some more examples

ropensci · Dec 14, 2018 · 046d070 · 046d070
1 parent e07d7a9
commit 046d070
Show file tree

Hide file tree

Showing 2 changed files with 93 additions and 28 deletions.
diff --git a/data-raw/itis.R b/data-raw/itis.R
@@ -1,7 +1,7 @@
 ## apt-get -y install mariadb-client postgresql-client
-library(taxizedb)
+#library(taxizedb)
 library(tidyverse)
-
+library(piggyback)
 #itis_store <- db_download_itis()
 ## Need to fix locale issue
 #db_load_itis(itis_store, user = "postgres", pwd = "password", host = "postgres")
@@ -96,25 +96,9 @@ itis_long <- itis %>%
          path_rank, path_id, path_rank_id, name_usage, update_date) %>%
   mutate(update_date = as_date(update_date))
 
-
-## encode language with common name(?)
-
-## Some langauage names have the same language code.
-## ISOcodes puts duplicates in same column, we need a tidy look-up table
-iso <- ISOcodes::ISO_639_2 %>%
-  select(language = Name, code = Alpha_2) %>%
-  na.omit() %>%
-  separate(language, c("name", "name2", "name3", "name4", "name5"),
-           sep = ";", extra="warn", fill = "right") %>%
-  gather(key, language, -code) %>%
-  select(-key) %>%
-  na.omit()
-
-itis_for_rdf <-
-  itis_long %>%
-  left_join(iso) %>%
-  unite("common_name", common_name, code, sep = "@")
-
+system.time({
+  write_tsv(itis_long, bzfile("data/itis_long.tsv.bz2", compression=9))
+})
 
 
 #itis_long <- read_tsv("data/itis_long.tsv.bz2")
@@ -127,9 +111,7 @@ itis_hierarchy <-
   spread(path_rank, path)
 
 
-system.time({
-  write_tsv(itis_long, bzfile("data/itis_long.tsv.bz2", compression=9))
-})
+
 ## write at compression 9 for best compression
 system.time({
   write_tsv(itis_hierarchy, bzfile("data/itis_hierarchy.tsv.bz2", compression=9))

diff --git a/notebook/taxa-rdf.R b/notebook/taxa-rdf.R
@@ -1,11 +1,41 @@
 remotes::install_github("cboettig/virtuoso")
 remotes::install_github("cboettig/rdftools")
 
+
 library(virtuoso)
 library(readr)
 library(rdftools)
 library(dplyr)
 library(taxald)
+library(ISOcodes)
+
+
+itis_long <- read_tsv("data/itis_long.tsv.bz2")
+
+## Some langauage names have the same language code.
+## ISOcodes puts duplicates in same column, we need a tidy look-up table
+iso <- ISOcodes::ISO_639_2 %>%
+  select(language = Name, code = Alpha_2) %>%
+  na.omit() %>%
+  separate(language, c("name", "name2", "name3", "name4", "name5"),
+           sep = ";", extra="warn", fill = "right") %>%
+  gather(key, language, -code) %>%
+  select(-key) %>%
+  na.omit()
+
+itis_for_rdf <-
+  itis_long %>%
+  left_join(iso) %>%
+  unite("common_name", common_name, code, sep = "@")
+
+species <- itis_long %>% select(id, name, rank, common_name, update_date) %>% distinct()
+classif <- itis_long %>% select(path_id, path, path_rank, path_rank_id, species_id = id) %>% distinct()
+#readr::write_tsv(species, "species.tsv")
+#readr::write_tsv(classif, "classif.tsv")
+rdftools::write_nquads(species, "itis_species.nq.gz", prefix = "taxald:", key = "id")
+rdftools::write_nquads(classif, "itis_classif.nq.gz", prefix = "taxald:", key = "path_id")
+
+
 
 taxa_tbl("itis", "taxonid") %>%
   collect() %>%
@@ -42,6 +72,8 @@ taxa_tbl("wd", "taxonid") %>% collect() %>%
                          key_column = "id",
                          prefix = "taxald:")
 #virtuoso:::vos_delete_db()
+
+library(virtuoso)
 vos_start()
 con <- vos_connect()
 files <- fs::dir_ls(glob="*.nq.gz")
@@ -65,18 +97,69 @@ vos_query(con,
 ")
 
 vos_query(con,
-'SELECT ?id ?rank
+'SELECT ?id ?rank ?common
  WHERE {
     ?id <taxald:name> "Homo sapiens" .
-    ?id <taxald:rank> ?rank
+    ?id <taxald:rank> ?rank .
+    OPTIONAL { ?id <taxald:common_name> ?common . }
+}
+')
+
+vos_query(con,
+           'SELECT ?id ?species ?rank
+  WHERE {
+     ?id <taxald:name> ?species .
+     ?id <taxald:rank> ?rank .
+     ?id <taxald:common_name> "Human" .
+ }
+ ')
+
+vos_query(con, 'SELECT ?id ?species ?rank
+ WHERE {
+    ?id <taxald:name> ?species .
+    ?id <taxald:rank> ?rank .
+    ?id <taxald:common_name> ?common_name .
+    ?id <taxald:common_name> "Human" .
+    FILTER langMatches(lang(?common_name), "en")
+}
+')
+
+vos_query(con, 'SELECT ?id ?species ?common_name ?rank
+ WHERE {
+    ?id <taxald:name> ?species .
+    ?id <taxald:rank> ?rank .
+    ?id <taxald:common_name> ?common_name .
+    FILTER(  ?common_name LIKE "%Cod%" ).
+}
+')
+
+
+vos_query(con,
+          'SELECT ?path ?rank
+ WHERE {
+    ?path_id <taxald:path> ?path .
+    ?path_id <taxald:path_rank> ?rank
+
 } LIMIT 20
 ')
 
 vos_query(con,
-          'SELECT ?id ?name
+          'SELECT ?id ?name ?common ?rank ?update_date
  WHERE {
     ?id <taxald:name> ?name .
-    ?id <taxald:rank> "kingdom"
+    ?id <taxald:common_name> ?common .
+    ?id <taxald:rank> ?rank .
+    ?id <taxald:update_date> ?update_date
+
 } LIMIT 20
 ')
 
+
+vos_query(con,
+          'SELECT DISTINCT ?p
+FROM <rdflib>
+ WHERE {
+    ?s ?p ?o
+  }
+LIMIT 20
+')