R client for Pleiades
More stable CRAN version
install.packages("pleiades")
Dev version
remotes::install_github("ropensci/pleiades")
library("pleiades")
pl_status()
#> $num_places
#> [1] 37283
#>
#> $num_locations
#> [1] 40274
#>
#> $num_names
#> [1] 33346
pl_places
uses the Pleiades API. Just a subst of output for brevity
pl_places(place_id = 579885)[1:2]
#> $features
#> list()
#>
#> $contributors
#> $contributors[[1]]
#> $contributors[[1]]$username
#> [1] "bkiesling"
#>
#> $contributors[[1]]$homepage
#> NULL
#>
#> $contributors[[1]]$name
#> [1] "Brady Kiesling"
#>
#> $contributors[[1]]$uri
#> [1] "https://pleiades.stoa.org/author/bkiesling"
#>
#>
#> $contributors[[2]]
#> $contributors[[2]]$username
#> [1] "sgillies"
#>
#> $contributors[[2]]$homepage
#> NULL
#>
#> $contributors[[2]]$name
#> [1] "Sean Gillies"
#>
#> $contributors[[2]]$uri
#> [1] "https://pleiades.stoa.org/author/sgillies"
#>
#>
#> $contributors[[3]]
#> $contributors[[3]]$username
#> NULL
#>
#> $contributors[[3]]$name
#> [1] "Chelsea Lee"
#>
#>
#> $contributors[[4]]
#> $contributors[[4]]$username
#> [1] "jbecker"
#>
#> $contributors[[4]]$homepage
#> NULL
#>
#> $contributors[[4]]$name
#> [1] "Jeffrey Becker"
#>
#> $contributors[[4]]$uri
#> [1] "https://pleiades.stoa.org/author/jbecker"
#>
#>
#> $contributors[[5]]
#> $contributors[[5]]$username
#> [1] "rbrowning"
#>
#> $contributors[[5]]$homepage
#> NULL
#>
#> $contributors[[5]]$name
#> [1] "Rachaelle L Browning"
#>
#> $contributors[[5]]$uri
#> [1] "https://pleiades.stoa.org/author/rbrowning"
#>
#>
#> $contributors[[6]]
#> $contributors[[6]]$username
#> NULL
#>
#> $contributors[[6]]$name
#> [1] "DARMC"
#>
#>
#> $contributors[[7]]
#> $contributors[[7]]$username
#> NULL
#>
#> $contributors[[7]]$name
#> [1] "R. Talbert"
#>
#>
#> $contributors[[8]]
#> $contributors[[8]]$username
#> [1] "jahlfeldt"
#>
#> $contributors[[8]]$homepage
#> NULL
#>
#> $contributors[[8]]$name
#> [1] "Johan Ã…hlfeldt"
#>
#> $contributors[[8]]$uri
#> [1] "https://pleiades.stoa.org/author/jahlfeldt"
#>
#>
#> $contributors[[9]]
#> $contributors[[9]]$username
#> [1] "alecpm"
#>
#> $contributors[[9]]$homepage
#> NULL
#>
#> $contributors[[9]]$name
#> [1] "Alec Mitchell"
#>
#> $contributors[[9]]$uri
#> [1] "https://pleiades.stoa.org/author/alecpm"
#>
#>
#> $contributors[[10]]
#> $contributors[[10]]$username
#> [1] "arabinowitz"
#>
#> $contributors[[10]]$homepage
#> NULL
#>
#> $contributors[[10]]$name
#> [1] "Adam Rabinowitz"
#>
#> $contributors[[10]]$uri
#> [1] "https://pleiades.stoa.org/author/arabinowitz"
#>
#>
#> $contributors[[11]]
#> $contributors[[11]]$username
#> [1] "thomase"
#>
#> $contributors[[11]]$homepage
#> NULL
#>
#> $contributors[[11]]$name
#> [1] "Tom Elliott"
#>
#> $contributors[[11]]$uri
#> [1] "https://pleiades.stoa.org/author/thomase"
#>
#>
#> $contributors[[12]]
#> $contributors[[12]]$username
#> [1] "pmotylewicz"
#>
#> $contributors[[12]]$homepage
#> NULL
#>
#> $contributors[[12]]$name
#> [1] "Pierre Motylewicz"
#>
#> $contributors[[12]]$uri
#> [1] "https://pleiades.stoa.org/author/pmotylewicz"
#>
#>
#> $contributors[[13]]
#> $contributors[[13]]$username
#> NULL
#>
#> $contributors[[13]]$name
#> [1] "April Kissinger"
#>
#>
#> $contributors[[14]]
#> $contributors[[14]]$username
#> NULL
#>
#> $contributors[[14]]$name
#> [1] "Eric Shea"
#>
#>
#> $contributors[[15]]
#> $contributors[[15]]$username
#> [1] "rmhorne"
#>
#> $contributors[[15]]$homepage
#> NULL
#>
#> $contributors[[15]]$name
#> [1] "Ryan Horne"
#>
#> $contributors[[15]]$uri
#> [1] "https://pleiades.stoa.org/author/rmhorne"
#>
#>
#> $contributors[[16]]
#> $contributors[[16]]$username
#> [1] "swright"
#>
#> $contributors[[16]]$homepage
#> NULL
#>
#> $contributors[[16]]$name
#> [1] "Sterling Wright"
#>
#> $contributors[[16]]$uri
#> [1] "https://pleiades.stoa.org/author/swright"
Pleiades nicely provides their bulk data (for locations, names, and places) in various formats, including .csv
. We’ve created three functions pl_search_loc()
, pl_search_names()
, and pl_search_places()
to search each of those datasets. As these are relatively large (approx 40K rows by 30 columns), dplyr
is a nice approach to dealing with big-ish data. At this time dplyr
is a dependency.
You can run pl_cache()
first to get the raw data files from Pleiades
pl_cache()
Or, that function is run internally in pl_search*()
functions for you.
Search across all tables in one database, or separately locations, names, or places. You can return the data (that is, a dplyr
representation of the data) if you don’t pass anything to the function call:
Gives each table in a list
pl_search()
#> [[1]]
#> # Source: table<locations> [?? x 25]
#> # Database: sqlite 3.33.0
#> # [/Users/sckott/Library/Caches/pleiades/pleiades_all.sqlite3]
#> authors bbox created creators currentVersion description featureType
#> <chr> <chr> <chr> <chr> <int> <chr> <chr>
#> 1 Becker… 13.4… 2013-0… jbecker 4 "The post-… settlement,
#> 2 Becker… 11.6… 2013-0… jbecker 3 "A major u… temple-2,
#> 3 Spann,… -3.6… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> 4 Spann,… -2.0… 2010-0… P.O. Sp… 0 "" unknown
#> 5 Spann,… -2.4… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> 6 Spann,… 0.10… 2011-0… P.O. Sp… 1 "1:1 milli… unknown
#> 7 Spann,… -3.9… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> 8 Spann,… 1.43… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> 9 Becker… 1.43… 2013-0… jbecker 1 "verified … island
#> 10 Spann,… -3.3… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> # … with more rows, and 18 more variables: geometry <chr>, id <chr>,
#> # locationPrecision <chr>, locationType <chr>, maxDate <int>, minDate <int>,
#> # modified <chr>, path <chr>, pid <chr>, reprLat <dbl>, reprLatLong <chr>,
#> # reprLong <dbl>, tags <chr>, timePeriods <chr>, timePeriodsKeys <chr>,
#> # timePeriodsRange <chr>, title <chr>, uid <chr>
#>
#> [[2]]
#> # Source: table<names> [?? x 26]
#> # Database: sqlite 3.33.0
#> # [/Users/sckott/Library/Caches/pleiades/pleiades_all.sqlite3]
#> authors bbox created creators currentVersion description extent id
#> <chr> <chr> <chr> <chr> <int> <chr> <chr> <chr>
#> 1 Spann,… "-3.… 2010-0… P.O. Sp… NA "" "{\"t… cons…
#> 2 Spann,… "-3.… 2010-0… P.O. Sp… NA "" "{\"t… cons…
#> 3 Spann,… "-3.… 2010-0… P.O. Sp… NA "" "{\"t… kond…
#> 4 Becker… "-3.… 2016-0… jbecker 1 "An ethnic… "{\"t… cons…
#> 5 Spann,… "-2.… 2010-0… P.O. Sp… 1 "" "{\"t… cont…
#> 6 Spann,… "-2.… 2010-0… P.O. Sp… NA "" "{\"t… cont…
#> 7 Spann,… "0.1… 2010-0… P.O. Sp… 0 "" "{\"t… dian…
#> 8 Spann,… "0.1… 2010-0… P.O. Sp… 1 "" "{\"t… heme…
#> 9 Becker… "" 2012-1… jbecker 1 "" "" denia
#> 10 Spann,… "-3.… 2010-0… P.O. Sp… NA "" "{\"t… ebura
#> # … with more rows, and 18 more variables: locationPrecision <chr>,
#> # maxDate <int>, minDate <int>, modified <chr>, nameAttested <chr>,
#> # nameLanguage <chr>, nameTransliterated <chr>, path <chr>, pid <chr>,
#> # reprLat <dbl>, reprLatLong <chr>, reprLong <dbl>, tags <chr>,
#> # timePeriods <chr>, timePeriodsKeys <chr>, timePeriodsRange <chr>,
#> # title <chr>, uid <chr>
#>
#> [[3]]
#> # Source: table<places> [?? x 26]
#> # Database: sqlite 3.33.0
#> # [/Users/sckott/Library/Caches/pleiades/pleiades_all.sqlite3]
#> authors bbox connectsWith created creators currentVersion description extent
#> <chr> <chr> <chr> <chr> <chr> <int> <chr> <chr>
#> 1 Becker… 13.4… "413005" 2016-1… jbecker… 1 The post-R… "{\"t…
#> 2 Becker… 11.6… "413393" 2016-1… jbecker… 2 A major ur… "{\"t…
#> 3 Spann,… -3.6… "" 2010-0… P.O. Sp… 15 An ancient… "{\"t…
#> 4 Spann,… -0.5… "" 2010-0… P.O. Sp… 4 Contestani… "{\"t…
#> 5 Spann,… -2.4… "" 2010-0… P.O. Sp… 6 An ancient… "{\"t…
#> 6 Spann,… 0.10… "269561" 2018-0… P.O. Sp… 8 An ancient… "{\"t…
#> 7 Spann,… -3.9… "" 2010-0… P.O. Sp… 3 An ancient… "{\"t…
#> 8 Spann,… 1.43… "265884" 2010-0… P.O. Sp… 5 An ancient… "{\"t…
#> 9 Spann,… 1.43… "266001" 2010-0… P.O. Sp… 7 Ebusus Ins… "{\"t…
#> 10 Spann,… -3.3… "" 2010-0… P.O. Sp… 5 An ancient… "{\"t…
#> # … with more rows, and 18 more variables: featureTypes <chr>,
#> # geoContext <chr>, hasConnectionsWith <chr>, id <dbl>,
#> # locationPrecision <chr>, maxDate <int>, minDate <int>, modified <chr>,
#> # path <chr>, reprLat <dbl>, reprLatLong <chr>, reprLong <dbl>, tags <chr>,
#> # timePeriods <chr>, timePeriodsKeys <chr>, timePeriodsRange <chr>,
#> # title <chr>, uid <chr>
Locations only
pl_search_loc()
#> # Source: table<locations> [?? x 25]
#> # Database: sqlite 3.33.0
#> # [/Users/sckott/Library/Caches/pleiades/pleiades_locations.sqlite3]
#> authors bbox created creators currentVersion description featureType
#> <chr> <chr> <chr> <chr> <int> <chr> <chr>
#> 1 Becker… 13.4… 2013-0… jbecker 4 "The post-… settlement,
#> 2 Becker… 11.6… 2013-0… jbecker 3 "A major u… temple-2,
#> 3 Spann,… -3.6… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> 4 Spann,… -2.0… 2010-0… P.O. Sp… 0 "" unknown
#> 5 Spann,… -2.4… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> 6 Spann,… 0.10… 2011-0… P.O. Sp… 1 "1:1 milli… unknown
#> 7 Spann,… -3.9… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> 8 Spann,… 1.43… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> 9 Becker… 1.43… 2013-0… jbecker 1 "verified … island
#> 10 Spann,… -3.3… 2011-0… P.O. Sp… 1 "1M scale … unknown
#> # … with more rows, and 18 more variables: geometry <chr>, id <chr>,
#> # locationPrecision <chr>, locationType <chr>, maxDate <int>, minDate <int>,
#> # modified <chr>, path <chr>, pid <chr>, reprLat <dbl>, reprLatLong <chr>,
#> # reprLong <dbl>, tags <chr>, timePeriods <chr>, timePeriodsKeys <chr>,
#> # timePeriodsRange <chr>, title <chr>, uid <chr>
Or you can submit a query:
pl_search_loc("SELECT * FROM locations limit 5")
#> # Source: SQL [?? x 25]
#> # Database: sqlite 3.33.0
#> # [/Users/sckott/Library/Caches/pleiades/pleiades_locations.sqlite3]
#> authors bbox created creators currentVersion description featureType geometry
#> <chr> <chr> <chr> <chr> <int> <chr> <chr> <chr>
#> 1 Becker… 13.4… 2013-0… jbecker 4 "The post-… settlement, "{\"typ…
#> 2 Becker… 11.6… 2013-0… jbecker 3 "A major u… temple-2, "{\"typ…
#> 3 Spann,… -3.6… 2011-0… P.O. Sp… 1 "1M scale … unknown "{\"typ…
#> 4 Spann,… -2.0… 2010-0… P.O. Sp… 0 "" unknown "{\"typ…
#> 5 Spann,… -2.4… 2011-0… P.O. Sp… 1 "1M scale … unknown "{\"typ…
#> # … with 17 more variables: id <chr>, locationPrecision <chr>,
#> # locationType <chr>, maxDate <int>, minDate <int>, modified <chr>,
#> # path <chr>, pid <chr>, reprLat <dbl>, reprLatLong <chr>, reprLong <dbl>,
#> # tags <chr>, timePeriods <chr>, timePeriodsKeys <chr>,
#> # timePeriodsRange <chr>, title <chr>, uid <chr>
Search names
pl_search_names("SELECT * FROM names limit 5")
#> # Source: SQL [?? x 26]
#> # Database: sqlite 3.33.0
#> # [/Users/sckott/Library/Caches/pleiades/pleiades_names.sqlite3]
#> authors bbox created creators currentVersion description extent id
#> <chr> <chr> <chr> <chr> <int> <chr> <chr> <chr>
#> 1 Spann,… -3.6… 2010-0… P.O. Sp… NA "" "{\"t… cons…
#> 2 Spann,… -3.6… 2010-0… P.O. Sp… NA "" "{\"t… cons…
#> 3 Spann,… -3.6… 2010-0… P.O. Sp… NA "" "{\"t… kond…
#> 4 Becker… -3.6… 2016-0… jbecker 1 "An ethnic… "{\"t… cons…
#> 5 Spann,… -2.0… 2010-0… P.O. Sp… 1 "" "{\"t… cont…
#> # … with 18 more variables: locationPrecision <chr>, maxDate <int>,
#> # minDate <int>, modified <chr>, nameAttested <chr>, nameLanguage <chr>,
#> # nameTransliterated <chr>, path <chr>, pid <chr>, reprLat <dbl>,
#> # reprLatLong <chr>, reprLong <dbl>, tags <chr>, timePeriods <chr>,
#> # timePeriodsKeys <chr>, timePeriodsRange <chr>, title <chr>, uid <chr>
Search places
pl_search_places("SELECT * FROM places limit 5")
#> # Source: SQL [?? x 26]
#> # Database: sqlite 3.33.0
#> # [/Users/sckott/Library/Caches/pleiades/pleiades_places.sqlite3]
#> authors bbox connectsWith created creators currentVersion description extent
#> <chr> <chr> <chr> <chr> <chr> <int> <chr> <chr>
#> 1 Becker… 13.4… "413005" 2016-1… jbecker… 1 The post-R… "{\"t…
#> 2 Becker… 11.6… "413393" 2016-1… jbecker… 2 A major ur… "{\"t…
#> 3 Spann,… -3.6… "" 2010-0… P.O. Sp… 15 An ancient… "{\"t…
#> 4 Spann,… -0.5… "" 2010-0… P.O. Sp… 4 Contestani… "{\"t…
#> 5 Spann,… -2.4… "" 2010-0… P.O. Sp… 6 An ancient… "{\"t…
#> # … with 18 more variables: featureTypes <chr>, geoContext <chr>,
#> # hasConnectionsWith <chr>, id <dbl>, locationPrecision <chr>, maxDate <int>,
#> # minDate <int>, modified <chr>, path <chr>, reprLat <dbl>,
#> # reprLatLong <chr>, reprLong <dbl>, tags <chr>, timePeriods <chr>,
#> # timePeriodsKeys <chr>, timePeriodsRange <chr>, title <chr>, uid <chr>
res <- pl_places(place_id = 462471)
pl_gist(res)
Which opens up the gist in your default browser, as long as browse = TRUE
(default).
thing