From 3d587d2f3241263a75cf1642be51dc77153e560a Mon Sep 17 00:00:00 2001
From: Jan Marvin Garbuszus <jan.garbuszus@rub.de>
Date: Sat, 30 Nov 2024 12:33:52 +0100
Subject: [PATCH] [load] read file with d3p1 namespace

---
 R/wb_load.R                        | 23 ++++++++++++++++++++++-
 tests/testthat/test-read_sources.R | 15 +++++++++++++++
 2 files changed, 37 insertions(+), 1 deletion(-)

diff --git a/R/wb_load.R b/R/wb_load.R
index 8998f0dba..4b2343232 100644
--- a/R/wb_load.R
+++ b/R/wb_load.R
@@ -106,8 +106,13 @@ wb_load <- function(
   wb <- wb_workbook()
   wb$path <- file
 
-  grep_xml <- function(pattern, perl = TRUE, value = TRUE, ...) {
+  # There is one known file in #1194. this file has lower case folders, while
+  # the references in the file are the usual camel case.
+  needs_lower <- ifelse(any(grepl("\\[content_types\\].xml$", xmlFiles)), TRUE, FALSE)
+
+  grep_xml <- function(pattern, perl = TRUE, value = TRUE, to_lower = needs_lower, ...) {
     # targets xmlFiles; has presents
+    if (to_lower) pattern <- tolower(pattern)
     grep(pattern, xmlFiles, perl = perl, value = value, ...)
   }
 
@@ -423,6 +428,22 @@ wb_load <- function(
     sheets <- xml_attr(workbook_xml, "workbook", "sheets", "sheet")
     sheets <- rbindlist(sheets)
 
+    if (any(grepl("d3p1", nams <- names(sheets)))) {
+      msg <- paste0(
+        "The `{%s}` namespace(s) has been removed from the xml files, for example:\n",
+        "\t<%s:ID> changed to:\n",
+        "\t<r:ID>\n",
+        "See 'Details' in ?openxlsx2::wb_load() for more information."
+      )
+      warning(sprintf(msg, "d3p1", "d3p1"))
+      nams <- stringi::stri_replace_all_fixed(
+        str         = nams,
+        pattern     = c("d3p1:", ":d3p1"),
+        replacement = c("r:", "")
+      )
+      names(sheets) <- nams
+    }
+
     ## Some veryHidden sheets do not have a sheet content and their rId is empty.
     ## Such sheets need to be filtered out because otherwise their sheet names
     ## occur in the list of all sheet names, leading to a wrong association
diff --git a/tests/testthat/test-read_sources.R b/tests/testthat/test-read_sources.R
index f631d5214..e6350d29c 100644
--- a/tests/testthat/test-read_sources.R
+++ b/tests/testthat/test-read_sources.R
@@ -470,3 +470,18 @@ test_that("file extension handling works", {
   expect_silent(wb_save(wb, file = tempfile(fileext = ".XLSM")))
 
 })
+
+test_that("loading d3p1 file works", {
+
+  fl  <- testfile_path("gh_issue_1194.xlsx")
+
+  expect_warning(
+    df <- wb_to_df(fl),
+    "d3p1"
+  )
+
+  exp <- c(1347, 31)
+  got <- dim(df)
+  expect_equal(exp, got)
+
+})