data_vector[[2*(i-1)]] <- rep(NA, length(time_vector[[2*(i-1)]]))

        time_vector[[2*(i-1) + 1]] <- t_start[[i]] + seq.int(0, (length(signal[[i]]) - 1)) / header$EBM_R_SAMPLING_RATE
        data_vector[[2*(i-1) + 1]] <- signal[[i]]

in the current case a a recording with 3 starts, the second part will be missing etc.

Cheers, steph

Unicode input file

When the .ebm file contains unicode character (i.e. Japanese, Korean etc) in the name field, the loading crashes.
i.e.

Warning: unable to translate ' <bd><c5><c8><f1><b8><f1>' to a wide stringError in gsub("~\\s+|\\s+$", "", out) : input string 1 is invalid

Is there any way to resolve this?

Problems reading files from "RemLogic PSG"

Hello!
I have EMB files from a polysomnography from a sleep lab. They were created with the "RemLogic PSG" software afaik. I'm trying to read them with your library, but am experiencing some problems.

This line:

time_vector[[i]] <- seq.int(from = t_stop[[i - 1]], to = t_start[[i]], by = (1 / header$EBM_R_SAMPLING_RATE))

Fails with:

Error during wrapup: wrong sign in 'by' argument

I think from is higher than to…

What software does your library support? Do you have any hints for me for debugging?
I can see my file has EBM version 4 according to the headers:

$EBM_R_VERSION
[1] "4.0"

Here's an example file: https://drive.google.com/open?id=0B05BdPZPk27DT084ZjZ1bnI3YUE