Merge branch 'r-0.0-1' into production

Author: krlmlr

.Rbuildignore

@@ -0,0 +1,8 @@
+^.*\.Rproj$
+^\.Rproj\.user$
+^API$
+^README\.Rmd$
+^README-.*\.png$
+^.travis.yml
+^appveyor\.yml$
+^tic\.R$

.gitignore

@@ -0,0 +1,4 @@
+.Rproj.user
+.Rhistory
+.RData
+inst/doc

.travis.yml

@@ -0,0 +1,7 @@
+language: r
+cache: packages
+warnings_are_errors: false
+r_packages:
+  - covr
+after_success:
+  - Rscript -e 'library(covr); codecov()'

API

@@ -0,0 +1,8 @@
+# API for styler package
+
+## Exported functions
+
+get_transformers(strict = TRUE)
+style_pkg(pkg = ".", transformers = get_transformers())
+style_src(path = ".", transformers = get_transformers(), recursive = TRUE)
+style_text(text, transformers = get_transformers())

DESCRIPTION

@@ -0,0 +1,30 @@
+Package: styler
+Title: Non-invasive Pretty Printing of R code
+Version: 0.0-1
+Authors@R: person("Kirill", "Müller", role = c("aut", "cre"), email = "[email protected]")
+Description:
+    Pretty-prints R code without changing the user's formatting intent.
+Imports:
+    rprojroot,
+    utf8,
+    withr,
+    dplyr,
+    tibble,
+    tidyr,
+    purrr,
+    magrittr
+Suggests: testthat,
+    knitr,
+    rmarkdown,
+    data.tree
+Remotes:
+    krlmlr/utf8
+License: GPL-3
+Encoding: UTF-8
+LazyData: true
+Date: 2017-06-15
+BugReports: https://github.com/krlmlr/styler/issues
+URL: https://github.com/krlmlr/styler, http://krlmlr.github.io/styler
+Roxygen: list(markdown = TRUE, roclets = c("rd", "namespace", "pkgapi::api_roclet"))
+RoxygenNote: 6.0.1.9000
+VignetteBuilder: knitr

NAMESPACE

@@ -0,0 +1,12 @@
+# Generated by roxygen2: do not edit by hand
+
+export(get_transformers)
+export(style_pkg)
+export(style_src)
+export(style_text)
+import(dplyr)
+import(tibble)
+import(tidyr)
+importFrom(purrr,map)
+importFrom(purrr,map2)
+importFrom(purrr,pmap)

NEWS.md

@@ -0,0 +1,7 @@
+## styler 0.0-1 (2017-06-15)
+
+Initial release, work in progress.
+
+- Create and serialize nested parse data.
+- Internal support for indention of expressions with parentheses.
+- Adding and removing spaces around operators on flat parse data.

R/modify_pd.R

@@ -0,0 +1,57 @@
+#' Update indention information of parse data
+#'
+#' @param pd A nested or flat parse table that is already enhanced with
+#'   line break and space information via [create_filler()] or
+#'   [create_filler_nested()].
+#' @param indent_by How many spaces should be added after the token of interest.
+#' @name update_indention
+NULL
+
+#' @rdname update_indention
+indent_round <- function(pd, indent_by) {
+  opening <- which(pd$token == "'('")
+  if (length(opening) > 0) {
+    start <- opening + 1
+    stop <- nrow(pd) - 1
+  } else {
+    start <- stop <- 0
+  }
+  pd <- pd %>%
+    mutate(indent = ifelse(seq_len(nrow(pd)) %in% start:stop, indent_by, 0)) %>%
+    select_(~indent, ~newlines, ~everything())
+  pd
+}
+
+
+
+#' Update indention information of nested parse data
+#'
+#' These functions apply the update functions of the same name but without
+#'   suffix nested to each level of nesting of the nested parse table.
+#' @param pd A nested parse table that is already enhanced with
+#'   line break and space information via [create_filler_nested].
+#' @name update_indention_nested
+NULL
+
+#' @rdname update_indention_nested
+indent_round_nested <- function(pd) {
+  if (is.null(pd)) return(pd)
+  pd <- pd %>%
+    indent_round(indent_by = 2) %>%
+    mutate(child = map(child, indent_round_nested))
+  pd
+}
+
+
+#' Strip EOL spaces
+#'
+#' Remove end-of-line spaces.
+#' @param pd_nested A nested parse table.
+#' @return A nested parse table.
+strip_eol_spaces_nested <- function(pd_nested) {
+  if (is.null(pd_nested)) return()
+  pd_nested <- pd_nested %>%
+    mutate(spaces = spaces * (newlines == 0),
+           child  = map(child, strip_eol_spaces_nested))
+  pd_nested
+}

R/nested.R

@@ -0,0 +1,135 @@
+#' Obtain a nested parse table from a character vector
+#'
+#' [utils::getParseData()] is used to obtain a flat parse table from `text`.
+#'   Subsequentially, it's representation is changed from a flat table into a
+#'   nested parse table with [nest_parse_data()].
+#' @param text A character vector to parse.
+#' @return A nested parse table. Apart from the columns provided by
+#'   `utils::getParseData()`, a column "short" with the first five characters of
+#'   "text" is added, the nested subtibbles are in column "child".
+#' TODO:
+#' - Implement enhance_parse_data_nested()
+#'     - Walk tree defined by `child`, compute whitespace information
+#'     - Store indention depth in a separate column, unaffected by
+#'       inter-token space
+#' - Implement compute_parse_data_nested_with_ws() as
+#'   compute_parse_data_nested() + enhance_parse_data_nested()
+#' - Implement serialization of nested parse data
+#' - Use compute_parse_data_nested_with_ws() instead of
+#'   compute_parse_data_flat_enhanced()
+#' - Perform all transformations on hierarchical structure
+#'     - Compute text for a sub-element
+#' - Compute indentation
+#'     - Braces
+#'     - Function calls
+#'     - Function definitions
+#' - Remove `includeText = TRUE`
+compute_parse_data_nested <- function(text) {
+  parsed <- parse(text = text, keep.source = TRUE)
+  parse_data <- tbl_df(utils::getParseData(parsed, includeText = TRUE))
+  pd_nested <-
+    parse_data %>%
+    mutate_(child = ~rep(list(NULL), length(text))) %>%
+    mutate_(short = ~substr(text, 1, 5)) %>%
+    select_(~short, ~everything()) %>%
+    nest_parse_data()
+
+  pd_nested
+}
+
+#' Nest a flat parse table
+#'
+#' `nest_parse_data` groups `pd_flat` into a parse table with tokens that are
+#'   a parent to other tokens (called internal) and such that are not (called
+#'   child). Then, the token in child are joined to their parents in internal
+#'   and all token information of the children is nested into a column "child".
+#'   This is done recursively until we are only left with a nested tibble that
+#'   contains one row: The nested parse table.
+#' @param pd_flat A flat parse table including both terminals and non-terminals.
+#' @seealso [compute_parse_data_nested()]
+#' @return A nested parse table.
+nest_parse_data <- function(pd_flat) {
+  if (nrow(pd_flat) <= 1) return(pd_flat)
+  split <-
+    pd_flat %>%
+    mutate_(internal = ~id %in% parent) %>%
+    nest_("data", names(pd_flat))
+
+  child <- split$data[!split$internal][[1L]]
+  internal <- split$data[split$internal][[1L]]
+
+  internal <- rename_(internal, internal_child = ~child)
+
+  nested <-
+    child %>%
+    mutate_(parent_ = ~parent) %>%
+    nest_(., "child", setdiff(names(.), "parent_")) %>%
+    left_join(internal, ., by = c("id" = "parent_")) %>%
+    mutate_(child = ~Map(bind_rows, child, internal_child)) %>%
+    mutate_(child = ~lapply(child, arrange_, ~line1, ~col1)) %>%
+    select_(~-internal_child) %>%
+    select_(~short, ~everything(), ~-text, ~text)
+
+  nest_parse_data(nested)
+}
+
+#' Enrich nested parse table with space and linebreak information
+#'
+#' Uses [create_filler()] in a recursion add space and line break information
+#'   separately on every level of nesting.
+#' @param pd_nested A nested parse table.
+#' @return A nested parse table with two new columns: newlines and spaces.
+#' @seealso [create_filler()]
+#' @importFrom purrr map
+create_filler_nested <- function(pd_nested) {
+  if (is.null(pd_nested$child)) return()
+  pd_nested <- pd_nested %>%
+    create_filler() %>%
+    mutate(child = map(child, create_filler_nested)) %>%
+    select_(~spaces, ~newlines, ~lag_newlines, ~short, ~everything())
+}
+
+#' Serialize a nested parse table
+#'
+#' Helper function that recursively extracts terminals from a nested tibble.
+#' @param pd_nested A nested parse table.
+#' @param pass_indent Level of indention of a token.
+#' @return A character vector with all terminal tokens in `pd_nested` plus
+#'   the appropriate amount of white spaces and line breaks are inserted between
+#'   them.
+#' @importFrom purrr pmap
+serialize_parse_data_nested_helper <- function(pd_nested, pass_indent) {
+  out <- pmap(list(pd_nested$terminal, pd_nested$text, pd_nested$child,
+             pd_nested$spaces, pd_nested$lag_newlines, pd_nested$indent),
+           function(terminal, text, child, spaces, lag_newlines, indent) {
+             total_indent <- pass_indent + indent
+             preceding_linebreak <- if_else(lag_newlines > 0, 1, 0)
+             if (terminal) {
+               c(add_newlines(lag_newlines),
+                 add_spaces(total_indent * preceding_linebreak),
+                 text,
+                 add_spaces(spaces))
+             } else {
+               c(add_newlines(lag_newlines),
+                 add_spaces(total_indent * preceding_linebreak),
+                 serialize_parse_data_nested_helper(child, total_indent),
+                 add_spaces(spaces))
+             }
+        }
+  )
+  out
+}
+
+#' Serialize a nested parse table
+#'
+#' Collapses a nested parse table into its character vector representation.
+#' @param pd_nested A nested parse table with line break, spaces and indention
+#'   information.
+#' @return A character string.
+serialize_parse_data_nested <- function(pd_nested) {
+  raw <- serialize_parse_data_nested_helper(pd_nested, pass_indent = 0) %>%
+    unlist() %>%
+    paste0(collapse = "") %>%
+    strsplit("\n", fixed = TRUE) %>%
+    .[[1L]]
+}

R/nested_to_tree.R

@@ -0,0 +1,39 @@
+#' Convert a nested tibble into a node tree
+#'
+#' This function is convenient to display all nesting levels of a nested tibble
+#'   at once.
+#' @param pd_nested A nested tibble.
+#' @return An object of class "Node" and "R6".
+#' @examples
+#' library("magrittr")
+#' code <- "a <- function(x) { if(x > 1) { 1+1 } else {x} }"
+#' l1 <- styler:::compute_parse_data_nested(code) %>%
+#'   styler:::create_filler_nested() %>%
+#'   styler:::create_node_from_nested_root()
+create_node_from_nested_root <- function(pd_nested) {
+  n <- data.tree::Node$new("xxx")
+  create_node_from_nested(pd_nested, n)
+  n
+}
+
+#' Create node from nested parse data
+#'
+#' @inheritParams create_node_from_nested_root
+#' @param parent The parent of the node to be created.
+#' @importFrom purrr map2
+create_node_from_nested <- function(pd_nested, parent) {
+  if (is.null(pd_nested))
+    return()
+
+  node_info <-
+    pd_nested %>%
+    transmute(formatted = paste0(token, ": ", short, " [", newlines, "/", spaces, "]")) %>%
+    .[["formatted"]]
+
+  child_nodes <-
+    node_info %>%
+    map(parent$AddChild)
+
+  map2(pd_nested$child, child_nodes, create_node_from_nested)
+  return()
+}