Reads one cross-sectional traffic volume ("type B") CSV file published by
the Japan Road Traffic Information Center. The files are CP932-encoded and
carry a header row; this function drops that row, supplies stable column
names, parses datetime in JST, and normalizes location_name with NFKC.
Value
A data.table of 10 columns, keyed on datetime:
- datetime
Observation time (
POSIXct,Asia/Tokyo).- source_code
Information source code.
- location_no
Observation location number.
- location_name
Observation location, as
from\U{2192}to.- meshcode10km
10 km mesh code.
- link_type
Link type.
- link_no
Link number.
- traffic
Traffic volume; missing observations are
NA.- to_link_end_10m
Distance to the link end, in units of 10 m. Kept as character so that whatever the source writes – padding included – survives the read.
- link_ver
Link version;
NAfor files published before 2018-02.
Details
The published layout has changed over time, so the header row is inspected before the file is read:
Files from 2018-02 onwards have 10 columns. Earlier files have 9 and no
link_ver; for those,link_veris filled withNAso that the returned columns are the same in either case.A file whose first row is already an observation (no header) is read in full; the first row is not consumed as a header.
datetimeis written as2026/06/01 00:00by most providers, but the pre-2018-02 files also use2017/8/1 0:00:00. Both are parsed, and values that survive asNAare reported with a warning rather than passed on silently.
Bytes that are not valid CP932 cannot be decoded and leave NA in
location_name. That is reported with a warning too, because the rest of
the row still looks intact.
Examples
read_jartic_traffic(
system.file("dummy", "type_b.csv", package = "jarticr")
)
#> Key: <datetime>
#> datetime source_code location_no location_name meshcode10km
#> <POSc> <char> <int> <char> <char>
#> 1: 2022-11-01 00:00:00 36 1 徳島駅前→県庁前 513376
#> 2: 2022-11-01 00:00:00 36 2 APAホテル前→南口 513376
#> 3: 2022-11-01 01:00:00 36 1 徳島駅前→県庁前 513376
#> 4: 2022-11-01 01:00:00 36 2 APAホテル前→南口 513376
#> link_type link_no traffic to_link_end_10m link_ver
#> <int> <int> <int> <char> <int>
#> 1: 1 101 120 0050 202200
#> 2: 1 102 45 0120 202200
#> 3: 1 101 80 0050 202200
#> 4: 1 102 NA 0120 202200