add xml_find_function_calls() helper to source expressions

DESCRIPTION

@@ -176,6 +176,7 @@ Collate:
     'settings_utils.R'
     'shared_constants.R'
     'sort_linter.R'
+    'source_utils.R'
     'spaces_inside_linter.R'
     'spaces_left_parentheses_linter.R'
     'sprintf_linter.R'

R/get_source_expressions.R

@@ -18,32 +18,36 @@
 #' \describe{
 #'   \item{expressions}{a `list` of
 #'   `n+1` objects. The first `n` elements correspond to each expression in
-#'   `filename`, and consist of a list of 9 elements:
+#'   `filename`, and consist of a list of 8 elements:
 #'   \itemize{
-#'     \item{`filename` (`character`)}
-#'     \item{`line` (`integer`) the line in `filename` where this expression begins}
-#'     \item{`column` (`integer`) the column in `filename` where this expression begins}
+#'     \item{`filename` (`character`) the name of the file.}
+#'     \item{`line` (`integer`) the line in the file where this expression begins.}
+#'     \item{`column` (`integer`) the column in the file where this expression begins.}
 #'     \item{`lines` (named `character`) vector of all lines spanned by this
-#'           expression, named with the line number corresponding to `filename`}
-#'     \item{`parsed_content` (`data.frame`) as given by [utils::getParseData()] for this expression}
-#'     \item{`xml_parsed_content` (`xml_document`) the XML parse tree of this
-#'          expression as given by [xmlparsedata::xml_parse_data()]}
-#'     \item{`content` (`character`) the same as `lines` as a single string (not split across lines)}
+#'           expression, named with the corresponding line numbers.}
+#'     \item{`parsed_content` (`data.frame`) as given by [utils::getParseData()] for this expression.}
+#'     \item{`xml_parsed_content` (`xml_document`) the XML parse tree of this expression as given by
+#'           [xmlparsedata::xml_parse_data()].}
+#'     \item{`content` (`character`) the same as `lines` as a single string (not split across lines).}
+#'     \item{`xml_find_function_calls(function_names)` (`function`) a function that returns all `SYMBOL_FUNCTION_CALL`
+#'           XML nodes from `xml_parsed_content` with specified function names.}
 #'   }
 #'
 #'   The final element of `expressions` is a list corresponding to the full file
-#'   consisting of 6 elements:
+#'   consisting of 7 elements:
 #'   \itemize{
-#'     \item{`filename` (`character`)}
-#'     \item{`file_lines` (`character`) the [readLines()] output for this file}
+#'     \item{`filename` (`character`) the name of this file.}
+#'     \item{`file_lines` (`character`) the [readLines()] output for this file.}
 #'     \item{`content` (`character`) for .R files, the same as `file_lines`;
-#'           for .Rmd or .qmd scripts, this is the extracted R source code (as text)}
+#'           for .Rmd or .qmd scripts, this is the extracted R source code (as text).}
 #'     \item{`full_parsed_content` (`data.frame`) as given by
-#'           [utils::getParseData()] for the full content}
+#'           [utils::getParseData()] for the full content.}
 #'     \item{`full_xml_parsed_content` (`xml_document`) the XML parse tree of all
-#'           expressions as given by [xmlparsedata::xml_parse_data()]}
+#'           expressions as given by [xmlparsedata::xml_parse_data()].}
 #'     \item{`terminal_newline` (`logical`) records whether `filename` has a terminal
-#'           newline (as determined by [readLines()] producing a corresponding warning)}
+#'           newline (as determined by [readLines()] producing a corresponding warning).}
+#'     \item{`xml_find_function_calls(function_names)` (`function`) a function that returns all `SYMBOL_FUNCTION_CALL`
+#'           XML nodes from `full_xml_parsed_content` with specified function names.}
 #'   }
 #'   }
 #'   \item{error}{A `Lint` object describing any parsing error.}
@@ -103,6 +107,7 @@ get_source_expressions <- function(filename, lines = NULL) {
       )
       for (i in seq_along(expressions)) {
         expressions[[i]]$xml_parsed_content <- expression_xmls[[i]]
+        expressions[[i]]$xml_find_function_calls <- build_xml_find_function_calls(expression_xmls[[i]])
       }
     }
 
@@ -113,6 +118,7 @@ get_source_expressions <- function(filename, lines = NULL) {
       content = source_expression$lines,
       full_parsed_content = parsed_content,
       full_xml_parsed_content = xml_parsed_content,
+      xml_find_function_calls = build_xml_find_function_calls(xml_parsed_content),
       terminal_newline = terminal_newline
     )
   }

R/source_utils.R

@@ -0,0 +1,16 @@
+#' Build the `xml_find_function_calls()` helper for a source expression
+#'
+#' @param xml The XML parse tree as an XML object (`xml_parsed_content` or `full_xml_parsed_content`)
+#'
+#' @return A fast function to query
+#' `xml_find_all(xml, glue::glue("//SYMBOL_FUNCTION_CALL[{ xp_text_in_table(function_names) }]"))`.
+#'
+#' @noRd
+build_xml_find_function_calls <- function(xml) {
+  function_call_cache <- xml_find_all(xml, "//SYMBOL_FUNCTION_CALL")
+  names(function_call_cache) <- get_r_string(function_call_cache)
+
+  function(function_names) {
+    unname(function_call_cache[names(function_call_cache) %in% function_names])
+  }
+}

tests/testthat/test-get_source_expressions.R

@@ -218,7 +218,10 @@ test_that("returned data structure is complete", {
 
   for (i in seq_along(lines)) {
     expr <- exprs$expressions[[i]]
-    expect_named(expr, c("filename", "line", "column", "lines", "parsed_content", "xml_parsed_content", "content"))
+    expect_named(expr, c(
+      "filename", "line", "column", "lines", "parsed_content", "xml_parsed_content", "content",
+      "xml_find_function_calls"
+    ))
     expect_identical(expr$filename, temp_file)
     expect_identical(expr$line, i)
     expect_identical(expr$column, 1L)
@@ -229,7 +232,8 @@ test_that("returned data structure is complete", {
   }
   full_expr <- exprs$expressions[[length(lines) + 1L]]
   expect_named(full_expr, c(
-    "filename", "file_lines", "content", "full_parsed_content", "full_xml_parsed_content", "terminal_newline"
+    "filename", "file_lines", "content", "full_parsed_content", "full_xml_parsed_content", "xml_find_function_calls",
+    "terminal_newline"
   ))
   expect_identical(full_expr$filename, temp_file)
   expect_identical(full_expr$file_lines, lines_with_attr)
@@ -245,6 +249,37 @@ test_that("returned data structure is complete", {
   expect_identical(exprs$lines, lines_with_attr)
 })
 
+test_that("xml_find_function_calls works as intended", {
+  lines <- c("foo()", "bar()", "foo()", "{ foo(); foo(); bar() }")
+  temp_file <- withr::local_tempfile(lines = lines)
+
+  exprs <- get_source_expressions(temp_file)
+
+  expect_length(exprs$expressions[[1L]]$xml_find_function_calls("foo"), 1L)
+  expect_length(exprs$expressions[[1L]]$xml_find_function_calls("bar"), 0L)
+  expect_identical(
+    exprs$expressions[[1L]]$xml_find_function_calls("foo"),
+    xml_find_all(exprs$expressions[[1L]]$xml_parsed_content, "//SYMBOL_FUNCTION_CALL[text() = 'foo']")
+  )
+
+  expect_length(exprs$expressions[[2L]]$xml_find_function_calls("foo"), 0L)
+  expect_length(exprs$expressions[[2L]]$xml_find_function_calls("bar"), 1L)
+
+  expect_length(exprs$expressions[[4L]]$xml_find_function_calls("foo"), 2L)
+  expect_length(exprs$expressions[[4L]]$xml_find_function_calls("bar"), 1L)
+  expect_length(exprs$expressions[[4L]]$xml_find_function_calls(c("foo", "bar")), 3L)
+
+  expect_length(exprs$expressions[[5L]]$xml_find_function_calls("foo"), 4L)
+  expect_length(exprs$expressions[[5L]]$xml_find_function_calls("bar"), 2L)
+  expect_length(exprs$expressions[[5L]]$xml_find_function_calls(c("foo", "bar")), 6L)
+
+  # Also check order is retained:
+  expect_identical(
+    exprs$expressions[[5L]]$xml_find_function_calls(c("foo", "bar")),
+    xml_find_all(exprs$expressions[[5L]]$full_xml_parsed_content, "//SYMBOL_FUNCTION_CALL")
+  )
+})
+
 test_that("#1262: xml_parsed_content gets returned as missing even if there's no parsed_content", {
   tempfile <- withr::local_tempfile(lines = '"\\R"')