From 089f38818abc32bee8d90513e2a927f2097735bb Mon Sep 17 00:00:00 2001
From: Hugo Gruson <10783929+Bisaloo@users.noreply.github.com>
Date: Mon, 29 Apr 2024 20:26:28 +0200
Subject: [PATCH 01/16] Try to get more info on failure

---
 R/extract_param.R | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index 965cb0fd8..55a37727f 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -228,14 +228,16 @@ extract_param <- function(type = c("percentiles", "range"),
     lower <- c(-1e5, 1e-10)
   }
 
-  optim_params <- stats::optim(
+  optim_params <- tryCatch(stats::optim(
     param,
     fit_func,
     method = "L-BFGS-B",
     val = values_in,
     dist = distribution,
     lower = lower
-  )
+  ), error = function(e) {
+    stop("found inf value in ", distribution)
+  })
 
   optim_params
 }

From 1adcd95ec532d1feb83d6a3328909b651521a324 Mon Sep 17 00:00:00 2001
From: Hugo Gruson <10783929+Bisaloo@users.noreply.github.com>
Date: Mon, 29 Apr 2024 20:57:21 +0200
Subject: [PATCH 02/16] Check if Inf occurs on lower or upper

---
 R/extract_param.R | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/R/extract_param.R b/R/extract_param.R
index 55a37727f..16f9857af 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -214,18 +214,22 @@ extract_param <- function(type = c("percentiles", "range"),
   if (distribution == "lnorm") {
     names(param) <- c("meanlog", "sdlog")
     lower <- c(-1e5, 1e-10)
+    upper <- c(1e10, 1e10)
   }
   if (distribution == "gamma") {
     names(param) <- c("shape", "scale")
     lower <- c(1e-10, 1e-10)
+    upper <- c(1e10, 1e10)
   }
   if (distribution == "weibull") {
     names(param) <- c("shape", "scale")
     lower <- c(1e-10, 1e-10)
+    upper <- c(1e10, 1e10)
   }
   if (distribution == "norm") {
     names(param) <- c("mean", "sd")
     lower <- c(-1e5, 1e-10)
+    upper <- c(1e10, 1e10)
   }
 
   optim_params <- tryCatch(stats::optim(

From a9bae5c78be08e4a22a2ea3af2ee5e374cac0d03 Mon Sep 17 00:00:00 2001
From: Hugo Gruson <10783929+Bisaloo@users.noreply.github.com>
Date: Mon, 29 Apr 2024 21:02:58 +0200
Subject: [PATCH 03/16] Actually use upper

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

diff --git a/R/extract_param.R b/R/extract_param.R
index 16f9857af..54ca082d5 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -238,7 +238,8 @@ extract_param <- function(type = c("percentiles", "range"),
     method = "L-BFGS-B",
     val = values_in,
     dist = distribution,
-    lower = lower
+    lower = lower,
+    upper = upper
   ), error = function(e) {
     stop("found inf value in ", distribution)
   })

From 1c18b6e22f194f60b6b521781b54d15021528db5 Mon Sep 17 00:00:00 2001
From: Hugo Gruson <10783929+Bisaloo@users.noreply.github.com>
Date: Tue, 30 Apr 2024 09:38:52 +0200
Subject: [PATCH 04/16] Increase weibull lower bounds

---
 R/extract_param.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index 54ca082d5..293b1a1ed 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -223,7 +223,7 @@ extract_param <- function(type = c("percentiles", "range"),
   }
   if (distribution == "weibull") {
     names(param) <- c("shape", "scale")
-    lower <- c(1e-10, 1e-10)
+    lower <- c(1e-7, 1e-7)
     upper <- c(1e10, 1e10)
   }
   if (distribution == "norm") {

From b97f0c7de6cb494d7ef4ca5706e30993bd09301f Mon Sep 17 00:00:00 2001
From: Hugo Gruson <10783929+Bisaloo@users.noreply.github.com>
Date: Tue, 30 Apr 2024 09:53:02 +0200
Subject: [PATCH 05/16] Increase lower even more?

---
 R/extract_param.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index 293b1a1ed..b286680fb 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -223,7 +223,7 @@ extract_param <- function(type = c("percentiles", "range"),
   }
   if (distribution == "weibull") {
     names(param) <- c("shape", "scale")
-    lower <- c(1e-7, 1e-7)
+    lower <- c(1e-3, 1e-3)
     upper <- c(1e10, 1e10)
   }
   if (distribution == "norm") {

From 82df9739a06a8e04b6ffb98de0cf853cda90b952 Mon Sep 17 00:00:00 2001
From: Hugo Gruson <10783929+Bisaloo@users.noreply.github.com>
Date: Tue, 30 Apr 2024 10:23:36 +0200
Subject: [PATCH 06/16] Adjust bounds

---
 R/extract_param.R | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index b286680fb..a5d007a0b 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -214,22 +214,18 @@ extract_param <- function(type = c("percentiles", "range"),
   if (distribution == "lnorm") {
     names(param) <- c("meanlog", "sdlog")
     lower <- c(-1e5, 1e-10)
-    upper <- c(1e10, 1e10)
   }
   if (distribution == "gamma") {
     names(param) <- c("shape", "scale")
     lower <- c(1e-10, 1e-10)
-    upper <- c(1e10, 1e10)
   }
   if (distribution == "weibull") {
     names(param) <- c("shape", "scale")
-    lower <- c(1e-3, 1e-3)
-    upper <- c(1e10, 1e10)
+    lower <- c(1e-5, 1e-5)
   }
   if (distribution == "norm") {
     names(param) <- c("mean", "sd")
     lower <- c(-1e5, 1e-10)
-    upper <- c(1e10, 1e10)
   }
 
   optim_params <- tryCatch(stats::optim(
@@ -238,8 +234,7 @@ extract_param <- function(type = c("percentiles", "range"),
     method = "L-BFGS-B",
     val = values_in,
     dist = distribution,
-    lower = lower,
-    upper = upper
+    lower = lower
   ), error = function(e) {
     stop("found inf value in ", distribution)
   })

From 7c7798374ca6f1b0cbdd24f6a6aa2ec900ee71e0 Mon Sep 17 00:00:00 2001
From: Joshua Lambert <joshua.lambert@lshtm.ac.uk>
Date: Wed, 1 May 2024 14:06:36 +0100
Subject: [PATCH 07/16] remove tryCatch in extract_param

---
 R/extract_param.R | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index a5d007a0b..deb9f36a4 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -228,16 +228,14 @@ extract_param <- function(type = c("percentiles", "range"),
     lower <- c(-1e5, 1e-10)
   }
 
-  optim_params <- tryCatch(stats::optim(
+  optim_params <- stats::optim(
     param,
     fit_func,
     method = "L-BFGS-B",
     val = values_in,
     dist = distribution,
     lower = lower
-  ), error = function(e) {
-    stop("found inf value in ", distribution)
-  })
+  )
 
   optim_params
 }

From 4143bed614aa73d0047569ed59b6de93e5404c77 Mon Sep 17 00:00:00 2001
From: Joshua Lambert <joshua.lambert@lshtm.ac.uk>
Date: Wed, 1 May 2024 14:17:07 +0100
Subject: [PATCH 08/16] print parameter index in extract-bias vignette

---
 vignettes/extract-bias.Rmd | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/vignettes/extract-bias.Rmd b/vignettes/extract-bias.Rmd
index b89e7f4bf..67e7aca90 100644
--- a/vignettes/extract-bias.Rmd
+++ b/vignettes/extract-bias.Rmd
@@ -206,7 +206,7 @@ estim_params <- vector("list", nrow(parameters_range))
 for (params_idx in seq_len(nrow(parameters_range))) {
   dist <- as.character(parameters_range[params_idx, "dist"])
   n_samples <- parameters_range[params_idx, "n_samples"]
-
+  print(params_idx)
   if (dist == "lnorm") {
     true_median <- do.call(
       paste0("q", dist),

From 73fd8908d417e700b2a457aaf4898aae231e537c Mon Sep 17 00:00:00 2001
From: Joshua Lambert <joshua.lambert@lshtm.ac.uk>
Date: Wed, 1 May 2024 14:25:17 +0100
Subject: [PATCH 09/16] add stop call to extract_param

---
 R/extract_param.R | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/R/extract_param.R b/R/extract_param.R
index deb9f36a4..8fb0e5e8f 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -81,6 +81,11 @@ extract_param <- function(type = c("percentiles", "range"),
   if (distribution == "norm") {
     checkmate::assert_numeric(values)
   } else {
+
+    if (any(values <= 1e-10)) {
+      stop("Distribution is ", distribution, " Values are ", values, " Samples are ", samples)
+    }
+
     checkmate::assert_numeric(values, lower = 1e-10)
   }
 

From 9fe4c10d0198a6b382c4d7cf1dd9b2d16f35e530 Mon Sep 17 00:00:00 2001
From: Joshua Lambert <joshua.lambert@lshtm.ac.uk>
Date: Wed, 1 May 2024 14:29:52 +0100
Subject: [PATCH 10/16] remove print from vignette and change lower bound in
 extract_param value assert to 0

---
 R/extract_param.R          | 8 ++++----
 vignettes/extract-bias.Rmd | 1 -
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index 8fb0e5e8f..b598b6e0b 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -82,11 +82,11 @@ extract_param <- function(type = c("percentiles", "range"),
     checkmate::assert_numeric(values)
   } else {
 
-    if (any(values <= 1e-10)) {
-      stop("Distribution is ", distribution, " Values are ", values, " Samples are ", samples)
-    }
+    # if (any(values <= 1e-10)) {
+    #   stop("Distribution is ", distribution, " Values are ", values, " Samples are ", samples)
+    # }
 
-    checkmate::assert_numeric(values, lower = 1e-10)
+    checkmate::assert_numeric(values, lower = 0)
   }
 
   if (type == "percentiles") {
diff --git a/vignettes/extract-bias.Rmd b/vignettes/extract-bias.Rmd
index 67e7aca90..cbe422314 100644
--- a/vignettes/extract-bias.Rmd
+++ b/vignettes/extract-bias.Rmd
@@ -206,7 +206,6 @@ estim_params <- vector("list", nrow(parameters_range))
 for (params_idx in seq_len(nrow(parameters_range))) {
   dist <- as.character(parameters_range[params_idx, "dist"])
   n_samples <- parameters_range[params_idx, "n_samples"]
-  print(params_idx)
   if (dist == "lnorm") {
     true_median <- do.call(
       paste0("q", dist),

From 3bb0e0d1dba04a381ac43a89feea5f55f2a85ee4 Mon Sep 17 00:00:00 2001
From: Joshua Lambert <joshua.lambert@lshtm.ac.uk>
Date: Wed, 1 May 2024 14:37:20 +0100
Subject: [PATCH 11/16] add stop to extract_param to check values

---
 R/extract_param.R | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/R/extract_param.R b/R/extract_param.R
index b598b6e0b..3819d8479 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -100,6 +100,11 @@ extract_param <- function(type = c("percentiles", "range"),
     checkmate::assert_numeric(percentiles, lower = 0, upper = 1, len = 2)
   }
   if (identical(type, "range")) {
+
+    if (values[2] < values[1] && values[1] < values[3]) {
+      stop("Distribution is ", distribution, " Values are ", values, " Type is ", type)
+    }
+
     stopifnot(
       "samples need to be given for type = 'range'" =
         !missing(samples),

From 59be14280dd79774984b3c8bb70f083e0688e5d4 Mon Sep 17 00:00:00 2001
From: Joshua Lambert <joshua.lambert@lshtm.ac.uk>
Date: Wed, 1 May 2024 14:45:35 +0100
Subject: [PATCH 12/16] correct logical expr in if in extract_param

---
 R/extract_param.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index 3819d8479..d28c997f9 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -101,7 +101,7 @@ extract_param <- function(type = c("percentiles", "range"),
   }
   if (identical(type, "range")) {
 
-    if (values[2] < values[1] && values[1] < values[3]) {
+    if (!(values[2] < values[1] && values[1] < values[3])) {
       stop("Distribution is ", distribution, " Values are ", values, " Type is ", type)
     }
 

From 1c6345e231762f6db7420972112122dcb860eff0 Mon Sep 17 00:00:00 2001
From: Joshua Lambert <joshua.lambert@lshtm.ac.uk>
Date: Wed, 1 May 2024 15:29:00 +0100
Subject: [PATCH 13/16] added while loop to extract-bias vignette to prevent
 incoherent values

---
 R/extract_param.R          |  14 +---
 vignettes/extract-bias.Rmd | 155 ++++++++++++++++++++-----------------
 2 files changed, 87 insertions(+), 82 deletions(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index d28c997f9..965cb0fd8 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -81,12 +81,7 @@ extract_param <- function(type = c("percentiles", "range"),
   if (distribution == "norm") {
     checkmate::assert_numeric(values)
   } else {
-
-    # if (any(values <= 1e-10)) {
-    #   stop("Distribution is ", distribution, " Values are ", values, " Samples are ", samples)
-    # }
-
-    checkmate::assert_numeric(values, lower = 0)
+    checkmate::assert_numeric(values, lower = 1e-10)
   }
 
   if (type == "percentiles") {
@@ -100,11 +95,6 @@ extract_param <- function(type = c("percentiles", "range"),
     checkmate::assert_numeric(percentiles, lower = 0, upper = 1, len = 2)
   }
   if (identical(type, "range")) {
-
-    if (!(values[2] < values[1] && values[1] < values[3])) {
-      stop("Distribution is ", distribution, " Values are ", values, " Type is ", type)
-    }
-
     stopifnot(
       "samples need to be given for type = 'range'" =
         !missing(samples),
@@ -231,7 +221,7 @@ extract_param <- function(type = c("percentiles", "range"),
   }
   if (distribution == "weibull") {
     names(param) <- c("shape", "scale")
-    lower <- c(1e-5, 1e-5)
+    lower <- c(1e-10, 1e-10)
   }
   if (distribution == "norm") {
     names(param) <- c("mean", "sd")
diff --git a/vignettes/extract-bias.Rmd b/vignettes/extract-bias.Rmd
index cbe422314..2174c333f 100644
--- a/vignettes/extract-bias.Rmd
+++ b/vignettes/extract-bias.Rmd
@@ -206,46 +206,54 @@ estim_params <- vector("list", nrow(parameters_range))
 for (params_idx in seq_len(nrow(parameters_range))) {
   dist <- as.character(parameters_range[params_idx, "dist"])
   n_samples <- parameters_range[params_idx, "n_samples"]
-  if (dist == "lnorm") {
-    true_median <- do.call(
-      paste0("q", dist),
-      list(
-        p = 0.5,
-        meanlog = parameters_range[params_idx, "param_1"],
-        sdlog = parameters_range[params_idx, "param_2"]
+  
+  # while loop to ensure values are min < median < max
+  resample_values <- TRUE
+  while (resample_values) {
+    if (dist == "lnorm") {
+      true_median <- do.call(
+        paste0("q", dist),
+        list(
+          p = 0.5,
+          meanlog = parameters_range[params_idx, "param_1"],
+          sdlog = parameters_range[params_idx, "param_2"]
+        )
       )
-    )
-    true_range <- do.call(
-      paste0("r", dist),
-      list(
-        n = n_samples,
-        meanlog = parameters_range[params_idx, "param_1"],
-        sdlog = parameters_range[params_idx, "param_2"]
+      true_range <- do.call(
+        paste0("r", dist),
+        list(
+          n = n_samples,
+          meanlog = parameters_range[params_idx, "param_1"],
+          sdlog = parameters_range[params_idx, "param_2"]
+        )
       )
-    )
-    true_range <- c(min(true_range), max(true_range))
-  } else {
-    true_median <- do.call(
-      paste0("q", dist),
-      list(
-        p = 0.5,
-        shape = parameters_range[params_idx, "param_1"],
-        scale = parameters_range[params_idx, "param_2"]
+      true_range <- c(min(true_range), max(true_range))
+    } else {
+      true_median <- do.call(
+        paste0("q", dist),
+        list(
+          p = 0.5,
+          shape = parameters_range[params_idx, "param_1"],
+          scale = parameters_range[params_idx, "param_2"]
+        )
       )
-    )
-    true_range <- do.call(
-      paste0("r", dist),
-      list(
-        n = n_samples,
-        shape = parameters_range[params_idx, "param_1"],
-        scale = parameters_range[params_idx, "param_2"]
+      true_range <- do.call(
+        paste0("r", dist),
+        list(
+          n = n_samples,
+          shape = parameters_range[params_idx, "param_1"],
+          scale = parameters_range[params_idx, "param_2"]
+        )
       )
-    )
-    true_range <- c(min(true_range), max(true_range))
+      true_range <- c(min(true_range), max(true_range))
+    }
+    
+    true_values <- c(true_median, true_range)
+    if (true_values[2] < true_values[1] && true_values[1] < true_values[3]) {
+      resample_values <- FALSE
+    }
   }
-
-  true_values <- c(true_median, true_range)
-
+  
   # message about stochastic optimisation suppressed
   estim_params[[params_idx]] <- suppressMessages(
     expr = extract_param(
@@ -392,46 +400,53 @@ for (params_idx in seq_len(nrow(parameters_range))) {
   dist <- as.character(parameters_range[params_idx, "dist"])
   n_samples <- parameters_range[params_idx, "n_samples"]
 
-  if (dist == "lnorm") {
-    true_median <- do.call(
-      paste0("q", dist),
-      list(
-        p = 0.5,
-        meanlog = parameters_range[params_idx, "param_1"],
-        sdlog = parameters_range[params_idx, "param_2"]
+  # while loop to ensure values are min < median < max
+  resample_values <- TRUE
+  while (resample_values) {
+    if (dist == "lnorm") {
+      true_median <- do.call(
+        paste0("q", dist),
+        list(
+          p = 0.5,
+          meanlog = parameters_range[params_idx, "param_1"],
+          sdlog = parameters_range[params_idx, "param_2"]
+        )
       )
-    )
-    true_range <- do.call(
-      paste0("r", dist),
-      list(
-        n = n_samples,
-        meanlog = parameters_range[params_idx, "param_1"],
-        sdlog = parameters_range[params_idx, "param_2"]
+      true_range <- do.call(
+        paste0("r", dist),
+        list(
+          n = n_samples,
+          meanlog = parameters_range[params_idx, "param_1"],
+          sdlog = parameters_range[params_idx, "param_2"]
+        )
       )
-    )
-    true_range <- c(min(true_range), max(true_range))
-  } else {
-    true_median <- do.call(
-      paste0("q", dist),
-      list(
-        p = 0.5,
-        shape = parameters_range[params_idx, "param_1"],
-        scale = parameters_range[params_idx, "param_2"]
+      true_range <- c(min(true_range), max(true_range))
+    } else {
+      true_median <- do.call(
+        paste0("q", dist),
+        list(
+          p = 0.5,
+          shape = parameters_range[params_idx, "param_1"],
+          scale = parameters_range[params_idx, "param_2"]
+        )
       )
-    )
-    true_range <- do.call(
-      paste0("r", dist),
-      list(
-        n = n_samples,
-        shape = parameters_range[params_idx, "param_1"],
-        scale = parameters_range[params_idx, "param_2"]
+      true_range <- do.call(
+        paste0("r", dist),
+        list(
+          n = n_samples,
+          shape = parameters_range[params_idx, "param_1"],
+          scale = parameters_range[params_idx, "param_2"]
+        )
       )
-    )
-    true_range <- c(min(true_range), max(true_range))
+      true_range <- c(min(true_range), max(true_range))
+    }
+    
+    true_values <- c(true_median, true_range)
+    if (true_values[2] < true_values[1] && true_values[1] < true_values[3]) {
+      resample_values <- FALSE
+    }
   }
-
-  true_values <- c(true_median, true_range)
-
+  
   # message about stochastic optimisation suppressed
   estim <- suppressMessages(
     replicate(

From 8983aab0902d73fe569391f23efc9e2ca7cc2045 Mon Sep 17 00:00:00 2001
From: Joshua Lambert <joshua.lambert@lshtm.ac.uk>
Date: Wed, 1 May 2024 15:33:00 +0100
Subject: [PATCH 14/16] increase weibull lower bound

---
 R/extract_param.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index 965cb0fd8..deb9f36a4 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -221,7 +221,7 @@ extract_param <- function(type = c("percentiles", "range"),
   }
   if (distribution == "weibull") {
     names(param) <- c("shape", "scale")
-    lower <- c(1e-10, 1e-10)
+    lower <- c(1e-5, 1e-5)
   }
   if (distribution == "norm") {
     names(param) <- c("mean", "sd")

From cbd2ec80ecfd7ef2773c36374c4dc383c5f30ce1 Mon Sep 17 00:00:00 2001
From: Joshua Lambert <joshua.lambert@lshtm.ac.uk>
Date: Wed, 1 May 2024 15:37:35 +0100
Subject: [PATCH 15/16] revert to lower value bound at 0

---
 R/extract_param.R | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/R/extract_param.R b/R/extract_param.R
index deb9f36a4..d8a359600 100644
--- a/R/extract_param.R
+++ b/R/extract_param.R
@@ -81,7 +81,7 @@ extract_param <- function(type = c("percentiles", "range"),
   if (distribution == "norm") {
     checkmate::assert_numeric(values)
   } else {
-    checkmate::assert_numeric(values, lower = 1e-10)
+    checkmate::assert_numeric(values, lower = 0)
   }
 
   if (type == "percentiles") {

From d13640d785d28149bd900f5036323ab80b07b738 Mon Sep 17 00:00:00 2001
From: Hugo Gruson <10783929+Bisaloo@users.noreply.github.com>
Date: Wed, 1 May 2024 17:14:10 +0200
Subject: [PATCH 16/16] Remove fail fast So we can see what happens in other
 runners

---
 .github/workflows/R-CMD-check.yaml | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/.github/workflows/R-CMD-check.yaml b/.github/workflows/R-CMD-check.yaml
index 62b0eff0b..251a9d100 100644
--- a/.github/workflows/R-CMD-check.yaml
+++ b/.github/workflows/R-CMD-check.yaml
@@ -85,10 +85,3 @@ jobs:
         with:
           upload-snapshots: true
           error-on: '"note"'
-
-      # fail-fast but only if rcmdcheck step fails
-      - name: Manual fail-fast
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        if: always() && steps.rcmdcheck.outcome == 'failure'
-        run: gh run cancel ${{ github.run_id }}