Question

我正在尝试通过折叠与第2列匹配的相应列值来重塑两列数据框-在这种情况下，将股票代码符号添加到其自己的唯一行，同时使第1列的内容成为数据的字段对应于那些行情记录器自己的列。下面是我的示例，其中包含一个小样本，因为它是一个具有500个行情自动报价和4个字段的数据框：

# Closed End Fund Selector

url<-"https://www.cefconnect.com/api/v3/DailyPricing?props=Ticker,Name,DistributionRateNAV,LastUpdated,Discount,DistributionRatePrice,ReturnOnNAV,CategoryId,CategoryName,IsManagedDistribution,Price,PriceChange,NAV,NAVPublished,Cusip/&_=1546832481302"

library(jsonlite)
library(rvest)
library(dplyr)

page<-html_session(url)
json<-readBin(page$response$content, what="json")
df<-fromJSON(json)

# Analyze and Group Closed End Funds by Investment Strategy and Average/Min/Max Discounts to NAV

df2 <- df %>% group_by(CategoryName) %>%
     summarize(Category_Fund_Avg_NAV_Difference= mean(Discount,na.rm=T), 
               Min_NAV_Fund = Name[which(Discount == min(Discount))],  Min_NAV_Ticker = Ticker[which(Discount == min(Discount))],
               Min_Nav_Fund_Difference = min(Discount),
               Max_NAV_Fund = Name[which(Discount == max(Discount))],
               Max_NAV_Ticker = Ticker[which(Discount == max(Discount))],Max_Nav_Fund_Difference = max(Discount))
df2 <- data.frame(df2)

tickers <- df$Ticker
tickers <- paste0("https://www.cefconnect.com/fund/", tickers)

lst_scraped_data <- lapply(tickers, FUN=function(URLLink){
CEF_Scrape <- read_html(URLLink)

test9 <- CEF_Scrape %>%
        html_nodes("#ContentPlaceHolder1_cph_main_cph_main_SummaryGrid") %>% 
        html_text() %>%
        strsplit(split = "\n") %>%
        unlist() %>%
        .[. != " "]


test9 <- str_replace_all(test9,pattern = "\t", replacement = "") 
test9 <- str_replace_all(test9,pattern = "\r", replacement = "")
test9 <- str_trim(test9,side="left")
test9 <- test9[test9 != ""]
test9 <- str_replace_all(test9,pattern="SharePriceNAVPremium/Discount",replacement = "SharePrice NAV Premium/Discount")
test9 <- str_replace_all(test9,pattern="Current",replacement = "Current ")
test9 <- str_replace_all(test9,pattern="52 Wk Avg",replacement = "52WkAvg ")
test9 <- str_replace_all(test9,pattern="52 Wk High",replacement = "52WkHigh ")
test9 <- str_replace_all(test9,pattern="52 Wk Low",replacement = "52WkLow ")
test9 <- str_replace_all(test9,pattern="-",replacement = " -")
#test9 <- str_replace_all(test9,pattern="$",replacement = " $")

# The below two lines are needed for edge cases where the premium/discount % is positive or negative
#test9 <- sub("\\s", "", gsub('(\\$.{2})', '\\1 ', test9))
#test9 <- gsub("(?<=[0-9])(?=[$])", " ", test9, perl = TRUE)
#test9 <- str_replace_all(test9,pattern="$ ",replacement = "$")

test9 <- gsub('(\\.\\d{2})', '\\1 ', test9, perl = T)
test9 <- trimws(gsub('\\s%', '% ', test9))


# This creates a space between prices based on a number followed by a '$'
test9 <- gsub("([0-9])([$])", "\\1 \\2", test9)

# This creates a space for the % values (usually a - sign)
#test9 <- gsub("(?=[-])", " ", test9, perl = TRUE)
#test9 <- gsub("(?<=\\.\\d{2})(?!%)", " ", test9, perl = TRUE)
#test <- sub("((?<=\\.\\d{2})|(?<=%))(?=[\\d$-])", " ", test9, perl = TRUE)

#gsub("(?<=[.])", " ", test9, perl = TRUE)

# Separate by whitespace into unique elements each word so as to create a column vector for indexing
test9 <- unlist(sapply(test9, strsplit, "\\s+", USE.NAMES = FALSE))
test10 <- paste(test9[4],test9[1],test9[5])
test11 <- paste(test9[4],test9[2],test9[6])
test12 <- paste(test9[4],test9[3],test9[7])
test13 <- paste(test9[8],test9[1],test9[9])
test14 <- paste(test9[8],test9[2],test9[10])
test15 <- paste(test9[8],test9[3],test9[11])
test16 <- paste(test9[12],test9[1],test9[13])
test17 <- paste(test9[12],test9[2],test9[14])
test18 <- paste(test9[12],test9[3],test9[15])
test19 <- paste(test9[16],test9[1],test9[17])
test20 <- paste(test9[16],test9[2],test9[18])
test21 <- paste(test9[16],test9[3],test9[19])

test22 <- c(test10, test11, test12, test13, test14, test15, test16, test17, test18, test19, test20, test21)

Ticker <- str_replace_all(URLLink,pattern="https://www.cefconnect.com/fund/",replacement = "")
Checker = data.frame(test22,Ticker)

})


df6 <- do.call(rbind, lst_scraped_data)

理想情况下，重新格式化转换后的行情栏是唯一的行，行名位于行名中，在这种情况下，包含12个对应的列，其中包含“ df6”的“ test22”列的内容，但它们各自的名称位于这个阶段还不重要。非常感谢您的帮助！

这是上述脚本输出的前19行的示例：

                             test22 Ticker
1          Current SharePrice $6.57    MFM
2                 Current NAV $7.11    MFM
3   Current Premium/Discount -7.59%    MFM
4          52WkAvg SharePrice $6.55    MFM
5                 52WkAvg NAV $7.21    MFM
6   52WkAvg Premium/Discount -9.19%    MFM
7         52WkHigh SharePrice $6.88    MFM
8                52WkHigh NAV $7.34    MFM
9  52WkHigh Premium/Discount -5.88%    MFM
10         52WkLow SharePrice $6.05    MFM
11                52WkLow NAV $7.03    MFM
12 52WkLow Premium/Discount -14.43%    MFM
13         Current SharePrice $4.84    CXE
14                Current NAV $5.21    CXE
15  Current Premium/Discount -7.10%    CXE
16         52WkAvg SharePrice $4.91    CXE
17                52WkAvg NAV $5.29    CXE
18  52WkAvg Premium/Discount -7.26%    CXE
19        52WkHigh SharePrice $5.31    CXE
20               52WkHigh NAV $5.37    CXE
21 52WkHigh Premium/Discount -1.12%    CXE
22         52WkLow SharePrice $4.58    CXE
23                52WkLow NAV $5.16    CXE
24 52WkLow Premium/Discount -11.92%    CXE
25         Current SharePrice $4.33    CMU

这就是我希望重新格式化后的输出的样子：

使用dput之后->

                         test22 Ticker

1当前股价$ 6.57 MFM 2当前资产净值$ 7.11 MFM 3当前的高级/折扣-7.59％MFM 4 52WkAvg股价$ 6.55 MFM 5 52WkAvg资产净值$ 7.21 MFM 6 52WkAvg Premium / Discount -9.19％MFM 7 52WkHigh SharePrice $ 6.88 MFM 8 52WkHigh资产净值$ 7.34 MFM 9 52WkHigh高级/折扣-5.88％MFM 10 52WkLow SharePrice $ 6.05 MFM 11 52WkLow NAV $ 7.03 MFM 12 52WkLow Premium / Discount -14.43％MFM 13当前股价$ 4.84 CXE 14当前资产净值$ 5.21 CXE 15当前的溢价/折扣-7.10％CXE 16 52WkAvg股价$ 4.91 CXE 17 52WkAvg资产净值$ 5.29 CXE 18 52WkAvg Premium / Discount -7.26％CXE 19 52WkHigh SharePrice $ 5.31 CXE

根据相应的列标识符代码R

0 个答案: