Fast and simple url parser for R. Initially developed for the
paws.common package.
urlparse::url_parse("https://user:pass@host.com:8000/path?query=1#fragment")
#> $scheme
#> [1] "https"
#>
#> $user
#> [1] "user"
#>
#> $password
#> [1] "pass"
#>
#> $host
#> [1] "host.com"
#>
#> $port
#> [1] "8000"
#>
#> $path
#> [1] "/path"
#>
#> $raw_path
#> [1] ""
#>
#> $query
#> $query$query
#> [1] "1"
#>
#>
#> $raw_query
#> [1] "query=1"
#>
#> $fragment
#> [1] "fragment"You can install the development version of urlparse like so:
remotes::install_github("dyfanjones/urlparse")r-universe installation:
install.packages("urlparse", repos = c("https://dyfanjones.r-universe.dev", "https://cloud.r-project.org"))This is a basic example which shows you how to solve a common problem:
library(urlparse)url_encoder("foo = bar + 5")
#> [1] "foo%20%3D%20bar%20%2B%205"
url_decoder(url_encoder("foo = bar + 5"))
#> [1] "foo = bar + 5"Similar to python’s from urllib.parse import quote,
urlparse::url_encoder supports the safe
parameter. The additional ASCII characters that should not be
encoded.
from urllib.parse import quote
quote("foo = bar + 5", safe = "+")
#> 'foo%20%3D%20bar%20+%205'url_encoder("foo = bar + 5", safe = "+")
#> [1] "foo%20%3D%20bar%20+%205"
url <- "http://example.com"
set_scheme(url, "https") |>
set_port(1234L) |>
set_path("foo/bar") |>
set_query("baz") |>
set_fragment("quux")
#> [1] "https://example.com:1234/foo/bar?baz#quux"
url_modify(url, scheme = "https", port = 1234, path = "foo/bar", query = "baz", fragment = "quux")
#> [1] "https://example.com:1234/foo/bar?baz#quux"Note: it is faster to use url_modify rather than piping
the set_* functions. This is because urlparse
has to parse the url within each set_* to modify the
url.
url <- "http://example.com"
bench::mark(
piping = {set_scheme(url, "https") |>
set_port(1234L) |>
set_path("foo/bar") |>
set_query("baz") |>
set_fragment("quux")},
single_function = url_modify(url, scheme = "https", port = 1234, path = "foo/bar", query = "baz", fragment = "quux")
)
#> # A tibble: 2 × 6
#> expression min median `itr/sec` mem_alloc `gc/sec`
#> <bch:expr> <bch:tm> <bch:tm> <dbl> <bch:byt> <dbl>
#> 1 piping 5.29µs 5.86µs 169576. 0B 0
#> 2 single_function 1.64µs 1.8µs 507863. 0B 0url <- "https://user:pass@host.com:8000/path?query=1#fragment"
(bm <- bench::mark(
urlparse = urlparse::url_parse(url),
httr2 = httr2::url_parse(url),
curl = curl::curl_parse_url(url),
urltools = urltools::url_parse(url),
check = F
))
#> # A tibble: 4 × 6
#> expression min median `itr/sec` mem_alloc `gc/sec`
#> <bch:expr> <bch:tm> <bch:tm> <dbl> <bch:byt> <dbl>
#> 1 urlparse 1.68µs 1.84µs 503156. 0B 0
#> 2 httr2 64.86µs 68.59µs 14312. 560.9KB 17.4
#> 3 curl 27.22µs 28.54µs 34390. 48.78KB 13.8
#> 4 urltools 124.35µs 129.03µs 7604. 2.17MB 20.9
show_relative(bm)
#> # A tibble: 4 × 6
#> expression min median `itr/sec` mem_alloc `gc/sec`
#> <bch:expr> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 urlparse 1 1 66.2 NaN NaN
#> 2 httr2 38.6 37.2 1.88 Inf Inf
#> 3 curl 16.2 15.5 4.52 Inf Inf
#> 4 urltools 74.0 69.9 1 Inf Inf
ggplot2::autoplot(bm)
#> Loading required namespace: tidyr
Note: urltools encode special characters to lower case
hex i.e.: “?” -> “%3f” instead of “%3F”
string <- "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-._~`!@#$%^&*()=+[{]}\\|;:'\",<>/? "
(bm <- bench::mark(
urlparse = urlparse::url_encoder(string),
curl = curl::curl_escape(string),
urltools = urltools::url_encode(string),
base = URLencode(string, reserved = T),
check = F
))
#> # A tibble: 4 × 6
#> expression min median `itr/sec` mem_alloc `gc/sec`
#> <bch:expr> <bch:tm> <bch:tm> <dbl> <bch:byt> <dbl>
#> 1 urlparse 1.48µs 1.56µs 623378. 208B 0
#> 2 curl 2.3µs 2.42µs 399842. 3.06KB 0
#> 3 urltools 2.42µs 2.67µs 370964. 2.48KB 0
#> 4 base 79.09µs 83.15µs 11703. 28.59KB 8.24
show_relative(bm)
#> # A tibble: 4 × 6
#> expression min median `itr/sec` mem_alloc `gc/sec`
#> <bch:expr> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 urlparse 1 1 53.3 1 NaN
#> 2 curl 1.56 1.55 34.2 15.0 NaN
#> 3 urltools 1.64 1.71 31.7 12.2 NaN
#> 4 base 53.6 53.4 1 141. Inf
ggplot2::autoplot(bm)
string <- "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-._~`!@#$%^&*()=+[{]}\\|;:'\",<>/? "
url <- paste0(sample(strsplit(string, "")[[1]], 1e4, replace = TRUE), collapse = "")
(bm <- bench::mark(
urlparse = urlparse::url_encoder(url),
curl = curl::curl_escape(url),
urltools = urltools::url_encode(url),
base = URLencode(url, reserved = T, repeated = T),
check = F,
filter_gc = F
))
#> # A tibble: 4 × 6
#> expression min median `itr/sec` mem_alloc `gc/sec`
#> <bch:expr> <bch:tm> <bch:tm> <dbl> <bch:byt> <dbl>
#> 1 urlparse 86.06µs 87.41µs 11291. 15.8KB 0
#> 2 curl 92.95µs 94.26µs 10209. 0B 0
#> 3 urltools 238.7µs 244.16µs 3950. 15.8KB 0
#> 4 base 6.72ms 6.84ms 141. 333.2KB 9.91
show_relative(bm)
#> # A tibble: 4 × 6
#> expression min median `itr/sec` mem_alloc `gc/sec`
#> <bch:expr> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 urlparse 1 1 80.2 Inf NaN
#> 2 curl 1.08 1.08 72.5 NaN NaN
#> 3 urltools 2.77 2.79 28.1 Inf NaN
#> 4 base 78.1 78.2 1 Inf Inf
ggplot2::autoplot(bm)