From 6431345140f822ee1356f30210e4c5912b5440f1 Mon Sep 17 00:00:00 2001
From: Valentine Rahier <valentine.rahier@inrae.fr>
Date: Wed, 7 Oct 2026 14:38:12 +0200
Subject: [PATCH 1/6] refactor: refactor plots generation

---
 R/generic_plotting.R          | 111 ++++---
 R/specific_plotting_dynamic.R | 402 +++++++------------------
 R/specific_plotting_scatter.R | 535 +++++++---------------------------
 man/specific_dynamic_plots.Rd |   3 -
 man/specific_scatter_plots.Rd |  33 ++-
 5 files changed, 295 insertions(+), 789 deletions(-)

diff --git a/R/generic_plotting.R b/R/generic_plotting.R
index 006e954..0c70891 100644
--- a/R/generic_plotting.R
+++ b/R/generic_plotting.R
@@ -183,68 +183,55 @@ plot_situations <- function(..., obs = NULL, obs_sd = NULL,
       i
     }
 
-    p[[i]] <- switch(item_case,
-      # Dynamic plots:
-      "mixture_versions_overlap" =
-        plot_dynamic_mixture_versions_overlap(sim_situation, i, successive,
-          title = plot_title
-        ),
-      "mixture_versions_no_overlap" =
-        plot_dynamic_mixture_versions(sim_situation, i, successive, title = plot_title),
-      "mixture_no_versions_overlap" =
-        plot_dynamic_mixture_overlap(sim_situation, i, successive, title = plot_title),
-      "mixture_no_versions_no_overlap" = plot_dynamic_mixture(sim_situation, i, successive,
-        title = plot_title
-      ),
-      "non_mixture_versions_overlap" =
-        plot_dynamic_versions_overlap(sim_situation, i, successive, title = plot_title),
-      "non_mixture_versions_no_overlap" =
-        plot_dynamic_versions(sim_situation, i, successive, title = plot_title),
-      "non_mixture_no_versions_overlap" =
-        plot_dynamic_overlap(sim_situation, i, successive, title = plot_title),
-      "non_mixture_no_versions_no_overlap" =
-        plot_dynamic(sim_situation, i, successive, title = plot_title),
-
-      # Scatter plots:
-      "mixture_versions" =
-        plot_scat_mixture_versions( # per sit and all sit share the same call
-          sim_situation, i, select_scat, shape_sit,
-          reference_var, is_obs_sd,
-          title = plot_title
-        ),
-      "mixture_no_versions" = # per sit and all sit share the same call
-        plot_scat_mixture_allsit(
-          sim_situation, i, select_scat, shape_sit,
-          reference_var, is_obs_sd,
-          title = plot_title
-        ),
-      "non_mixture_versions_situations" =
-        plot_scat_versions_allsit(
-          sim_situation, i, select_scat, shape_sit,
-          reference_var, is_obs_sd,
-          title = plot_title
-        ),
-      "non_mixture_versions_per_situations" =
-        plot_scat_versions_per_sit(
-          sim_situation, i, select_scat, shape_sit,
-          reference_var, is_obs_sd,
-          title = plot_title
-        ),
-      "non_mixture_no_versions_situations" =
-        plot_scat_allsit(
-          sim_situation, i, select_scat, shape_sit,
-          reference_var, is_obs_sd,
-          title = plot_title, has_distinct_situations = has_distinct_situations,
-          one_version = one_version, mixture = mixture
-        ),
-      "non_mixture_no_versions_per_situations" =
-        plot_scat_allsit(
-          sim_situation, i, select_scat, shape_sit,
-          reference_var, is_obs_sd,
-          title = plot_title, has_distinct_situations = has_distinct_situations,
-          one_version = one_version, mixture = mixture
-        )
-    )
+    if (type == "dynamic") {
+      p[[i]] <- build_dynamic_plot(
+        sim_situation, i, successive,
+        title = plot_title,
+        mixture, one_version, overlap
+      )
+    } else {
+      p[[i]] <- switch(item_case,
+            # Scatter plots:
+            "mixture_versions" =
+              plot_scat_mixture_versions( # per sit and all sit share the same call
+                sim_situation, i, select_scat, shape_sit,
+                reference_var, is_obs_sd,
+                title = plot_title
+              ),
+            "mixture_no_versions" = # per sit and all sit share the same call
+              plot_scat_mixture_allsit(
+                sim_situation, i, select_scat, shape_sit,
+                reference_var, is_obs_sd,
+                title = plot_title
+              ),
+            "non_mixture_versions_situations" =
+              plot_scat_versions_allsit(
+                sim_situation, i, select_scat, shape_sit,
+                reference_var, is_obs_sd,
+                title = plot_title
+              ),
+            "non_mixture_versions_per_situations" =
+              plot_scat_versions_per_sit(
+                sim_situation, i, select_scat, shape_sit,
+                reference_var, is_obs_sd,
+                title = plot_title
+              ),
+            "non_mixture_no_versions_situations" =
+              plot_scat_allsit(
+                sim_situation, i, select_scat, shape_sit,
+                reference_var, is_obs_sd,
+                title = plot_title, has_distinct_situations = has_distinct_situations,
+                one_version = one_version, mixture = mixture
+              ),
+            "non_mixture_no_versions_per_situations" =
+              plot_scat_allsit(
+                sim_situation, i, select_scat, shape_sit,
+                reference_var, is_obs_sd,
+                title = plot_title, has_distinct_situations = has_distinct_situations,
+                one_version = one_version, mixture = mixture
+              )
+          )
+    }
   }
 
   names(p) <- common_situations_models
diff --git a/R/specific_plotting_dynamic.R b/R/specific_plotting_dynamic.R
index 8fae437..377239f 100644
--- a/R/specific_plotting_dynamic.R
+++ b/R/specific_plotting_dynamic.R
@@ -94,329 +94,151 @@ make_multiline_title <- function(title, max_char = 120) {
   paste(lines, collapse = "\n")
 }
 
-#' @keywords internal
-#' @rdname specific_dynamic_plots
-plot_dynamic <- function(df_data, sit, successive, title = NULL) {
-  p <- ggplot2::ggplot(
-    df_data,
-    ggplot2::aes(x = .data$Date)
-  ) +
-    ggplot2::geom_line(ggplot2::aes(y = .data$Simulated))
-
-  p <- add_vertical_lines(df_data, successive, p)
-
-  if ("Observed" %in% colnames(df_data)) {
-    p <- p + ggplot2::geom_point(ggplot2::aes(y = .data$Observed), na.rm = TRUE)
-
-    if ("Obs_SD" %in% colnames(df_data)) {
-      p <- p +
-        ggplot2::geom_errorbar(
-          ggplot2::aes(
-            ymin = .data$Observed - 2 * .data$Obs_SD,
-            ymax = .data$Observed + 2 * .data$Obs_SD
-          ),
-          na.rm = TRUE
-        )
-    }
-  }
-  title <- make_multiline_title(title)
-  p <- p +
-    ggplot2::ggtitle(title) +
-    ggplot2::theme_get()
-  p <- add_facet_wrap(p, var = "var", scales = "free_y")
-
-  return(p)
-}
-
-plot_dynamic_mixture <- function(df_data, sit, successive, title = NULL) {
-  p <- ggplot2::ggplot(
-    df_data,
-    ggplot2::aes(
-      x = .data$Date,
-      colour = paste(.data$Dominance, ":", .data$Plant)
-    )
-  ) +
-    ggplot2::geom_line(ggplot2::aes(y = .data$Simulated))
-
-  p <- add_vertical_lines(df_data, successive, p)
-
-  if ("Observed" %in% colnames(df_data)) {
-    p <- p + ggplot2::geom_point(ggplot2::aes(y = .data$Observed), na.rm = TRUE)
-
-    if ("Obs_SD" %in% colnames(df_data)) {
-      p <- p +
-        ggplot2::geom_errorbar(
-          ggplot2::aes(
-            ymin = .data$Observed - 2 * .data$Obs_SD,
-            ymax = .data$Observed + 2 * .data$Obs_SD
-          ),
-          na.rm = TRUE
-        )
-    }
-  }
-
-  title <- make_multiline_title(title)
-  p <- p +
-    ggplot2::ggtitle(title) +
-    ggplot2::labs(colour = "Plant") +
-    ggplot2::theme_get()
-  p <- add_facet_wrap(
-    p,
-    var = "var", scales = "free_y",
-    legend_labels = unique(paste(df_data$Dominance, ":", df_data$Plant))
+dynamic_plot_spec <- function(df_data, mixture, one_version, overlap) {
+  spec_name <- paste(
+    if (mixture) "mixture" else "sole",
+    if (one_version) "one_version" else "versions",
+    if (is.null(overlap)) "no_overlap" else "overlap",
+    sep = "_"
   )
-  return(p)
-}
 
-plot_dynamic_mixture_overlap <- function(df_data, sit, successive, title = NULL) {
-  p <- ggplot2::ggplot(
-    df_data,
-    ggplot2::aes(
-      x = .data$Date,
-      colour = .data$var,
-      linetype = paste(.data$Dominance, ": ", .data$Plant),
-      shape = paste(.data$Dominance, ": ", .data$Plant)
-    )
-  ) +
-    ggplot2::geom_line(ggplot2::aes(y = .data$Simulated))
-
-  p <- add_vertical_lines(df_data, successive, p)
-
-  if ("Observed" %in% colnames(df_data)) {
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(
-        y = .data$Observed,
-        shape = paste(
-          .data$Dominance,
-          ": ", .data$Plant
-        ),
-        color = .data$var,
+  switch(
+    spec_name,
+    mixture_one_version_no_overlap = list(
+      mapping = ggplot2::aes(colour = paste(.data$Dominance, ":", .data$Plant)),
+      scales = "free_y", facets = "var",
+      extra = list(ggplot2::labs(colour = "Plant")),
+      legend_labels = unique(paste(df_data$Dominance, ":", df_data$Plant))
+    ),
+    mixture_one_version_overlap = list(
+      mapping = ggplot2::aes(
+        colour = .data$var,
+        linetype = paste(.data$Dominance, ": ", .data$Plant),
+        shape = paste(.data$Dominance, ": ", .data$Plant)
       ),
-      na.rm = TRUE
-    )
-    if ("Obs_SD" %in% colnames(df_data)) {
-      p <- p +
-        ggplot2::geom_errorbar(
-          ggplot2::aes(
-            ymin = .data$Observed - 2 * .data$Obs_SD,
-            ymax = .data$Observed + 2 * .data$Obs_SD
-          ),
-          na.rm = TRUE
+      facets = "group_var",
+      obs_mapping = ggplot2::aes(
+        shape = paste(.data$Dominance, ": ", .data$Plant),
+        color = .data$var
+      ),
+      extra = list(
+        ggplot2::guides(
+          colour = ggplot2::guide_legend(title = "Variable"),
+          linetype = ggplot2::guide_legend(title = "Plant", order = 1),
+          shape = ggplot2::guide_legend(title = "Plant", order = 1)
         )
-    }
-  }
-
-  title <- make_multiline_title(title)
-  p <- p +
-    ggplot2::ggtitle(title) +
-    ggplot2::guides(
-      colour = ggplot2::guide_legend(title = "Variable"),
-      # add override.aes = list(shape = NA) in prev guide_legend?
-      linetype = ggplot2::guide_legend(title = "Plant", order = 1),
-      shape = ggplot2::guide_legend(title = "Plant", order = 1)
-    ) +
-    ggplot2::theme_get()
-  p <- add_facet_wrap(
-    p, "group_var",
-    scales = "free",
-    legend_labels = c(
-      unique(as.character(df_data$var)),
-      unique(paste(df_data$Dominance, ": ", df_data$Plant))
-    )
-  )
-  return(p)
-}
-
-plot_dynamic_versions <- function(df_data, sit, successive, title = NULL) {
-  df_data$Observed_Legend <- "Observed Value"
-  p <- ggplot2::ggplot(
-    df_data,
-    ggplot2::aes(x = .data$Date, colour = .data$version)
-  ) +
-    ggplot2::geom_line(ggplot2::aes(y = .data$Simulated))
-
-  p <- add_vertical_lines(df_data, successive, p)
-
-  if ("Observed" %in% colnames(df_data)) {
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(y = .data$Observed, shape = .data$Observed_Legend),
-      # NB: the shape is constant, but used to have a legend entry
-      color = "black",
-      na.rm = TRUE
-    )
-    if ("Obs_SD" %in% colnames(df_data)) {
-      p <- p +
-        ggplot2::geom_errorbar(
-          ggplot2::aes(
-            ymin = .data$Observed - 2 * .data$Obs_SD,
-            ymax = .data$Observed + 2 * .data$Obs_SD,
-            shape = .data$version
+      ),
+      legend_labels = c(
+        unique(as.character(df_data$var)),
+        unique(paste(df_data$Dominance, ": ", df_data$Plant))
+      )
+    ),
+    sole_one_version_no_overlap = list(scales = "free_y", facets = "var"),
+    sole_one_version_overlap = list(
+      mapping = ggplot2::aes(colour = .data$var),
+      facets = "group_var",
+      obs_mapping = ggplot2::aes(shape = .data$var),
+      extra = list(
+        ggplot2::labs(colour = "Variable", shape = "Variable")
+      ),
+      legend_labels = unique(as.character(df_data$var))
+    ),
+    sole_versions_no_overlap = list(
+      mapping = ggplot2::aes(colour = .data$version),
+      facets = "var",
+      obs_mapping = ggplot2::aes(shape = "Observed Value"),
+      obs_params = list(color = "black"),
+      extra = list(
+        ggplot2::guides(
+          colour = ggplot2::guide_legend(
+            title = "Version",
+            override.aes = list(shape = NA)
           ),
-          na.rm = TRUE
+          shape = ggplot2::guide_legend(title = "Observations")
         )
-    }
-  }
-
-  title <- make_multiline_title(title)
-  p <- p +
-    ggplot2::ggtitle(title) +
-    ggplot2::guides(
-      colour = ggplot2::guide_legend(
-        title = "Version",
-        override.aes = list(shape = NA)
       ),
-      shape = ggplot2::guide_legend(title = "Observations")
-    ) +
-    ggplot2::theme_get()
-  p <- add_facet_wrap(
-    p,
-    var = "var", scales = "free",
-    legend_labels = c(unique(df_data$version), "Observed Value")
-  )
-  return(p)
-}
-
-plot_dynamic_overlap <- function(df_data, sit, successive, title = NULL) {
-  p <- ggplot2::ggplot(
-    df_data,
-    ggplot2::aes(x = .data$Date, colour = .data$var)
-  ) +
-    ggplot2::geom_line(ggplot2::aes(y = .data$Simulated))
-
-  p <- add_vertical_lines(df_data, successive, p)
-
-  if ("Observed" %in% colnames(df_data)) {
-    p <- p +
-      ggplot2::labs(shape = "Variable") +
-      ggplot2::geom_point(
-        ggplot2::aes(y = .data$Observed, shape = .data$var),
-        na.rm = TRUE
+      legend_labels = c(unique(df_data$version), "Observed Value")
+    ),
+    sole_versions_overlap = list(
+      mapping = ggplot2::aes(
+        colour = .data$var,
+        linetype = .data$version
+      ),
+      facets = "group_var",
+      obs_mapping = ggplot2::aes(colour = .data$var),
+      extra = list(
+        ggplot2::labs(colour = "Variable", linetype = "Version")
+      ),
+      legend_labels = c(
+        unique(as.character(df_data$var)),
+        unique(df_data$version)
       )
-    if ("Obs_SD" %in% colnames(df_data)) {
-      p <- p +
-        ggplot2::geom_errorbar(
-          ggplot2::aes(
-            ymin = .data$Observed - 2 * .data$Obs_SD,
-            ymax = .data$Observed + 2 * .data$Obs_SD,
-            shape = .data$version
-          ),
-          na.rm = TRUE
+    ),
+    mixture_versions_no_overlap = list(
+      mapping = ggplot2::aes(
+        colour = paste(.data$Dominance, ":", .data$Plant),
+        linetype = .data$version
+      ),
+      facets = "var",
+      extra = list(
+        ggplot2::labs(
+          colour = "Plant",
+          linetype = "Version"
         )
-    }
-  }
-  title <- make_multiline_title(title)
-  p <- p +
-    ggplot2::labs(colour = "Variable") +
-    ggplot2::ggtitle(title) +
-    ggplot2::theme_get()
-  p <- add_facet_wrap(
-    p,
-    var = "group_var", scales = "free",
-    legend_labels = unique(as.character(df_data$var))
-  )
-  return(p)
-}
-
-plot_dynamic_mixture_versions_overlap <- function(df_data, sit, successive, title = NULL) {
-  stop(
-    "Too many cases to consider at a time: mixture + versions + overlap. ",
-    "Please use only a maximum of two combinations of: ",
-    "mixture, versions, overlap."
+      ),
+      legend_labels = c(
+        unique(paste(df_data$Dominance, ":", df_data$Plant)),
+        unique(df_data$version)
+      )
+    ),
+    mixture_versions_overlap = stop(
+      "Too many cases to consider at a time: mixture + versions + overlap. ",
+      "Please use only a maximum of two combinations of: ",
+      "mixture, versions, overlap."
+    )
   )
 }
 
-
-plot_dynamic_versions_overlap <- function(df_data, sit, successive, title = NULL) {
+build_dynamic_plot <- function(df_data, sit, successive, title = NULL,
+                               mixture, one_version, overlap) {
+  spec <- dynamic_plot_spec(df_data,mixture, one_version, overlap)
   p <- ggplot2::ggplot(
     df_data,
-    ggplot2::aes(
-      x = .data$Date, colour = .data$var,
-      linetype = .data$version
-    )
+    ggplot2::aes(x = .data$Date, !!!spec$mapping)
   ) +
     ggplot2::geom_line(ggplot2::aes(y = .data$Simulated))
 
   p <- add_vertical_lines(df_data, successive, p)
 
   if ("Observed" %in% colnames(df_data)) {
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(y = .data$Observed, colour = .data$var),
-      na.rm = TRUE
+    p <- p + do.call(
+      ggplot2::geom_point,
+      c(
+        list(
+          mapping = ggplot2::aes(y = .data$Observed, !!!spec$obs_mapping),
+          na.rm = TRUE
+        ),
+        spec$obs_params
+      )
     )
     if ("Obs_SD" %in% colnames(df_data)) {
       p <- p +
         ggplot2::geom_errorbar(
           ggplot2::aes(
             ymin = .data$Observed - 2 * .data$Obs_SD,
-            ymax = .data$Observed + 2 * .data$Obs_SD,
-            colour = .data$var
+            ymax = .data$Observed + 2 * .data$Obs_SD
           ),
           na.rm = TRUE
         )
     }
   }
 
-  title <- make_multiline_title(title)
   p <- p +
-    ggplot2::ggtitle(title) +
-    ggplot2::labs(colour = "Variable", linetype = "Version") +
-    ggplot2::theme_get()
-
-  p <- add_facet_wrap(
-    p,
-    var = "group_var", scales = "free",
-    legend_labels = c(
-      unique(as.character(df_data$var)),
-      unique(df_data$version)
-    )
-  )
-  return(p)
-}
-
-plot_dynamic_mixture_versions <- function(df_data, sit, successive, title = NULL) {
-  p <- ggplot2::ggplot(
-    df_data,
-    ggplot2::aes(
-      x = .data$Date,
-      colour = paste(.data$Dominance, ":", .data$Plant),
-      linetype = .data$version
-    )
-  ) +
-    ggplot2::geom_line(ggplot2::aes(y = .data$Simulated))
+    ggplot2::ggtitle(make_multiline_title(title)) +
+    ggplot2::theme_get() +
+    spec$extra
 
-  p <- add_vertical_lines(df_data, successive, p)
+  p <- add_facet_wrap(p, spec$facets, spec$scales, legend_labels = spec$legend_labels)
 
-  if ("Observed" %in% colnames(df_data)) {
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(y = .data$Observed),
-      na.rm = TRUE
-    )
-    if ("Obs_SD" %in% colnames(df_data)) {
-      p <- p +
-        ggplot2::geom_errorbar(
-          ggplot2::aes(
-            ymin = .data$Observed - 2 * .data$Obs_SD,
-            ymax = .data$Observed + 2 * .data$Obs_SD
-          ),
-          na.rm = TRUE
-        )
-    }
-  }
-  title <- make_multiline_title(title)
-  p <- p +
-    ggplot2::ggtitle(title) +
-    ggplot2::labs(
-      colour = "Plant",
-      linetype = "Version"
-    ) +
-    ggplot2::theme_get()
-  p <- add_facet_wrap(
-    p,
-    var = "var", scales = "free",
-    legend_labels = c(
-      unique(paste(df_data$Dominance, ":", df_data$Plant)),
-      unique(df_data$version)
-    )
-  )
-  return(p)
+  p
 }
diff --git a/R/specific_plotting_scatter.R b/R/specific_plotting_scatter.R
index 4756dc2..15c19ff 100644
--- a/R/specific_plotting_scatter.R
+++ b/R/specific_plotting_scatter.R
@@ -230,28 +230,24 @@ give_y_var_type <- function(select_scat) {
 
 
 #' @keywords internal
-#' @description Add error bars on observed values in given scatterplot
+#' @description Build a scatter plot shared by all the specific scatter plot
+#' functions.
 #' @rdname specific_scatter_plots
-#' @param p A ggplot to modify`
-#' @param colour_factor The factor to use for colouring the error bars
-#' @return The modified ggplot
-add_obs_error_bars <- function(p, colour_factor = NULL) {
-  p <- p +
-    ggplot2::geom_linerange(
-      ggplot2::aes(
-        xmin = .data$Observed - 2 * .data$Obs_SD,
-        xmax = .data$Observed + 2 * .data$Obs_SD,
-        colour = .data[[colour_factor]],
-      ),
-      na.rm = TRUE
-    )
-  return(p)
-}
-
-#' @keywords internal
-#' @rdname specific_scatter_plots
-plot_scat_mixture_allsit <- function(df_data, sit, select_scat, shape_sit,
-                                     reference_var, is_obs_sd, title = NULL) {
+#' @param mapping Aesthetic mapping of the points (colour and/or shape). When
+#' `shape_sit` is "symbol" or "group", the situation is added on the shape, or
+#' on the colour if `mapping` has no colour (it replaces any existing shape).
+#' Error bars and text labels use the same colour as the points.
+#' @param smooth_by_colour If TRUE, one regression line per point colour,
+#' otherwise a single blue regression line.
+#' @param extra List of additional ggplot components (labs, scales...).
+#' @param legend_labels Named list (by aesthetic) of the labels displayed in the
+#' legend, passed to `add_facet_wrap`.
+#' @return A ggplot object
+build_scatter_plot <- function(
+  df_data, select_scat, shape_sit, reference_var, is_obs_sd, title = NULL,
+  mapping = ggplot2::aes(), smooth_by_colour = FALSE, extra = NULL,
+  legend_labels = list()
+) {
   tmp <- give_reference_var(reference_var)
   reference_var <- tmp$reference_var
   reference_var_name <- tmp$reference_var_name
@@ -261,6 +257,20 @@ plot_scat_mixture_allsit <- function(df_data, sit, select_scat, shape_sit,
     df_data %>%
     dplyr::filter(!is.na(.data[[reference_var]]) & !is.na(.data[[y_var_type]]))
 
+  if (shape_sit %in% c("symbol", "group")) {
+    sit_aes <- if (is.null(mapping$colour)) "colour" else "shape"
+    mapping[[sit_aes]] <- rlang::quo(as.factor(.data$sit_name))
+    extra <- c(extra, list(ggplot2::labs(!!sit_aes := "Situation")))
+    legend_labels[[sit_aes]] <- unique(df_data$sit_name)
+  }
+
+  smooth_aes <- ggplot2::aes(y = .data[[y_var_type]], x = .data[[reference_var]])
+  smooth_params <- list(colour = "blue")
+  if (smooth_by_colour) {
+    smooth_aes$colour <- mapping$colour
+    smooth_params <- list()
+  }
+
   p <-
     ggplot2::ggplot(
       df_data,
@@ -268,68 +278,56 @@ plot_scat_mixture_allsit <- function(df_data, sit, select_scat, shape_sit,
         y = .data[[y_var_type]], x = .data[[reference_var]],
         label = .data$sit_name
       )
-    )
-
-  if (shape_sit == "none" || shape_sit == "txt") {
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(
-        colour = as.factor(paste(.data$Dominance, ":", .data$Plant))
-      ),
-      na.rm = TRUE
-    )
-  } else if (shape_sit == "symbol" || shape_sit == "group") {
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(
-        colour = as.factor(paste(.data$Dominance, ":", .data$Plant)),
-        shape = as.factor(paste(.data$sit_name))
-      ),
-      na.rm = TRUE
     ) +
-      ggplot2::scale_shape_discrete(name = "Situation")
-  }
-
-  p <- p +
+    ggplot2::geom_point(mapping, na.rm = TRUE) +
     ggplot2::geom_abline(
       intercept = 0, slope = ifelse(select_scat == "sim", 1, 0),
       color = "grey30", linetype = 2
     ) +
-    ggplot2::geom_smooth(
-      ggplot2::aes(y = .data[[y_var_type]], x = .data[[reference_var]]),
-      inherit.aes = FALSE,
-      method = lm, color = "blue",
-      se = FALSE, linewidth = 0.6, formula = y ~ x,
-      fullrange = TRUE, na.rm = TRUE
+    do.call(
+      ggplot2::geom_smooth,
+      c(
+        list(
+          mapping = smooth_aes,
+          inherit.aes = FALSE,
+          method = lm,
+          se = FALSE, linewidth = 0.6, formula = y ~ x,
+          fullrange = TRUE, na.rm = TRUE
+        ),
+        smooth_params
+      )
     ) +
-    ggplot2::xlab(reference_var_name)
-
-  p <- p +
+    ggplot2::xlab(reference_var_name) +
     ggplot2::ggtitle(title)
 
   if (is_obs_sd && reference_var == "Observed") {
-    p$data$colour_factor <- as.factor(paste(p$data$Dominance, ":", p$data$Plant))
-    p <- add_obs_error_bars(p, colour_factor = "colour_factor")
+    error_aes <- ggplot2::aes(
+      xmin = .data$Observed - 2 * .data$Obs_SD,
+      xmax = .data$Observed + 2 * .data$Obs_SD
+    )
+    error_aes$colour <- mapping$colour
+    p <- p + ggplot2::geom_linerange(error_aes, na.rm = TRUE)
   }
 
   p <- p + ggplot2::theme(aspect.ratio = 1)
 
   if (shape_sit == "txt") {
+    text_aes <- ggplot2::aes()
+    text_aes$colour <- mapping$colour
     p <- p +
       ggrepel::geom_text_repel(
-        ggplot2::aes(
-          colour = as.factor(paste(.data$Dominance, ":", .data$Plant))
-        ),
+        text_aes,
         show.legend = FALSE,
         max.overlaps = 100
       )
   }
 
+  p <- p + extra
+
   p <- add_facet_wrap(
     p,
     var = "var", scales = "free",
-    legend_labels = c(
-      unique(paste(df_data$Dominance, ":", df_data$Plant)),
-      if (shape_sit %in% c("symbol", "group")) unique(df_data$sit_name)
-    )
+    legend_labels = unlist(legend_labels, use.names = FALSE)
   )
 
   # Set same limits for x and y axis for sim VS obs scatter plots
@@ -340,9 +338,23 @@ plot_scat_mixture_allsit <- function(df_data, sit, select_scat, shape_sit,
     p <- force_y_axis(df_data, reference_var, y_var_type, is_obs_sd, p)
   }
 
-  p <- p + ggplot2::scale_color_discrete(name = "Plant")
+  p
+}
 
-  return(p)
+#' @keywords internal
+#' @rdname specific_scatter_plots
+plot_scat_mixture_allsit <- function(df_data, sit, select_scat, shape_sit,
+                                     reference_var, is_obs_sd, title = NULL) {
+  build_scatter_plot(
+    df_data, select_scat, shape_sit, reference_var, is_obs_sd, title,
+    mapping = ggplot2::aes(
+      colour = as.factor(paste(.data$Dominance, ":", .data$Plant))
+    ),
+    extra = list(ggplot2::labs(colour = "Plant")),
+    legend_labels = list(
+      colour = unique(paste(df_data$Dominance, ":", df_data$Plant))
+    )
+  )
 }
 
 
@@ -350,115 +362,24 @@ plot_scat_mixture_allsit <- function(df_data, sit, select_scat, shape_sit,
 #' @rdname specific_scatter_plots
 plot_scat_mixture_versions <- function(df_data, sit, select_scat, shape_sit,
                                        reference_var, is_obs_sd, title = NULL) {
-  tmp <- give_reference_var(reference_var)
-  reference_var <- tmp$reference_var
-  reference_var_name <- tmp$reference_var_name
-  y_var_type <- give_y_var_type(select_scat)
-
-  df_data <-
-    df_data %>%
-    dplyr::filter(!is.na(.data[[reference_var]]) & !is.na(.data[[y_var_type]]))
-
-  p <-
-    ggplot2::ggplot(
-      df_data,
-      ggplot2::aes(
-        y = .data[[y_var_type]], x = .data[[reference_var]],
-        label = .data$sit_name
-      )
-    )
-
-  if (shape_sit == "none" || shape_sit == "txt") {
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(
-        shape = as.factor(paste(.data$Dominance, ":", .data$Plant)),
-        colour = as.factor(.data$version)
-      ),
-      na.rm = TRUE
-    ) +
-      ggplot2::labs(color = "Version", shape = "Plant")
-  } else if (shape_sit == "symbol" || shape_sit == "group") {
-    # ! In this case we loose the colour by species for mixtures, because
-    # there would be three aesthetics to handle (situation, version and
-    # species). We made this decision because the user explicitly asks
-    # for shape to be the situation name. If they want to color by species,
+  build_scatter_plot(
+    df_data, select_scat, shape_sit, reference_var, is_obs_sd, title,
+    # ! With shape_sit = "symbol" or "group", we loose the shape by species for
+    # mixtures, because there would be three aesthetics to handle (situation,
+    # version and species). We made this decision because the user explicitly
+    # asks for shape to be the situation name. If they want the species,
     # they can put shape_sit = "none" or shape_sit = "txt" to have it all.
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(
-        colour = as.factor(.data$version),
-        shape = as.factor(.data$sit_name)
-      ),
-      na.rm = TRUE
-    ) +
-      ggplot2::labs(color = "Version", shape = "Situation")
-  }
-
-  p <- p +
-    ggplot2::geom_abline(
-      intercept = 0, slope = ifelse(select_scat == "sim", 1, 0),
-      color = "grey30", linetype = 2
-    ) +
-    ggplot2::geom_smooth(
-      ggplot2::aes(
-        y = .data[[y_var_type]], x = .data[[reference_var]],
-        colour = as.factor(.data$version)
-      ),
-      inherit.aes = FALSE,
-      method = lm,
-      se = FALSE, linewidth = 0.6, formula = y ~ x,
-      fullrange = TRUE, na.rm = TRUE
-    ) +
-    ggplot2::xlab(reference_var_name)
-
-  p <- p + ggplot2::ggtitle(title)
-
-  if (is_obs_sd && reference_var == "Observed") {
-    p <- p +
-      ggplot2::geom_linerange(
-        ggplot2::aes(
-          xmin = .data$Observed - 2 * .data$Obs_SD,
-          xmax = .data$Observed + 2 * .data$Obs_SD,
-          colour = as.factor(.data$version),
-        ),
-        na.rm = TRUE
-      )
-  }
-
-  p <- p + ggplot2::theme(aspect.ratio = 1)
-
-  if (shape_sit == "txt") {
-    p <- p +
-      ggrepel::geom_text_repel(
-        ggplot2::aes(
-          colour = as.factor(.data$version)
-        ),
-        show.legend = FALSE,
-        max.overlaps = 100
-      )
-  }
-
-  p <- add_facet_wrap(
-    p,
-    var = "var", scales = "free",
-    legend_labels = c(
-      unique(df_data$version),
-      if (shape_sit %in% c("none", "txt")) {
-        unique(paste(df_data$Dominance, ":", df_data$Plant))
-      } else {
-        unique(df_data$sit_name)
-      }
+    mapping = ggplot2::aes(
+      colour = as.factor(.data$version),
+      shape = as.factor(paste(.data$Dominance, ":", .data$Plant))
+    ),
+    smooth_by_colour = TRUE,
+    extra = list(ggplot2::labs(colour = "Version", shape = "Plant")),
+    legend_labels = list(
+      colour = unique(df_data$version),
+      shape = unique(paste(df_data$Dominance, ":", df_data$Plant))
     )
   )
-
-  # Set same limits for x and y axis for sim VS obs scatter plots
-  if (select_scat == "sim" && reference_var == "Observed") {
-    p <- make_axis_square(df_data, reference_var, y_var_type, is_obs_sd, p)
-  }
-  if (select_scat == "res") {
-    p <- force_y_axis(df_data, reference_var, y_var_type, is_obs_sd, p)
-  }
-
-  return(p)
 }
 
 
@@ -468,88 +389,10 @@ plot_scat_allsit <- function(df_data, sit, select_scat, shape_sit,
                              reference_var, is_obs_sd, title = NULL,
                              has_distinct_situations = FALSE,
                              one_version = FALSE, mixture = FALSE) {
-  tmp <- give_reference_var(reference_var)
-  reference_var <- tmp$reference_var
-  reference_var_name <- tmp$reference_var_name
-  y_var_type <- give_y_var_type(select_scat)
-  df_data <-
-    df_data %>%
-    dplyr::filter(!is.na(.data[[reference_var]]) & !is.na(.data[[y_var_type]]))
-  p <-
-    ggplot2::ggplot(
-      df_data,
-      ggplot2::aes(
-        y = .data[[y_var_type]], x = .data[[reference_var]],
-        label = .data$sit_name
-      )
-    )
-  if (shape_sit == "none" || shape_sit == "txt") {
-    p <- p + ggplot2::geom_point(na.rm = TRUE)
-  } else if (shape_sit == "symbol" || shape_sit == "group") {
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(
-        colour = as.factor(paste(.data$sit_name))
-      ),
-      na.rm = TRUE
-    ) +
-      ggplot2::scale_color_discrete(name = "Situation")
-  }
-
-  p <- p +
-    ggplot2::geom_abline(
-      intercept = 0, slope = ifelse(select_scat == "sim", 1, 0),
-      color = "grey30", linetype = 2
-    ) +
-    ggplot2::geom_smooth(
-      ggplot2::aes(y = .data[[y_var_type]], x = .data[[reference_var]]),
-      inherit.aes = FALSE,
-      method = lm, color = "blue",
-      se = FALSE, linewidth = 0.6, formula = y ~ x,
-      fullrange = TRUE, na.rm = TRUE
-    ) +
-    ggplot2::xlab(reference_var_name)
-
-  p <- p +
-    ggplot2::ggtitle(title)
-
-  if (is_obs_sd && reference_var == "Observed") {
-    line_aes <- NULL
-    if (shape_sit == "symbol" || shape_sit == "group") {
-      line_aes <- ggplot2::aes(
-        xmin = .data$Observed - 2 * .data$Obs_SD,
-        xmax = .data$Observed + 2 * .data$Obs_SD,
-        colour = as.factor(paste(.data$sit_name))
-      )
-    } else {
-      line_aes <- ggplot2::aes(
-        xmin = .data$Observed - 2 * .data$Obs_SD,
-        xmax = .data$Observed + 2 * .data$Obs_SD
-      )
-    }
-    p <- p + ggplot2::geom_linerange(line_aes, na.rm = TRUE)
-  }
-
-  p <- p + ggplot2::theme(aspect.ratio = 1)
-
-  if (shape_sit == "txt") {
-    p <- p + ggrepel::geom_text_repel(max.overlaps = 100)
-  }
-
-  p <- add_facet_wrap(
-    p,
-    var = "var", scales = "free",
-    legend_labels = if (shape_sit %in% c("symbol", "group")) {
-      unique(df_data$sit_name)
-    }
+  p <- build_scatter_plot(
+    df_data, select_scat, shape_sit, reference_var, is_obs_sd, title
   )
 
-  # Set same limits for x and y axis for sim VS obs scatter plots
-  if (select_scat == "sim" && reference_var == "Observed") {
-    p <- make_axis_square(df_data, reference_var, y_var_type, is_obs_sd, p)
-  }
-  if (select_scat == "res") {
-    p <- force_y_axis(df_data, reference_var, y_var_type, is_obs_sd, p)
-  }
   if (
     has_distinct_situations == FALSE &&
       one_version == TRUE &&
@@ -558,7 +401,7 @@ plot_scat_allsit <- function(df_data, sit, select_scat, shape_sit,
     p <- p + ggplot2::theme(legend.position = "none")
   }
 
-  return(p)
+  p
 }
 
 #' @keywords internal
@@ -566,87 +409,16 @@ plot_scat_allsit <- function(df_data, sit, select_scat, shape_sit,
 plot_scat_versions_per_sit <- function(df_data,
                                        sit, select_scat, shape_sit,
                                        reference_var, is_obs_sd, title = NULL) {
-  tmp <- give_reference_var(reference_var)
-  reference_var <- tmp$reference_var
-  reference_var_name <- tmp$reference_var_name
-  y_var_type <- give_y_var_type(select_scat)
-
-  df_data <-
-    df_data %>%
-    dplyr::filter(!is.na(.data[[reference_var]]) & !is.na(.data[[y_var_type]]))
-
-  p <-
-    ggplot2::ggplot(
-      df_data,
-      ggplot2::aes(
-        y = .data[[y_var_type]], x = .data[[reference_var]],
-        label = .data$sit_name,
-      )
-    )
-
-  p <- p + ggplot2::geom_point(
-    ggplot2::aes(colour = as.factor(.data$version)),
-    na.rm = TRUE
-  ) +
-    ggplot2::labs(color = "Version")
-  p <- p +
-    ggplot2::geom_abline(
-      intercept = 0, slope = ifelse(select_scat == "sim", 1, 0),
-      color = "grey30", linetype = 2
-    ) +
-    ggplot2::geom_smooth(
-      ggplot2::aes(
-        y = .data[[y_var_type]], x = .data[[reference_var]],
-        colour = as.factor(.data$version)
-      ),
-      method = lm,
-      inherit.aes = FALSE,
-      se = FALSE, linewidth = 0.6, formula = y ~ x,
-      fullrange = TRUE, na.rm = TRUE
-    ) +
-    ggplot2::xlab(reference_var_name)
-
-  p <- p + ggplot2::ggtitle(title)
-  if (shape_sit == "txt") {
-    p <- p +
-      ggrepel::geom_text_repel(
-        ggplot2::aes(
-          colour = as.factor(.data$version)
-        ),
-        show.legend = FALSE,
-        max.overlaps = 100
-      )
-  }
-
-  if (is_obs_sd && reference_var == "Observed") {
-    p <- p +
-      ggplot2::geom_linerange(
-        ggplot2::aes(
-          xmin = .data$Observed - 2 * .data$Obs_SD,
-          xmax = .data$Observed + 2 * .data$Obs_SD,
-          colour = as.factor(.data$version),
-        ),
-        na.rm = TRUE
-      )
-  }
-
-  p <- p + ggplot2::theme(aspect.ratio = 1)
-
-  p <- add_facet_wrap(
-    p,
-    var = "var", scales = "free",
-    legend_labels = unique(df_data$version)
+  build_scatter_plot(
+    df_data, select_scat,
+    # Only one situation per plot: no need for the situation in the legend
+    shape_sit = if (shape_sit == "txt") "txt" else "none",
+    reference_var, is_obs_sd, title,
+    mapping = ggplot2::aes(colour = as.factor(.data$version)),
+    smooth_by_colour = TRUE,
+    extra = list(ggplot2::labs(colour = "Version")),
+    legend_labels = list(colour = unique(df_data$version))
   )
-
-  # Set same limits for x and y axis for sim VS obs scatter plots
-  if (select_scat == "sim" && reference_var == "Observed") {
-    p <- make_axis_square(df_data, reference_var, y_var_type, is_obs_sd, p)
-  }
-  if (select_scat == "res") {
-    p <- force_y_axis(df_data, reference_var, y_var_type, is_obs_sd, p)
-  }
-
-  return(p)
 }
 
 
@@ -655,106 +427,11 @@ plot_scat_versions_per_sit <- function(df_data,
 plot_scat_versions_allsit <- function(df_data,
                                       sit, select_scat, shape_sit,
                                       reference_var, is_obs_sd, title = NULL) {
-  tmp <- give_reference_var(reference_var)
-  reference_var <- tmp$reference_var
-  reference_var_name <- tmp$reference_var_name
-  y_var_type <- give_y_var_type(select_scat)
-
-  df_data <-
-    df_data %>%
-    dplyr::filter(!is.na(.data[[reference_var]]) & !is.na(.data[[y_var_type]]))
-
-  p <-
-    ggplot2::ggplot(
-      df_data,
-      ggplot2::aes(
-        y = .data[[y_var_type]], x = .data[[reference_var]],
-        label = .data$sit_name
-      )
-    )
-
-  if (shape_sit == "none" || shape_sit == "txt") {
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(
-        colour = as.factor(.data$version)
-      ),
-      na.rm = TRUE
-    ) +
-      ggplot2::labs(color = "Version")
-  } else if (shape_sit == "symbol" || shape_sit == "group") {
-    # ! In this case we loose the colour by species for mixtures, because
-    # there would be three aesthetics to handle (situation, version and
-    # species). We made this decision because the user explicitly asks
-    # for shape to be the situation name. If they want to color by species,
-    # they can put shape_sit = "none" or shape_sit = "txt" to have it all.
-    p <- p + ggplot2::geom_point(
-      ggplot2::aes(
-        colour = as.factor(.data$version),
-        shape = as.factor(.data$sit_name)
-      ),
-      na.rm = TRUE
-    ) +
-      ggplot2::labs(color = "Version", shape = "Situation")
-  }
-
-  p <- p +
-    ggplot2::geom_abline(
-      intercept = 0, slope = ifelse(select_scat == "sim", 1, 0),
-      color = "grey30", linetype = 2
-    ) +
-    ggplot2::geom_smooth(
-      ggplot2::aes(
-        y = .data[[y_var_type]], x = .data[[reference_var]],
-        colour = as.factor(.data$version)
-      ),
-      inherit.aes = FALSE,
-      method = lm,
-      se = FALSE, linewidth = 0.6, formula = y ~ x,
-      fullrange = TRUE, na.rm = TRUE
-    ) +
-    ggplot2::xlab(reference_var_name)
-
-  p <- p + ggplot2::ggtitle(title)
-
-  if (is_obs_sd && reference_var == "Observed") {
-    p <- p +
-      ggplot2::geom_linerange(
-        ggplot2::aes(
-          xmin = .data$Observed - 2 * .data$Obs_SD,
-          xmax = .data$Observed + 2 * .data$Obs_SD,
-          colour = as.factor(.data$version),
-        ),
-        na.rm = TRUE
-      )
-  }
-
-  p <- p + ggplot2::theme(aspect.ratio = 1)
-
-  if (shape_sit == "txt") {
-    p <- p +
-      ggrepel::geom_text_repel(
-        ggplot2::aes(colour = as.factor(.data$version)),
-        show.legend = FALSE,
-        max.overlaps = 100
-      )
-  }
-
-  p <- add_facet_wrap(
-    p,
-    var = "var", scales = "free",
-    legend_labels = c(
-      unique(df_data$version),
-      if (shape_sit %in% c("symbol", "group")) unique(df_data$sit_name)
-    )
+  build_scatter_plot(
+    df_data, select_scat, shape_sit, reference_var, is_obs_sd, title,
+    mapping = ggplot2::aes(colour = as.factor(.data$version)),
+    smooth_by_colour = TRUE,
+    extra = list(ggplot2::labs(colour = "Version")),
+    legend_labels = list(colour = unique(df_data$version))
   )
-
-  # Set same limits for x and y axis for sim VS obs scatter plots
-  if (select_scat == "sim" && reference_var == "Observed") {
-    p <- make_axis_square(df_data, reference_var, y_var_type, is_obs_sd, p)
-  }
-  if (select_scat == "res") {
-    p <- force_y_axis(df_data, reference_var, y_var_type, is_obs_sd, p)
-  }
-
-  return(p)
 }
diff --git a/man/specific_dynamic_plots.Rd b/man/specific_dynamic_plots.Rd
index 28b1ea9..a00c4bb 100644
--- a/man/specific_dynamic_plots.Rd
+++ b/man/specific_dynamic_plots.Rd
@@ -4,14 +4,11 @@
 \alias{specific_dynamic_plots}
 \alias{add_vertical_lines}
 \alias{make_multiline_title}
-\alias{plot_dynamic}
 \title{Specific functions to generate dynamic plots}
 \usage{
 add_vertical_lines(df_data, successive, p)
 
 make_multiline_title(title, max_char = 120)
-
-plot_dynamic(df_data, sit, successive, title = NULL)
 }
 \arguments{
 \item{df_data}{A named list of data frame including the data to plot (one df
diff --git a/man/specific_scatter_plots.Rd b/man/specific_scatter_plots.Rd
index 405ccf4..0cccab8 100644
--- a/man/specific_scatter_plots.Rd
+++ b/man/specific_scatter_plots.Rd
@@ -6,7 +6,7 @@
 \alias{compute_axis_bounds}
 \alias{make_axis_square}
 \alias{force_y_axis}
-\alias{add_obs_error_bars}
+\alias{build_scatter_plot}
 \alias{plot_scat_mixture_allsit}
 \alias{plot_scat_mixture_versions}
 \alias{plot_scat_allsit}
@@ -22,7 +22,18 @@ make_axis_square(df_data, reference_var, y_var_type, is_obs_sd, p)
 
 force_y_axis(df_data, reference_var, y_var_type, is_obs_sd, p)
 
-add_obs_error_bars(p, colour_factor = NULL)
+build_scatter_plot(
+  df_data,
+  select_scat,
+  shape_sit,
+  reference_var,
+  is_obs_sd,
+  title = NULL,
+  mapping = ggplot2::aes(),
+  smooth_by_colour = FALSE,
+  extra = NULL,
+  legend_labels = list()
+)
 
 plot_scat_mixture_allsit(
   df_data,
@@ -90,7 +101,18 @@ per situation, or only one df if sit==all_situations)}
 
 \item{is_obs_sd}{TRUE if error standard deviation of observations is provided}
 
-\item{colour_factor}{The factor to use for colouring the error bars}
+\item{mapping}{Aesthetic mapping of the points (colour and/or shape). When
+\code{shape_sit} is "symbol" or "group", the situation is added on the shape, or
+on the colour if \code{mapping} has no colour (it replaces any existing shape).
+Error bars and text labels use the same colour as the points.}
+
+\item{smooth_by_colour}{If TRUE, one regression line per point colour,
+otherwise a single blue regression line.}
+
+\item{extra}{List of additional ggplot components (labs, scales...).}
+
+\item{legend_labels}{Named list (by aesthetic) of the labels displayed in the
+legend, passed to \code{add_facet_wrap}.}
 
 \item{sit}{The name of the situation to plot (or all_situations)}
 
@@ -114,7 +136,7 @@ The modified ggplot
 
 The modified ggplot
 
-The modified ggplot
+A ggplot object
 }
 \description{
 Generate scatter plots for the different cases handled in
@@ -135,7 +157,8 @@ Make axis square
 Ensure that the Y axis includes zero when all values in a facet
 are strictly positive or strictly negative.
 
-Add error bars on observed values in given scatterplot
+Build a scatter plot shared by all the specific scatter plot
+functions.
 }
 \details{
 List of the different specific functions:

From 7c7a315a3f4ca1092904c29c1f5c6557e6b897fc Mon Sep 17 00:00:00 2001
From: vrahier <vrahier@users.noreply.github.com>
Date: Wed, 7 Oct 2026 12:40:33 +0000
Subject: [PATCH 2/6] Style code (GHA)

---
 R/generic_plotting.R          | 78 +++++++++++++++++------------------
 R/specific_plotting_dynamic.R |  5 +--
 2 files changed, 41 insertions(+), 42 deletions(-)

diff --git a/R/generic_plotting.R b/R/generic_plotting.R
index 0c70891..b0d40ca 100644
--- a/R/generic_plotting.R
+++ b/R/generic_plotting.R
@@ -191,46 +191,46 @@ plot_situations <- function(..., obs = NULL, obs_sd = NULL,
       )
     } else {
       p[[i]] <- switch(item_case,
-            # Scatter plots:
-            "mixture_versions" =
-              plot_scat_mixture_versions( # per sit and all sit share the same call
-                sim_situation, i, select_scat, shape_sit,
-                reference_var, is_obs_sd,
-                title = plot_title
-              ),
-            "mixture_no_versions" = # per sit and all sit share the same call
-              plot_scat_mixture_allsit(
-                sim_situation, i, select_scat, shape_sit,
-                reference_var, is_obs_sd,
-                title = plot_title
-              ),
-            "non_mixture_versions_situations" =
-              plot_scat_versions_allsit(
-                sim_situation, i, select_scat, shape_sit,
-                reference_var, is_obs_sd,
-                title = plot_title
-              ),
-            "non_mixture_versions_per_situations" =
-              plot_scat_versions_per_sit(
-                sim_situation, i, select_scat, shape_sit,
-                reference_var, is_obs_sd,
-                title = plot_title
-              ),
-            "non_mixture_no_versions_situations" =
-              plot_scat_allsit(
-                sim_situation, i, select_scat, shape_sit,
-                reference_var, is_obs_sd,
-                title = plot_title, has_distinct_situations = has_distinct_situations,
-                one_version = one_version, mixture = mixture
-              ),
-            "non_mixture_no_versions_per_situations" =
-              plot_scat_allsit(
-                sim_situation, i, select_scat, shape_sit,
-                reference_var, is_obs_sd,
-                title = plot_title, has_distinct_situations = has_distinct_situations,
-                one_version = one_version, mixture = mixture
-              )
+        # Scatter plots:
+        "mixture_versions" =
+          plot_scat_mixture_versions( # per sit and all sit share the same call
+            sim_situation, i, select_scat, shape_sit,
+            reference_var, is_obs_sd,
+            title = plot_title
+          ),
+        "mixture_no_versions" = # per sit and all sit share the same call
+          plot_scat_mixture_allsit(
+            sim_situation, i, select_scat, shape_sit,
+            reference_var, is_obs_sd,
+            title = plot_title
+          ),
+        "non_mixture_versions_situations" =
+          plot_scat_versions_allsit(
+            sim_situation, i, select_scat, shape_sit,
+            reference_var, is_obs_sd,
+            title = plot_title
+          ),
+        "non_mixture_versions_per_situations" =
+          plot_scat_versions_per_sit(
+            sim_situation, i, select_scat, shape_sit,
+            reference_var, is_obs_sd,
+            title = plot_title
+          ),
+        "non_mixture_no_versions_situations" =
+          plot_scat_allsit(
+            sim_situation, i, select_scat, shape_sit,
+            reference_var, is_obs_sd,
+            title = plot_title, has_distinct_situations = has_distinct_situations,
+            one_version = one_version, mixture = mixture
+          ),
+        "non_mixture_no_versions_per_situations" =
+          plot_scat_allsit(
+            sim_situation, i, select_scat, shape_sit,
+            reference_var, is_obs_sd,
+            title = plot_title, has_distinct_situations = has_distinct_situations,
+            one_version = one_version, mixture = mixture
           )
+      )
     }
   }
 
diff --git a/R/specific_plotting_dynamic.R b/R/specific_plotting_dynamic.R
index 377239f..104bfa7 100644
--- a/R/specific_plotting_dynamic.R
+++ b/R/specific_plotting_dynamic.R
@@ -102,8 +102,7 @@ dynamic_plot_spec <- function(df_data, mixture, one_version, overlap) {
     sep = "_"
   )
 
-  switch(
-    spec_name,
+  switch(spec_name,
     mixture_one_version_no_overlap = list(
       mapping = ggplot2::aes(colour = paste(.data$Dominance, ":", .data$Plant)),
       scales = "free_y", facets = "var",
@@ -201,7 +200,7 @@ dynamic_plot_spec <- function(df_data, mixture, one_version, overlap) {
 
 build_dynamic_plot <- function(df_data, sit, successive, title = NULL,
                                mixture, one_version, overlap) {
-  spec <- dynamic_plot_spec(df_data,mixture, one_version, overlap)
+  spec <- dynamic_plot_spec(df_data, mixture, one_version, overlap)
   p <- ggplot2::ggplot(
     df_data,
     ggplot2::aes(x = .data$Date, !!!spec$mapping)

From bd63d80f7a4156a69091f26aaa2af824e9a5e2a6 Mon Sep 17 00:00:00 2001
From: Valentine Rahier <valentine.rahier@inrae.fr>
Date: Wed, 7 Oct 2026 14:58:45 +0200
Subject: [PATCH 3/6] fix scales

---
 R/specific_plotting_dynamic.R | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/R/specific_plotting_dynamic.R b/R/specific_plotting_dynamic.R
index 104bfa7..4c7ae6f 100644
--- a/R/specific_plotting_dynamic.R
+++ b/R/specific_plotting_dynamic.R
@@ -237,7 +237,8 @@ build_dynamic_plot <- function(df_data, sit, successive, title = NULL,
     ggplot2::theme_get() +
     spec$extra
 
-  p <- add_facet_wrap(p, spec$facets, spec$scales, legend_labels = spec$legend_labels)
+  scales <- if (is.null(spec$scales)) "free" else spec$scales
+  p <- add_facet_wrap(p, spec$facets, scales, legend_labels = spec$legend_labels)
 
   p
 }

From 02ad3be584c7413d53ba28f758cefc061e17dcc8 Mon Sep 17 00:00:00 2001
From: Valentine Rahier <valentine.rahier@inrae.fr>
Date: Wed, 7 Oct 2026 15:19:22 +0200
Subject: [PATCH 4/6] clean code

---
 R/aesthetics.R                |  29 -----
 R/generic_plotting.R          |  48 +-------
 R/specific_plotting_scatter.R | 204 ++++++++++++++--------------------
 man/detect_item_case.Rd       |  36 ------
 man/specific_scatter_plots.Rd | 109 +++++-------------
 5 files changed, 114 insertions(+), 312 deletions(-)
 delete mode 100644 man/detect_item_case.Rd

diff --git a/R/aesthetics.R b/R/aesthetics.R
index 5b0a0d4..8c30066 100644
--- a/R/aesthetics.R
+++ b/R/aesthetics.R
@@ -89,32 +89,3 @@ detect_mixture_version_situations <- function(is_mixture, one_version, has_disti
 
   return(case)
 }
-
-#' Detect items cases
-#'
-#' This function returns a unique string based on the type of plot, and
-#' whether the situation is a mixture or not, if there is one or multiple
-#' versions to plot, and if there is one or several situations to plot
-#' into the same plot.
-#' The output is used to choose the right plotting function in a switch.
-#'
-#' @param type The type of plot required, either 'dynamic' or 'scatter
-#' @param is_mixture A logical value indicating whether the crop is a mixture or not.
-#' @param one_version A logical value indicating whether the plot has one or multiple versions (e.g. of the model).
-#' @param has_distinct_situations A logical value indicating whether there are one or several situations to plot.
-#' @param overlap A logical value indicating whether there is any overlapping variables in the plot.
-#'
-#' @return A unique character string for the plot.
-#'
-#' @keywords internal
-detect_item_case <- function(type, is_mixture, one_version, has_distinct_situations, overlap) {
-  if (type == "dynamic") {
-    item_case <- detect_mixture_version_overlap(is_mixture, one_version, overlap)
-  } else if (type == "scatter") {
-    item_case <- detect_mixture_version_situations(is_mixture, one_version, has_distinct_situations)
-  } else {
-    stop("type must be either 'dynamic' or 'scatter'")
-  }
-
-  return(item_case)
-}
diff --git a/R/generic_plotting.R b/R/generic_plotting.R
index b0d40ca..c6b47be 100644
--- a/R/generic_plotting.R
+++ b/R/generic_plotting.R
@@ -171,9 +171,6 @@ plot_situations <- function(..., obs = NULL, obs_sd = NULL,
   for (i in common_situations_models) {
     sim_situation <- formated_situation_list[[i]]
     mixture <- detect_mixture(sim_situation)
-    item_case <- detect_item_case(
-      type, mixture, one_version, has_distinct_situations, overlap
-    )
 
     plot_title <- if (!is.null(title)) {
       title[[i]]
@@ -190,46 +187,11 @@ plot_situations <- function(..., obs = NULL, obs_sd = NULL,
         mixture, one_version, overlap
       )
     } else {
-      p[[i]] <- switch(item_case,
-        # Scatter plots:
-        "mixture_versions" =
-          plot_scat_mixture_versions( # per sit and all sit share the same call
-            sim_situation, i, select_scat, shape_sit,
-            reference_var, is_obs_sd,
-            title = plot_title
-          ),
-        "mixture_no_versions" = # per sit and all sit share the same call
-          plot_scat_mixture_allsit(
-            sim_situation, i, select_scat, shape_sit,
-            reference_var, is_obs_sd,
-            title = plot_title
-          ),
-        "non_mixture_versions_situations" =
-          plot_scat_versions_allsit(
-            sim_situation, i, select_scat, shape_sit,
-            reference_var, is_obs_sd,
-            title = plot_title
-          ),
-        "non_mixture_versions_per_situations" =
-          plot_scat_versions_per_sit(
-            sim_situation, i, select_scat, shape_sit,
-            reference_var, is_obs_sd,
-            title = plot_title
-          ),
-        "non_mixture_no_versions_situations" =
-          plot_scat_allsit(
-            sim_situation, i, select_scat, shape_sit,
-            reference_var, is_obs_sd,
-            title = plot_title, has_distinct_situations = has_distinct_situations,
-            one_version = one_version, mixture = mixture
-          ),
-        "non_mixture_no_versions_per_situations" =
-          plot_scat_allsit(
-            sim_situation, i, select_scat, shape_sit,
-            reference_var, is_obs_sd,
-            title = plot_title, has_distinct_situations = has_distinct_situations,
-            one_version = one_version, mixture = mixture
-          )
+      p[[i]] <- build_scatter_plot(
+        sim_situation, select_scat, shape_sit, reference_var, is_obs_sd,
+        title = plot_title,
+        mixture = mixture, one_version = one_version,
+        has_distinct_situations = has_distinct_situations
       )
     }
   }
diff --git a/R/specific_plotting_scatter.R b/R/specific_plotting_scatter.R
index 15c19ff..fa83919 100644
--- a/R/specific_plotting_scatter.R
+++ b/R/specific_plotting_scatter.R
@@ -17,10 +17,9 @@
 #'
 #' @details List of the different specific functions:
 #' \itemize{
-#'   \item `plot_scat_mixture_allsit`: Generate a scatter plot for the case of
-#' mixture of crops, single simulation version and all_situations in same plot
-#'   \item `plot_scat_allsit`: Generate a scatter plot for the case of
-#' sole crops, single simulation version and all_situations in same plot
+#'   \item `scatter_plot_spec`: Give the aesthetics, labels and legend of the
+#' scatter plot according to the case (mixture or not, one or several versions)
+#'   \item `build_scatter_plot`: Generate a scatter plot for all the cases
 #' }
 #'
 #' @return A list of ggplot objects
@@ -228,25 +227,68 @@ give_y_var_type <- function(select_scat) {
   return(y_var_type)
 }
 
+#' @keywords internal
+#' @description Give the plot specification (aesthetics, labels, legend) of a
+#' scatter plot according to the case: mixture or not, one or several versions.
+#' @rdname specific_scatter_plots
+#' @return A list with `mapping` (aesthetic mapping of the points),
+#' `smooth_by_colour` (one regression line per point colour if TRUE, otherwise
+#' a single blue line), `extra` (additional ggplot components) and
+#' `legend_labels` (named list by aesthetic of the legend labels).
+scatter_plot_spec <- function(df_data, mixture, one_version) {
+  plant_labels <- unique(paste(df_data$Dominance, ":", df_data$Plant))
+
+  if (mixture && one_version) {
+    list(
+      mapping = ggplot2::aes(
+        colour = as.factor(paste(.data$Dominance, ":", .data$Plant))
+      ),
+      smooth_by_colour = FALSE,
+      extra = list(ggplot2::labs(colour = "Plant")),
+      legend_labels = list(colour = plant_labels)
+    )
+  } else if (mixture) {
+    list(
+      mapping = ggplot2::aes(
+        colour = as.factor(.data$version),
+        shape = as.factor(paste(.data$Dominance, ":", .data$Plant))
+      ),
+      smooth_by_colour = TRUE,
+      extra = list(ggplot2::labs(colour = "Version", shape = "Plant")),
+      legend_labels = list(
+        colour = unique(df_data$version),
+        shape = plant_labels
+      )
+    )
+  } else if (!one_version) {
+    list(
+      mapping = ggplot2::aes(colour = as.factor(.data$version)),
+      smooth_by_colour = TRUE,
+      extra = list(ggplot2::labs(colour = "Version")),
+      legend_labels = list(colour = unique(df_data$version))
+    )
+  } else {
+    list(
+      mapping = ggplot2::aes(),
+      smooth_by_colour = FALSE,
+      extra = NULL,
+      legend_labels = list()
+    )
+  }
+}
+
 
 #' @keywords internal
-#' @description Build a scatter plot shared by all the specific scatter plot
-#' functions.
+#' @description Build a scatter plot for all the cases handled in CroPlotR.
+#' When `shape_sit` is "symbol" or "group", the situation is added on the
+#' shape of the points, or on their colour if no colour is used (it replaces
+#' any existing shape). Error bars and text labels use the same colour as the
+#' points.
 #' @rdname specific_scatter_plots
-#' @param mapping Aesthetic mapping of the points (colour and/or shape). When
-#' `shape_sit` is "symbol" or "group", the situation is added on the shape, or
-#' on the colour if `mapping` has no colour (it replaces any existing shape).
-#' Error bars and text labels use the same colour as the points.
-#' @param smooth_by_colour If TRUE, one regression line per point colour,
-#' otherwise a single blue regression line.
-#' @param extra List of additional ggplot components (labs, scales...).
-#' @param legend_labels Named list (by aesthetic) of the labels displayed in the
-#' legend, passed to `add_facet_wrap`.
 #' @return A ggplot object
 build_scatter_plot <- function(
   df_data, select_scat, shape_sit, reference_var, is_obs_sd, title = NULL,
-  mapping = ggplot2::aes(), smooth_by_colour = FALSE, extra = NULL,
-  legend_labels = list()
+  mixture = FALSE, one_version = TRUE, has_distinct_situations = FALSE
 ) {
   tmp <- give_reference_var(reference_var)
   reference_var <- tmp$reference_var
@@ -257,17 +299,30 @@ build_scatter_plot <- function(
     df_data %>%
     dplyr::filter(!is.na(.data[[reference_var]]) & !is.na(.data[[y_var_type]]))
 
+  spec <- scatter_plot_spec(df_data, mixture, one_version)
+
+  # Sole crop, versions, one situation per plot: no need for the situation in
+  # the legend
+  if (!mixture && !one_version && !has_distinct_situations) {
+    shape_sit <- if (shape_sit == "txt") "txt" else "none"
+  }
+
   if (shape_sit %in% c("symbol", "group")) {
-    sit_aes <- if (is.null(mapping$colour)) "colour" else "shape"
-    mapping[[sit_aes]] <- rlang::quo(as.factor(.data$sit_name))
-    extra <- c(extra, list(ggplot2::labs(!!sit_aes := "Situation")))
-    legend_labels[[sit_aes]] <- unique(df_data$sit_name)
+    # ! For mixtures with versions, the situation replaces the species on the
+    # shape, because there would be three aesthetics to handle (situation,
+    # version and species). We made this decision because the user explicitly
+    # asks for shape to be the situation name. If they want the species,
+    # they can put shape_sit = "none" or shape_sit = "txt" to have it all.
+    sit_aes <- if (is.null(spec$mapping$colour)) "colour" else "shape"
+    spec$mapping[[sit_aes]] <- rlang::quo(as.factor(.data$sit_name))
+    spec$extra <- c(spec$extra, list(ggplot2::labs(!!sit_aes := "Situation")))
+    spec$legend_labels[[sit_aes]] <- unique(df_data$sit_name)
   }
 
   smooth_aes <- ggplot2::aes(y = .data[[y_var_type]], x = .data[[reference_var]])
   smooth_params <- list(colour = "blue")
-  if (smooth_by_colour) {
-    smooth_aes$colour <- mapping$colour
+  if (spec$smooth_by_colour) {
+    smooth_aes$colour <- spec$mapping$colour
     smooth_params <- list()
   }
 
@@ -279,7 +334,7 @@ build_scatter_plot <- function(
         label = .data$sit_name
       )
     ) +
-    ggplot2::geom_point(mapping, na.rm = TRUE) +
+    ggplot2::geom_point(spec$mapping, na.rm = TRUE) +
     ggplot2::geom_abline(
       intercept = 0, slope = ifelse(select_scat == "sim", 1, 0),
       color = "grey30", linetype = 2
@@ -305,7 +360,7 @@ build_scatter_plot <- function(
       xmin = .data$Observed - 2 * .data$Obs_SD,
       xmax = .data$Observed + 2 * .data$Obs_SD
     )
-    error_aes$colour <- mapping$colour
+    error_aes$colour <- spec$mapping$colour
     p <- p + ggplot2::geom_linerange(error_aes, na.rm = TRUE)
   }
 
@@ -313,7 +368,7 @@ build_scatter_plot <- function(
 
   if (shape_sit == "txt") {
     text_aes <- ggplot2::aes()
-    text_aes$colour <- mapping$colour
+    text_aes$colour <- spec$mapping$colour
     p <- p +
       ggrepel::geom_text_repel(
         text_aes,
@@ -322,12 +377,12 @@ build_scatter_plot <- function(
       )
   }
 
-  p <- p + extra
+  p <- p + spec$extra
 
   p <- add_facet_wrap(
     p,
     var = "var", scales = "free",
-    legend_labels = unlist(legend_labels, use.names = FALSE)
+    legend_labels = unlist(spec$legend_labels, use.names = FALSE)
   )
 
   # Set same limits for x and y axis for sim VS obs scatter plots
@@ -338,100 +393,9 @@ build_scatter_plot <- function(
     p <- force_y_axis(df_data, reference_var, y_var_type, is_obs_sd, p)
   }
 
-  p
-}
-
-#' @keywords internal
-#' @rdname specific_scatter_plots
-plot_scat_mixture_allsit <- function(df_data, sit, select_scat, shape_sit,
-                                     reference_var, is_obs_sd, title = NULL) {
-  build_scatter_plot(
-    df_data, select_scat, shape_sit, reference_var, is_obs_sd, title,
-    mapping = ggplot2::aes(
-      colour = as.factor(paste(.data$Dominance, ":", .data$Plant))
-    ),
-    extra = list(ggplot2::labs(colour = "Plant")),
-    legend_labels = list(
-      colour = unique(paste(df_data$Dominance, ":", df_data$Plant))
-    )
-  )
-}
-
-
-#' @keywords internal
-#' @rdname specific_scatter_plots
-plot_scat_mixture_versions <- function(df_data, sit, select_scat, shape_sit,
-                                       reference_var, is_obs_sd, title = NULL) {
-  build_scatter_plot(
-    df_data, select_scat, shape_sit, reference_var, is_obs_sd, title,
-    # ! With shape_sit = "symbol" or "group", we loose the shape by species for
-    # mixtures, because there would be three aesthetics to handle (situation,
-    # version and species). We made this decision because the user explicitly
-    # asks for shape to be the situation name. If they want the species,
-    # they can put shape_sit = "none" or shape_sit = "txt" to have it all.
-    mapping = ggplot2::aes(
-      colour = as.factor(.data$version),
-      shape = as.factor(paste(.data$Dominance, ":", .data$Plant))
-    ),
-    smooth_by_colour = TRUE,
-    extra = list(ggplot2::labs(colour = "Version", shape = "Plant")),
-    legend_labels = list(
-      colour = unique(df_data$version),
-      shape = unique(paste(df_data$Dominance, ":", df_data$Plant))
-    )
-  )
-}
-
-
-#' @keywords internal
-#' @rdname specific_scatter_plots
-plot_scat_allsit <- function(df_data, sit, select_scat, shape_sit,
-                             reference_var, is_obs_sd, title = NULL,
-                             has_distinct_situations = FALSE,
-                             one_version = FALSE, mixture = FALSE) {
-  p <- build_scatter_plot(
-    df_data, select_scat, shape_sit, reference_var, is_obs_sd, title
-  )
-
-  if (
-    has_distinct_situations == FALSE &&
-      one_version == TRUE &&
-      mixture == FALSE
-  ) {
+  if (!has_distinct_situations && one_version && !mixture) {
     p <- p + ggplot2::theme(legend.position = "none")
   }
 
   p
 }
-
-#' @keywords internal
-#' @rdname specific_scatter_plots
-plot_scat_versions_per_sit <- function(df_data,
-                                       sit, select_scat, shape_sit,
-                                       reference_var, is_obs_sd, title = NULL) {
-  build_scatter_plot(
-    df_data, select_scat,
-    # Only one situation per plot: no need for the situation in the legend
-    shape_sit = if (shape_sit == "txt") "txt" else "none",
-    reference_var, is_obs_sd, title,
-    mapping = ggplot2::aes(colour = as.factor(.data$version)),
-    smooth_by_colour = TRUE,
-    extra = list(ggplot2::labs(colour = "Version")),
-    legend_labels = list(colour = unique(df_data$version))
-  )
-}
-
-
-#' @keywords internal
-#' @rdname specific_scatter_plots
-plot_scat_versions_allsit <- function(df_data,
-                                      sit, select_scat, shape_sit,
-                                      reference_var, is_obs_sd, title = NULL) {
-  build_scatter_plot(
-    df_data, select_scat, shape_sit, reference_var, is_obs_sd, title,
-    mapping = ggplot2::aes(colour = as.factor(.data$version)),
-    smooth_by_colour = TRUE,
-    extra = list(ggplot2::labs(colour = "Version")),
-    legend_labels = list(colour = unique(df_data$version))
-  )
-}
diff --git a/man/detect_item_case.Rd b/man/detect_item_case.Rd
deleted file mode 100644
index cefa32c..0000000
--- a/man/detect_item_case.Rd
+++ /dev/null
@@ -1,36 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/aesthetics.R
-\name{detect_item_case}
-\alias{detect_item_case}
-\title{Detect items cases}
-\usage{
-detect_item_case(
-  type,
-  is_mixture,
-  one_version,
-  has_distinct_situations,
-  overlap
-)
-}
-\arguments{
-\item{type}{The type of plot required, either 'dynamic' or 'scatter}
-
-\item{is_mixture}{A logical value indicating whether the crop is a mixture or not.}
-
-\item{one_version}{A logical value indicating whether the plot has one or multiple versions (e.g. of the model).}
-
-\item{has_distinct_situations}{A logical value indicating whether there are one or several situations to plot.}
-
-\item{overlap}{A logical value indicating whether there is any overlapping variables in the plot.}
-}
-\value{
-A unique character string for the plot.
-}
-\description{
-This function returns a unique string based on the type of plot, and
-whether the situation is a mixture or not, if there is one or multiple
-versions to plot, and if there is one or several situations to plot
-into the same plot.
-The output is used to choose the right plotting function in a switch.
-}
-\keyword{internal}
diff --git a/man/specific_scatter_plots.Rd b/man/specific_scatter_plots.Rd
index 0cccab8..4873845 100644
--- a/man/specific_scatter_plots.Rd
+++ b/man/specific_scatter_plots.Rd
@@ -6,12 +6,8 @@
 \alias{compute_axis_bounds}
 \alias{make_axis_square}
 \alias{force_y_axis}
+\alias{scatter_plot_spec}
 \alias{build_scatter_plot}
-\alias{plot_scat_mixture_allsit}
-\alias{plot_scat_mixture_versions}
-\alias{plot_scat_allsit}
-\alias{plot_scat_versions_per_sit}
-\alias{plot_scat_versions_allsit}
 \title{Specific functions to generate scatter plots}
 \usage{
 get_facet_order(p, facet_var = "var")
@@ -22,70 +18,18 @@ make_axis_square(df_data, reference_var, y_var_type, is_obs_sd, p)
 
 force_y_axis(df_data, reference_var, y_var_type, is_obs_sd, p)
 
-build_scatter_plot(
-  df_data,
-  select_scat,
-  shape_sit,
-  reference_var,
-  is_obs_sd,
-  title = NULL,
-  mapping = ggplot2::aes(),
-  smooth_by_colour = FALSE,
-  extra = NULL,
-  legend_labels = list()
-)
+scatter_plot_spec(df_data, mixture, one_version)
 
-plot_scat_mixture_allsit(
-  df_data,
-  sit,
-  select_scat,
-  shape_sit,
-  reference_var,
-  is_obs_sd,
-  title = NULL
-)
-
-plot_scat_mixture_versions(
-  df_data,
-  sit,
-  select_scat,
-  shape_sit,
-  reference_var,
-  is_obs_sd,
-  title = NULL
-)
-
-plot_scat_allsit(
+build_scatter_plot(
   df_data,
-  sit,
   select_scat,
   shape_sit,
   reference_var,
   is_obs_sd,
   title = NULL,
-  has_distinct_situations = FALSE,
-  one_version = FALSE,
-  mixture = FALSE
-)
-
-plot_scat_versions_per_sit(
-  df_data,
-  sit,
-  select_scat,
-  shape_sit,
-  reference_var,
-  is_obs_sd,
-  title = NULL
-)
-
-plot_scat_versions_allsit(
-  df_data,
-  sit,
-  select_scat,
-  shape_sit,
-  reference_var,
-  is_obs_sd,
-  title = NULL
+  mixture = FALSE,
+  one_version = TRUE,
+  has_distinct_situations = FALSE
 )
 }
 \arguments{
@@ -101,26 +45,13 @@ per situation, or only one df if sit==all_situations)}
 
 \item{is_obs_sd}{TRUE if error standard deviation of observations is provided}
 
-\item{mapping}{Aesthetic mapping of the points (colour and/or shape). When
-\code{shape_sit} is "symbol" or "group", the situation is added on the shape, or
-on the colour if \code{mapping} has no colour (it replaces any existing shape).
-Error bars and text labels use the same colour as the points.}
-
-\item{smooth_by_colour}{If TRUE, one regression line per point colour,
-otherwise a single blue regression line.}
-
-\item{extra}{List of additional ggplot components (labs, scales...).}
-
-\item{legend_labels}{Named list (by aesthetic) of the labels displayed in the
-legend, passed to \code{add_facet_wrap}.}
+\item{mixture}{TRUE if the plot is for a mixture of crops}
 
-\item{sit}{The name of the situation to plot (or all_situations)}
+\item{one_version}{TRUE if the plot is for one version}
 
 \item{has_distinct_situations}{TRUE if the plot is for several situations}
 
-\item{one_version}{TRUE if the plot is for one version}
-
-\item{mixture}{TRUE if the plot is for a mixture of crops}
+\item{sit}{The name of the situation to plot (or all_situations)}
 }
 \value{
 A ggplot object
@@ -136,6 +67,11 @@ The modified ggplot
 
 The modified ggplot
 
+A list with \code{mapping} (aesthetic mapping of the points),
+\code{smooth_by_colour} (one regression line per point colour if TRUE, otherwise
+a single blue line), \code{extra} (additional ggplot components) and
+\code{legend_labels} (named list by aesthetic of the legend labels).
+
 A ggplot object
 }
 \description{
@@ -157,16 +93,21 @@ Make axis square
 Ensure that the Y axis includes zero when all values in a facet
 are strictly positive or strictly negative.
 
-Build a scatter plot shared by all the specific scatter plot
-functions.
+Give the plot specification (aesthetics, labels, legend) of a
+scatter plot according to the case: mixture or not, one or several versions.
+
+Build a scatter plot for all the cases handled in CroPlotR.
+When \code{shape_sit} is "symbol" or "group", the situation is added on the
+shape of the points, or on their colour if no colour is used (it replaces
+any existing shape). Error bars and text labels use the same colour as the
+points.
 }
 \details{
 List of the different specific functions:
 \itemize{
-\item \code{plot_scat_mixture_allsit}: Generate a scatter plot for the case of
-mixture of crops, single simulation version and all_situations in same plot
-\item \code{plot_scat_allsit}: Generate a scatter plot for the case of
-sole crops, single simulation version and all_situations in same plot
+\item \code{scatter_plot_spec}: Give the aesthetics, labels and legend of the
+scatter plot according to the case (mixture or not, one or several versions)
+\item \code{build_scatter_plot}: Generate a scatter plot for all the cases
 }
 
 The function relies on \code{ggplot2::ggplot_build()} to compute the plot

From 36285e38faecaf2aff10023ecab40fa7ad477f74 Mon Sep 17 00:00:00 2001
From: Valentine Rahier <valentine.rahier@inrae.fr>
Date: Wed, 7 Oct 2026 15:56:48 +0200
Subject: [PATCH 5/6] remove dead code

---
 R/aesthetics.R                           | 55 ------------------------
 man/detect_mixture_version_overlap.Rd    | 24 -----------
 man/detect_mixture_version_situations.Rd | 28 ------------
 3 files changed, 107 deletions(-)
 delete mode 100644 man/detect_mixture_version_overlap.Rd
 delete mode 100644 man/detect_mixture_version_situations.Rd

diff --git a/R/aesthetics.R b/R/aesthetics.R
index 8c30066..e5b6280 100644
--- a/R/aesthetics.R
+++ b/R/aesthetics.R
@@ -34,58 +34,3 @@ detect_mixture <- function(sim_situation) {
   return(is_mixture)
 }
 
-#' Detect items cases for dynamic plots
-#'
-#' This function detects the cases for computing the aesthetics of a plot based on
-#' whether it is a mixture or not, whether it has one or multiple versions, and
-#' whether there is any overlap.
-#'
-#' @param is_mixture A logical value indicating whether the crop is a mixture or not.
-#' @param one_version A logical value indicating whether the plot has one or multiple versions (e.g. of the model).
-#' @param overlap A logical value indicating whether there is any overlapping variables in the plot.
-#'
-#' @return A character string indicating the case for computing the aesthetics of the plot.
-#'
-#' @keywords internal
-detect_mixture_version_overlap <- function(is_mixture, one_version, overlap) {
-  case <- switch(paste(is_mixture, !one_version, !is.null(overlap)),
-    "TRUE TRUE TRUE" = "mixture_versions_overlap",
-    "TRUE TRUE FALSE" = "mixture_versions_no_overlap",
-    "TRUE FALSE TRUE" = "mixture_no_versions_overlap",
-    "TRUE FALSE FALSE" = "mixture_no_versions_no_overlap",
-    "FALSE TRUE TRUE" = "non_mixture_versions_overlap",
-    "FALSE TRUE FALSE" = "non_mixture_versions_no_overlap",
-    "FALSE FALSE TRUE" = "non_mixture_no_versions_overlap",
-    "FALSE FALSE FALSE" = "non_mixture_no_versions_no_overlap"
-  )
-
-  return(case)
-}
-
-#' Detect items cases for scatter plots
-#'
-#' This function detects the cases for computing the aesthetics of a plot based on
-#' whether it is a mixture or not, whether it has one or multiple versions, and
-#' whether there are one or several situations to plot into the same plot.
-#'
-#' @param is_mixture A logical value indicating whether the crop is a mixture or not.
-#' @param one_version A logical value indicating whether the plot has one or multiple versions (e.g. of the model).
-#' @param has_distinct_situations A logical value indicating whether there are one or several situations to plot.
-#'
-#' @return A character string indicating the case for computing the aesthetics of the plot.
-#'
-#' @keywords internal
-detect_mixture_version_situations <- function(is_mixture, one_version, has_distinct_situations) {
-  case <- switch(paste(is_mixture, !one_version, has_distinct_situations),
-    "TRUE TRUE TRUE" = "mixture_versions",
-    "TRUE TRUE FALSE" = "mixture_versions",
-    "TRUE FALSE TRUE" = "mixture_no_versions",
-    "TRUE FALSE FALSE" = "mixture_no_versions",
-    "FALSE TRUE TRUE" = "non_mixture_versions_situations",
-    "FALSE TRUE FALSE" = "non_mixture_versions_per_situations",
-    "FALSE FALSE TRUE" = "non_mixture_no_versions_situations",
-    "FALSE FALSE FALSE" = "non_mixture_no_versions_per_situations"
-  )
-
-  return(case)
-}
diff --git a/man/detect_mixture_version_overlap.Rd b/man/detect_mixture_version_overlap.Rd
deleted file mode 100644
index de6c8be..0000000
--- a/man/detect_mixture_version_overlap.Rd
+++ /dev/null
@@ -1,24 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/aesthetics.R
-\name{detect_mixture_version_overlap}
-\alias{detect_mixture_version_overlap}
-\title{Detect items cases for dynamic plots}
-\usage{
-detect_mixture_version_overlap(is_mixture, one_version, overlap)
-}
-\arguments{
-\item{is_mixture}{A logical value indicating whether the crop is a mixture or not.}
-
-\item{one_version}{A logical value indicating whether the plot has one or multiple versions (e.g. of the model).}
-
-\item{overlap}{A logical value indicating whether there is any overlapping variables in the plot.}
-}
-\value{
-A character string indicating the case for computing the aesthetics of the plot.
-}
-\description{
-This function detects the cases for computing the aesthetics of a plot based on
-whether it is a mixture or not, whether it has one or multiple versions, and
-whether there is any overlap.
-}
-\keyword{internal}
diff --git a/man/detect_mixture_version_situations.Rd b/man/detect_mixture_version_situations.Rd
deleted file mode 100644
index fe342d6..0000000
--- a/man/detect_mixture_version_situations.Rd
+++ /dev/null
@@ -1,28 +0,0 @@
-% Generated by roxygen2: do not edit by hand
-% Please edit documentation in R/aesthetics.R
-\name{detect_mixture_version_situations}
-\alias{detect_mixture_version_situations}
-\title{Detect items cases for scatter plots}
-\usage{
-detect_mixture_version_situations(
-  is_mixture,
-  one_version,
-  has_distinct_situations
-)
-}
-\arguments{
-\item{is_mixture}{A logical value indicating whether the crop is a mixture or not.}
-
-\item{one_version}{A logical value indicating whether the plot has one or multiple versions (e.g. of the model).}
-
-\item{has_distinct_situations}{A logical value indicating whether there are one or several situations to plot.}
-}
-\value{
-A character string indicating the case for computing the aesthetics of the plot.
-}
-\description{
-This function detects the cases for computing the aesthetics of a plot based on
-whether it is a mixture or not, whether it has one or multiple versions, and
-whether there are one or several situations to plot into the same plot.
-}
-\keyword{internal}

From 1935eb932cbbfe8ad8178e389742e6f1f63b82ec Mon Sep 17 00:00:00 2001
From: vrahier <vrahier@users.noreply.github.com>
Date: Wed, 7 Oct 2026 13:58:12 +0000
Subject: [PATCH 6/6] Style code (GHA)

---
 R/aesthetics.R | 1 -
 1 file changed, 1 deletion(-)

diff --git a/R/aesthetics.R b/R/aesthetics.R
index e5b6280..55e4ef0 100644
--- a/R/aesthetics.R
+++ b/R/aesthetics.R
@@ -33,4 +33,3 @@ detect_mixture <- function(sim_situation) {
 
   return(is_mixture)
 }
-