
Convert .bvdr files to .marp files
convert_bvdr_marp.Rd
If you mistakenly delete the marport data for a haul, you can retrieve that data through this converter. Before using this script, 1. Open the .bvdr file in Notepad ++ or a similar text editor. 2. Find the uninterpretable character symbol. Often, depending on the editor, this will look like a box or the highlighted letters "SUB". Find and delete (via replace) these characters for the whole document. An error will appear and only part of the file will be read (stopping at the line before where this unsupported symbol is) if you do not edit the data ahead of time. 3. Save the .bvdr file with these changes and use the link to that file below for path_bvdr For an example of what a proper .marp file looks like, refer to system.file("exdata/convert_bvdr_marp/HAUL0001.marp", package = "GAPsurvey")
Usage
convert_bvdr_marp(
path_bvdr = NULL,
make_btd_bth = TRUE,
sort_by_path = TRUE,
verbose = TRUE,
...
)
Arguments
- path_bvdr
Character string. The full path of the .bvdr file you want to convert. For example, path_bvdr <- system.file("exdata/convert_bvdr_marp/20220811-00Za.bvdr", package = "GAPsurvey")
- make_btd_bth
Logical. Should a .btd and bth file be generated?
- sort_by_path
Logical. Should .bvdr files be read in alphabetical order. Note: Keep this TRUE when the original file names are used to ensure NMEA strings are read in chronological order.
- verbose
Logical. Default = FALSE. If you would like a readout of what the file looks like in the console, set to TRUE.
- ...
Optional additional arguments passed to convert_nmea_btd().
Examples
# readLines(system.file("exdata/convert_bvdr_marp/20220811-00Za.bvdr",
# package = "GAPsurvey"))[1:5] # input file
# head(convert_bvdr_marp(
# path_bvdr = system.file("exdata/convert_bvdr_marp/20220811-00Za.bvdr",
# package = "GAPsurvey"),
# verbose = TRUE), 20)
# convert_bvdr_marp(
# path_bvdr = system.file("exdata/convert_bvdr_marp/20220811-00Za.bvdr",
# package = "GAPsurvey"))
# readLines(system.file("exdata/convert_bvdr_marp/20220811-00Za.marp",
# package = "GAPsurvey")) # output file