From c805c0c43eb4f2be057b71b74ac46b177bbc1759 Mon Sep 17 00:00:00 2001
From: RichardPatterson <richard-patterson@hotmail.co.uk>
Date: Mon, 4 Mar 2024 17:18:06 +0000
Subject: [PATCH 1/5] Remove dplyr Group lines using igraph not looping

---
 .Rbuildignore          |  2 ++
 .gitignore             |  2 ++
 DESCRIPTION            |  6 ++--
 R/midlines_functions.R |  6 ++--
 R/midlines_polish.R    | 15 ++++----
 R/midlines_refine.R    | 81 +++++++++++++-----------------------------
 vignettes/midlines.Rmd |  7 ++--
 7 files changed, 48 insertions(+), 71 deletions(-)

diff --git a/.Rbuildignore b/.Rbuildignore
index acc4601..c8ea19e 100644
--- a/.Rbuildignore
+++ b/.Rbuildignore
@@ -6,3 +6,5 @@
 ^working/examples\.R$
 ^README\.Rmd$
 ^data-raw$
+^doc$
+^Meta$
diff --git a/.gitignore b/.gitignore
index e67f93b..c48f0ba 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,4 @@
 .Rproj.user
 inst/doc
+/doc/
+/Meta/
diff --git a/DESCRIPTION b/DESCRIPTION
index f273c54..1f1ea8e 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -12,13 +12,13 @@ License: MIT + file LICENSE
 Encoding: UTF-8
 LazyData: true
 Roxygen: list(markdown = TRUE)
-RoxygenNote: 7.1.1
+RoxygenNote: 7.2.3
 Imports: 
     sf,
-    dplyr,
     zoo,
     lwgeom,
-    stats
+    stats,
+    igraph
 Depends: 
     R (>= 2.10)
 Suggests: 
diff --git a/R/midlines_functions.R b/R/midlines_functions.R
index a77de8c..04930d2 100644
--- a/R/midlines_functions.R
+++ b/R/midlines_functions.R
@@ -168,7 +168,7 @@ midlines_clean = function(x, n_removed = 1, border_line = NULL){
 
   }#for loop
 
-  removed_mid_points = dplyr::bind_rows(removed_mid_points)
+  removed_mid_points = do.call("rbind", removed_mid_points)
 
   removed_mid_points$removed_flag = factor(1)
   trimmed_mid_points$removed_flag = factor(0)
@@ -179,7 +179,7 @@ midlines_clean = function(x, n_removed = 1, border_line = NULL){
                        sf::st_drop_geometry(trimmed_points),
                        FUN = unique)
 
-  return = dplyr::inner_join(x, rem_flag, by = "line_id")
+  return = merge(x, rem_flag, by = "line_id")
 
   # Remove line_id var if it wasn't present in input (x)
   if(!line_id_present){
@@ -190,7 +190,7 @@ midlines_clean = function(x, n_removed = 1, border_line = NULL){
 }
 
 
-group_id = line_id = geometry = NULL
+removed_flag = group_id = line_id = geometry = NULL
 
 
 
diff --git a/R/midlines_polish.R b/R/midlines_polish.R
index ecc6a68..02f1393 100644
--- a/R/midlines_polish.R
+++ b/R/midlines_polish.R
@@ -42,14 +42,16 @@ midlines_debit = function(x, length) {
 midlines_smooth = function(x, width = 3){
 
   dat = midlines_group(x)
-  dat = dplyr::select(dat, geometry)  # stop warning about repeating attributes
-  dat = sf::st_cast(dat,"LINESTRING")
+  dat = subset(dat, select =  geometry)  # stop warning about repeating attributes
+  #dat = sf::st_cast(dat,"LINESTRING")
+  dat = sf::st_sf(geometry = sf::st_cast(sf::st_line_merge(sf::st_union(x)), "LINESTRING"))
+
 
   s = function(x){
     l = length(dat$geometry[[x]])
 
-    a = zoo::rollapply(dat$geometry[[x]][1:(l/2)], width = width, mean )
-    b = zoo::rollapply(dat$geometry[[x]][(l/2+1):l], width = width, mean )
+    a = zoo::rollmean(dat$geometry[[x]][1:(l/2)], k = width)
+    b = zoo::rollmean(dat$geometry[[x]][(l/2+1):l], k = width)
 
     sx = dat$geometry[[x]][1]
     ex = dat$geometry[[x]][(l/2)]
@@ -106,8 +108,9 @@ midlines_smooth = function(x, width = 3){
 midlines_dedensify = function(x, density){
 
   x = midlines_group(x)
-  x = dplyr::select(x, geometry)
-  ls = sf::st_cast(x,"LINESTRING")
+  x = subset(x, select = geometry)
+  ls = sf::st_sf(geometry = sf::st_cast(sf::st_line_merge(sf::st_union(x)), "LINESTRING"))
+
   ls$line_id = 1:nrow(ls)
 
   de_densified = sf::st_as_sf(sf::st_line_sample(ls, density = density))
diff --git a/R/midlines_refine.R b/R/midlines_refine.R
index c0c5cd8..aad8f05 100644
--- a/R/midlines_refine.R
+++ b/R/midlines_refine.R
@@ -1,4 +1,5 @@
 # An internal function
+# Largely replicates the answer here: https://stackoverflow.com/questions/69175360/is-there-a-way-to-merge-contiguous-multilinestrings-sf-object-in-r
 
 #' Groups line segments into contiguous groups, returns these as multilinestrings.
 #'
@@ -7,50 +8,31 @@
 #' @export
 midlines_group = function(x) {
 
-  lines = sf::st_as_sf(sf::st_cast(sf::st_line_merge(sf::st_union(x)), "LINESTRING"))
+  touches = sf::st_touches(x)
+  graph = igraph::graph_from_adj_list(touches)
 
-  colnames(lines)[colnames(lines) == colnames(lines)] = "geometry" #this only works cos there is one column
-  sf::st_geometry(lines) <- "geometry"
+  groups <- igraph::components(graph)$membership
 
-  lines$group_id = NA
-  lines$n_ = 1:nrow(lines)
+  grouped = stats::aggregate(x, by = list(group_id = groups), FUN = unique)
 
-  inter = sf::st_intersects(lines$geometry, lines$geometry)
-  group_index = 1
+  grouped$n_lines = lengths(grouped$line_id)
+  grouped$length = sf::st_length(grouped)
 
-  for(l in 1:nrow(lines)) {
+  #grouped = subset(grouped, select = -c(removed_flag))
+  grouped = grouped[, names(grouped) != "removed_flag"]
 
-    lines$group_id[lines$n_ %in%
-                     inter[[lines$n_[l]]]] = group_index
-
-    lines = lines[order(lines$group_id, na.last = TRUE),]
-
-    if(anyNA(lines$group_id[l+1])) group_index = group_index +1
-    #print(group_index)
-    l = l+1
-  }
-
-  multilines = sf::st_cast(
-    dplyr::summarise(
-      dplyr::group_by(lines, group_id),
-      do_union = FALSE)
-    ,"MULTILINESTRING")
-
-  multilines$n_lines =lengths(lapply(multilines$geometry, unlist))/4
-  multilines$length = sf::st_length(multilines)
-
-  #return(list(multilines,lines))
-  #}
-  grouped_multilinestring = multilines
-  ##############
-
-
-  # this reokaces the variable from the group_lines_a as applying that to that grouped line does something odd.
-  grouped_multilinestring$n_lines = (lengths(lapply(grouped_multilinestring$geometry, unlist)) - 2)/ 2
-
-  return(grouped_multilinestring)
+  return(grouped)
 }
 
+# This seems to be okay. It returns a df rather than a tibble, but that's fine because that's what it it passed.
+# d = midlines_group(removed)
+# e = midlines::midlines_group(removed)
+# d
+# e
+# class(d)
+# class(e)
+# all.equal(d,e)
+# class(removed)
 
 
 
@@ -107,10 +89,8 @@ midlines_check = function(x, n_removed = NULL, length = NULL, border_line = NULL
   removed = x[x$removed_flag==1,]
   cleaned = x[x$removed_flag==0,]
 
-  #moving this within this cleaning function
   x_multilines = midlines_group(removed)
 
-
   # using n_lines as the number of cycles of removing
   if(!(is.null(n_removed))){
     add_back_groups1 = x_multilines$group_id[x_multilines$n_lines >= n_removed]
@@ -127,7 +107,7 @@ midlines_check = function(x, n_removed = NULL, length = NULL, border_line = NULL
 
   #the first line finds lines touching border and then those removed groups that hit these
   if(!(is.null(border_line))) {
-    #touch_border = cleaned[sf::st_is_within_distance(cleaned, border_line, dist = border_distance, sparse = FALSE),]
+
     touch_border = removed[sf::st_intersects(removed, border_line, sparse = FALSE),]
     add_back_groups3 = x_multilines$group_id[sf::st_intersects(x_multilines, sf::st_union(touch_border), sparse = FALSE)]
   } else {
@@ -136,25 +116,14 @@ midlines_check = function(x, n_removed = NULL, length = NULL, border_line = NULL
 
   add_back_groups = unique(c(add_back_groups1, add_back_groups2, add_back_groups3))
 
-  add_back = x_multilines[x_multilines$group_id %in% add_back_groups,"geometry"]
-
-  # identify the lines to add back from the multilines
-  add_back_index = lengths(sf::st_covered_by(removed, add_back)) != 0
+  add_back_line_ids = unlist(x_multilines$line_id[x_multilines$group_id %in% add_back_groups])
 
-  add_back_lines = removed[add_back_index,]
-  still_removed = removed[!add_back_index,]
+  x$removed_flag2 = x$removed_flag
+  x$removed_flag2[x$line_id %in% add_back_line_ids] = 0
 
-  cleaned$added_flag = factor(0)
-  still_removed$added_flag = factor(0)
-  add_back_lines$added_flag = factor(1)
-
-  cleaned$removed_flag2 = factor(0)
-  add_back_lines$removed_flag2 = factor(0)
-  still_removed$removed_flag2 = factor(1)
-
-  rbind(cleaned, add_back_lines, still_removed)
+  x$added_flag = factor(as.integer(x$removed_flag != x$removed_flag2)) # should probably remove this as its unnecessary
 
+  return(x)
 }
 
 
-
diff --git a/vignettes/midlines.Rmd b/vignettes/midlines.Rmd
index 2054207..27daf43 100644
--- a/vignettes/midlines.Rmd
+++ b/vignettes/midlines.Rmd
@@ -4,7 +4,7 @@ output: rmarkdown::html_vignette
 vignette: >
   %\VignetteIndexEntry{midlines}
   %\VignetteEngine{knitr::rmarkdown}
-  %\VignetteEncoding{UTF-8}
+  %\VignetteEncoding{UTF-8}{inputenc}
 ---
 
 ```{r, include = FALSE}
@@ -25,7 +25,7 @@ The goal of `midlines` is to estimate the midline of one or more polygons, takin
 
 ## Installation
 
-You can install the package from [GitHub](https://github.com/) with:
+You can install the current version of the package from [GitHub](https://github.com/) with:
 
 ``` r
 # install.packages("devtools")
@@ -167,7 +167,7 @@ The lack of lines is due to the narrowness of the channel, or more precisely, th
 plot(side1)
 plot(bbox_line_s, add = TRUE)
 
-m_s1 = midlines_draw(side1, dfMaxLength = units::set_units(8,"m"), border_line = bbox_line_s)
+m_s1 = midlines_draw(side1, dfMaxLength = set_units(8,"m"), border_line = bbox_line_s)
 #plot(ml2$geometry, add = TRUE, col = "BLUE")
 
 # using the border_line to prevent removal of lines of interest
@@ -301,3 +301,4 @@ plot(pols)
 
 * The very comprehensive R package [`cmgo`](https://github.com/AntoniusGolly/cmgo) does many of the same things as this package and much more. If your aim to to estimate the midline of a river, you should probably start with [`cmgo`](https://github.com/AntoniusGolly/cmgo)
 * [centerline](https://centerline.readthedocs.io/en/latest/) is a Python library with a similar aim
+

From a261db7752c9d4d453bf1e4f2682684220f92a6d Mon Sep 17 00:00:00 2001
From: RichardPatterson <richard-patterson@hotmail.co.uk>
Date: Thu, 7 Mar 2024 17:28:31 +0000
Subject: [PATCH 2/5] Update readme plots Also a small simplification of
 midlines_smooth

---
 R/midlines_polish.R                  |  25 ++++++---------
 README.Rmd                           |  24 +++++++-------
 README.md                            |  45 ++++++++++++++-------------
 man/figures/README-clean1-1.png      | Bin 3424 -> 3413 bytes
 man/figures/README-device_off1-1.png | Bin 5986 -> 5978 bytes
 man/figures/README-draw-1.png        | Bin 5986 -> 5978 bytes
 man/figures/README-gaps3-1.png       | Bin 4298 -> 4315 bytes
 man/figures/README-smooth-1.png      | Bin 3535 -> 3526 bytes
 8 files changed, 47 insertions(+), 47 deletions(-)

diff --git a/R/midlines_polish.R b/R/midlines_polish.R
index 02f1393..91d5e81 100644
--- a/R/midlines_polish.R
+++ b/R/midlines_polish.R
@@ -62,25 +62,20 @@ midlines_smooth = function(x, width = 3){
 
   }
 
-  nrow = nrow(dat)
+  smoothed = sf::st_sfc(lapply(seq_len(nrow(dat)), s))
 
-  smoothed = sf::st_as_sf(sf::st_as_sfc(lapply(1:nrow, s)))
+  smoothed = sf::st_sf(geometry =
+                          sf::st_collection_extract(
+                            lwgeom::st_split(smoothed,
+                                             sf::st_union(
+                                               sf::st_cast(smoothed, "MULTIPOINT"))), type = "LINESTRING"),
+                        crs = sf::st_crs(x))
 
-  smoothed = sf::st_as_sf(
-    sf::st_collection_extract(
-      lwgeom::st_split(smoothed$x,
-               sf::st_union(
-                 sf::st_cast(smoothed, "MULTIPOINT"))), type = "LINESTRING"))
-
-  colnames(smoothed)[colnames(smoothed) == "x"] = "geometry"
-  sf::st_geometry(smoothed) <- "geometry"
-
-  smoothed$line_id = 1:nrow(smoothed)
-
-  sf::st_crs(smoothed) = sf::st_crs(dat)
+  smoothed$line_id = seq_len(nrow(smoothed))
 
   return(smoothed)
-}
+
+  }
 
 
 
diff --git a/README.Rmd b/README.Rmd
index 7ebddd6..6b5d78d 100644
--- a/README.Rmd
+++ b/README.Rmd
@@ -50,12 +50,14 @@ library(midlines)
 library(sf)
 library(units)
 
-plot(thames)
+plot(thames$geometry)
+
 ```
 
 To estimate the midline of this stretch of the Thames, use the `midline_draw` function.
 
 ```{r draw}
+
 # create a sf collection for the midline
 m1 = midlines_draw(thames)
 
@@ -115,7 +117,7 @@ bbox_line = st_cast(st_as_sfc(st_bbox(c(xmin = 535070, ymin = 177800, xmax = 542
 
 m1 = midlines_draw(thames, border_line = bbox_line)
 
-plot(thames)
+plot(thames$geometry)
 plot(m1$geometry, col = "BLUE", add = TRUE)
 plot(bbox_line, add = TRUE)
 ```
@@ -142,7 +144,7 @@ bbox_poly = st_as_sfc(st_bbox(c(xmin = 536070, ymin = 180700, xmax = 537800, yma
 # a slightly smaller area to use as a border line
 bbox_line_s = st_cast(st_as_sfc(st_bbox(c(xmin = 536120, ymin = 180760, xmax = 537750, ymax = 181800), crs = st_crs(thames))),"LINESTRING")
 
-plot(thames)
+plot(thames$geometry)
 plot(bbox_poly, add = TRUE)
 ```
 
@@ -151,7 +153,7 @@ plot(bbox_poly, add = TRUE)
 # crop the larger thames polygon to the area we want
 side1 = st_intersection(thames, bbox_poly)
 
-plot(side1)
+plot(side1$geometry)
 plot(bbox_line_s, add = TRUE)
 
 # estimating midline of the original polygon
@@ -163,7 +165,7 @@ The lack of lines is due to the narrowness of the channel, or more precisely, th
 
 ```{r gaps3}
 # some trial and error revealed 8 meters to be the optimal max length between points
-plot(side1)
+plot(side1$geometry)
 plot(bbox_line_s, add = TRUE)
 
 m_s1 = midlines_draw(side1, dfMaxLength = set_units(8,"m"), border_line = bbox_line_s)
@@ -174,7 +176,7 @@ m_s2 = midlines_clean(m_s1, n_removed = 20, border_line = bbox_line_s)
 plot(m_s2$geometry, col = c("BLUE", "RED")[m_s2$removed_flag], add = TRUE)
 ```
 
-Conversely, if the points on the perimeter are too dense and result in a very complicated tessellation or slow computations times then sf::st_line_sample might also be useful.  
+Conversely, if the points on the perimeter are too dense and result in a very complicated tessellation or slow computation times, then sf::st_line_sample might also be useful.  
 
 ## Zig-zagging
 
@@ -227,12 +229,12 @@ bbox_poly = st_as_sfc(st_bbox(c(xmin = 538700, ymin = 180750, xmax = 539800, yma
 # a slightly smaller area to use as a border line
 bbox_line_s = st_cast(st_as_sfc(st_bbox(c(xmin = 538750, ymin = 180800, xmax = 539750, ymax = 181800), crs = st_crs(thames))),"LINESTRING")
 
-plot(thames)
+plot(thames$geometry)
 plot(bbox_poly, add = TRUE)
 
 # crop the larger thames polygon to the area we want
 side2 = st_intersection(thames,bbox_poly)
-plot(side2)
+plot(side2$geometry)
 
 
 plot(bbox_line_s, add = TRUE)
@@ -245,7 +247,7 @@ plot(ml1$geometry, add = TRUE, col = "RED")
 One option would be to use the dfMaxLength option with midlines_draw like we did on the left side channel. 
 
 ```{r smooth}
-plot(side2)
+plot(side2$geometry)
 m_s1a = midlines_draw(side2, border_line = bbox_line_s, dfMaxLength = set_units(8,"m"))
 
 # clean this to remove extraneous lines
@@ -260,7 +262,7 @@ An alternative is to smooth the midline using a rolling average - basically a wr
 # smooth with a rolling average 
 m_s2b = midlines_smooth(m_s1)
 
-plot(side2)
+plot(side2$geometry)
 plot(m_s2b$geometry, add = TRUE, col = "BLUE")
 ```
 
@@ -283,7 +285,7 @@ plot(m2$geometry)
 
 ### De-densifying midlines
 
-In addition to manipulating the density of points on the perimeter of the initial polygon, it is sometimes desirable to manipulate the density of points the form the midline. This results in fewer longer line segments, rather than many shorter line segments. This can be done using `midlines_dedensify` either before or after cleaning. Like with the `midlines_smooth` function, the points where lines join are not affected but between line joins the number of points can be reduced. This function calls sf::st_line_sample so see this for the relevant options.
+In addition to manipulating the density of points on the perimeter of the initial polygon, it is sometimes desirable to manipulate the density of points that form the midline. This results in fewer longer line segments, rather than many shorter line segments. This can be done using `midlines_dedensify` either before or after cleaning. Like with the `midlines_smooth` function, the points where lines join are not affected but between line joins the number of points can be reduced. This function calls sf::st_line_sample so see this for the relevant options.
 
 ## Voronoi polygons
 
diff --git a/README.md b/README.md
index 5827f35..ec584c0 100644
--- a/README.md
+++ b/README.md
@@ -42,9 +42,11 @@ copyright is retained by OpenStreetMap contributors.
 # Load libraries
 library(midlines)
 library(sf)
+#> Warning: package 'sf' was built under R version 4.3.3
 library(units)
+#> Warning: package 'units' was built under R version 4.3.3
 
-plot(thames)
+plot(thames$geometry)
 ```
 
 <img src="man/figures/README-thames-1.png" width="100%" />
@@ -53,6 +55,7 @@ To estimate the midline of this stretch of the Thames, use the
 `midline_draw` function.
 
 ``` r
+
 # create a sf collection for the midline
 m1 = midlines_draw(thames)
 
@@ -160,10 +163,10 @@ linestring, as in this example.
 bbox_line = st_cast(st_as_sfc(st_bbox(c(xmin = 535070, ymin = 177800, xmax = 542560, ymax = 181550), crs = st_crs(thames))), "LINESTRING")
 
 m1 = midlines_draw(thames, border_line = bbox_line)
-#> Warning: attribute variables are assumed to be spatially constant throughout all
-#> geometries
+#> Warning: attribute variables are assumed to be spatially constant throughout
+#> all geometries
 
-plot(thames)
+plot(thames$geometry)
 plot(m1$geometry, col = "BLUE", add = TRUE)
 plot(bbox_line, add = TRUE)
 ```
@@ -204,7 +207,7 @@ bbox_poly = st_as_sfc(st_bbox(c(xmin = 536070, ymin = 180700, xmax = 537800, yma
 # a slightly smaller area to use as a border line
 bbox_line_s = st_cast(st_as_sfc(st_bbox(c(xmin = 536120, ymin = 180760, xmax = 537750, ymax = 181800), crs = st_crs(thames))),"LINESTRING")
 
-plot(thames)
+plot(thames$geometry)
 plot(bbox_poly, add = TRUE)
 ```
 
@@ -214,13 +217,13 @@ plot(bbox_poly, add = TRUE)
 # crop the larger thames polygon to the area we want
 side1 = st_intersection(thames, bbox_poly)
 
-plot(side1)
+plot(side1$geometry)
 plot(bbox_line_s, add = TRUE)
 
 # estimating midline of the original polygon
 m_s1 = midlines_draw(side1, border_line = bbox_line_s)
-#> Warning: attribute variables are assumed to be spatially constant throughout all
-#> geometries
+#> Warning: attribute variables are assumed to be spatially constant throughout
+#> all geometries
 plot(m_s1$geometry, add = TRUE, col = "RED")
 ```
 
@@ -237,12 +240,12 @@ sf::st_segmentize so see that help file for more details.
 
 ``` r
 # some trial and error revealed 8 meters to be the optimal max length between points
-plot(side1)
+plot(side1$geometry)
 plot(bbox_line_s, add = TRUE)
 
 m_s1 = midlines_draw(side1, dfMaxLength = set_units(8,"m"), border_line = bbox_line_s)
-#> Warning: attribute variables are assumed to be spatially constant throughout all
-#> geometries
+#> Warning: attribute variables are assumed to be spatially constant throughout
+#> all geometries
 #plot(ml2$geometry, add = TRUE, col = "BLUE")
 
 # using the border_line to prevent removal of lines of interest
@@ -253,7 +256,7 @@ plot(m_s2$geometry, col = c("BLUE", "RED")[m_s2$removed_flag], add = TRUE)
 <img src="man/figures/README-gaps3-1.png" width="100%" />
 
 Conversely, if the points on the perimeter are too dense and result in a
-very complicated tessellation or slow computations times then
+very complicated tessellation or slow computation times, then
 sf::st_line_sample might also be useful.
 
 ## Zig-zagging
@@ -321,7 +324,7 @@ bbox_poly = st_as_sfc(st_bbox(c(xmin = 538700, ymin = 180750, xmax = 539800, yma
 # a slightly smaller area to use as a border line
 bbox_line_s = st_cast(st_as_sfc(st_bbox(c(xmin = 538750, ymin = 180800, xmax = 539750, ymax = 181800), crs = st_crs(thames))),"LINESTRING")
 
-plot(thames)
+plot(thames$geometry)
 plot(bbox_poly, add = TRUE)
 ```
 
@@ -331,15 +334,15 @@ plot(bbox_poly, add = TRUE)
 
 # crop the larger thames polygon to the area we want
 side2 = st_intersection(thames,bbox_poly)
-plot(side2)
+plot(side2$geometry)
 
 
 plot(bbox_line_s, add = TRUE)
 
 # estimating midline of the original polygon
 m_s1 = midlines_draw(side2, border_line = bbox_line_s)
-#> Warning: attribute variables are assumed to be spatially constant throughout all
-#> geometries
+#> Warning: attribute variables are assumed to be spatially constant throughout
+#> all geometries
 plot(ml1$geometry, add = TRUE, col = "RED")
 ```
 
@@ -349,10 +352,10 @@ One option would be to use the dfMaxLength option with midlines_draw
 like we did on the left side channel.
 
 ``` r
-plot(side2)
+plot(side2$geometry)
 m_s1a = midlines_draw(side2, border_line = bbox_line_s, dfMaxLength = set_units(8,"m"))
-#> Warning: attribute variables are assumed to be spatially constant throughout all
-#> geometries
+#> Warning: attribute variables are assumed to be spatially constant throughout
+#> all geometries
 
 # clean this to remove extraneous lines
 m_s2a = midlines_clean(m_s1a, border_line = bbox_line_s, n_removed = 5)
@@ -373,7 +376,7 @@ basically a wrapper round the rollapply function in the excellent
 # smooth with a rolling average 
 m_s2b = midlines_smooth(m_s1)
 
-plot(side2)
+plot(side2$geometry)
 plot(m_s2b$geometry, add = TRUE, col = "BLUE")
 ```
 
@@ -413,7 +416,7 @@ plot(m2$geometry)
 
 In addition to manipulating the density of points on the perimeter of
 the initial polygon, it is sometimes desirable to manipulate the density
-of points the form the midline. This results in fewer longer line
+of points that form the midline. This results in fewer longer line
 segments, rather than many shorter line segments. This can be done using
 `midlines_dedensify` either before or after cleaning. Like with the
 `midlines_smooth` function, the points where lines join are not affected
diff --git a/man/figures/README-clean1-1.png b/man/figures/README-clean1-1.png
index 9de81ccd455a5bbd26dfb03024fc0299e5aef15c..25e80673cce77f656411f04626da9924d4e3fee7 100644
GIT binary patch
literal 3413
zcmeH~=|2<-7sux^nJI&cZe)-(vX0#tOxEm6mL}P^Nm(+=7M{s%Bs5QCNti)d5-Pit
ztYc{yQF9rF#xk<U*e7%KFFbFaH}~^7zt8!8zAw&;^Xi-w2RjSC6X#9<002JA%cf2M
z02dJe_+uK(ePnE%RC$itaR(b`GXQ|Y`5n%ds*1{Q8Gads0RZ?rf9DTUSh+s{0NJoK
zz2qF7zcNX?&(J<;=J=e$SrEyy7X7d2*u@IFitX6Tff<m7e*O8tg*h&;dutS6mltb%
z?%(=fHgFB~A<Ekh7#t`&%nu%=#1$)&wWS@bY2t~CG89)}@PdW~SBM!AcS5f^C+k$B
zcP!2dTb<T)e5AH6fz9Z6NK-So(x^67*s7ZI07mGZwW%6vyTu;L{LM;d4%qGa_P*iS
z;8AFjM*Zr?nPw%#NrR-KvHeZ5EiN9aA&PEZzjr7<uXTEeDd6W>0{dwqOhr!&NpzIs
zW)cp;=5yEzH$P_;;W%Zvx%+RazD_F9?iH@{dd&%?%A3>pUi}M9`|I;!Y}n^M-prbg
z+*Y-{i)noWPP=Ce)u0z>hA~!c_+ed?XF|Kr-<|&XN;Qd0-DPhzPkR3>@+s`>g@J>L
zOlilitJe{m6Q2ntjvda|g~VebWF85p@^V#mU%lsmmMvu@Ar)a~1SIUg^R{&g;21>q
z_=V)+zbbUkQ(|qpO@4?M{vLsNYF7kJk?K*WuhA^>joW-LcCK8n`20lPSLq1EJAT8=
z8A2E4e1GmhboUi`P;pu_#4Q2RkCpdRDG8|ocv!V{{fQA-+2JolDB8fDIT*28Sm4M<
zB@_~*<oMU1pO$QhMN#H)bHUd`OgxLB;FP|b8MBuSr-nHrrdMdD(U6v#xW3mqt5fCC
zz!=xO0PVLM_p}CJ3%=RkP184nv;8RDN}Vat1x-*<GAd9|pMrX94603&izLdl;M>te
zir8nf?srOT)(d+ZQ{MECKG0xYDOKU*bB}v@dnA+f_o@~;!G{}8C{~iM!oCtXGNL^Q
z7C`upfAn&?Bv->h{K4S~!$%6-d<izUj!E!f2UW(^vb}OF4<J!CEg9lH?ulV*E*#F}
z#SaH%<mOkHiNM<&;o&I)hIUxFp^>bh`X)EbjM+R`%t$hAOyor2YGvWH*u34r;(1J{
z7PD7&zCO77E6Ls5jXtt_4L!$Cx33G)*SMxf8ICI=$3aW6=GJv}yAHg<pW?;6P8fV+
z%DNT<_X`dXhmg9KhLx%&)e1V3!M(y3{=!``)tn<8KLGDY-39JvZ6NuVX(a1?$$c5h
znkndAgJJG<e{K(YYa#o?wpI3`X?0A4mR$3@FRAskM49H1YSRl5w5n2L2}MD8*6O4J
zg44V(y#7Kj<G>S@Zyd1CKA0RQd%cy_%VgTspMCX(*=R9JO;n9hYE|G5P(IuM7F^go
z<M)`w--(R90;{@cNBGKdQ4cnOB|P!)jj3mvrxdtIYYT0b(Ycl?r;cF@vZS>IOUf)B
zh(==k1=u^TkH1~kj?=I%A~g(Yyo_?<{_*qh#SWK1=u}+PoIKx)y-D&1_bm)NZFd`F
z>WN#I7D5wJXRjMQyXQ-*NtB5v7-R3o1Vwm!aE<opt*})U6MqPj6GY>0>aP+|+O6Gw
zqDIXDn>p%Inc&Ek2nTV(jl1rQF+N_;<T(o&3V+ni=R?CHlMKhQM#Z~g*@`JhukDV*
zCf%yHKDAa<AH5N9<m9jP;{>yYTQ2kJN>Qh=DjGcF4UYS*SHP@ziBkExJDDdy=GaeA
zp5=~RmxeIM>0bGDiOz=x=P7R{NNWBdDW1hdv`n<U4`o|ILa=Jj*lB<dKGg-hSEjbS
zL7Lkxn8|Up$J>50D>^$R32B8#VHTkgr|VH74_XGjXG^^rw1^HR^^WB;U|eVnOa1t$
zRg-nR)Gcx-M(EtC6#?Kb?O<X@bu=|7;hXZ@H|F-VIOUy$qYZeLlLCY!r>J_fdrsf>
z*(vxkS=GqHt5EG0M89UtCB>+s&NVi}1F`(Eew0eZ8$^UkI0=ZM7nhbS#G8HGg6};%
zEWSiHB#E`$K70C1q}SY%iQNHtE*=UGp8LrBk56T*q=P>2oFC=}uUK998p!c2L1wFk
z3<hiVegoG@4>2utesYtqEj$c(&(v%8ql7lTQX#!MuK5~TStv!xH}a9}%VWxG4h)Z5
zy_>b<B1bf7XS~iumh``6d~zEVcpu=X_b0-h_@hvP+%{q87RIhx+MYM^ch_^(r<SoI
zptbLzsRfD@2T;{kNY~|?24M|&o*unmau|k$3Qis4nrw^VEQx#8t?^@fCT8hmQJ$9N
z%GP74m4AclGU<f&aa0D`yv=2Ie370w(8%Tdd=1jFd4J)DoOa@yL&)BNSB~9lb52d}
z#G@7wZspKoWs1qp)Jw7PeAlW~Bv^A@&d3k$Zgw+ILMLc@7l$+n8)1ZQWtVPA2UgAr
zqSr${%Eju|GGj*A3Sv<|K{n8Z8OZL8w-7Dv&H)t~o#7m|h)mSJOS^VQum`U<5dEwo
zZ{Q**adhvMO3L94I!efjcqN1t7u&hm$|!hYWmw7DoAN~sL`{+CFY7K5vrZq|FK3)u
z9Nrm<TeeOW!oromXgNILecm)tjiX9=G`vq>cepIM7Xsn5yD#mCZfv2#HT6l7AQ}l=
zthiwhj%-+2JU0FpGKjV+<70RdZ8X!S#!&Yl0LRHgONui(HbkG-Z4<7R#40)d$j-2=
z&PE&JsXidw-SF3`g<jhcrVkJvKEjHDV)-YuvEmlQnFd?Jg_8?!(npvgKzQ&HE5_<R
z(!i^{ZEvqyYRUtAa}gO9ChaldW=Uj{MHjYt;UC<ZDNGbSxtfq8)0$&R2#S$tPI$q!
zHH9tDfDFbzTh%}5lF@G9mC9<Vu~iR>v*g*S{Y*EaP9KO*zfr=tv+WA;VxL4%^Yi&Z
zg_qn1)OB|5w5%Z=jzrtwL8tfH<ij4$CqBkU$^*nuxKU`yuIf|7(E1TH)r`i2jQcgT
zI9dB^myjvA`47%(E2O9BV7m1dJ;CN#sa*J&`s030EiOL!ucPvsh0Z3@l;{XP+wU_Z
zT^T9`teJoP$b@I`*)sIQ!;$ZbPN6mXaOT8ef6=_2v#{c84M#@p2TpDyWV-xmC|ccU
zo!e!@R(V)pEALWAg`TqB_ZP}CP*JRT#_d}wHwSgmydq3<BGy0uIcdnbS2sv-k0<Ka
z;a@*12X%s^$9SSRvH0%;gW5qFmLQ$jF-y{|^GmZlHTp}rw3h`Fe?KS_!_>BBioV?-
z?TpP8MAUYbh+aiA6dL;xEshUVo+VFQ+*<At%8TtHN7#pTEn@7OP^&-ruEwwa(P(Yh
zb~s>fI2|~~q{jUkW73p<0S@@j|B8MNYcEvgf=vddy_tashw2AL|Ih*AXpcoax0;@N
zMRWIYnY6+P+)4HA(6uyB`;QSY=H45dM+mZ7pO-7@Ntt9|a{!EQL$h1cOB0`jNQVu%
zg<zkQ>B;R^ORD2;;GZSHw@x2)@NRmjjTZaWp9(U9+_m?>@ZR3;4&RAS$7>(>b$BhC
zTNRk7&7QDs-~WYg2Vh-MMu`?hCMVHZ#_2#f5bLV&^>UH)N%R?CmNC%<Jo@hq{r`F)
a#*K7DDLk@TX+uXZ5x~;S&a~Rt`~E+>YZy2H

literal 3424
zcmeH~`9BkmAIE1Dxkb)N<qTmGmNm27xh7{sjzvVdLX6Ma2a%&luB<ttixsiRkz1}U
z7R{LZSf$CxG3={<;rq+?m+#~8dOzN;_w)VJ`}Ny9)xpj}h+mc;000PCS(-Wm06;td
zz%|XseL!p-wGSV(BMvrBW&i+(^S3w~+FDwF;cZLz2mnC1<8N{G+<M^;00{0`nVLAo
zJYAWj-7C`<HA60QILk8La<tMj4pvdtbz>KGbNmI>@4U^>Zr%m7OETpF`w09)^Z)C=
zNx;R0DW5NnDRj4b&ka1cj$5{_9ps}7*Y^Y$^Ngp#h0u$Z_$oe%!^xi9l!}l8bDEqh
zX5e%8$!M1MuFkB!4T*AQfEc&25a2Tq0ES%t!P1uFu>Ak}OI<KRKFdmQ=qRSp?{>xL
zq$@NpiMKRBN6#bCYs3Fr54aZdL;*)q=xQ>b-QM*X2-Z_c;G2`pU~gZP{(5_Yq|o*P
zs<RQ4-gn54&{giK^XH09-yv;{uYFL}G17a~@m;alrTJZCdK1<-84~NW<}95frQ)3g
zj&<j6ZRbs%^QWoWGd97ps{Q@ihJtbV={XA<Oi+rvV0XVq0qDN&;;etHaa6ccn)HJQ
z0l1>=2sP{cv<K~q_L-()eqko~I_U=@{8gP!@uGps6|*yQr{ia$KlrK%_l?0I63RAR
zT$oWm_2br?tDw|V-LoZ;{0o*6;3vNmykWBvJI5}|gf12L{axu!4ACwc9yLoxHO=P4
z2%E}8$Pi-q7vPV<c{6=kCJ<570P6jU5j<~L`(2l(IlCFFO5(q4qe3BxYh3^q&#JA1
zI|30K?(s*BJCMu%_rCD_U~MZl$D}za?)!VSJRjPSJ0=pt))zaD{Z;6r`y+#Gpnlv0
zmwUay_6^i~FkG-PA7zM~Ysz1cjs*ALKE-=}&}7-?3AIx=@<|!FO|$HXyjtonKOI@M
zNZttKJ52%O3k-4bFNm^R(kZf;y#n1O$%<?D!g^=y5;S&n?s<EuKS*%UG<$XkW2d#H
zS@HAiF7lm3yhOy|zB00Rn*|t!Z_7%pn(jD3rkcrhul)!ZhIPp2RF|jx9-X2Ef8`_+
zt(p5bBC5ZGi0%`{=IfhF6ZM=|y8}Lb25tkyQkG``^0!GTt(9ZcK|NX=a=|t}0Lg+7
z>t;c*4YeXM>SwB9blt`^T=y74BYHKVG=x_0r4N%y7YL!x=?LjvDSh6Ra^(}?7i-IL
z`5i3LSJ@gr)j*mf!Cwc{=xOZi?`SO;c_{1Tg@-%Eagbr+8%57fRWji*Z}_z-m=+^t
zIS1Y<F5k;hB<d-T{;7-`Z*q%kHvDsUZF5}1{grq_(|zh_{(PUcR4|ca8D4oc)o#At
zBV>upo=UQN9!J{7aXi-pV}?PW(F9Gq5?%EEUeL-S;am38&k>d50?rH;?NDuhYaUdR
zE|F3rnHYO-C}%qjY;iZq=S1_%e3xUDr0*PcnYt2bZ~r_uuayR*Z@}04LS4HriJx}9
zN1I#=Shzpfxy(S$9G(PGxjD_bW#_U7=>rGHuCQzv#T5A6CK#jn6PC6~zY4MDzL5|W
z!F1UJ87sEzaXso6P52kM!cu{ZDKMja8wrU@=w}GbomkXbBF;{uNpqf3ktZzD$(?iL
zeNUXxuV?Q1O<jI+#tkUeC7DN7JdJlCYFS3nt;`C;?IoV6lm!bVvT$^85$TJ}<Seq1
zsPPA;>3S5?0uo@)&A#tW9#n4@=`ZlYk4TKyI2RCo8glp|3&7<1jNzqX1fI#(F)OJm
zax&E6=ULRbCe=--K|cUmm}5B7=0CYG$qTULlSZOo2hK&swm5!lJ!dO_`K{~X5z2k0
zL~`?9SOa=#i}a$hN99K;Sz+p=V0`qiMG9nIDX3PLKdQK`*S7U$FucS(xXhtPOmK2-
z8Flo<F``)d<mxo4FeDSv(LAF1X_SxhaD=T^iu;m$OB-FnHDw`o4Vz5s_j=JoQ$w^=
zNA$8BM|XBpvnrLfBlq1v`i;;&y#{>lNAgT`!f!hCXG<N0Nl=KJpMaGgKTrPaW9N5U
zE}q;m!#1HajUwmhmQG}JE<pscVvx5Pw)`TOjpI(P8eVPqA^==DV}ak(&#Nb!2y7;Z
z9%EZM5@_lFz_B+o;6~wl{GG42FxSsScGgU?o=$@Ui5|{*eqVWHT{4~R#R7+(y?y9E
z7$H%(5E!EDzNUn^4)K(;a<Rc97u>?8qbVXEQAmg+)E>{kogQqRfHCP?9%C<u5_{{z
z-`uITXDk?AkA_GZx#CNNuly)>Ew95F&@8y;Ky{0}u3=Bcee2<yx9A+Si1v7N3fwO=
zoVh5(ef5ts{a#njW#CHr=5uVZGWF)VM-rjx*B&um175d1gg25hwKsUTjok?7o#He`
z`kvcK?3#C>3`z%viHBg0H*&OA$cyiTqIeN_jKQ|d2t(m*(;M3x3G}B_TldL_dRflG
z)QYaj?<<c3_U==Iws~aZiwLR5hknM=^%AXnuI_S)^X?dOd&#H!#t}jj2QjKv8x_Lq
zjqhlquZjZE=be)JZN7TCK(@+;q@t|W?O8*m4#47|vZ)a{m!(4lCFeEw=Om8gn1Fc_
ze^#^g4J=$knSUMGrAR3wJlp)1CwUqeb)`HQf9lrF*_Auznj{%DlN)nV)mOaNl*rV2
zjVNIAbmeM%9s`DhbDwgsw~a&KYeQVM$tpbOX?IkQurP3T*!E|moDD1dq~{M%s}E4f
zZ*p~<7Z`Fb3@?}P8q|UUW>~6_scVcZxJ76oOhl24DXJjN@kLJ8A6Yj9$`eC&pcr^c
z$Ws{U1R3d3NxFQL=X-3){^HJ6vzCvdqZACx+brBm8G|#pg{QG9Hr1tep#5y}1}ppD
zO<e{{lc_kC<%Hii5_QgQYtSA+^OjocepsrrB(TSoBI@(}Z;zoGy+W;Z10&oQ>$i%b
z3P%5&tr-|(UqKg|c=YQUY(_OQ;STPLi>*WKj%&=x``w&RLbEO8Bq-^&acJ3=_G2h%
z^}u`}P|&=HzO&~zxjRe~cxv)uvlDZr=W$>V{n7?K!KR=>Wwjm!%DoZ9BkS<$B{C5D
zGT9vedtSD%y(cTovbsbp(6Hfv%qYnfc1BN6|KopAksO#UlmRdMtgn}$?Q2?h^MG6%
z7P6hEeL|j|B>N{!^S+udDx{^cGR6_@FCk|jGv#NLKyn!K?68|!!9#|D0y0bHc#Qwk
zq8?t#*&(WFOWV~oQ0w>j2>tAk+ncpZtEhmY+JbVdT=UOzKmCVM;n=a_sDYJXyU?HQ
zBbg2E^v@Y?+jHS2^wS2HppBR$6<ql1figMg<?PtrDo0YNP034(3C+!QbC__~X&%``
z@ynNBt^57<u<7g6rLnj@>Jm+J55N{au@_SWZo^4o@<Y#>e+xS-NR&Uj%P`<cre(@_
zt$!%;j^X+Yyx0VO#*M9O12NNi+ZZE!5%F(r(x8JnpS@k&A61^b-hcuNt?6~?dFy3i
zp&4M3YY661B}8SX`lN31jjV!%^HHhjnf7=o!MbI)kWQ)ycJlJr1mY<gJc0I}D;wO;
zUzPg!$Z<(ju%W%?P5<aZbi7-HAh4<?VLua*a)PX$Bzq*14_M{+>Jh@?VC-B@g1;7h
kAq7m*`@jAd3A|5i_?ImEgNhSA_>KTpW_G64h->%$1HVFOr2qf`

diff --git a/man/figures/README-device_off1-1.png b/man/figures/README-device_off1-1.png
index a27fa792ca29ef9ae97f0c5427cac93a094e556c..90832b4a9809af53af70483c55d500339c98cd6b 100644
GIT binary patch
literal 5978
zcmeHL={FPr_m*s7qOyxZ7=&aL!kFx`jJ+&l-?yxju}nq{vZbi(GxjC>zJ!v5n9-2z
zB-@0EnX!%K?fn~mUw$9n5BJ>r+~=O>!#(GobI+3qHPHjG3b4}A(E$wfwaw}1=rMG3
z=N4Jc|LYi=%QO8;f>0w%oqt*M?JDrsiZk9cKR<tjt!u%;mpZ_Cc6Rp4#Wdj`ec`eG
zqYyf}i+%r%a|40pZgg}!D+byS%O|;8c{ZrH>s)jS{vl^q0&gZhbt+PKUUf}k{H^kl
zC4mUI&S!N*NB2V$$_YTByg0eE<1p6+7#PQ{0~44a5D*J%A>@D5|79DKW;FWZ@Bn7L
ziW@bFFX|WWO$6E*+|n_7U}W(%=xbYyUGqbF?D%DE?hiMZWj=SaPNX?=w!B&ADbD89
zy+b)*oZ;LhPj1bH$#Vfp^67NLmcH#9C?WNa99y?fVkWp2Z(DNpJVEDo&SmvHudHV}
zz4;_4kE6JBUB`TBbL+qqN<Ga&^el8%y5#0nqImo9WJiy(&f+QDPQoqFvkn*DkNBrK
z)j00xO6}%l=rxE;Qn0l&n?oH|B<0+R&1glKL15s~><GMd=vQvq5IyhoPp)ynVwj^!
ziuTcsz-C5O!RCJ1_pE@~NIxsY%|cZHTf1^H)NULlH-^1fF*DNLYbrC3TO)Jmm_i-C
znb>hU*Tn~i4Voq*V~!_yw<wmp4UJY_$@TiK0^6aM^6NCxrCbBo>+5Yc?}ux_Slo-A
zp8dE##|lKY$t)N}O-m)-NSGU~CQjA))8@MqPmBmLPmPRvSk&z#BrNe3hZAWXSRR|;
zrMivtR>q+!X8{?8WA~6xk3_>BUyq8-;n5)-$5gV7W7)#7YqMNd>jD43IS4t1Hm-O|
zTlQTysZ1iAs@7lg$k^5Y(O4t*H?dsSq*F$xq$Bv%<ASD~=}%3q^jm-GQ0`@{a}V5|
zr|n`r0_|;ogV}~MZunaAWLFuL6wed*Xt?yQ`H4Jq!Z>$UG`ClfI&BBLz^v(58^dm}
zasJf<hx93TSj89dWzKv;UqS5>lAUU9<rH2a*^7IGCyk5o5sNC@Hp?~LXVarDE>-B7
zmdRh`BVtSI{Hko;lfJs}b*iN{oN4HuN&f2nZD#TC4r3TtKI59wJWs{pUOAG_@-sHE
zsIc7cu)^?e%qg^Wr|&ig%sRPgA8KZKXY)#t=K_oBeANa@qe_)`r(6;^6XT)!()cLi
z>X$&lkmf^Qe#X}bRD*PRa?XC@f_m3p1q8}5!Qeh;YOu0d#(y}i0{<FvjZ~VAd4EBA
z2-k`jZ7!1l%3@7?E)2aWD3l|SPhP470a1u@aiGi<M8fxwW2O$yrv1k3kV~0{s9zrv
zhC4d;k2}&RgC8eou#xtJ2jY;O2;dm5f~Nl=@a`5Xjx+LZrgnb_|J(2g<o;Z)1-@*%
zILP_zIKNGc8zOENlms-z1<HS)Szo7&NV85n>#r!(1joaQpJRq2ww4}<I|sG@ll$)k
z-!&JCR4hc?x&0-kr!3t=3fu#`%wbKh^jSw`Mpfr%3N98c10BIlR=v2Aj`}m86D@ON
zE!aD1fFkI+^+c+nV3^qh*c<XA=v21Bi;%M(3TfA|T_(n69Y>k^=aZXZ9*nBd0U3t?
z%k_8z2Aeb^<Z+J*v8lB;L4ECnYtg5Byk}rW9Lp^8G)b=%e0)j`0Be*XZAF^h46}Ht
zr#z=tlmu-tDrnnX`BcwQhU$43Q@}GwC95~9ew+I{`>8hQ5Wmhn7Tp1pmcsRrp024(
zw#WEo{1~NNA*(tPIYaZLT@)wY5hXV=4NC%Sj?~=CUVZ(^fpK7=3OoKvDK?Q#9P0Nm
zKCh(oTeKp{Ry26UqN;Q?9(&=}rxpGCvO8^|D&v^-Xg>}evpB=TAgW7L?RM>{3zGUt
zdj;|FF`%XsCUup)Um%`4LxT%qFU4cakvzd~vH4Q8=u7I1&(M|Fq|cx3N<5O<&#t<i
zEbAnokiz!NrLVUG(LigsWmS%*==)ixHP0PLGbp_jN)%fCPzvyEAD5XwaK5g>75Pk2
z&m3l8hFeSY{%ph$?=p#<(u%G%Pf3kCbCC5Q(7c_Boq!psz9*?5d#y2dnGRsB)su;(
zGL!m;56k+S`#TFHzO3#X<c_WNE-Wv_Sl_QK?5-U|*RuFP<#;OpG$HK;Dd148K-?PT
zb9zg9CmlVv!^y4!Tywn<ecucME-Ms7>}e=r!>&%dKaRSw9n*6r>q`ilnY_j2RsZl+
zQKT(jjN+_<S-Hs|33HORcJRY+S^53zh|YU}@%Dv#1LpjxM=MPal;)JpVu=~-x&{0D
zfXtr8b>%!e#6@<~aJg{Vn3_oqvC$Wfy$VT%O&HRK%C}Z^I=3%YDsucibmcRIbR89I
zGW^1j>?@ryukwxf-4v#cskODJyVPSlH?w5wtPm=ECQ?av>B>(}Fm7u#m{MLq>aN0%
z7INA>k_YHObH0B(P_AI}solOBBLFyFP``?*jG%eNM<(~7!fAWH7tYncBc<_F<sww~
z?@VJ(x2=QS4coJ-OraM%by!jNorMh@0n;%@N7`@MNbtrbY0A&h#Su!_wT+|mli)`o
z5sZ0nK7CF6+~Bdf$>`fTT}8@qs7MTQKNP^<HP^M7I+`~YmEiml%kGdpT``N8X|eTl
z-4F_WEcw`NJ>NYD<`ZRnDa3_UO#Gr8*i)-DAY_#$;=km!PRN4zyqP+bJa<x~KnpI(
z&HBW0Auf!sr!)So@7DD8bgAi<=Qc_%II=oC9uVEW_wc+(S?R6Yg-+M{XN4%i)i7h9
z`2+9bjK<<{*o*}b$>qR~P7exT6%%7wH4B}iZ8%0=uV1q;1-hD5zgJtXyL0zdQ1Fnb
z2-A<d-<CRy0@lY|?d?}_;vxBpF&%=@eQ7Ji43607^B@>N%SR(R9*o<~$R!h|Dr8)G
zoy=Y^JMDz|ADwizs|2M|)M4BG7lQ8UZ}zHeV#0JD@Gh*ak+$ryiSu(wgrI5DeX*zg
zIH)r#@4_gIcD7`T6rK8p-rgbHYr!i%Yj=0MfN15()ll^xC`4%p33^?j$XeT?(NX#`
z*uPJp<b~7J)#ip%`%e*iO0w}?@2c*iS@_TzYJ$z$W8d!`gBSqHZ3{K6c9$T`W2b&#
z__waAt=+!?<DBAbq+PP7*P@#fp{7r3SOsw6{<!B}B`$Ldsm6o6?zQutcC;F)3J>og
z8J>8QWp=c6OZ;@sQw*uarZxLdW9_F>a;e1C$NQb~hYP+h$^1k<W2(dQa76tv_Jrd1
zowI*vySsb8>y59eIFfL-<0hEJ4Tf(ies?q#7MuZTRTR<&<TuhHz8HdS;IC}(&Grsx
zxl)WO(ulVVI=JN*SkI^=Uy;Q?b!*Su7)trx&EQjNG{_qxAY+xgFY4v3?bE6Hav8ff
zyF%1yjcP>mNF8((tT83LUk!%Kr>3ll)#t6a#I3ZYTLWax0y>e3V~xtf5CZoj(}P9L
z@qUTOX=o0n3*+POUL4iH;!_+MS`0!XfY8k_F#=n5<+pR(<*rrFF)69+BQP(SZ+pvb
z3TXq~U*PJ4Fv+WyvD+2*XDmVR^pWF$q}H!DhI}7~9Fq@?LQ11i+ao?{SO9EI@$JuV
zay$Ds9iY*N1O-DypMbLbiJ7q<&oC8=!^E<D`y<aa`r5yv00mHB?8zKyl&U};mRaA1
znIJC@hhCJlVLlx4MUoyYq?#MyUsv09R^orKk678;;ra#JFJDcx!?ULtm15EXI(V-3
zpHnxlg&RdWIzaJaFI?79H%pooA9cTzF<^5vfuvX>S2_V#zXqjYS+Z4M`zaK>9+C0d
zHw%T5pP>1C(u?Xmgj0Fp#Uvrq)2C_aKD|YR)auUX#j^a_cKEkiIQG6IbGFJZl<>%?
z)$<lLRZuM^HPPN;Pk!z4xUl1*2Kqyy>>E&YQHB}5r>6_CdYNK@?n(ECEFPS!$jRdT
zb4>9<@+*d~NSgTqp}U}X%m|+?6#qMJpy7f~rZ^>iZ$+cT^Fv5i<ZZhj?Yj@VmE=sj
zeUYVuGO?H}fYqJ+%)yJAI5`_(Mq9%gX)gmUj>NZG1f#--{a;U)WF)+7wOrq<H2Hg1
zjjN&F;0Zvx-Bu@J1-j=uzWFuTfqe;^ENw$?9OCE`5;M|WgBkrWi#zv>Thh3MmweUp
zl4S}wWPu*`5IPx<pJBqRv~ig5CS2H^6=d3#(olk2%EjRcN1L`FIdFzvOXOeWi1FF^
zNN-3U8ZDDe6XnTu>_-uGhW`$k><@@kVq_`+qJA6DBUNiB_@KQ#bU*)eQle^bEE8aI
z&5$#;H7Gsx{&kD77;@sfBn$V+E4wwTDe5+k_rTlkKVJ7A0h_geEscI^3#^JP0w)ft
z0yF6DMz~(UQ%t~+!!NN9Dc?S<$!WRY3JVk~kADx}@#x^2R&;3P9PgtterUm_w$ceP
zLoD2Qjq!2l4vEh)Ur3Q)6aonJ>YBWx42N|wP&(Calx5BQ={wPALe2*KdxLk{XNYHK
z$K>O``>hVeIb<cP0dH+QMKS526u7B9+8o5tF{9PTctWE^sty=G0#f;`97df%oF$5~
zPdt}?U%lC&uQKmT+!sXrz)8iY`FolljMqk|(1=a^94^Uk6YClu2md=L=V8}%-yJKE
zCUgrsPN(5We>4^A+TF?sA-M5T?P}r43Dh&G{-YszbS^l5kC_>wu73%H#jd{S%cchJ
z^*E4UGC2t`qdEp>C%j99tnkfdMzs+&y{;&q6HGc|%3;l_F2l&!H;M^x9dY?}A@gyU
z_;v|WJ917P1rr-ixv{-J5CFNBHjs}l*32N|FrhwT4q+{l!#@iWM?k`9LL+OnZ_DOx
zYYgq!KyOc&Y2pjf-SL6NzVxaRyhhBZfTi}YZdX9zq=_1BSf6;;ncr+cUp3BAev>=}
zH(MPiR}0CbI;2F(^AT(I0<ZQ`-6Sp)Xt}ygh836&ijCw@TMn7%Aah<i4>ZoV$sDeu
zwhh*79vJeyzDtc3EzkLm;!&{7z*x`}|A~lt{gMt<#MWXIj+6gFub^t2l4C;TjB|b0
zU{~LrA`I&p@+R!_Zic>|hcZwX-`{`Ds~p7LQR7uCb6oj6K3NCPo!h#j!kBW$XL9sY
z&cim0q}(6D8gyot3uMm!0Y77O>1E1R-e-yT6^h3yh__WN#4m)sEMt5(Rbj=hH&5<4
zi$LV+l(L21_|5e{dR=Ssd%y490;SHNz6fdk*5As&1g~BXx=yM4q`+#eiNCbGG&&#q
zb!x^%h}ZIc)|G}dIl4wCIBs`b#qLw6RKq0T9<V(0-U2t9PfSU6ZF2;p5~;LPd#Zq9
z)SXAlW>gw_V*wRobHK|v?|EjUzh5<Qs|m_jdpG>XeVz#}oQ;)YR4-h2|Gn8hnl4=S
z4ug<r#5fQ)_vca^^+)r4g|-y5S&kGzU)-{2w{4b9aYn;~<h0?5)P0r}hOp%KX}1yZ
zz->nk_mQTk7sJ8_7h~N2bR;ity=kYQ!$+0s4EmP@l70X}Y~8clKIXBZn4p1e<2Iwh
z4tO;iH3A-Vo@9%eeq7OdM-X#Tx%e#wd&yJGFV&~0BtyeotOuhjoDRz!7v<J_(DPC+
zw=NjIcLYYxE7h4ye9UIs7t&T3tu}jz?}!u5s&sYdAZL}v3|e~ZJ7wnTLC=T950&3X
zDqi8{%#fzs42w6cz3*64(jI&_@U?Ur!jgC~)4Ib;u`+*12!b<*m;YvYft9SdXO7so
zVTo-8*4FyH7!nh&WKE%l_m-BcnIm?-8yD3{!X0qLKH>BsF)M4RVh+946Gls6sfB36
zbC7G@+;zHJ>AN5*7pLW+b+16P-cXtTsXUei@q#E;l8fXizki9sHKAuHzc~Cs*<zbf
z;~F;<S-dG?oK%u^TMutBeyyU=s$7!H^hr30NwkDpVGWd=WY0-q2WT70yLf1&sJ-f<
z6d`OAEZe2IAaULz<_LbW1|}8r5p+qa`XviOAv+<%`fHb%E+^>S9nqCLTo8Fp!Sr3V
zbGaRVJ&ux@)wTX1^}_r%%IZ^hlE6c3x_$(c$0}FGVV235e({i+`Wh)Z@W$pPFL85m
zb_af^NyDFg-v4UhZVY{jrXp6_35e3g<UvQot@0ux8rXNpYJh$pX&-R4{~L*G3((F|
zl(3Ojz4<hbg0AV3=)}H@CqU~cg%SmqXN_KE^_dXN>$=bqyctqNQ-XQZZCL1zB9HO&
zVc_Omr-ZHUf@KzURn7mojLllx?U-dxx|Dw3etbeY^o!#G|9#(cWI+LWE)TYozRru`
zc-vm<ZJ0~{CnF|spO>7AM=VL^&t*3}tS!!4-lYnSj_)&!Wd*%t(rSVAKWF?rF9g=|
zR`B>o06TtWLX19w_(6^M+J0gP<k2>|V5y0}8K8v&5`Qv+#qeG~TiDgh%Y*%525@{$
z0(Xa@Wrg=#m*-#Zl_-M*m8%xzX)y1fy8e%dbBi8<ZD(IK%jS9?iMOU?w}!;)S5MKR
zY}uv(AwYoM=$%j<Gs5TU7hx5ieP$O+R`65q`!Yw>w`{6L<XiHM2bABH#<c>Et3&+5
z6MDdr1_Y6drCaN=4;G$F9X+|p(=4>)B*0OyjF^+H?ENr%711KR=#-sSpf$E)h2TiC
zD^fUAH2BDVF7$b?Y)%^OWIk+Loh1;cjh|zsgoeqf9MFD70G*YYL^gwTo#SD?j!fb3
z%ufOgLA?Y!%Hz2`quEJo7tfms%U=y^SGM{>`OoIc;B&T%b&A5zc87IIE4`h88&T!m
z^`Y**E&`WODviMocXCKHnj~_O#oH*MG}hEi@9dT5136NVNG?yn$34~;`Sl_gHhYn&
z2`yZ8Bt6T<WgF&`qrFb_(ntP;$|SKC5fAf(n)nZ?mwTHy9l?=uGoR%Z7@p(-^e#C1
zB>a_#EBtZAV3v0U#YyVDbxEP|e)<0_(ErDV@Cf*v76S}^0xtUJg3}r3m}uABcS8LK
DP>v);

literal 5986
zcmeHL=RX^cw^n<%R+MUK)vi(0)`(Fvb}5RY_NuR0E1KFWO4JO3+AXzrRDDq^iB&}j
zwIWDL?0I|d{S)qs`{wuJe4g{1=fydn=X2hilVobF%fiUZNI^ltqOYfIPC-G5q@cLA
zL{Ig%0ydYX{R_8DjVyKk;=Sy3(cf!MSd+rS!ZD^UoQ3zn;Ek)Rs|;t8#J@7bD?OV)
z3JTV~|Bh>e{*^B&C;&V9+91m?#7;p{R*g6(%|nZ`E5~2sIn-?LsQLgkKf;fKRb}{U
z%&tv0QGCmpprOd20)cp`X*_Rmegv7mr~ME4KY0*zr}2S?+j7X4EQ8;9mAoU}atkdV
zs4bFho%1wdPX~hM^M=@e8cKrX$GsaGK-+nyoy^@|nzdU6rekFqinmOD3f?UO+Rv4@
z7vmiW3@<FXTb`Tbk@)qGdJHF9TJtEM$0z^h#$bJ0RdF?@WUn_xOn!)sP;-m@dsSZ!
zLe2#%J%QzqUt4mmY+G`5iN^c&<lH7^prTO*vBeRU5+Q8N3jvL;bA#B#OP&ZPUo*6U
zYw^4<z7QRQ?yseQKuRym4|l_!OQZxo7H9gkh%UD;1+9-?#hojg?PPTQ0N{P50IgAA
zJi6OagzSl%7{ShI79`A-v}|#9&)f}cZ$igRd@1R6sAEa?Lw4l&wg`NCd41(p20uIG
za&n4$XWx>$;WOCdUAtbL{P|{%F}9_B8@>8c0PkJ*tOEB+6T=0bbWp<7m7-qx^5|gX
zJ~U!h&7Hnmm^(K8=mbEj9b_u}GA!aAn&?IK+Ny)4ii)EAvY9GQxTd(PKj`9CiAZ^2
z7kE@8A!TYf=?{%<qlPu-aix@LjctV{rhTouT1jD=#;d<xdOgXL*DdYA#||Y)e~z>0
zA=w-Nh}+x_0Z>d{pMA$w)8!OCY3{`|IYGNW8=1AeUCdx_;Ob;M7vn~Hp0|3%m1&(Z
zYG&gB+Y>LVy#~o%t@})r(f(Nb{eibe%GPY&(fh8=!t0(n3Ebs>YqBOtjy*|_t4O^O
zqj8>U#?gBR)aU1nr_s?2T3VJ|*)%Mb1Gk47>AA@SUtTFC8F>w`)5+(;;hUwNq@%+}
zBPf489Pn#w-nTsq(+~Q*>s1vz@eT3$C)RLA4)Nq+X&`tLjk(@O9$Hs@Ju5T$SVPC`
z#Hr|AA*prwXXBQmb*XDgnZ7#cWXHW6dcJD&ep$#E=LR1?p*Wy+wK7f3KYHV?Un>nF
z(VwPxL1jr5=IWg@@4ATTsOV@vd~b0ca+E=Q&E30pXWqolG+eWLianDDSY}MZo!ylq
z9B5)GnyFpa<L;P;qiQuopdry#fe%W|^-ZB70iPQq2HlUpOCgru3S1LsF!-*e158*h
zC60|8s%JdYs?!uQW8kxIo;>3qKhDlih>Ew)6%x=DGTTk5tZ+ZOD2G16&gMlt(!Hnf
zA-2D!1X*+;I-!jXNaOfv$qpz=NsdP8rq7L^zaP(Ge^+_`0UdkL&55Qm`BG_#H(-{4
zhl?m<ePE~4$51M-6XnYW_Gf%Rsr3bc*V#2rL=x@(JlOx2v_x-qA6<Ff*Nd9c=|GEe
z4f6ic)=I#Azw~vb@E5ee`C<$?qISkd!%+d&^U%9ZD{mQ}zBzpLTDRbU>*hFZ)0Tb5
z^Fr9;J63kMMI9!Iuf2kkTvyJJ2z*E<i-e`OTL&OeK6z?E5JWNE*OrMoH<9pPpA`<4
zdo%BZv}<!7Q2XRZ*C%<-mvB9G`i6RZQdU&mq?z~t8A0X)Dhv?ZPZPz$Q&z8mHmdF^
zXr*{Q&bU)$)n~olYa*1+6BU7f7ED-=YYFGLdRlp3oX-1^?mi}FA1JVyX?yRVRW0w1
z=){ey&MVmSwEI)+X<9RVIo5Q3IR4}kLGh%4@tWPoombte|2Qp$_5K0Zsyaxj?N1Zx
zp43zt--mHBk%K4S$*Y?Sx$)FKsi`udW@S&oB3`<jd&3N@;qxn^x<Yhx^!?YVwp0-I
z_IDDL9^{&){fk-O9sd!$;(a41Hsn4a=jf)wr!`A|+eJ>VQodxNZ|YX0V$CH|QudoP
zHrD=<5_JSFZQ>nD`06Zt6wh$mOonq{pfwuATU%5Y6bg7C20vB#phye!P454u`~W36
zN0gwFYMm!1E=FAXDHb6+lH;Y|gQ$Cz;Pz)Z<@M3KzSs+wTjA}v1QFP$)N`&ImO+QV
zS5ztPUbY#8Ffz`yfG2@6=ROPt|2(Vcoe>O}**fr_%R{w9J+DLc8^EMCosQ)T3}~cH
zhP<)FqO(7GrS_~F^!Bw5RKB${1C5q_<6jVqjlFRq8c#Brp8#tWIhO)C)f33sh3XgZ
z0=Rt&|5kesn>7}eqLEuBD(RuBmN47Fp@BN)9Tcl$q6ln{Xb>IrxqJ+2*ZD`3mqqT}
zMY9p4l67*@o!^nR2sJ(Qz-$oNrXnxjaK~L=3iFG_j5kL==DB5iDjpGERZLGXSGj2-
z;tN`sP94OU(OpGcPeZPEMm7Deb!kSr?E^i8oOio{d7t(On+p+BbO#$~LureVqV{;|
zr%K}D^l{G5Gi!N*E|^o3jJX{K<)Fz^vN@Z$9UY+M?jPdapl6Jw97&!dKiinrLxQax
zA*^M>1KDeiY0XGaAJFHbV3?uPOUh-=kHE88I-gqPy1Mdr2Ry#J=GCn~FY%C(-$y&?
zrAaihbmcPQ0e?<T)L-`a1)`3$ethf2b}GJ|GHPV%q2B*oDcAbQ?+q+j?<T@IasYM*
z)g0Jqm%sQK5@;QiF<(??ec9J7vG&r-dzA7vAXT)j?3k5KJ+j6fe(D#=UzXWe?&r|s
zSRC02-e~dS%m3y<?esi(V{AvgD@4Z`6TeBcD@0%!oLDhaR;h0-0C6{!taC1!OMUTk
zvc|^b`HWdK0NA}|p`;Qg#!q=fi=WQe_B`3XeMvcD(J*Fm^>Jr!utKJvY6-9uRetvC
zmygnu!#gKufC3*%@+SyM>A{2O*kSXI)4M>R<HlS*e!55LcK$_>TRZiH>aDmMlAHfU
zH2E{`y^ot#_Aw8_?q<&2aJMGvU|dywd(aP<4$Z-_XAjc9Vd7y^Zh4r(@Q=lzZq@#N
z9um_i=aA|AW@h#-f!~7qNxZ_FRJ8W^T*Nu^C~>xMf6ZJQdkKMGp3wL&HivjZZxtkF
zd^kEec^I@Y@-^Q6%!!DIkCqJsM9)Q{eXf=#;}w>DjkSA+RcDe&eeJJ~qs$20il3*Z
zS%P~Hx6T*z;6iVUfezjQfZ^$=O!Bk}q9y^SH3SiESd){-Qq7}-a==+V)V}Kq%541v
zBCsb^t=wBt2vQGWczG+A2gtocE|F{c<70K(;HkPAL=1f^evo06*O-w}CA$WL9qN;s
z9|?NJIQ?tvklhZuEA|ME44G@iCZ8-g7~KJAV;kCGr;VL-=`Nfj27>0e>14hcdV-C*
z%|EtUDVC9o7+(%;x4`yO?!kfEmG+1=N`8!&M^bp_N`7{GvsptsZPsAFaoC#n)Q~V@
zm2Q~HYu)O}ut*BBG-w8$Ie?|0saE%Y84j|TeyvMInDsg}{}6~~9&_*%8NI$?hR9h9
z!0dtt)&CJj{c@)7SZxNn3xs|cJh9Js?By-=qx0mnKH^`+{&ZG`4*43P_`*MGMePGC
z+gYRKiSSwc3ml1<l^!_1|7o%0Rwy+%-^L2KvpNA(Z6!047H(*ft(4TxiT3<*JzY?R
zVy+Q5JuzHCi#$EkO*ACBD}xjdy(G5^fVsu()msGRlY7?CjANQH?2tS_zM+^t?;f{e
z{@3|g!Hv_4dL+`DvmeV~<F(}`dm&?;aR-Hf!ptwGNE40s28H79c@T`$-rlbwk4bot
zEt{+ZqvCH8-j_D4sHyQ2^?)kTWJx3V)6T7N&nq$Yg;!tH+G(Oz`CQN<l?;p>3aXB8
z+bgfC%=Z`xu&!3m!R$Z=0$Tb{wv(qbQx@Vo_8s){Vr^P38@a&orxX*qjpzX`tc{Xd
z#zZ_WXO*o8uWhltlYG#^0W1_MT%?u-Hm#5f5g}#$DnC139LWiOAAY(s<YTij>~cEY
zrgnn9$PuoihPo+CwNYCMOT-k|*Q)Vi8u<QKSb%292Ym=rmLI#HY>~PRiR`BadwYy_
z__Rp+!t->o?BOs%$gUI^KC>!Dh^p$|T@cfuceQM1_sBc5%qPp9wL!y=_8^ik9fK$2
z!yl#v06=3~xL>U?bbzT=yQM(p{I_}gVcLeOX|gmOB6DU76OdO@FY#_s<tdu{%r@Nf
zK1WLddzI}yM_6b$%R-$85&n~SE)JNgwT(9kIlba5VN@YcmpQ~bs$@P+<N9fz`<97-
zlRN*niHEOX_C=ssNIh9W%3O?^r~2@9(f6=`>TQ+7$eNSE!XEz3P^pXLR5M#0(UwMq
zH#Ogwp---|U~!}<!O7b+P3KT#^l-s%!Kl>Ho{f82k8UO>Bd;|q_NPtuV`0l(amZ^h
zX?#7)K_rcb`rqy7^njio9Yk@qyN|W@z{z6FP+Cz@k1mYDF$;?*uRgZCvol<+_BMF(
z{NiH4ua$VTA$_qj#k6=mY&W{xOI?Ib{vuxokxBaT&}f_upC84og>`S2Hk7(c>@LTo
zQq?};F`DmgR8kg6F*1i8<({s^`yB*&q>+w#B>jz}h#{Q^;O(O(HW!ppY(f8!q(?RE
zb4)P=PCjP^k&bTG9Yi_M29eehPZj{wV7Sv%&7vt~t$9B6POV}oDzGC!r1qW%1s8~e
zR|VkRj3gBcUp>c<DXnRk0&^2o3Yu1;ja73rA=i#CT?>${Bv?94a9#-Y0sM94q`vCj
z(V{#2rjgz9#sPp(mE5O3$3P9TaGgEo*43Ns+X=CGfM5G3X~8)>ak@T~6;$eQLuy2}
z0ogN&R}qF(I71uvT*!|&K^L4inktJ;3b{W9Dz4gy;O^q3-yID`LI+R}?J~dD7&e~G
zw>g|elO2XfR>tMQ>n<1V-*rrZLe$L`X^L6<#;zdQw_MV<_|p<LjU?q0R((aGr}foY
z=PXO8)l#O8`vpp4R`=rzDX9-EY_a#^4H#51DgQQQlMmoHl`9slm=5JV<M-0U@@!Wx
zbQyANgda#R=3Yh`)KgznE1xxBahuzchW*gfeC#Th@vm641n(v@(0f_jg)NAd<Qr)W
z61*N7J;^BC;)rFbIN+~+(xz9*0l}F+0g<Yxecwvg&nkXJT`7kb)zA|}ut&LIte>99
zn)TOisU0@?!+iIZE|n5D5I_IKbLRNHolRXnd(aAMvkyKe3<_QZ{RTBXJL7r!DF!M1
z1ZRQgOGI=_{0IlRj`3lo7FY+kI%%^i&?oD)2JgFle;R<McYGuOL6Q1go?c@O?>3_K
zGcKNB_IG@|0YR3n2q<ku`8JUQF%qaXW15$ty?(Ro*FdYE&EotLrf*Y2FZ-f{<S!)g
z#0O?EalpwD<d!JldN@KHxPIS#u->4Pl>r0<=5JPO=k{E(Yx>(77eb!HtQH?-a!koJ
zE{6TMKXS2C{3Ab0G2kauddZ@-d`JoBmp4*YfY@{r<4V|i+DznNoOSrMgrNb2n)<hm
z+M!wPuxjpI^~K5}R8u<5%d);MVBzGG7G&9pop!|X-btl`?Z8jDM|(^v!=((LV;n8{
zB+s<(?8bdaX>5fg9cj)qngxGBeTXP>ag@F$^}u5Rlx8$t-ylAdqTE%gVo^{LRlN9A
z#&DsS_fc_?btU(WOyMAj;Dlj6O@35a`#+(NQBOmsO3gznDJ@<z$rsaiq<3hQ3I09R
zKN9ug3Q?ldvFj-2+L&QMq3Kva8}sAGA9Tg1rs0r*#Lw;rPHB2;>EaT2rF<XK-MZvq
zK`z;kF~}n<V@~Oz9u23M0N&AqX4EX=BjxpHfHry^EZCC2>Z|4QIan0xS8}R1%N2Rd
z?LMr&RzoC|UFx-4<Ov(fxx$e7l|XZyeC#(Z8?IMSJ)5VaC%gmtQuG~h9NNRR&Td+7
zP0$4b+v!SLABP|P9XVf4TU<uEakX*Koj!ukm`$uEgFam5fAF>YI{~~OnoX>ym_D2W
z%~quWc}*Bjn>5+mL=%*CKQgl5-(~dB&U|GSNwP%tsZ}7miaa^~=DBPIOFo^AvFeoW
zNtO&+x}Jh~1vGord!!hZ1{Lp{$=K?cVtU>;<XD281$-l#2I%st<QdD;9S86HwWx+(
zg)<`%ZY;^!^%G?BmI^s{o1;GxD8iU!KfPMF#V0b9-=`2*{(^=DEG1f3i5tx=;X-t<
z0U3{ZQY_*m=6@c)|8+JDZ(a1+c)rX#tx&8$n)0iXt1g1~dk<g}(~pIN^IvjEX9vD&
ziHT!L<xUQu^C&Swf19OLsP)gdRu~|A+@3b6^FVDx-CSlMNp0Qgn?IVn-bU*vp-=w}
zZIFlv^DLq`f-5+55Yvr#=++odGz)5wfain+4kO1Lg#E07I&MW1HxOoLPI!Qhdgqcm
zwJ~ptV|=aURG31~Z+B)iui~I-T?v!R3J%+SVw(kMLnyp5?&l|VoA;!ixO%>NJM?*G
zFXPSFYpRW@^WQk}um;7zLxt4Fr7ow%ktD$lUX@b(&wm|Rwk$Cmy04|(SqtorzLbz?
z16yVzR|u~OH%aSLz4?}xvgzKBAP{1PMM9Y325hnE5I#v(@oOC_3DgZ!6uDls5xHQL
z_qFV~T6m%Q*G}cA^GHFBe9mrMl%WYfuKR|DBmH?pT$@83cm?<N4>6AZG~x{r)hl!z
zVxnp-#JnXCdwWsz0oI%7{9@NsY!q}Og0fUc5$B8f8tf+&G5}RV(K813I2JNj>|&zA
zCr4>mRaZcwxKZ0+$asi)f#PIv7TO1FB&aFVGm6L_TsT^|uuysxD=(}asI6y)IW~V`
zXiN&Yi(rlJ&0<V0tiOxoh`ehfuZj5_K}o;KiKDnUKTIiK?gb=Pr#))9>t>$l7gGjf
z|Jrl|Cu3+l|F85z>aaq77&~g}#&XeqvWm(50LM%w&90dAq{Lh<O_m7(QN6cF0Tw7!
jo}TeP<p0-$G2q@D72$P^G4J1RIEB8Bv39NI3+R6V8Ffz2

diff --git a/man/figures/README-draw-1.png b/man/figures/README-draw-1.png
index a27fa792ca29ef9ae97f0c5427cac93a094e556c..90832b4a9809af53af70483c55d500339c98cd6b 100644
GIT binary patch
literal 5978
zcmeHL={FPr_m*s7qOyxZ7=&aL!kFx`jJ+&l-?yxju}nq{vZbi(GxjC>zJ!v5n9-2z
zB-@0EnX!%K?fn~mUw$9n5BJ>r+~=O>!#(GobI+3qHPHjG3b4}A(E$wfwaw}1=rMG3
z=N4Jc|LYi=%QO8;f>0w%oqt*M?JDrsiZk9cKR<tjt!u%;mpZ_Cc6Rp4#Wdj`ec`eG
zqYyf}i+%r%a|40pZgg}!D+byS%O|;8c{ZrH>s)jS{vl^q0&gZhbt+PKUUf}k{H^kl
zC4mUI&S!N*NB2V$$_YTByg0eE<1p6+7#PQ{0~44a5D*J%A>@D5|79DKW;FWZ@Bn7L
ziW@bFFX|WWO$6E*+|n_7U}W(%=xbYyUGqbF?D%DE?hiMZWj=SaPNX?=w!B&ADbD89
zy+b)*oZ;LhPj1bH$#Vfp^67NLmcH#9C?WNa99y?fVkWp2Z(DNpJVEDo&SmvHudHV}
zz4;_4kE6JBUB`TBbL+qqN<Ga&^el8%y5#0nqImo9WJiy(&f+QDPQoqFvkn*DkNBrK
z)j00xO6}%l=rxE;Qn0l&n?oH|B<0+R&1glKL15s~><GMd=vQvq5IyhoPp)ynVwj^!
ziuTcsz-C5O!RCJ1_pE@~NIxsY%|cZHTf1^H)NULlH-^1fF*DNLYbrC3TO)Jmm_i-C
znb>hU*Tn~i4Voq*V~!_yw<wmp4UJY_$@TiK0^6aM^6NCxrCbBo>+5Yc?}ux_Slo-A
zp8dE##|lKY$t)N}O-m)-NSGU~CQjA))8@MqPmBmLPmPRvSk&z#BrNe3hZAWXSRR|;
zrMivtR>q+!X8{?8WA~6xk3_>BUyq8-;n5)-$5gV7W7)#7YqMNd>jD43IS4t1Hm-O|
zTlQTysZ1iAs@7lg$k^5Y(O4t*H?dsSq*F$xq$Bv%<ASD~=}%3q^jm-GQ0`@{a}V5|
zr|n`r0_|;ogV}~MZunaAWLFuL6wed*Xt?yQ`H4Jq!Z>$UG`ClfI&BBLz^v(58^dm}
zasJf<hx93TSj89dWzKv;UqS5>lAUU9<rH2a*^7IGCyk5o5sNC@Hp?~LXVarDE>-B7
zmdRh`BVtSI{Hko;lfJs}b*iN{oN4HuN&f2nZD#TC4r3TtKI59wJWs{pUOAG_@-sHE
zsIc7cu)^?e%qg^Wr|&ig%sRPgA8KZKXY)#t=K_oBeANa@qe_)`r(6;^6XT)!()cLi
z>X$&lkmf^Qe#X}bRD*PRa?XC@f_m3p1q8}5!Qeh;YOu0d#(y}i0{<FvjZ~VAd4EBA
z2-k`jZ7!1l%3@7?E)2aWD3l|SPhP470a1u@aiGi<M8fxwW2O$yrv1k3kV~0{s9zrv
zhC4d;k2}&RgC8eou#xtJ2jY;O2;dm5f~Nl=@a`5Xjx+LZrgnb_|J(2g<o;Z)1-@*%
zILP_zIKNGc8zOENlms-z1<HS)Szo7&NV85n>#r!(1joaQpJRq2ww4}<I|sG@ll$)k
z-!&JCR4hc?x&0-kr!3t=3fu#`%wbKh^jSw`Mpfr%3N98c10BIlR=v2Aj`}m86D@ON
zE!aD1fFkI+^+c+nV3^qh*c<XA=v21Bi;%M(3TfA|T_(n69Y>k^=aZXZ9*nBd0U3t?
z%k_8z2Aeb^<Z+J*v8lB;L4ECnYtg5Byk}rW9Lp^8G)b=%e0)j`0Be*XZAF^h46}Ht
zr#z=tlmu-tDrnnX`BcwQhU$43Q@}GwC95~9ew+I{`>8hQ5Wmhn7Tp1pmcsRrp024(
zw#WEo{1~NNA*(tPIYaZLT@)wY5hXV=4NC%Sj?~=CUVZ(^fpK7=3OoKvDK?Q#9P0Nm
zKCh(oTeKp{Ry26UqN;Q?9(&=}rxpGCvO8^|D&v^-Xg>}evpB=TAgW7L?RM>{3zGUt
zdj;|FF`%XsCUup)Um%`4LxT%qFU4cakvzd~vH4Q8=u7I1&(M|Fq|cx3N<5O<&#t<i
zEbAnokiz!NrLVUG(LigsWmS%*==)ixHP0PLGbp_jN)%fCPzvyEAD5XwaK5g>75Pk2
z&m3l8hFeSY{%ph$?=p#<(u%G%Pf3kCbCC5Q(7c_Boq!psz9*?5d#y2dnGRsB)su;(
zGL!m;56k+S`#TFHzO3#X<c_WNE-Wv_Sl_QK?5-U|*RuFP<#;OpG$HK;Dd148K-?PT
zb9zg9CmlVv!^y4!Tywn<ecucME-Ms7>}e=r!>&%dKaRSw9n*6r>q`ilnY_j2RsZl+
zQKT(jjN+_<S-Hs|33HORcJRY+S^53zh|YU}@%Dv#1LpjxM=MPal;)JpVu=~-x&{0D
zfXtr8b>%!e#6@<~aJg{Vn3_oqvC$Wfy$VT%O&HRK%C}Z^I=3%YDsucibmcRIbR89I
zGW^1j>?@ryukwxf-4v#cskODJyVPSlH?w5wtPm=ECQ?av>B>(}Fm7u#m{MLq>aN0%
z7INA>k_YHObH0B(P_AI}solOBBLFyFP``?*jG%eNM<(~7!fAWH7tYncBc<_F<sww~
z?@VJ(x2=QS4coJ-OraM%by!jNorMh@0n;%@N7`@MNbtrbY0A&h#Su!_wT+|mli)`o
z5sZ0nK7CF6+~Bdf$>`fTT}8@qs7MTQKNP^<HP^M7I+`~YmEiml%kGdpT``N8X|eTl
z-4F_WEcw`NJ>NYD<`ZRnDa3_UO#Gr8*i)-DAY_#$;=km!PRN4zyqP+bJa<x~KnpI(
z&HBW0Auf!sr!)So@7DD8bgAi<=Qc_%II=oC9uVEW_wc+(S?R6Yg-+M{XN4%i)i7h9
z`2+9bjK<<{*o*}b$>qR~P7exT6%%7wH4B}iZ8%0=uV1q;1-hD5zgJtXyL0zdQ1Fnb
z2-A<d-<CRy0@lY|?d?}_;vxBpF&%=@eQ7Ji43607^B@>N%SR(R9*o<~$R!h|Dr8)G
zoy=Y^JMDz|ADwizs|2M|)M4BG7lQ8UZ}zHeV#0JD@Gh*ak+$ryiSu(wgrI5DeX*zg
zIH)r#@4_gIcD7`T6rK8p-rgbHYr!i%Yj=0MfN15()ll^xC`4%p33^?j$XeT?(NX#`
z*uPJp<b~7J)#ip%`%e*iO0w}?@2c*iS@_TzYJ$z$W8d!`gBSqHZ3{K6c9$T`W2b&#
z__waAt=+!?<DBAbq+PP7*P@#fp{7r3SOsw6{<!B}B`$Ldsm6o6?zQutcC;F)3J>og
z8J>8QWp=c6OZ;@sQw*uarZxLdW9_F>a;e1C$NQb~hYP+h$^1k<W2(dQa76tv_Jrd1
zowI*vySsb8>y59eIFfL-<0hEJ4Tf(ies?q#7MuZTRTR<&<TuhHz8HdS;IC}(&Grsx
zxl)WO(ulVVI=JN*SkI^=Uy;Q?b!*Su7)trx&EQjNG{_qxAY+xgFY4v3?bE6Hav8ff
zyF%1yjcP>mNF8((tT83LUk!%Kr>3ll)#t6a#I3ZYTLWax0y>e3V~xtf5CZoj(}P9L
z@qUTOX=o0n3*+POUL4iH;!_+MS`0!XfY8k_F#=n5<+pR(<*rrFF)69+BQP(SZ+pvb
z3TXq~U*PJ4Fv+WyvD+2*XDmVR^pWF$q}H!DhI}7~9Fq@?LQ11i+ao?{SO9EI@$JuV
zay$Ds9iY*N1O-DypMbLbiJ7q<&oC8=!^E<D`y<aa`r5yv00mHB?8zKyl&U};mRaA1
znIJC@hhCJlVLlx4MUoyYq?#MyUsv09R^orKk678;;ra#JFJDcx!?ULtm15EXI(V-3
zpHnxlg&RdWIzaJaFI?79H%pooA9cTzF<^5vfuvX>S2_V#zXqjYS+Z4M`zaK>9+C0d
zHw%T5pP>1C(u?Xmgj0Fp#Uvrq)2C_aKD|YR)auUX#j^a_cKEkiIQG6IbGFJZl<>%?
z)$<lLRZuM^HPPN;Pk!z4xUl1*2Kqyy>>E&YQHB}5r>6_CdYNK@?n(ECEFPS!$jRdT
zb4>9<@+*d~NSgTqp}U}X%m|+?6#qMJpy7f~rZ^>iZ$+cT^Fv5i<ZZhj?Yj@VmE=sj
zeUYVuGO?H}fYqJ+%)yJAI5`_(Mq9%gX)gmUj>NZG1f#--{a;U)WF)+7wOrq<H2Hg1
zjjN&F;0Zvx-Bu@J1-j=uzWFuTfqe;^ENw$?9OCE`5;M|WgBkrWi#zv>Thh3MmweUp
zl4S}wWPu*`5IPx<pJBqRv~ig5CS2H^6=d3#(olk2%EjRcN1L`FIdFzvOXOeWi1FF^
zNN-3U8ZDDe6XnTu>_-uGhW`$k><@@kVq_`+qJA6DBUNiB_@KQ#bU*)eQle^bEE8aI
z&5$#;H7Gsx{&kD77;@sfBn$V+E4wwTDe5+k_rTlkKVJ7A0h_geEscI^3#^JP0w)ft
z0yF6DMz~(UQ%t~+!!NN9Dc?S<$!WRY3JVk~kADx}@#x^2R&;3P9PgtterUm_w$ceP
zLoD2Qjq!2l4vEh)Ur3Q)6aonJ>YBWx42N|wP&(Calx5BQ={wPALe2*KdxLk{XNYHK
z$K>O``>hVeIb<cP0dH+QMKS526u7B9+8o5tF{9PTctWE^sty=G0#f;`97df%oF$5~
zPdt}?U%lC&uQKmT+!sXrz)8iY`FolljMqk|(1=a^94^Uk6YClu2md=L=V8}%-yJKE
zCUgrsPN(5We>4^A+TF?sA-M5T?P}r43Dh&G{-YszbS^l5kC_>wu73%H#jd{S%cchJ
z^*E4UGC2t`qdEp>C%j99tnkfdMzs+&y{;&q6HGc|%3;l_F2l&!H;M^x9dY?}A@gyU
z_;v|WJ917P1rr-ixv{-J5CFNBHjs}l*32N|FrhwT4q+{l!#@iWM?k`9LL+OnZ_DOx
zYYgq!KyOc&Y2pjf-SL6NzVxaRyhhBZfTi}YZdX9zq=_1BSf6;;ncr+cUp3BAev>=}
zH(MPiR}0CbI;2F(^AT(I0<ZQ`-6Sp)Xt}ygh836&ijCw@TMn7%Aah<i4>ZoV$sDeu
zwhh*79vJeyzDtc3EzkLm;!&{7z*x`}|A~lt{gMt<#MWXIj+6gFub^t2l4C;TjB|b0
zU{~LrA`I&p@+R!_Zic>|hcZwX-`{`Ds~p7LQR7uCb6oj6K3NCPo!h#j!kBW$XL9sY
z&cim0q}(6D8gyot3uMm!0Y77O>1E1R-e-yT6^h3yh__WN#4m)sEMt5(Rbj=hH&5<4
zi$LV+l(L21_|5e{dR=Ssd%y490;SHNz6fdk*5As&1g~BXx=yM4q`+#eiNCbGG&&#q
zb!x^%h}ZIc)|G}dIl4wCIBs`b#qLw6RKq0T9<V(0-U2t9PfSU6ZF2;p5~;LPd#Zq9
z)SXAlW>gw_V*wRobHK|v?|EjUzh5<Qs|m_jdpG>XeVz#}oQ;)YR4-h2|Gn8hnl4=S
z4ug<r#5fQ)_vca^^+)r4g|-y5S&kGzU)-{2w{4b9aYn;~<h0?5)P0r}hOp%KX}1yZ
zz->nk_mQTk7sJ8_7h~N2bR;ity=kYQ!$+0s4EmP@l70X}Y~8clKIXBZn4p1e<2Iwh
z4tO;iH3A-Vo@9%eeq7OdM-X#Tx%e#wd&yJGFV&~0BtyeotOuhjoDRz!7v<J_(DPC+
zw=NjIcLYYxE7h4ye9UIs7t&T3tu}jz?}!u5s&sYdAZL}v3|e~ZJ7wnTLC=T950&3X
zDqi8{%#fzs42w6cz3*64(jI&_@U?Ur!jgC~)4Ib;u`+*12!b<*m;YvYft9SdXO7so
zVTo-8*4FyH7!nh&WKE%l_m-BcnIm?-8yD3{!X0qLKH>BsF)M4RVh+946Gls6sfB36
zbC7G@+;zHJ>AN5*7pLW+b+16P-cXtTsXUei@q#E;l8fXizki9sHKAuHzc~Cs*<zbf
z;~F;<S-dG?oK%u^TMutBeyyU=s$7!H^hr30NwkDpVGWd=WY0-q2WT70yLf1&sJ-f<
z6d`OAEZe2IAaULz<_LbW1|}8r5p+qa`XviOAv+<%`fHb%E+^>S9nqCLTo8Fp!Sr3V
zbGaRVJ&ux@)wTX1^}_r%%IZ^hlE6c3x_$(c$0}FGVV235e({i+`Wh)Z@W$pPFL85m
zb_af^NyDFg-v4UhZVY{jrXp6_35e3g<UvQot@0ux8rXNpYJh$pX&-R4{~L*G3((F|
zl(3Ojz4<hbg0AV3=)}H@CqU~cg%SmqXN_KE^_dXN>$=bqyctqNQ-XQZZCL1zB9HO&
zVc_Omr-ZHUf@KzURn7mojLllx?U-dxx|Dw3etbeY^o!#G|9#(cWI+LWE)TYozRru`
zc-vm<ZJ0~{CnF|spO>7AM=VL^&t*3}tS!!4-lYnSj_)&!Wd*%t(rSVAKWF?rF9g=|
zR`B>o06TtWLX19w_(6^M+J0gP<k2>|V5y0}8K8v&5`Qv+#qeG~TiDgh%Y*%525@{$
z0(Xa@Wrg=#m*-#Zl_-M*m8%xzX)y1fy8e%dbBi8<ZD(IK%jS9?iMOU?w}!;)S5MKR
zY}uv(AwYoM=$%j<Gs5TU7hx5ieP$O+R`65q`!Yw>w`{6L<XiHM2bABH#<c>Et3&+5
z6MDdr1_Y6drCaN=4;G$F9X+|p(=4>)B*0OyjF^+H?ENr%711KR=#-sSpf$E)h2TiC
zD^fUAH2BDVF7$b?Y)%^OWIk+Loh1;cjh|zsgoeqf9MFD70G*YYL^gwTo#SD?j!fb3
z%ufOgLA?Y!%Hz2`quEJo7tfms%U=y^SGM{>`OoIc;B&T%b&A5zc87IIE4`h88&T!m
z^`Y**E&`WODviMocXCKHnj~_O#oH*MG}hEi@9dT5136NVNG?yn$34~;`Sl_gHhYn&
z2`yZ8Bt6T<WgF&`qrFb_(ntP;$|SKC5fAf(n)nZ?mwTHy9l?=uGoR%Z7@p(-^e#C1
zB>a_#EBtZAV3v0U#YyVDbxEP|e)<0_(ErDV@Cf*v76S}^0xtUJg3}r3m}uABcS8LK
DP>v);

literal 5986
zcmeHL=RX^cw^n<%R+MUK)vi(0)`(Fvb}5RY_NuR0E1KFWO4JO3+AXzrRDDq^iB&}j
zwIWDL?0I|d{S)qs`{wuJe4g{1=fydn=X2hilVobF%fiUZNI^ltqOYfIPC-G5q@cLA
zL{Ig%0ydYX{R_8DjVyKk;=Sy3(cf!MSd+rS!ZD^UoQ3zn;Ek)Rs|;t8#J@7bD?OV)
z3JTV~|Bh>e{*^B&C;&V9+91m?#7;p{R*g6(%|nZ`E5~2sIn-?LsQLgkKf;fKRb}{U
z%&tv0QGCmpprOd20)cp`X*_Rmegv7mr~ME4KY0*zr}2S?+j7X4EQ8;9mAoU}atkdV
zs4bFho%1wdPX~hM^M=@e8cKrX$GsaGK-+nyoy^@|nzdU6rekFqinmOD3f?UO+Rv4@
z7vmiW3@<FXTb`Tbk@)qGdJHF9TJtEM$0z^h#$bJ0RdF?@WUn_xOn!)sP;-m@dsSZ!
zLe2#%J%QzqUt4mmY+G`5iN^c&<lH7^prTO*vBeRU5+Q8N3jvL;bA#B#OP&ZPUo*6U
zYw^4<z7QRQ?yseQKuRym4|l_!OQZxo7H9gkh%UD;1+9-?#hojg?PPTQ0N{P50IgAA
zJi6OagzSl%7{ShI79`A-v}|#9&)f}cZ$igRd@1R6sAEa?Lw4l&wg`NCd41(p20uIG
za&n4$XWx>$;WOCdUAtbL{P|{%F}9_B8@>8c0PkJ*tOEB+6T=0bbWp<7m7-qx^5|gX
zJ~U!h&7Hnmm^(K8=mbEj9b_u}GA!aAn&?IK+Ny)4ii)EAvY9GQxTd(PKj`9CiAZ^2
z7kE@8A!TYf=?{%<qlPu-aix@LjctV{rhTouT1jD=#;d<xdOgXL*DdYA#||Y)e~z>0
zA=w-Nh}+x_0Z>d{pMA$w)8!OCY3{`|IYGNW8=1AeUCdx_;Ob;M7vn~Hp0|3%m1&(Z
zYG&gB+Y>LVy#~o%t@})r(f(Nb{eibe%GPY&(fh8=!t0(n3Ebs>YqBOtjy*|_t4O^O
zqj8>U#?gBR)aU1nr_s?2T3VJ|*)%Mb1Gk47>AA@SUtTFC8F>w`)5+(;;hUwNq@%+}
zBPf489Pn#w-nTsq(+~Q*>s1vz@eT3$C)RLA4)Nq+X&`tLjk(@O9$Hs@Ju5T$SVPC`
z#Hr|AA*prwXXBQmb*XDgnZ7#cWXHW6dcJD&ep$#E=LR1?p*Wy+wK7f3KYHV?Un>nF
z(VwPxL1jr5=IWg@@4ATTsOV@vd~b0ca+E=Q&E30pXWqolG+eWLianDDSY}MZo!ylq
z9B5)GnyFpa<L;P;qiQuopdry#fe%W|^-ZB70iPQq2HlUpOCgru3S1LsF!-*e158*h
zC60|8s%JdYs?!uQW8kxIo;>3qKhDlih>Ew)6%x=DGTTk5tZ+ZOD2G16&gMlt(!Hnf
zA-2D!1X*+;I-!jXNaOfv$qpz=NsdP8rq7L^zaP(Ge^+_`0UdkL&55Qm`BG_#H(-{4
zhl?m<ePE~4$51M-6XnYW_Gf%Rsr3bc*V#2rL=x@(JlOx2v_x-qA6<Ff*Nd9c=|GEe
z4f6ic)=I#Azw~vb@E5ee`C<$?qISkd!%+d&^U%9ZD{mQ}zBzpLTDRbU>*hFZ)0Tb5
z^Fr9;J63kMMI9!Iuf2kkTvyJJ2z*E<i-e`OTL&OeK6z?E5JWNE*OrMoH<9pPpA`<4
zdo%BZv}<!7Q2XRZ*C%<-mvB9G`i6RZQdU&mq?z~t8A0X)Dhv?ZPZPz$Q&z8mHmdF^
zXr*{Q&bU)$)n~olYa*1+6BU7f7ED-=YYFGLdRlp3oX-1^?mi}FA1JVyX?yRVRW0w1
z=){ey&MVmSwEI)+X<9RVIo5Q3IR4}kLGh%4@tWPoombte|2Qp$_5K0Zsyaxj?N1Zx
zp43zt--mHBk%K4S$*Y?Sx$)FKsi`udW@S&oB3`<jd&3N@;qxn^x<Yhx^!?YVwp0-I
z_IDDL9^{&){fk-O9sd!$;(a41Hsn4a=jf)wr!`A|+eJ>VQodxNZ|YX0V$CH|QudoP
zHrD=<5_JSFZQ>nD`06Zt6wh$mOonq{pfwuATU%5Y6bg7C20vB#phye!P454u`~W36
zN0gwFYMm!1E=FAXDHb6+lH;Y|gQ$Cz;Pz)Z<@M3KzSs+wTjA}v1QFP$)N`&ImO+QV
zS5ztPUbY#8Ffz`yfG2@6=ROPt|2(Vcoe>O}**fr_%R{w9J+DLc8^EMCosQ)T3}~cH
zhP<)FqO(7GrS_~F^!Bw5RKB${1C5q_<6jVqjlFRq8c#Brp8#tWIhO)C)f33sh3XgZ
z0=Rt&|5kesn>7}eqLEuBD(RuBmN47Fp@BN)9Tcl$q6ln{Xb>IrxqJ+2*ZD`3mqqT}
zMY9p4l67*@o!^nR2sJ(Qz-$oNrXnxjaK~L=3iFG_j5kL==DB5iDjpGERZLGXSGj2-
z;tN`sP94OU(OpGcPeZPEMm7Deb!kSr?E^i8oOio{d7t(On+p+BbO#$~LureVqV{;|
zr%K}D^l{G5Gi!N*E|^o3jJX{K<)Fz^vN@Z$9UY+M?jPdapl6Jw97&!dKiinrLxQax
zA*^M>1KDeiY0XGaAJFHbV3?uPOUh-=kHE88I-gqPy1Mdr2Ry#J=GCn~FY%C(-$y&?
zrAaihbmcPQ0e?<T)L-`a1)`3$ethf2b}GJ|GHPV%q2B*oDcAbQ?+q+j?<T@IasYM*
z)g0Jqm%sQK5@;QiF<(??ec9J7vG&r-dzA7vAXT)j?3k5KJ+j6fe(D#=UzXWe?&r|s
zSRC02-e~dS%m3y<?esi(V{AvgD@4Z`6TeBcD@0%!oLDhaR;h0-0C6{!taC1!OMUTk
zvc|^b`HWdK0NA}|p`;Qg#!q=fi=WQe_B`3XeMvcD(J*Fm^>Jr!utKJvY6-9uRetvC
zmygnu!#gKufC3*%@+SyM>A{2O*kSXI)4M>R<HlS*e!55LcK$_>TRZiH>aDmMlAHfU
zH2E{`y^ot#_Aw8_?q<&2aJMGvU|dywd(aP<4$Z-_XAjc9Vd7y^Zh4r(@Q=lzZq@#N
z9um_i=aA|AW@h#-f!~7qNxZ_FRJ8W^T*Nu^C~>xMf6ZJQdkKMGp3wL&HivjZZxtkF
zd^kEec^I@Y@-^Q6%!!DIkCqJsM9)Q{eXf=#;}w>DjkSA+RcDe&eeJJ~qs$20il3*Z
zS%P~Hx6T*z;6iVUfezjQfZ^$=O!Bk}q9y^SH3SiESd){-Qq7}-a==+V)V}Kq%541v
zBCsb^t=wBt2vQGWczG+A2gtocE|F{c<70K(;HkPAL=1f^evo06*O-w}CA$WL9qN;s
z9|?NJIQ?tvklhZuEA|ME44G@iCZ8-g7~KJAV;kCGr;VL-=`Nfj27>0e>14hcdV-C*
z%|EtUDVC9o7+(%;x4`yO?!kfEmG+1=N`8!&M^bp_N`7{GvsptsZPsAFaoC#n)Q~V@
zm2Q~HYu)O}ut*BBG-w8$Ie?|0saE%Y84j|TeyvMInDsg}{}6~~9&_*%8NI$?hR9h9
z!0dtt)&CJj{c@)7SZxNn3xs|cJh9Js?By-=qx0mnKH^`+{&ZG`4*43P_`*MGMePGC
z+gYRKiSSwc3ml1<l^!_1|7o%0Rwy+%-^L2KvpNA(Z6!047H(*ft(4TxiT3<*JzY?R
zVy+Q5JuzHCi#$EkO*ACBD}xjdy(G5^fVsu()msGRlY7?CjANQH?2tS_zM+^t?;f{e
z{@3|g!Hv_4dL+`DvmeV~<F(}`dm&?;aR-Hf!ptwGNE40s28H79c@T`$-rlbwk4bot
zEt{+ZqvCH8-j_D4sHyQ2^?)kTWJx3V)6T7N&nq$Yg;!tH+G(Oz`CQN<l?;p>3aXB8
z+bgfC%=Z`xu&!3m!R$Z=0$Tb{wv(qbQx@Vo_8s){Vr^P38@a&orxX*qjpzX`tc{Xd
z#zZ_WXO*o8uWhltlYG#^0W1_MT%?u-Hm#5f5g}#$DnC139LWiOAAY(s<YTij>~cEY
zrgnn9$PuoihPo+CwNYCMOT-k|*Q)Vi8u<QKSb%292Ym=rmLI#HY>~PRiR`BadwYy_
z__Rp+!t->o?BOs%$gUI^KC>!Dh^p$|T@cfuceQM1_sBc5%qPp9wL!y=_8^ik9fK$2
z!yl#v06=3~xL>U?bbzT=yQM(p{I_}gVcLeOX|gmOB6DU76OdO@FY#_s<tdu{%r@Nf
zK1WLddzI}yM_6b$%R-$85&n~SE)JNgwT(9kIlba5VN@YcmpQ~bs$@P+<N9fz`<97-
zlRN*niHEOX_C=ssNIh9W%3O?^r~2@9(f6=`>TQ+7$eNSE!XEz3P^pXLR5M#0(UwMq
zH#Ogwp---|U~!}<!O7b+P3KT#^l-s%!Kl>Ho{f82k8UO>Bd;|q_NPtuV`0l(amZ^h
zX?#7)K_rcb`rqy7^njio9Yk@qyN|W@z{z6FP+Cz@k1mYDF$;?*uRgZCvol<+_BMF(
z{NiH4ua$VTA$_qj#k6=mY&W{xOI?Ib{vuxokxBaT&}f_upC84og>`S2Hk7(c>@LTo
zQq?};F`DmgR8kg6F*1i8<({s^`yB*&q>+w#B>jz}h#{Q^;O(O(HW!ppY(f8!q(?RE
zb4)P=PCjP^k&bTG9Yi_M29eehPZj{wV7Sv%&7vt~t$9B6POV}oDzGC!r1qW%1s8~e
zR|VkRj3gBcUp>c<DXnRk0&^2o3Yu1;ja73rA=i#CT?>${Bv?94a9#-Y0sM94q`vCj
z(V{#2rjgz9#sPp(mE5O3$3P9TaGgEo*43Ns+X=CGfM5G3X~8)>ak@T~6;$eQLuy2}
z0ogN&R}qF(I71uvT*!|&K^L4inktJ;3b{W9Dz4gy;O^q3-yID`LI+R}?J~dD7&e~G
zw>g|elO2XfR>tMQ>n<1V-*rrZLe$L`X^L6<#;zdQw_MV<_|p<LjU?q0R((aGr}foY
z=PXO8)l#O8`vpp4R`=rzDX9-EY_a#^4H#51DgQQQlMmoHl`9slm=5JV<M-0U@@!Wx
zbQyANgda#R=3Yh`)KgznE1xxBahuzchW*gfeC#Th@vm641n(v@(0f_jg)NAd<Qr)W
z61*N7J;^BC;)rFbIN+~+(xz9*0l}F+0g<Yxecwvg&nkXJT`7kb)zA|}ut&LIte>99
zn)TOisU0@?!+iIZE|n5D5I_IKbLRNHolRXnd(aAMvkyKe3<_QZ{RTBXJL7r!DF!M1
z1ZRQgOGI=_{0IlRj`3lo7FY+kI%%^i&?oD)2JgFle;R<McYGuOL6Q1go?c@O?>3_K
zGcKNB_IG@|0YR3n2q<ku`8JUQF%qaXW15$ty?(Ro*FdYE&EotLrf*Y2FZ-f{<S!)g
z#0O?EalpwD<d!JldN@KHxPIS#u->4Pl>r0<=5JPO=k{E(Yx>(77eb!HtQH?-a!koJ
zE{6TMKXS2C{3Ab0G2kauddZ@-d`JoBmp4*YfY@{r<4V|i+DznNoOSrMgrNb2n)<hm
z+M!wPuxjpI^~K5}R8u<5%d);MVBzGG7G&9pop!|X-btl`?Z8jDM|(^v!=((LV;n8{
zB+s<(?8bdaX>5fg9cj)qngxGBeTXP>ag@F$^}u5Rlx8$t-ylAdqTE%gVo^{LRlN9A
z#&DsS_fc_?btU(WOyMAj;Dlj6O@35a`#+(NQBOmsO3gznDJ@<z$rsaiq<3hQ3I09R
zKN9ug3Q?ldvFj-2+L&QMq3Kva8}sAGA9Tg1rs0r*#Lw;rPHB2;>EaT2rF<XK-MZvq
zK`z;kF~}n<V@~Oz9u23M0N&AqX4EX=BjxpHfHry^EZCC2>Z|4QIan0xS8}R1%N2Rd
z?LMr&RzoC|UFx-4<Ov(fxx$e7l|XZyeC#(Z8?IMSJ)5VaC%gmtQuG~h9NNRR&Td+7
zP0$4b+v!SLABP|P9XVf4TU<uEakX*Koj!ukm`$uEgFam5fAF>YI{~~OnoX>ym_D2W
z%~quWc}*Bjn>5+mL=%*CKQgl5-(~dB&U|GSNwP%tsZ}7miaa^~=DBPIOFo^AvFeoW
zNtO&+x}Jh~1vGord!!hZ1{Lp{$=K?cVtU>;<XD281$-l#2I%st<QdD;9S86HwWx+(
zg)<`%ZY;^!^%G?BmI^s{o1;GxD8iU!KfPMF#V0b9-=`2*{(^=DEG1f3i5tx=;X-t<
z0U3{ZQY_*m=6@c)|8+JDZ(a1+c)rX#tx&8$n)0iXt1g1~dk<g}(~pIN^IvjEX9vD&
ziHT!L<xUQu^C&Swf19OLsP)gdRu~|A+@3b6^FVDx-CSlMNp0Qgn?IVn-bU*vp-=w}
zZIFlv^DLq`f-5+55Yvr#=++odGz)5wfain+4kO1Lg#E07I&MW1HxOoLPI!Qhdgqcm
zwJ~ptV|=aURG31~Z+B)iui~I-T?v!R3J%+SVw(kMLnyp5?&l|VoA;!ixO%>NJM?*G
zFXPSFYpRW@^WQk}um;7zLxt4Fr7ow%ktD$lUX@b(&wm|Rwk$Cmy04|(SqtorzLbz?
z16yVzR|u~OH%aSLz4?}xvgzKBAP{1PMM9Y325hnE5I#v(@oOC_3DgZ!6uDls5xHQL
z_qFV~T6m%Q*G}cA^GHFBe9mrMl%WYfuKR|DBmH?pT$@83cm?<N4>6AZG~x{r)hl!z
zVxnp-#JnXCdwWsz0oI%7{9@NsY!q}Og0fUc5$B8f8tf+&G5}RV(K813I2JNj>|&zA
zCr4>mRaZcwxKZ0+$asi)f#PIv7TO1FB&aFVGm6L_TsT^|uuysxD=(}asI6y)IW~V`
zXiN&Yi(rlJ&0<V0tiOxoh`ehfuZj5_K}o;KiKDnUKTIiK?gb=Pr#))9>t>$l7gGjf
z|Jrl|Cu3+l|F85z>aaq77&~g}#&XeqvWm(50LM%w&90dAq{Lh<O_m7(QN6cF0Tw7!
jo}TeP<p0-$G2q@D72$P^G4J1RIEB8Bv39NI3+R6V8Ffz2

diff --git a/man/figures/README-gaps3-1.png b/man/figures/README-gaps3-1.png
index 05b4132551f045b32f4e9bd2ac45a4b8bfa15410..cd6abeac7f579956e6ca5895db2fe0254d68f3e0 100644
GIT binary patch
literal 4315
zcmeHL`#)6M|JIn8AtWI~h!4^Ul}m2TOpaS&W<)wgq7vpPN}XKBXqR+3$E}<kh1rr^
zO5}c<92|5KGct}#F-IC^ltzr^Za#bF`zL&V{k&d#zxG~xt@ra@>sjyTS@W7Z1e}kC
z8cIz`NlD|7ujdgZB^aorwBcWbGV~?jh@A@b*hKIT_JW>DN-~-Jmi<`1oCLLy3BJb@
zm6SAl<@bhx_{y_NN}I17@;n%vRJ1fba=Lnto_C}JS2nzG`^d=l9l)1Q_@9`4{tJu;
zHqVh`r;A~xal`{kbvLm)vi^5r^P;N5HUVb6!5w&*SA7+^1b#O-$Jvmvj|i5{)Oyc1
zHFS$ymI0&h*^bM*8-|Rxe|L5gc8PC?W0IPuD09Z%YBrIRzYhH}|EJH(zB9w!lBMon
zQZIcv<E)X!l~7mXT1|Y=8Rtv%4f@ZbV_f@g@8q=}Dx-C4ncg&B8*ccf$-t0Dt@+}%
z+O9f5$uqLx=lDJEtB@55+<$GWlfS$aANcD59_Bb4;9ENR^eQC}A3}NhU8SM8X!+tj
zJNwsOal0u1)_fQzNY@zay1&n<fJTum`pPDXwemgg54kNatQa61KJqdo+a3~A_`+4o
zvCP<}?zjc3Cv*pzJNFSSZA)cfgoj!{ZXM|>Y53;)#UJ+WxcT+)akMhvy;UvWW50zB
zMNF~9%PMsXb8oFYnm@Ywi=SY~4M0U%e)>;W<jmfd-1=H1W2eI`PljMKC-vSH(6FF(
zx#bgj>u)UkHE#=!N&RV^BQbC5NtRzxFvq(pQ{B><#pD#N*b2T--Mu1jBBJ?{4lbwM
zL#BBwzl}8ip>C^m?r1`fb;5&FgD2%P$uG)nki0P;>}1)mHhZy6Y98yXU^%pU()^MZ
z%<&D*w-nE1*~iCkt_*Z1s2yl08MjJfMFZ8#8Bmi}oKT5}H66ykRxTIvUXgOeO??%C
z?<1N&>V!*orwe#w-p{xZ41s0U$wQ^^Mn+axHwQ-EM>yqRYeQq!x!GUFkCmtYVPx6)
z_0akUIiWpgdVWRcV|X29wef)$P$L@W5(}5_VKR*cW2wB<;PUibl4WN^X?j|T)bc3D
z#Jm=37s9@jf)VPYk&J|g%cn7!ddJb~B<i|KLvlb;`F&C;%YJiA#`}!%wz?|D%JKQo
z+wprIF*A-?H%=%LUhLth0k#F1`E3<&Vv<S<VA1$V(UI80`DK*3Av*sg+K+5L3C*N9
zVfj93R)LoYD%+Lz70`N^eu+0GNy^&=74((CXj3|Zi5xgE;I^B>>MH(MrNKqjmA%C^
zKvU5LJMd?;@Uk$QX!v&Cn4NSp#YSFYMOH=$$Sa{}>Z>Y-O2bh^;|*%=@I~P!IBM3I
zP4KH6NTirBnH=$A+@8=ZtlNY}Y*>j{GKD=U^65|Q0fxWL3|4Sk^d=*mU;&QB<iLyX
zoEG0<_UXd5f4oKSVvP!~_WsT>F$ke=<+glIn91h0=ubpA@sSl9Lg+5|aUY*&8b{01
z&q|b+){~#|9~p?^-=wM6?Ej<28x*BR4&IB-x^W0p0WJ)t*iEd4KC8=WHhdd0+c<U+
z3>soJOog1du*VR%@P}&cgcWcCE*d2YpI^~JFihztC4t6dw~QNxHO=KA@|a}0=J)GO
zt{umWC|C_xRx)M!vnC#b_HI;&f+0BBxgIW{6NPbgR87sSRFJyXMow@b+ldA;u^1}t
ztIfP6p{(cEdEV8Y&_Ve=z=0PI63(jQpUV)X+hg#@Za9U@t6G4Y7+d4nkgsetjAUz^
zGpBn=nVkdC+fr{#N|eW#s#;BU&&7ND<Wa7X81>3OIn=-V*R--zK~ab4Yu+L7K|ENz
z8&t!P*ERM*u8@qI7&IV%O%#IGFvm{ir3fuwclQ)CLa!s;Sdacsm%9iHX3k}tESf!P
zY<xhwa@MF)pc%71$hU(N1pyP?Fi|+#OT~iFyPv=jGkNbmIl~-3SO%ZMRx}q_4ro_i
z9K%xXTNlrmAQ;OJ2_3Yxl`01d09CMwXO1u5A+PL&Eo5=wSfbyF<Sj87p<s2|3*(Mw
ztV5tST4&J~{5<eAWfe-RTX~Lh9iK)vCG<x1RP~288-W7Gf-1~W0vqe#epFOp4$@Jd
z(4_w3X#^udJ$UKGXFGWn8d>I(5X^2x<y#Qi(6X7IgOcS%7)bW$-q@I<sCpmpUfQsU
zZ<g*s(EA&bB`RWasJG-jo*X0#w=7jy<{Q;r<I#5VRJVI{?%IH(j1f8N<0SA%no41p
z!M#juR=v)!`h&N{C3tSr%pw&5bL`jPaYs5{yIUg|J6%hIDYdMisFMG{iCZ;;SypuZ
ziuGRHE|}wCTxG+xU-)eHO$Y)Y?grBfGI>oT6<F3z=Td)4DNQk#S?qh}?AY&DvaDRL
zVY4pjK;yWq2H{Rs$itlI|L|KrD~7#9>gv%i*!r(SV2^=jY6rFR#TjJ)Rc;w4Gm7CW
z7H|n@UP`>C7+#7rQI8o#o5)6lisk$S2I$jYQNmKp31x3H&WpcrqAtEiFt(Y}caTLS
zGeX(x3>&dgw1PrQdS+T?<O8{TTUf@mT+{r&Iq!+=MfdtD5|t<~N#5i_L=596s%Z-x
z>a%0ITr<KPFQewg=|w@ikO>j9C8igxdO}{`Ma6YPNBy}4ePc8Ac2JP&)MMJ7e>qYB
zn{^-ZdjgR;*u@G;=bwWhK4jS7mJh5Cl?TI#m-X4LihMyjxHVbW5}1=6j$k;O#0-g@
zR}|s)Iw7a9x6o9~=E{q%^!BvuPC32Kdm~vJdD2WoQ}a(I0BRcUn)R)|^*o$dq|tbe
zQY$mw2A~G4p0{1It>Bj`PAMzp8pgsUR`l;Rj-H^mDqxEzXTMTFxkuL6OW3rbbW_4y
z7^o}SRG?6Q5u&jNmQp3>`mZYNwAhHN;M3VP@}|9HOu^lN@)dM%j-dMENjR}mrQO|m
zF+yL@6YTFs7Soms!sNvGs-z4lAZD@KY0LVGBhKg%eQn%#Jau{f7FL6bwfg}1lof34
zvwhe%-a4e)2(X~F&NdsHRR*MZ@g&0{@sKKjIt@4pL-DQ%23;+9_5d{$Vv?Q_uJv)@
zhbKIKp$W_}TXm7XP>$zn!HI~Ct|!HEtp0B7_7e|n$tlp{ezYb?sjMulMj@6XU(_tt
zk8gm;$C`V>%4r4w3I#NidXp{S#MkiZu4lWT05gIfT^|DW|4EW8x`U=57z9NA#>)r6
zOiz#}^Z+xlphR<BOCGw&t$?QXx+LM@(|X`i-n1crO4h**u6eT_e$;-a2B0`PxKS~&
z^?Gfn6<rtRxWy{0&)};C^W|;yID+9@m>}xR{R2C_x$-5ZwC$~w>?e8iWN35&3hdCC
zGW@WmYhUcDyK1{P=uHM&7`6UWuIg+QfJ%jmi{gUWs*%5u2T8?EXPC0SllT2w)qTSS
z7ZDy#d;%8*vfb#@eOF}zs(N(jOB}wH^!+6JymT_QtXoEDmq|XYOdrQ$v@2h_%!>XN
zRy=Au)HxzN#Z4Wwj}}0F85hvH{^1!xC+4=zlhMyU<5X6my#%5~4erXkJjPy$JPeB5
zDj_#;vXn>|VeDG;!Xu1%aR{d6EBp;il!eX8@^LvO69=+Gz}0b5SWtZny5SnxoM8R~
zor><dfCJkQ41r6j^e<`xSTw5vb0i0B<E)Yg5xiH@rQ_H#h#w>mol43NXOqF*pdNt2
zSQdpJT`zBQAIg8gUTW6gsUEWlK;5YjjB#E0hG`!@HYbkl1o}5u<_SzWtJj|n`4-To
zolb4*F+f9R*H}^bCCX&PGv0)AVdEPZ$_w<KiobVy+=6<h)8<wDH?g?knzwdk9<M_{
z86WcMAf<ojs|XVJ2SV~vzr2uKLGgnU8UZNl@f#Pyk~&v!uOGWLy0pTb)P<nl=&t#|
z(o_UzkgNlFknx}4KISJ>(tJu1ALsl$-*q12cO@21{1VVA6(+?Da#HV3e@ZWUPU(WS
z-fi|D^*M4U?9*!R+9j9TVv$?90nG7;r9`?IV8|Y~IdunZUM#!s<6{U>QoP5Qa}~^c
z-?<)jCpbN)5mNPVhSR(E;|p`A9A)EDzD=De!Tf3u%@-s4t#{t(d-&rF$nShC!=&Y+
zv99yvgFU~x9|W}rEG3!_vs{O17b@4BFy{{*PiN!Ok0DG6WuBmjU#|}*8Wkl-{mi6p
zlhED_t5&C*h!2AOcd(oW=+qomsZQ5Q^#J?7L2Y-ay|Fr7Ch~xOMeN!^m}6&1q!%DT
z8foUEw?htQMv%IVZ-*p>^Q9O=Mkg}8qyH)yGM<Ty{US(M{iR5kE;lHUuDkY}f`dlv
z*^$=1gJ7QSQ4+E8Bj~;U%#}ZLT?aa%^IT@<rZGHja0{fnM+sAD>Su}$f(_%8gCug%
z+Re`PX?<WgJg8><Z1l?b%WLgs^vSbm++sd@TATe=rCpL!gX4^v0K;VHJv7*qa24+Q
zUY&Q#EiJ{IKKT-xti;X1M&E)(NuOWCMhgKn(;<9y=C=%N^dvAd8h_UwaruEE`x$=7
z&3-GaI7&st$0h#Vb*bkd+;vc$H&?tlmZ1v_qhZrmm|e)jv4z*O8-g4s@|H`JP1XIq
z%FfW;??yd+JhCCixaV|Glb{M|5o`SFagcT_%|Ywrx2lKkN?sct`V2+yGF!@>*Is>e
z*VnBqxsg<1_AWcd*jiR~yHV$%Pt|a+=hTKF^WgYhDBX(Wq-tEBN45VdJ-E{~dPHPv
lo~^7L3s?K!_5Wo+rZo40g0z_ej>-R|IOK)%eBlvs{l6(e$;to#

literal 4298
zcmeHL`9D<q8`lt~j7U;rjiOxJ#ZY1#dyB?gw1~8*h%|~qgK@4&b%osAHp?8eOl8SZ
zvQ7w<m=O_Su52@QW;AADzGvq9CwzbXzFwct>%5-xdCs%EpXYg>^E$WjI1ki1^g0<C
z8PxuLuHG^-2uMa|&A*DW@SQ{6=JN1G3GeCa247`l#A3-7yF8tbfhA<bKEFsA8RaI)
zUeijTo|TbNIlJH0$u}l@W^m@ue2b0lr>6(R4)k03r*5PJ#<6N0oS@3+{`=k?$wxYM
z5PBD>4l=KA0NYakzkrOhI|GzJ{RVus7oz#J3*R2O$-`l!5Pz@){B^y?cx!LRx~bsS
zy=6Z%k2JjcV#f=q6qbG#4-&bn&E-8SeRG$dFtnZ#OZ#*B76J<w9+q2Ch_8nn`Z!pv
z>iVmz^|424b%{eauI6g6TTaze?cCCLa%-Im=pE5|H*%NS%NMn&G;#7lo%AqAQqTgi
zczM>muHozhL#@UFFpD}Ko9>{q?QPsf>g90fb%$)?gm>b~m#v@LKAiqZrYFVaI$b&w
z#8;9-U%uKX4x_H7W|mgP6|cTBSc#8zz^`+-SX*%90Udx;9~h1HG76Fjlvf(NY&?rl
zhBer2H*`3kIi>soBM`mS@Cid!*@Uu}8JS08Uh&+QSCeK!0RaQcwFmODluao@W``H4
z<lLjXHy5+8US>1(nfoF4T;!gQgx%F~I@PY{1$#}q$eXIimlwAxTI=_cH=!TAj?<~b
z;{-J(KVYex^Y#4nPdD~lX(4D6K}@->J=9d%BwUuw?b>(I!ml0~TgE>6bMeuYh5#<_
zXbY0!WrK-GN!fn_D+{`hX5TFkeX(k*7_(X^^JC2w6{;-)?4aY=uGtL;%pz`Io;h>O
zYGz0eZ$N}LUBAY+zszsr@(zz9W1&rf(Ty*#$nMV$+cspjGzqzGll1cx_wb}oWnMA7
z0uj4O*hi^yC3%cd=6k|eeI6cN>A^Z4grraQh&n<ZyF|N>Vt#;RyT>zznIw8MW~CBB
zZYA5s?<7sWDWJUiXq(ECzgOiM({|FutCp)%mzK|s{+mv=HQ{zmeEV(J0g`U)+Uw=c
z&zpqJUd!9@O&1!oR?v^n@xRP-1e%`XnBg4d#YgugHM*oWH1XkCKwEI`D;4G`OB!;{
zJLIj@-w3%$8<B%UeTqDY?4DQR${!5rkRD8s2T6(XLt|DG>YpxIJenfs=#E;MOA41Q
zI}?L!ua7%lZ`O`7<e(}>n9dyONu1F%Z_C{4xN(|Y1IjQ`xuR=F)D|si1ajB^q%X3s
zq=v$Q>4cDOTsM`PGMbJ+UiHV;75j!JeOf2i|Gq9Fdqh&_Mh8ameSG=^R-<<=s?JW+
z^N2n}el|HMAXg~r1N;QU7Gh&s+QLxV^SQ$*7cI7eI+%b2pM|{e%%`+q*if^19%ZbH
z|7N(6%TxM|rOI^d9j?`8zS8)9%94k?seFX-yH~G=M>$Fz))&c!uf{%`)3xF<GliQ&
z2^N00fu*?{rarklt{H-8xoT53p&^iY91QOsJXMCcS(GfReZY*;{MI`%>62nq)9#oC
zxDE6fhs(#xyG`}*Nx;$CeEXFN72KtY@ApH)E&RX<C(U;ng?*iM|1Qggs)T&uu>!Bf
zATWy7`5lsU^FM?w2zZUoMPucSztG<?v*6pF2~ooz;yTuxCgOxZXE5x`A;=xHZD%u)
zCo$5&le6<#sHF2`1_>6Aq9M~9)bUu!ybkX@?K9EU8^hjd<J_z1=?>jnEU9e`W)YJ<
ziDbXgHsl_l9ILc+<H3cnJH2;3_5E<&=a!}DG$JI3cDl>H2beMeO+P?AVMV0$l_!5M
z7We(LYQl7+Am12)sg(`?SLbd_4A&V-4uechpf@M&kL%%^E;kyc%^Pn>4^-@qQ1%rm
zQ&)cq;rS66UOcsC#f+k#w2?ZjN#)LzTo1jX1)?vhy)=hh27T_0D0aUKgGxSSD$jYz
zg~lAB4o<}JeINe~r25NOfAqPq4}r-x)FO*TU#6>e6xk!K8%^=wu)*nr-xjws6e-B7
z8Vr6=bRA!@dw!j7xmUw;tOgvxqlldgM*cW_++-vOplHC-r5Y)DGzK}hul1ezaCZv=
zGk`c;!xpp!>oF8>B6sLE+P;pHN2Ep|eT(UIX1{NslpHAdh&AC-793Wko~4Xvcn7(u
zP&4A4tFWy!?ogwYk$B`Q3d`4|UCx*NXR7=k$^5vEBbzrfPKSg@N|ND#e}PT<5}d5%
z0=W+%Th|^OcTc~`^CnF?tVdu3C@zO};%u&gV)sY8JYUM(n9sRf7g*4uekMT7F`qNp
z1+cUyuLIFPP*7kC(~|tl(!>ND>Oz>*`eo-wSeGnGD!0CXJ+y{$t0`lzD+mzHH|#zJ
z(At8R{=mWfMS$KA2whUOi)zLuP<!^n+GHTL5bUX|Dyl6Bk_FLKhJ|9S5ROzQ4(ypJ
zPm<>H3`toZ@(oK7cm9`y{h`AejJK2kfj-`VPQEI<$PT~!L$RBs$1qC}#%;qJw2?8w
z>0oJaMvTAWlP^ip*<)B(nRxAu1m}+yr<<`ZUQ$Q*Wy3{Z?kXZM#a_o=(R+vMUPP<U
z>ElNh$!|K&gf7?eugZm<MOK`lXiGVx7<X*K<k{Shp9cti8X&q?wx~0hH9x=V3biyN
zZwBvtN*D{~chiV!s_>hw5v{Pi*)fv66M=cWz5-@)m5>ilkvA80kN#jwp)6K7zG(9+
zE`|g<{7i1vU=(kVbYg&V#YKvPl+?^TkHxH6S#=_N`<tW#spk>ef*Ah+8#r9~<7IRn
zDf$kO%2B%?06cFZ_DT~<0XI)Z+HuN1_89Bq3qatxkt6Y>q@S{LUw5n4^CrgOCVjU#
z1jIIlOM%fqn7&RN3YFk>c^96C6%wTmI&jDSRU}nPgXRM{>GKZtuVUboUxK&|Y3O(B
zt>BnP$sHcc{$>&~-7Z@<A!*2;QBa!4a+dU7$03%$^6QFDElGu%r@=}UZpTkKLkaQ8
zdv%5=irp$o+|5Qk%`j3n_>x<Jy@eziUk~qM!Z4;RJxKykMS%M1v*<%|q1lfaQ%<6o
z<jmFh@tbhIpdB}Ms!6B@qMboy9=6sIHerdsc|SQzS7H~x&R0CN=nf@&LfrpGR3%iB
zI{>NbgI_0Z0jN|UY_%bFhg|4KxtFA)>Hw7N3YkCz*v}&iA%`m)Uk`_?vU5F)44`fU
zY)&eG0)bf6DjZTO%*@X%6}$K4Mr=v01+V-)2qSVqO}UHj&VT$<^{WzyZZ*<Mp6iR)
z%B;jeR&t?Q*%6zAm)rOnDXhpeAk{!kxlQdhvKs}Wg=&wx2E;lFq(#O!I})CSLFCNZ
zRpr&uympQj3+5>{h)si~qVmKJRcfj2@VwveMt-vA)|}A82L(4S{77-uO(C2Z*iG*=
zhXL5maLZ!?(HhK^le))%2a4U_b0Y+^m5f$VO_O_)n6^|YXlM8PNl&sPR!2o`O#$17
zSj#0OVw;u6vaVd{#Y4NrLp6A{kXxotnuJGt>&6}#h+rPIl(P$r>S1xmGT`i?eAx2C
z2!2s1;|;bog{H54o>n<4_ZcBb0a8;v_Be`#eyk(V2;HP(SD!z{4<M;hGd8gTDUKRO
z3}3m>LfjT23QK^rhEWJikKr9IpZ7q8HM@0(38V_6h2#>rqRE&%Ay}4p9M<;SUHeM(
zWRF98lu2hRGZ3!R@35Br2YTSt0D&8Oo9#W->jiBF(F(fTO}#?;??efPf;g$$w1$He
zyLbPxzUMGw@R${pF?$4l^bhz^;*MR{0>9d}w2tyNPW93eXg8=OulMIsdREFc{>rRC
ze)(M%Jg>!Tt2wn@lN&!H&>m1r2FJ-3XFn9{O6wV(f49P%{tIsYk6V|55%Fu^m5~>1
zGmfZGOQXLMcYm9(qW|7Gv4=x@OL<z2L12Cv4y-2FSFq<DKA0^!;(F*9xzK6cyQP+Q
z|3oL6(6Y@RjP$=CJ@<8iy&!v;br)n;sz&#A9(^)GT4aU5tmt8M$3vhQGHN+TiWP;^
zxVMcE7;nRh3~WNZuK2|QZMDZ7M1trSYGwu}67DpO+x!~JC&OP|!G_uK-Ye5g8`8h;
zv+SK9)d@opiu-=TdgXEq6VB-$;2xHR6Bnxhkb4@iL^$^w4A`Dwk(>UCW}gy_fWk|X
z<_{;R<p#jE+AATPm!$Rn&QJ><mK`Msy(9_yogsP9cpH9XnjDu<ser)X6b?|7VSZ78
zzuCi-L3FUu6R!>MWX0e8pvkmXp<ce!s1RSRhac%CYcX?RdGgNn9xo$U<KbQGApCa2
zZu7F>!z8|PB8SU{hy4RmLFYXYruuMa?C~KDFl|UTX<1;$R@me6K_dCIW6L=6^wPv?
zeMaZ0dTjTu_-X@21}aLR>!0)Wxg$cmNZxM6-&YSGEoaxiKB_qhTc^rm1wi(RJP)YA
z2P{k23_t~3;H0i)No?44XQ*Hl;NQ)1R*cF-vueEBHM`Ui+B@X!gj(m?e{^j?Xg^(B
zy?ju|dh##8BL#?d_+|cQK<G1QWYSlCcHc5%AY<#Z{7!7hY_q*jXNZDz^y#-RLPb5~
zZ0qQgJuhxmoO@hjCiFe5lc+D-UN>%~o?){uGM#_O?MZ6qfPcM@>ID~TUZ&XBwO_Hl
z&iA~ret6=zj<KzJ>R(+04sOrSY&-m{Dr{YDj@66!d#bHzpf%#g|JVPBKn2OIKexwq
TJkvz-Ph!6t&b8S2)Sv$WA+XVu

diff --git a/man/figures/README-smooth-1.png b/man/figures/README-smooth-1.png
index 821c2250ff7b846912ef64099374cbd633d439cb..9083bdb321822136df94f1758f1b8e2cbdead7a7 100644
GIT binary patch
delta 2613
zcmV-53d;4*8^#-uP61=FPr3nrAQJjOR`W?NqHOch%%<~O?kn|0$Z;D^I-zb|co4hM
zBope_hR5)m%t=E18u1K<bNS8Ui$zqP<ycmWC_jmc^Bl^mYW8_;rXx8?sB1ffxlZIX
zp{@;4Wjl{0eKc>}5lzP9NFOS(h-z^j#(JA4qN1!vu`Zr{UX%AE78X%|b?$T6ETWuo
z(qlh{O#@g&<uHWh?35@GRlo=`NAtAjeoWw<`$3dN^ltS3!}$-Qbm(TZ|I@-EDvIWR
z+b*J<wpT&#|J!zgMN|Zx|FN)$3Zn5VyG4}K_5x`8+O8Yy^Y^3a8%u5xrCB$i=R4Vm
zvWWgUI)0Gf=l}4H;7!ASQ&>cW(C(Do8K>QYfB8=A(e1R|IZ+nTyU^<soVM3Mt55AC
z%1v_aK%+DE$Mf6_@9)v)oc%({BKqg(a$fF~6TSNbH2F$qpJx$`M~g49&)<m#7hs>i
z5B)8Yi71O`D7srH7tud{D^_T3v3*2OAHhJhw!-<xpR_Rwovm_z8c`O}AoR7;_7)H8
z2mbn<c%ZG-&hC0kD<ja=1}ArHWSb8_Pn)d&hz*NqBwE@i{dbmV=V)}aS#l9&n;(pR
zwo4x@(b1u3XNUB`5-lBxZgydtk47*1u+0afm7O@6k3}cDvCT)KkNr+QSNi)mG9%hJ
zK>BkfS~~(=93u69gD9;Xfi8|gLW9u6L4LP+5;_t+9Ojr%n%j>aj`P2lETIF@!g1K-
z1JJ>N*yP8fgCmj9;b`DU1avqWI1~vTjsA`Gb0kkf$Dn;<9Y0Z`!(C|KU~KYZ(Y?_~
z=uk9oxWflZ^tc1f8;&J(BziX<NAd&Fx&UnQ<IuVQETQ9n(76aKp~KL*5c`DE=2mnr
z1OXj|&c*odD*^3A=VB1hLFinJZ9wUAGg=pefDS_ILhKHb=yEe!7osPi%fhYke$k2$
zEyqa9x8fIApc5gg0=f$3{u_WsgeVT^>3+^&zR-sdr2);fpUHfp4Iv5xn(aQL`9K#!
zlm+yn`OM~jO%r02T`O|gdxk>~Vhjo>)1d`1iq6xozxsNu|EhQLDW+&agpz=+?m_Yu
z&e!W0Q^rFFA`}F4ozL~xcM-Lpk9`9|Tpz?A@8NXv1@{5D*B|3DpqYH`7g6&GnU^2r
zDxj+`q>hc=B=m=O^+B%AGF|0!w}@KL$h!C_7XiJ0&gXj@QRg8!*B<8LAU<^ff1Cqr
z=qKaS<KzRHIzU<ye_);^^n-Efak9rq$qoMFi-(Bvr@i0Kl?TfGO6O@l->HZuUS-C&
z@;JF2K9|q;F`}=`xqc6}g$L>g=y4$Kzqo8$c%0lxrptUD7EuC<!O9lU%rW{lqUH&8
ztvk+tPe&8UN?yX0x|HO&(z*`HweC1&JN(rdq-CFT*GZX{9jIuBFE|Om<i3#))3W0f
z?eJwE66H>(+Au9UPSFlu_7TzbWzPmYtB#XfKo8;FKYsGPtd~&Fs^bh?Kv_aRSQZ^8
zd+bd>8(0<{Cl}D_d|qCe@4&I<Ky^F(Rr3jdsAtV_igx&lt327`Eo+W*$Y6GdzwZ4m
zppMfrEICf)AyRQbotIF<lH=qS&>;ae9G73kak3GtKZZ}e$ZuSJ6~`IA!?VXbRvf43
z5Q%`+@vAsY*#b&H>-be1hJZ%!t2j*IF?`(-x&5hnzlx*ekGPu-;@^kV`;E)5;wS`v
zG=yKpQSv9%31}m~ih~p`pw%A=ZGQ;$8<$_jK?)a80$Rzh;vfVxhF`@oE<QCUpsoBW
zj&X6_QR`_?4wC#Tj&bpkPwPQE0rjgmMsPs+Bne%vt}G^?t^6tuadlBg+tr|4EaF&k
z2m%_#ui^+-J3Ikx=2vlq%N!BVR(=(KN2uT7>n;@`py<)C!xPX#eia93*x{QlwsRbp
zU&R3kXdJ(a$Gbj~X*@<^hqo+wI072Tu;kGS_TkQe8t?FqC688cu9AQ@Gc0+uuz<3|
zzddUnt7MrYpv@d>9;@Wj$U}&$0_wcOTh=^O+X6~JOF7m&6gzw@%c4iB+2Q+t?ri5d
zE6bt>Dq7|UXgkZI2P(Q?vnimhmyk}zvgm>01L~bm)2fFlOHoTeSwO#dRy|D70!lzT
zcvd}1;hCVGfY$yZ{cJp|9;Gs%ETHvF%O0d`hbN#NOv@gl^s^oU+Rn7>F@AzK{72Bn
zPwyvWTJ{j7!5b1#*Fm|~Jwok&4qu)&7SJEAb&pW`qh<YNNi3j0YzrTtHlX!8JPW9A
z;RA#Plm+ycZ{__8&w>)r9=?_Lt2|9#{3Gb^<JQ;Ge=#n-SMxzU3up!7(t8!|+|}zL
z`CN0Zy;JKBPe5xp*WRh}TF}zGk<Sh5;`;<$tV}>Rtc&kc6wvksltcJ`Pu|sc*#)$8
z0cD4;;$3~0@7C2Wp!t^n=92~VlX>|)3Ie+NT6I8+o<2YG@_RG}lzdikufM}Kpe&#<
z-0SbK3uvN~)d3}+2lfqUcKoDs-^*k^31}Vr1~fZfRJ@O9>JXlMR-yx~5&_*tlm&DI
zI?yU{3jXuUKN(Ffpt**Blg|WN(CQ?h@Bhnabv_Ab3N2`r4&e7=%SnLC^2vutzvw};
zRYG4UtLWFGsh8xw9}DOTJ!rS?-EkF7JW`fVE@N7w2^}^GJ&Py{XcwB$qadN}3n;Vs
zKoh!b+LwxGX+Bv%Ptb)v+x{IQTAa_sd4ERpfi85~HSqQ6=Rc=^&pZ7#ETCJop;xMd
zuT4(Ro|XJA1aylwbSr6LaXvZ6v_l{IUH9<v87!b@=)(Zz`MZ3afbP(T5lWAME}uX?
z_h`frh3DWepFu$PXv7#Lmw8@4gM1#)i9t4vT-GXHOF$3k#3*%{+`RJtKVAg%fKJ?x
z1vG<J+>d}}(2CoCk<T1@aXSK<Loe=@`sF32@`F~~E&bbd%;g)6xLy4(&^w_Iw@cq5
zhsk`Q4fi{_1q9pt2bv(@>?S1)haN~cy^9Xhp#>7wchT8CU_T%G252}rP<k^U_xeSg
z2h@E+=H<&+Cv?AtTF=P3cqN&HzN@b=orrJcVlM9T)N&zz^b{Ox)^u@~r?2Pu#nh`>
zX}g4Gp6C=)mpUI{e;cx^r+b9dp~j~;+$it*74wg8{t6#tKMW#DK&2dqK}6@S?r9QX
zOZ<tOf>+IB=GvYp6}H3OaF4HwdOP=<xU`2lTMx4_op%wPwy+1wrR{aRyZH5so@~;x
z9_~7A`M<t@IFTufnm|F?JXX;OOPWN<D%$X(r<tyxi4;UM@uG-#a--SmnagEFQ;TTK
zTe-_*<xJ=D!lmP}Qrw%l!(`>mDR<xI#L;~C`}s?ARWrAfoX(xlsfzJ%2<g-bov9Wh
z`QHzjX{a;R;v|3h{X|ops1z^xU4EXi&QpmSlU;rennL_!M>ow}rzymcMMAR-46gwc
XW!B8eoQ&2N00000NkvXXu0mjfoaZJ(

delta 2622
zcmV-E3c>Zp8_ye%P62GOPr3nrpx;_wsV_p2)qH*z(O2q=5M(x;--NP=K7-qE(%ZaS
z7aqiJG|7bewc#=RCUcTdzeYTR;aryF52B8pcmm6@tQJvz5*6n;lvUO2^V&>Da*|Nj
zb_jEw$Z0}d8=}f~9!vUY-nb*0jK`5aRALd;;yjG?Hcv!FS&w2}Jo~(VChtisETZb%
z=df8sIpd_qehixiu!zcG2+P?iQ6j2<5oC_$Y0v$bz&rPYD2wRb=>LcFA4KWU&1nCp
zg+){p&HuJtL^*A*g5LkP?F5Uc2s-~`VG$KX<5zZzD5vcO(Dt=mH`wRzN7FZ!+#*V|
zZa~j>vJqtw{d08uAivLl|KS_Kn}(;bhzg<IDZ4XHy9fXBo!Fz>X}fcxETVUz*C#k_
zuYp#d+DDX|<lKQqXY7yXxf$Nyqt7|}g_1?|&(Y<)+$kq|_XlY5mCQcRA{vhtUt*uX
z6AdoFK7SwjTO<=v7ST|2w@@ykfBaUg(A;ACh@L)zfoN@o^N&A&X=4;RTjeyOETTc^
zYo+Zi9@Y>1^*ixETdSSj^_EsfpsNi|?%2pSAAp`VS^p6m7STwwv{Cx+EYZ%<=xDR#
zBFZ*D82xOQK3JloL($F->4PO&IuhON!Zsg`UiM*|4@N6HaWo%`PIhCPk3=8)oqVqJ
z_itoIv~hs+=SsAHb_BXOMCu1oS~~(=9D{@gp^1b1Zu2B`Bzic^F`+cKA3YrBe=k`=
z2cm`Ju*nCYg9EY2k4FbbBB8_4z>x^(a5Qiz5;_|F8|&vto`jA;`^GwcqC|(g(7wUg
z<j10Wqmj^|Xx?y#50vO}2bwn=OXx`SZaj|U2cmTW*yP85p>+XRLdT(V5m-Wpp>rYj
z38l@g=v)W_ItrbO@!eMf+KbM`AfSWLxft7k(&c8fE(QS|gw}=F9VF4^X0$FuPe7N2
zTjTwr6(L%Vk(O`8FR(x-LR1BG70mrN0F4My9MIGKoWXpd4<SkenrT0i`9vE+6b3Zg
zeMa+vE`%t53+P4jna!If#3;K~<g)h+haSWj6i}u^3t|+Vr(b{d^;-W`@8nZV(Sisi
z0bSjL<SU%7*D<DyhYmz22<SSW>#y%3YCj+Q286gih(F%L>EsLU19GoF#$`Y=`P?s}
z<`Xh6Kgd-;S6@gS8@);B5AW)OT%BdQ%I9towVsiGb@5Ry0(za#_co%=LvpS?%*8=`
z>Hz*Y2iDL}#-+!}2Q+nnv?Tt(JWJ>Y<I>|~kCBoa{Kpp$5#>*Nznv=&l>3#=(|o>D
z5ly_xjBn*}ayxu3pYLNtUzv0L9&8H_)Dh6*K-_<E*|zXFxsyzn`8+J51QdgnEufiW
z^le0c%@gWccbuP&CX$uBgei3?$#JE19h7U`amsf1t20Q;KIg8JGA%n$(GFj55`M{j
zBOj(^$0^$3%RVH^oldo3T6UbG9lq=%qU+0^4R}@^C%1qe!n=R`<a=2!p`KO88MuJ5
zgnqCrI!^Z3n}9a3EILjupw;=jyfWW`W6gno>UQ|6<`YoQn&TAh@D*2ivd3H29Osb1
z><)k3`&~dCr)5}joXSI_;($6Yp@t>L$t|Em0%|xezl!5zBUpb7pL&tsxcn-PGkk|<
zk9Vv%PSGI}0j=X#ahS3Nlz`Upt2hh+jo?>tn8IWDx+8M?Q}uooN68;?Hyy;k52^Qm
z8<$_jQ3z-Vzlx*ePpT8pMt&6sDO^CSKNQ;j5b8HBzlwttE}#Uol3&F^2xttyiep@S
zYED2~`Bfa_;<}^O)1VwA`Bfa_;v=8dgLne!S8<HsfbvNax?EjZOh8-tRUG2#qK>w!
zLAhAOvEmQ}G>Tuv5w3Q40@}>4;s}?2IU=B~{3?!6zr)vEDndZfqhW_9poRP@4$!c}
zH(hM!I4-}60}#+Seie^*eI(O(jKmIaS@LiMG>~D*qZRDKodGr8;T=mJt>9cG0c~bj
z@@QcJWru%z);w0pGDkq0Io3Q@$)}Nr5LX4%d55>Gd8oDplz^6Uta&JQ_*j;IMUPao
z!}s0U&U03lMGsW8%n{IbmPHR#birm*KwB>%os4DC1H}i_JD;Xi4^x(+mVmN=e(|h&
zn4$%gfOhb#dX&O5K|KMj{YCoOcvd}1Wk6X#>zS54NZAfgKs%V0Jx1wgJp{C!Y1w1^
z1aJ6{ppBp2Psp_FAxeWcB%rQ;gL18VgxVdxJZ&tXKV0h`q4Y<~`pc47K!4a4K0s|i
z>vwn-P~XA_2n#3+=r7;O`xTxAC7?ZgEALl%n!fl)(BH?cuciNDTzaqOgLoFu3dW`P
zD%`oN*F*BT=3INH)*YUJ)^M)9Q{}awrFkQt8`j153A$LBfNoe9-=`>lpzRANhwz`g
ztM9T4Xz2pV4qwH)`YzwCt6f0zE&t6Y3+N~F@_Q5nboI6BfEGP{e&*%(XbdR%tm0mO
zhiyPvKx4Sq-(eTfL?^2QN<I(l8_?|dN$0+o$$S#fI`$1{cD$%~AJNnyJo&6d2U;Zp
zx{W9c=m>P6RpJ!<=a+wfGMZdKa}6h-3ACWqNkHHKm(l8c63`S{&?+6k@5h#t0GH*H
z50QS+gJ!FQzD`!ruSZia$$dW-&=q>nZr!`%Dw=qtET3G)v_=y;Y!Z4FQ5MiHG@(a9
zLfaQmX7hn2blJ2o717dsvVfkT3w^fzJ4CcNpNaGSjOGJf=(KBp;Oo=Re@>rw`fXT1
zw`fDJR0m(1oSr=^`CSO;7H#NO(!%0=a*k<-KJ>fp;pH<}K+n*J0m}1t`8WaHp${XJ
z9sylGfqd@Kh#?Bk!CyXufbP+VF-k7;ynY7xJfIVUY#O<&RlJsf9?*$V>N2@`<^O-Y
z2<QQwxE~8>2CcY%9|6sv6}KawIrQRo1T=?U+%5IXOHAblt+-qIx9gb8HyUxf`d^@T
zLLY9IzC{j``9d4+cXA5|w)qb<LBQEfN*E43kZ^hz9i~GIB&_eEvwgsRKK2dJaB`sZ
zW<c)si#QLc`-IHPm$6Rhehsyrk#+G(G6{WGUtu~C-^#^*T-@cU<wEEwIM%G`;x12L
z&+&_?SGCf13C%pwDW)!UKEVDqWLHo32&qGjPjR?W-t{ZyAK&~HKFEF;M3jI^ISzw}
z&RyNpB*K>X6E_8~n#atwJy9xbhr8h(UlsLs?l*C14|TR4W@9?<B06nh50*>Y>v(tZ
z>lZ!Qq-8yS+;!UWe|>QxQx-LWg0y+8q7#-hiIP>c;YCj~T|pBmh-l(P5%1(iv(+<~
z%ZR2H(U`Y#m&wYR&gF$m$77|qH*<%{%9&H{zRiiF`SADim*%QwZYMdNJE2n*<KYm}
zsS`R=Ek^RcA2QQWXR5_X{_^{YraDn6Uh=#AJY$^~rxG_NyZkhT_{r|_)68|6LJVbR
gH_KqN0}QVL6qw@7#Ws(a3;+NC07*qoM6N<$f^<k900000


From 174b5fcbf6f6e9b6b6c8d603a78c9a58d1a714eb Mon Sep 17 00:00:00 2001
From: RichardPatterson <richard-patterson@hotmail.co.uk>
Date: Tue, 12 Mar 2024 17:13:34 +0000
Subject: [PATCH 3/5] Add date to description

---
 DESCRIPTION | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/DESCRIPTION b/DESCRIPTION
index 1f1ea8e..18b98ce 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -26,3 +26,5 @@ Suggests:
     rmarkdown,
     units
 VignetteBuilder: knitr
+Date: 2024-03-07
+

From 3148d7280b35ca8d8ce28e1b33b274391f5115d0 Mon Sep 17 00:00:00 2001
From: RichardPatterson <richard-patterson@hotmail.co.uk>
Date: Mon, 18 Mar 2024 16:15:37 +0000
Subject: [PATCH 4/5] Update description and NEWS

---
 DESCRIPTION         |  4 ++--
 NEWS.md             |  6 ++++++
 R/midlines_refine.R |  2 +-
 README.Rmd          | 10 ++++++++++
 README.md           | 18 ++++++++++++++++++
 5 files changed, 37 insertions(+), 3 deletions(-)
 create mode 100644 NEWS.md

diff --git a/DESCRIPTION b/DESCRIPTION
index 18b98ce..e8a2ca6 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,6 +1,6 @@
 Package: midlines
 Title: Estimate Polygon Midlines
-Version: 0.0.0.9001
+Version: 0.0.0.9002
 Authors@R: 
     person(given = "Richard",
            family = "Patterson",
@@ -26,5 +26,5 @@ Suggests:
     rmarkdown,
     units
 VignetteBuilder: knitr
-Date: 2024-03-07
+Date: 2024-03-12
 
diff --git a/NEWS.md b/NEWS.md
new file mode 100644
index 0000000..e59280d
--- /dev/null
+++ b/NEWS.md
@@ -0,0 +1,6 @@
+# midlines 0.0.0.9002
+
+- removed dplyr dependency
+- added igraph dependency
+- replaced wonky bespoke code that grouped intersecting lines with igraph function. This effects midlines_group(), midlines_check(), midlines_dedensify(), midlines_debit(), midlines_smooth(). The new code is ~2.5x fast than the previous version.
+- Vignette added
diff --git a/R/midlines_refine.R b/R/midlines_refine.R
index aad8f05..741016b 100644
--- a/R/midlines_refine.R
+++ b/R/midlines_refine.R
@@ -11,7 +11,7 @@ midlines_group = function(x) {
   touches = sf::st_touches(x)
   graph = igraph::graph_from_adj_list(touches)
 
-  groups <- igraph::components(graph)$membership
+  groups = igraph::components(graph)$membership
 
   grouped = stats::aggregate(x, by = list(group_id = groups), FUN = unique)
 
diff --git a/README.Rmd b/README.Rmd
index 6b5d78d..8d9c47a 100644
--- a/README.Rmd
+++ b/README.Rmd
@@ -298,6 +298,16 @@ plot(pols)
 
 ```
 
+## Another use case
+
+An alternative use case is to merge two versions of a road network that are measured differently and therefore do not exactly match. Use buffers to creating a polygon around both networks and then estimate the midline of this polygon to give a combined network against which each can be compared. See this [blog](https://richardpatterson.github.io/2023/08/06/network_merge.html) post for more details.
+
+## Citation
+
+If you use this package in your work it would be helpful if you would cite it. A suggested citation is:
+
+_Patterson, R. midlines: Estimate Polygon Midlines. R package version 0.0.0.9002. DOI: 10.17863/CAM.100113_
+
 ## Alternatives
 
 * The very comprehensive R package [`cmgo`](https://github.com/AntoniusGolly/cmgo) does many of the same things as this package and much more. If your aim to to estimate the midline of a river, you should probably start with [`cmgo`](https://github.com/AntoniusGolly/cmgo)
diff --git a/README.md b/README.md
index ec584c0..6f56460 100644
--- a/README.md
+++ b/README.md
@@ -436,6 +436,24 @@ perimeter of the river Thames polygon as the starting point.
 
 <img src="man/figures/README-voronoi-1.png" width="100%" />
 
+## Another use case
+
+An alternative use case is to merge two versions of a road network that
+are measured differently and therefore do not exactly match. Use buffers
+to creating a polygon around both networks and then estimate the midline
+of this polygon to give a combined network against which each can be
+compared. See this
+[blog](https://richardpatterson.github.io/2023/08/06/network_merge.html)
+post for more details.
+
+## Citation
+
+If you use this package in your work it would be helpful if you would
+cite it. A suggested citation is:
+
+*Patterson, R. midlines: Estimate Polygon Midlines. R package version
+0.0.0.9002. DOI: 10.17863/CAM.100113*
+
 ## Alternatives
 
 - The very comprehensive R package

From 987019d7890327bd0e05fbe6b83c6ceacfa0b6b6 Mon Sep 17 00:00:00 2001
From: RichardPatterson <richard-patterson@hotmail.co.uk>
Date: Mon, 18 Mar 2024 16:15:37 +0000
Subject: [PATCH 5/5] Update description and NEWS

---
 DESCRIPTION         |  4 ++--
 NEWS.md             |  6 ++++++
 R/midlines_refine.R | 11 +----------
 README.Rmd          | 10 ++++++++++
 README.md           | 18 ++++++++++++++++++
 5 files changed, 37 insertions(+), 12 deletions(-)
 create mode 100644 NEWS.md

diff --git a/DESCRIPTION b/DESCRIPTION
index 18b98ce..e8a2ca6 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,6 +1,6 @@
 Package: midlines
 Title: Estimate Polygon Midlines
-Version: 0.0.0.9001
+Version: 0.0.0.9002
 Authors@R: 
     person(given = "Richard",
            family = "Patterson",
@@ -26,5 +26,5 @@ Suggests:
     rmarkdown,
     units
 VignetteBuilder: knitr
-Date: 2024-03-07
+Date: 2024-03-12
 
diff --git a/NEWS.md b/NEWS.md
new file mode 100644
index 0000000..e59280d
--- /dev/null
+++ b/NEWS.md
@@ -0,0 +1,6 @@
+# midlines 0.0.0.9002
+
+- removed dplyr dependency
+- added igraph dependency
+- replaced wonky bespoke code that grouped intersecting lines with igraph function. This effects midlines_group(), midlines_check(), midlines_dedensify(), midlines_debit(), midlines_smooth(). The new code is ~2.5x fast than the previous version.
+- Vignette added
diff --git a/R/midlines_refine.R b/R/midlines_refine.R
index aad8f05..637886d 100644
--- a/R/midlines_refine.R
+++ b/R/midlines_refine.R
@@ -11,7 +11,7 @@ midlines_group = function(x) {
   touches = sf::st_touches(x)
   graph = igraph::graph_from_adj_list(touches)
 
-  groups <- igraph::components(graph)$membership
+  groups = igraph::components(graph)$membership
 
   grouped = stats::aggregate(x, by = list(group_id = groups), FUN = unique)
 
@@ -24,15 +24,6 @@ midlines_group = function(x) {
   return(grouped)
 }
 
-# This seems to be okay. It returns a df rather than a tibble, but that's fine because that's what it it passed.
-# d = midlines_group(removed)
-# e = midlines::midlines_group(removed)
-# d
-# e
-# class(d)
-# class(e)
-# all.equal(d,e)
-# class(removed)
 
 
 
diff --git a/README.Rmd b/README.Rmd
index 6b5d78d..8d9c47a 100644
--- a/README.Rmd
+++ b/README.Rmd
@@ -298,6 +298,16 @@ plot(pols)
 
 ```
 
+## Another use case
+
+An alternative use case is to merge two versions of a road network that are measured differently and therefore do not exactly match. Use buffers to creating a polygon around both networks and then estimate the midline of this polygon to give a combined network against which each can be compared. See this [blog](https://richardpatterson.github.io/2023/08/06/network_merge.html) post for more details.
+
+## Citation
+
+If you use this package in your work it would be helpful if you would cite it. A suggested citation is:
+
+_Patterson, R. midlines: Estimate Polygon Midlines. R package version 0.0.0.9002. DOI: 10.17863/CAM.100113_
+
 ## Alternatives
 
 * The very comprehensive R package [`cmgo`](https://github.com/AntoniusGolly/cmgo) does many of the same things as this package and much more. If your aim to to estimate the midline of a river, you should probably start with [`cmgo`](https://github.com/AntoniusGolly/cmgo)
diff --git a/README.md b/README.md
index ec584c0..6f56460 100644
--- a/README.md
+++ b/README.md
@@ -436,6 +436,24 @@ perimeter of the river Thames polygon as the starting point.
 
 <img src="man/figures/README-voronoi-1.png" width="100%" />
 
+## Another use case
+
+An alternative use case is to merge two versions of a road network that
+are measured differently and therefore do not exactly match. Use buffers
+to creating a polygon around both networks and then estimate the midline
+of this polygon to give a combined network against which each can be
+compared. See this
+[blog](https://richardpatterson.github.io/2023/08/06/network_merge.html)
+post for more details.
+
+## Citation
+
+If you use this package in your work it would be helpful if you would
+cite it. A suggested citation is:
+
+*Patterson, R. midlines: Estimate Polygon Midlines. R package version
+0.0.0.9002. DOI: 10.17863/CAM.100113*
+
 ## Alternatives
 
 - The very comprehensive R package