How to extract osm data using osmar

Question

Similar to this question I get an error from get_osm

library(maptools)
library(osmar)
url <- "http://osmar.r-forge.r-project.org/"
file <- "muenchen.osm.gz"
# download.file(sprintf("%s%s", url, file), file)
# gzip is linux only, on windows I unzipped this manually with 7zip!
# unzip("gzip -d muenchen.osm.gz") 

src <- osmsource_osmosis(file = "140-muenchen.osm")
muc_bbox <- center_bbox(11.575278, 48.137222, 3000, 3000)
muc <- get_osm(x = muc_bbox, source = src)

The error is

Fehler in file(con, "r") : kann Verbindung nicht öffnen Zusätzlich: Warnmeldung: In file(con, "r") : kann Datei 'C:\Users\sqc\AppData\Local\Temp\RtmpQtndOZ\file3bc03a7b5127' nicht öffnen: No such file or directory

I ran RStudio as administrator, I installed gzip, but it still does not work. Any ideas? I am on windows.

(The example is from here)

---

Edit:

My path looks like C:/07 Rprogress/00 Erste Testprogramme/140-muenchen.osm.gz but I hope this is not the problem

file.exists("muenchen.osm.gz")
[1] TRUE
> shell("7z e muenchen.osm.gz")
Der Befehl "7z" ist entweder falsch geschrieben oder
konnte nicht gefunden werden.
Warnmeldung:
In shell("7z e muenchen.osm.gz") :
  '7z e muenchen.osm.gz' Ausführung mit Fehlerkode 1 fehlgeschlagen
> shell.exec("7z e muenchen.osm.gz")
Fehler in shell.exec("7z e muenchen.osm.gz") : 
  '7z e 140-muenchen.osm.gz' nicht gefunden
> system("7z e muenchen.osm.gz")
[1] 127

I follow the instructions from here and set

set PATH=%PATH%;C:\Program Files\7-Zip\
echo %PATH%

Edit 2:

st_layers("myfile.osm", do_count = TRUE)
Driver: OSM 
Available layers:
        layer_name       geometry_type features fields
1           points               Point       61     10
2            lines         Line String        0      9
3 multilinestrings   Multi Line String        0      4
4    multipolygons       Multi Polygon        0     25
5  other_relations Geometry Collection        0      4
Warnmeldungen:
1: In CPL_get_layers(dsn, options, do_count) :
  GDAL Error 1: Non increasing node id. Use OSM_USE_CUSTOM_INDEXING=NO
2: ...

Answer 1

The following code should work for reading into R the muenchen.osm.gz file.

# packages
library(sf)
#> Linking to GEOS 3.8.0, GDAL 3.0.4, PROJ 6.3.1
options(width = 120)

# download muenchen.osm.gz file
download.file(
  url = "http://osmar.r-forge.r-project.org/muenchen.osm.gz", 
  destfile = "muenchen.osm.gz", 
  mode = "wb"
)
# Unzip using 7zip. You can also do this manually. 
shell("7z e muenchen.osm.gz")

I'm coding now with the sf approach. I'm quite sure that it's possible to use osmdata to read in R the same .osm file but I can't make it work for the moment.

# 1. List all the layers
st_layers("muenchen.osm", do_count = TRUE)
#> Driver: OSM 
#> Available layers:
#>         layer_name       geometry_type features fields
#> 1           points               Point    37583     10
#> 2            lines         Line String    56325      9
#> 3 multilinestrings   Multi Line String       96      4
#> 4    multipolygons       Multi Polygon    32328     25
#> 5  other_relations Geometry Collection      337      4

# 2. Read the lines layer (or any other layer)
muenchen_lines <- read_sf("muenchen.osm", layer = "lines")

The printing of the object is quite messy here in SO but it's more clear if you copy and paste the code in R.

# 3. Print the result
muenchen_lines
#> Simple feature collection with 56325 features and 9 fields
#> geometry type:  LINESTRING
#> dimension:      XY
#> bbox:           xmin: 11.30704 ymin: 48.0155 xmax: 11.82806 ymax: 48.28529
#> geographic CRS: WGS 84
#> # A tibble: 56,325 x 10
#>    osm_id name    highway waterway aerialway barrier man_made z_order other_tags                                geometry
#>    <chr>  <chr>   <chr>   <chr>    <chr>     <chr>   <chr>      <int> <chr>                             <LINESTRING [°]>
#>  1 98     <NA>    tertia~ <NA>     <NA>      <NA>    <NA>           4 "\"junction\"=>\"r~ (11.36604 48.1764, 11.36607 4~
#>  2 99     <NA>    tertia~ <NA>     <NA>      <NA>    <NA>           4 "\"junction\"=>\"r~ (11.36614 48.17414, 11.36617 ~
#>  3 100    <NA>    second~ <NA>     <NA>      <NA>    <NA>           6 "\"junction\"=>\"r~ (11.33037 48.18013, 11.33031 ~
#>  4 101    <NA>    second~ <NA>     <NA>      <NA>    <NA>           6 "\"junction\"=>\"r~ (11.32986 48.18367, 11.32973 ~
#>  5 102    <NA>    primary <NA>     <NA>      <NA>    <NA>           7 "\"junction\"=>\"r~ (11.45455 48.1672, 11.45424 4~
#>  6 680    <NA>    motorw~ <NA>     <NA>      <NA>    <NA>          29 "\"bridge\"=>\"yes~ (11.54136 48.22608, 11.543 48~
#>  7 1939   Gerani~ reside~ <NA>     <NA>      <NA>    <NA>           3  <NA>               (11.48282 48.10904, 11.48291 ~
#>  8 1945   Sonnen~ reside~ <NA>     <NA>      <NA>    <NA>           3  <NA>               (11.48714 48.10564, 11.48732 ~
#>  9 122956 Berber~ reside~ <NA>     <NA>      <NA>    <NA>           3 "\"maxspeed\"=>\"3~ (11.42325 48.15542, 11.42308 ~
#> 10 122957 Otto-K~ reside~ <NA>     <NA>      <NA>    <NA>           3 "\"maxspeed\"=>\"3~ (11.42276 48.15559, 11.42251 ~
#> # ... with 56,315 more rows

# 4. Plot
par(mar = rep(0, 4))
plot(st_geometry(muenchen_lines))

^{Created on 2020-05-26 by the reprex package (v0.3.0)}

You should also check the osmdata vignettes if you want to build your own queries for downloading data from Open Street Map.

References

If you want to check a few references on sf I would suggest reading the vignettes (see the Articles) of the package and Chapters 1 to 7 of Geocomputation with R. If you want to check how to use osmdata to create your own queries to Open Street Map I suggest reading the vignettes. There are a few differences between the two approaches but you can read them in the vignettes.

AFAIK the suggested way (see here, here and here) for reading .osm data into R using sf is 1) transform the .osm data into .gpkg format (you can use sf::gdal_utils) and 2) read the .gpkg file. If you can share the .osm file I can check it from my laptop.

The error that you show in the OP is mentioned here and if you can share the .osm file I can test reading it.