为什么geom_pattern=梯度没有在RMarkdown子页中分割数字？

Question

我想在我的.Rmd文档中自动生成页面。我没有办法做这件事，但我终于找到了一个可以接受的选择(https://somtom.github.io/post/using-dynamically-rendered-r-markdown-childs-for-reports/)。

这个例子一切都很好，但是当我修改它以介绍我的数据时，会出现一些问题。(一些数字未出现在最后文件中)。几个小时后，我发现唯一的问题出现在我试图绘制一个包含pattern=gradient的ggplot时。在这个链接(https://somtom.github.io/post/using-dynamically-rendered-r-markdown-childs-for-reports/)中，您可以找到原始代码。

在这里，您可以找到相同的代码，包括仅在子页面代码中进行的小修改:我用一些随机图形(包括梯度模式)替换了原始代码中的图形。

请注意我放了一些。在块‘之前，为了被识别为单个块代码(您必须删除它们才能运行代码)。

我研究了html代码，发现这个图形绘制得很好，但是它没有出现在相应的图表中。没有渐变的任何其他图形看起来都是正常的。

---
title: "Example Analytics"
output: 
  flexdashboard::flex_dashboard:
    orientation: rows
    vertical_layout: fill
---
# Overview

`r shiny::tags$h1("Overview", style="text-align:center")`

.```{r setup, include=FALSE}
library(flexdashboard) 
library(tidyverse)

# Create dummy data for example
set.seed(345)
date_start <- as.Date("2018-01-01")
product_groups <- c("PG1", "PG2", "PG3")
products <- list(
  PG1 = c("A", "B", "C"),
  PG2 = c("D", "E", "F"),
  PG3 = c("G", "H")
)
data <- data.frame(
  product_group = sample(product_groups, 100, replace = T),
  trade_volume = rnorm(100, mean = 3448, sd = 200),
  date = date_start + 1:100
) %>% 
  group_by(id = row_number()) %>% 
  mutate(product = sample(unlist(products[[product_group]]), 1, replace = TRUE)) %>% 
  ungroup()
.```


Row
-----------------------------------------------------------------------

### Trade Volume Development

.```{r fig.width=15, fig.height=5}
data %>% 
  group_by(date) %>% 
  summarise(trade_volume = sum(trade_volume)) %>% 
  ggplot() +
  geom_line(aes(x = date, y = trade_volume), 
           color = "#2780e3",
           alpha = 0.8) +
  scale_y_continuous(labels = scales::dollar) +
  theme_minimal() +
  ylab("Trade Volume") +
  xlab("Date")
.```


Row
-----------------------------------------------------------------------

### Product Group Share on Trade Volume

.```{r}
data %>% 
  group_by(product_group) %>% 
  summarise(trade_volume = sum(trade_volume)) %>%
  mutate(trade_volume_share = trade_volume / sum(trade_volume)) %>% 
  ggplot() +
  geom_col(aes(x = reorder(product_group, -trade_volume_share), y = trade_volume_share), 
           fill = "#2780e3",
           alpha = 0.8) +
  scale_y_continuous(labels = scales::percent) +
  theme_minimal() +
  ylab("Share on Trade Volume") +
  xlab("Product Group")
.```

### Product Group Share on Order Count

.```{r}
data %>% 
  group_by(product_group) %>% 
  summarise(order_count = n()) %>% 
  mutate(order_count_share = order_count / sum(order_count)) %>% 
  ggplot() +
  geom_col(aes(x = reorder(product_group, -order_count_share), y = order_count_share), 
           fill = "#2780e3",
           alpha = 0.8) +
  scale_y_continuous(labels = scales::percent) +
  theme_minimal() +
  ylab("Share on Order Count") +
  xlab("Product Group")
.```

.```{r render subpages, include=FALSE}
# Get all unique product groups for the subpages
product_groups <- unique(data$product_group)

# Create variable which stores all subpages outputs
out = NULL

# Set knitr options to allow duplicate labels (needed for the subpages)
options(knitr.duplicate.label = 'allow')

# Create temporary environment which we use for knitting subpages.RMD 
subpage_env <- new.env()

for (pg in product_groups) {
  # Filter data for product group 
  subpage_data <- data %>% 
    filter(product_group == pg)
  
  # Assign filtered data and product group to subpage_env 
  assign("subpage_data", subpage_data, subpage_env)
  assign("product_group", pg, subpage_env)
  
  # Knit subpage.RMD using the subpage_env and add result to out vector
  out = c(out, knitr::knit_child('subpage.RMD', envir = subpage_env))
}
.```

`r paste(knitr::knit_child(text = out), collapse = '')`

`r paste0('# ', product_group, '{data-navmenu="Product Groups"}')`

`r shiny::tags$h1(product_group, style="text-align:center")`

Row
-----------------------------------------------------------------------

### Product Share on Trade Volume

.```{r fig.width=10}
#------------------------------------------! MODIFICATION INCLUDING GRADIENT PLOT-------------------------------------
df <- data.frame(
  trt         = c("a", "b", "c"),
  outcome     = c(2.3, 1.9, 3.2)
)

ggplot(df, aes(trt, outcome)) +
  geom_col_pattern(
    aes(
      pattern_fill        = trt,
      pattern_orientation = trt
    ),
    pattern       = 'gradient',
    pattern_fill2 = 'white',
    colour        = 'black'
  ) +
  theme_bw(15) +
  labs(
    title    = "ggpattern::geom_col_pattern()",
    subtitle = "pattern = 'gradient'"
  ) +
  theme(
    legend.key.size = unit(1.5, 'cm')
  ) +
  coord_fixed(ratio = 1/2)
.```


### Products with Most Trade Volume

.```{r}
subpage_data %>% 
  group_by(Product = product) %>% 
  summarise(`Trade Volume [€]` = sum(trade_volume)) %>% 
  arrange(-`Trade Volume [€]`) %>% 
  knitr::kable()
.```

Answer 1

geom_col_pattern是ggpattern 套餐的一部分--它不在ggplot2(tidyverse)中，您需要将包添加到您的文件中，如果您希望该函数工作的话。在您的问题中，您问了关于geom_pattern的问题，但是看起来这个函数实际上是geom_col_pattern()或ggpattern::geom_col_pattern()

Answer 2

尝试使用{ggpattern}. 这解决了另一个用户的类似问题。显式禁用R4.1梯度支持。