Merge pull request #28 from sailthru/df_to_json

Implement as.tbl_json.data.frame, closes #27
sailthru · Apr 3, 2015 · 2a65060 · 2a65060
2 parents e2f6ce4 + fd10eb6
commit 2a65060
Show file tree

Hide file tree

Showing 28 changed files with 85 additions and 28 deletions.
diff --git a/NAMESPACE b/NAMESPACE
@@ -3,6 +3,7 @@
 S3method("[",tbl_json)
 S3method(arrange_,tbl_json)
 S3method(as.tbl_json,character)
+S3method(as.tbl_json,data.frame)
 S3method(as.tbl_json,tbl_json)
 S3method(filter_,tbl_json)
 S3method(mutate_,tbl_json)

diff --git a/R/tbl_json.r b/R/tbl_json.r
@@ -60,6 +60,27 @@ as.tbl_json.character <- function(x, ...) {
   tbl_json(ids, json)
 }
 
+#' Turns a data.frame into a tbl_json object
+#' @param x data.frame that has a column of JSON data
+#' @param json.column the name of the JSON column of data in x, if x is a data.frame
+#' @rdname tbl_json
+#' @export
+as.tbl_json.data.frame <- function(x, json.column, ...) {
+
+  assert_that(is.character(json.column))
+  assert_that(json.column %in% names(x))
+
+  # Parse the json
+  json <- lapply(x[[json.column]], fromJSON, simplifyVector = FALSE)
+
+  # Remove json column
+  x <- x[, setdiff(names(x), json.column), drop = FALSE]
+
+  # Construct tbl_json
+  tbl_json(x, json)
+
+}
+
 #' @rdname tbl_json
 #' @export
 is.tbl_json <- function(x) inherits(x, "tbl_json")

diff --git a/man/allowed_json_types.Rd b/man/allowed_json_types.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/json_types.r
 \docType{data}
 \name{allowed_json_types}
 \alias{allowed_json_types}

diff --git a/man/append_values.Rd b/man/append_values.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/append_values.r
 \name{append_values}
 \alias{append_values}
 \alias{append_values_logical}

diff --git a/man/append_values_factory.Rd b/man/append_values_factory.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/append_values.r
 \name{append_values_factory}
 \alias{append_values_factory}
 \title{Creates the append_values_* functions}

diff --git a/man/append_values_type.Rd b/man/append_values_type.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/append_values.r
 \name{append_values_type}
 \alias{append_values_type}
 \title{get list of values from json}

diff --git a/man/commits.Rd b/man/commits.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/data-commits.r
 \docType{data}
 \name{commits}
 \alias{commits}

diff --git a/man/companies.Rd b/man/companies.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/data-companies.r
 \docType{data}
 \name{companies}
 \alias{companies}

diff --git a/man/determine_types.Rd b/man/determine_types.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/json_types.r
 \name{determine_types}
 \alias{determine_types}
 \title{Determines the types of a list of parsed JSON}

diff --git a/man/enter_object.Rd b/man/enter_object.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/enter_object.r
 \name{enter_object}
 \alias{enter_object}
 \title{Selects an object by key and filters rows to just those with matching keys}

diff --git a/man/gather_array.Rd b/man/gather_array.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/gather_array.r
 \name{gather_array}
 \alias{gather_array}
 \title{Expands a tbl_json to span the indices of a JSON array}

diff --git a/man/gather_keys.Rd b/man/gather_keys.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/gather_keys.r
 \name{gather_keys}
 \alias{gather_keys}
 \title{Gathers every key from the top level of the json and stacks them}

diff --git a/man/issues.Rd b/man/issues.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/data-issues.r
 \docType{data}
 \name{issues}
 \alias{issues}

diff --git a/man/jfactory.Rd b/man/jfactory.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/spread_values.r
 \name{jfactory}
 \alias{jfactory}
 \title{Factory that creates the j* functions below}

diff --git a/man/jfunctions.Rd b/man/jfunctions.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/spread_values.r
 \name{jfunctions}
 \alias{jfunctions}
 \alias{jlogical}

diff --git a/man/json_lengths.Rd b/man/json_lengths.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/json_lengths.r
 \name{json_lengths}
 \alias{json_lengths}
 \title{Add a column that tells the 'length' of the data in the root of the JSON}

diff --git a/man/json_types.Rd b/man/json_types.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/json_types.r
 \name{json_types}
 \alias{json_types}
 \title{Add a column that tells the 'type' of the data in the root of the JSON}

diff --git a/man/list_path.Rd b/man/list_path.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/path.r
 \name{list_path}
 \alias{list_path}
 \title{Recursively access a path}

diff --git a/man/my_unlist.Rd b/man/my_unlist.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/append_values.r
 \name{my_unlist}
 \alias{my_unlist}
 \title{Unlists while preserving NULLs and only unlisting lists with one value}

diff --git a/man/prep_path.Rd b/man/prep_path.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/path.r
 \name{prep_path}
 \alias{prep_path}
 \title{Prepare a path from ...}

diff --git a/man/replace_nulls.Rd b/man/replace_nulls.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/path.r
 \name{replace_nulls}
 \alias{replace_nulls}
 \title{Replace nulls with something else}

diff --git a/man/spread_values.Rd b/man/spread_values.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/spread_values.r
 \name{spread_values}
 \alias{spread_values}
 \title{Extracts values from JSON refereced by a sequence of keys}

diff --git a/man/sub-.tbl_json.Rd b/man/sub-.tbl_json.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/tbl_json.r
 \name{[.tbl_json}
 \alias{[.tbl_json}
 \title{Extract subsets of a tbl_json object (not replace)}

diff --git a/man/tbl_json.Rd b/man/tbl_json.Rd
@@ -1,8 +1,8 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/tbl_json.r
 \name{tbl_json}
 \alias{as.tbl_json}
 \alias{as.tbl_json.character}
+\alias{as.tbl_json.data.frame}
 \alias{as.tbl_json.tbl_json}
 \alias{is.tbl_json}
 \alias{tbl_json}
@@ -16,6 +16,8 @@ as.tbl_json(x, ...)
 
 \method{as.tbl_json}{character}(x, ...)
 
+\method{as.tbl_json}{data.frame}(x, json.column, ...)
+
 is.tbl_json(x)
 }
 \arguments{
@@ -25,9 +27,13 @@ is.tbl_json(x)
 
 \item{drop.null.json}{drop NULL json entries from data.frame and json}
 
+\item{...}{other arguments}
+
 \item{x}{character vector of json}
 
-\item{...}{other arguments}
+\item{x}{data.frame that has a column of JSON data}
+
+\item{json.column}{the name of the JSON column of data in x, if x is a data.frame}
 }
 \description{
 Combines structured JSON (as a data.frame) with remaining JSON
@@ -38,5 +44,7 @@ note that "..JSON" is a reserved column name used internally for filtering
 tbl_json objects, and so is not allowed in the data.frame names.
 
 Turns a character vector into a tbl_json object
+
+Turns a data.frame into a tbl_json object
 }
 
diff --git a/man/tidyjson.Rd b/man/tidyjson.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/tidyjson-package.r
 \docType{package}
 \name{tidyjson}
 \alias{tidyjson}

diff --git a/man/worldbank.Rd b/man/worldbank.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/data-worldbank.r
 \docType{data}
 \name{worldbank}
 \alias{worldbank}

diff --git a/man/wrap_dplyr_verb.Rd b/man/wrap_dplyr_verb.Rd
@@ -1,5 +1,4 @@
 % Generated by roxygen2 (4.1.0): do not edit by hand
-% Please edit documentation in R/tbl_json.r
 \name{wrap_dplyr_verb}
 \alias{wrap_dplyr_verb}
 \title{Wrapper for extending dplyr verbs to tbl_json objects}

diff --git a/tests/testthat/test-tbl_json.r b/tests/testthat/test-tbl_json.r
@@ -68,6 +68,59 @@ test_that("throws error on invalid json", {
   }
 )
 
+context("tbl_json: as.tbl_json.data.frame")
+
+test_that("works for a data.frame and data_frame created objects", {
+
+    df <- data.frame(
+      document.id = 1:2,
+      json = c('{"name": "bob"}', '{"name": "susan"}'),
+      stringsAsFactors = FALSE)
+    # data.frame
+    expect_identical(
+      as.tbl_json(df, json.column = "json"),
+      as.tbl_json(df$json)
+    )
+    # data_frame
+    df <- data_frame(
+      document.id = 1:2,
+      json = c('{"name": "bob"}', '{"name": "susan"}'))
+    expect_identical(
+      as.tbl_json(df, json.column = "json"),
+      as.tbl_json(df$json)
+    )
+
+  }
+)
+
+test_that("works in a pipeline", {
+
+    df <- data_frame(
+      age = c(32, 45),
+      json = c('{"name": "bob"}', '{"name": "susan"}')
+    )
+
+    expect_identical(
+      df %>% as.tbl_json(json.column = "json") %>% 
+        spread_values(name = jstring("name")) %>%
+        filter(age == 32) %>%
+        `[[`("name"),
+      "bob"
+    )
+
+  }
+)
+
+test_that("throws an error without json.column specified", {
+    expect_error(as.tbl_json(iris))
+  }
+)
+
+test_that("throws an error if json column doesn't exist", {
+    expect_error(as.tbl_json(iris, json.column = "json"))
+  }
+)
+
 context("tbl_json")
 
 test_that("tbl_json constructor works with no data", {
@@ -164,7 +217,6 @@ test_that("dplyr::arrange works with a simple example", {
   }
 )
 
-
 test_that("dplyr::mutate works with a simple example", {
 
     x <- as.tbl_json(c('{"name": "bob"}', '{"name": "susan"}'))
@@ -185,7 +237,6 @@ test_that("dplyr::mutate works with a simple example", {
   }
 )
 
-
 test_that("dplyr::mutate works in a more complex pipeline", {
 
     json <- c(