From b1754e3d1112ea04e2ee4c393581362fb8f605f7 Mon Sep 17 00:00:00 2001 From: Ranke Johannes Date: Mon, 7 Aug 2023 11:06:17 +0200 Subject: Typos --- R/read_spreadsheet.R | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/R/read_spreadsheet.R b/R/read_spreadsheet.R index 1c5508da..bad93d43 100644 --- a/R/read_spreadsheet.R +++ b/R/read_spreadsheet.R @@ -12,7 +12,7 @@ #' to contain name and acronym of the parent compound. #' #' The dataset sheets should be named using the dataset numbers read in from -#' the 'Datasets' sheet, i.e. '1', '2', ... . In each dataset sheet, name +#' the 'Datasets' sheet, i.e. '1', '2', ... . In each dataset sheet, the name #' of the observed variable (e.g. the acronym of the parent compound or #' one of its transformation products) should be in the first column, #' the time values should be in the second colum, and the observed value @@ -26,7 +26,7 @@ #' Their names should preferably not contain special characters like spaces, #' so they can be easily used for specifying covariate models. #' -#' An similar data structure is defined as the R6 class [mkindsg], but +#' A similar data structure is defined as the R6 class [mkindsg], but #' is probably more complicated to use. #' #' @param path Absolute or relative path to the spreadsheet file -- cgit v1.2.1