Merge pull request #218 from lorenzwalthert/start_token2

- Remove outdated line and col information (#218).

Author: lorenzwalthert

R/initialize.R

@@ -9,6 +9,7 @@ initialize_attributes <- function(pd_flat) {
   init_pd <-
     initialize_newlines(pd_flat) %>%
     initialize_spaces() %>%
+    remove_line_col() %>%
     initialize_multi_line() %>%
     initialize_indention_ref_id() %>%
     initialize_indent() %>%
@@ -37,6 +38,11 @@ initialize_spaces <- function(pd_flat) {
   pd_flat
 }
 
+remove_line_col <- function(pd_flat) {
+  pd_flat[c("line1", "line2", "col1", "col2")] <- NULL
+  pd_flat
+}
+
 #' @describeIn initialize_attributes Initializes `multi_line`.
 initialize_multi_line <- function(pd_flat) {
   nrow <- nrow(pd_flat)

R/nested.R

@@ -176,8 +176,8 @@ combine_children <- function(child, internal_child) {
 #' Get the start right
 #'
 #' On what line does the first token occur?
-#' @param pd A parse table.
+#' @param pd_nested A nested parse table.
 #' @return The line number on which the first token occurs.
-start_on_line <- function(pd) {
-  pd$line1[1]
+find_start_line <- function(pd_nested) {
+  pd_nested$line1[1]
 }

R/serialize.R

@@ -2,9 +2,9 @@
 #'
 #' Collapses a flattened parse table into character vector representation.
 #' @param flattened_pd A flattened parse table.
-serialize_parse_data_flattened <- function(flattened_pd) {
-  flattened_pd$lag_newlines[1] <- flattened_pd$line1[1] - 1
-
+#' @param start_line The line number on which the code starts.
+serialize_parse_data_flattened <- function(flattened_pd, start_line = 1) {
+  flattened_pd$lag_newlines[1] <- start_line - 1
   res <- with(flattened_pd,
     paste0(collapse = "",
       map(lag_newlines, add_newlines), map(lag_spaces, add_spaces), text)

R/token-create.R

@@ -45,10 +45,6 @@ create_tokens <- function(tokens,
     spaces = spaces,
     multi_line = rep(FALSE, len_text),
     indention_ref_id = indention_ref_ids,
-    line1 = rep(NA, len_text),
-    line2 = rep(NA, len_text),
-    col1 = rep(NA, len_text),
-    col2 = rep(NA, len_text),
     indent = indents,
     child = rep(list(NULL), len_text)
   )

R/transform.R

@@ -84,7 +84,7 @@ make_transformer <- function(transformers) {
 #' @inheritParams apply_transformers
 parse_transform_serialize <- function(text, transformers) {
   pd_nested <- compute_parse_data_nested(text)
-
+  start_line <- find_start_line(pd_nested)
   if (nrow(pd_nested) == 0) {
     warning(
       "Text to style did not contain any tokens. Returning empty string.",
@@ -98,7 +98,8 @@ parse_transform_serialize <- function(text, transformers) {
     enrich_terminals(transformers$use_raw_indention) %>%
     apply_ref_indention()
 
-  serialized_transformed_text <- serialize_parse_data_flattened(flattened_pd)
+  serialized_transformed_text <-
+    serialize_parse_data_flattened(flattened_pd, start_line = start_line)
   serialized_transformed_text
 }
 

R/visit.R

@@ -138,7 +138,7 @@ enrich_terminals <- function(flattened_pd, use_raw_indention = FALSE) {
   flattened_pd$spaces <- NULL # depreciate spaces
   flattened_pd <- choose_indention(flattened_pd, use_raw_indention)
   flattened_pd$line1 <-
-    cumsum(flattened_pd$lag_newlines) + flattened_pd$line1[1]
+    cumsum(flattened_pd$lag_newlines)
 
   flattened_pd$newlines <- lead(flattened_pd$lag_newlines, default = 0L)
   flattened_pd$nchar <- nchar(flattened_pd$text, type = "width")

man/start_on_line.Rd renamed to man/find_start_line.Rd

@@ -4,8 +4,7 @@ test_that("can create a token that has relevant columns", {
   pd_names <- c(
     "token", "text", "short", "lag_newlines", "newlines", "pos_id",
     "parent", "token_before", "token_after", "id", "terminal", "internal",
-    "spaces", "multi_line", "indention_ref_id", "line1", "line2",
-    "col1", "col2", "indent", "child"
+    "spaces", "multi_line", "indention_ref_id", "indent", "child"
   )
 
   expect_equal(