Merge pull request #196 from Crunch-io/master

Recreating develop

.github/workflows/ci.yml

@@ -10,7 +10,7 @@ jobs:
     steps:
     - uses: actions/checkout@v2
     - name: Add kableExtra
-      run: Rscript -e 'install.packages("kableExtra")'
+      run: Rscript -e 'devtools::install_github("haozhu233/kableExtra")'
     - name: CRAN crunch
       run: Rscript -e 'install.packages("crunch")'
     - name: Build

DESCRIPTION

@@ -6,7 +6,7 @@ Description: In order to generate custom survey reports, this package provides
     'banners' (cross-tabulations) of datasets in the Crunch
     (<https://crunch.io/>) web service. Reports can be written in 'PDF' format
     using 'LaTeX' or in Microsoft Excel '.xlsx' files.
-Version: 1.2.7
+Version: 1.2.9
 Authors@R: c(
     person("Persephone", "Tsebelis", role="aut"),
     person("Kamil", "Sedrowicz", role="aut"),
@@ -17,9 +17,10 @@ BugReports: https://github.com/Crunch-io/crunchtabs/issues
 License: LGPL (>= 3)
 Depends:
     R (>= 3.5.0),
-    crunch,
-    kableExtra
+    crunch
 Imports:
+    kableExtra,
+    rlang,
     openxlsx,
     digest,
     methods,
@@ -31,7 +32,8 @@ Suggests:
     jsonlite,
     knitr,
     rmarkdown,
-    testthat (>= 2.1.0)
-RoxygenNote: 7.1.0
+    testthat (>= 2.1.0),
+    kableExtra (>= 1.1.0.9000)
+RoxygenNote: 7.1.1
 VignetteBuilder: knitr
 Encoding: UTF-8

NAMESPACE

@@ -54,7 +54,6 @@ export(themeHuffPoToplines)
 export(themeNew)
 export(themeUKPolitical)
 export(with_api_fixture)
-export(writeCodeBook)
 export(writeCodeBookLatex)
 export(writeExcel)
 export(writeLatex)
@@ -89,6 +88,7 @@ importFrom(kableExtra,column_spec)
 importFrom(kableExtra,kable_styling)
 importFrom(magrittr,`%>%`)
 importFrom(methods,is)
+importFrom(rlang,.data)
 importFrom(stats,ave)
 importFrom(stats,median)
 importFrom(stats,pnorm)

NEWS.md

@@ -1,9 +1,23 @@
+## crunchtabs 1.2.9
+
+- Codebook question descriptions now appropriately escape special characters (#187)
+- Added option enforce_onehundred which allows one to avoid rounding errors in totals rows (#189)
+- Codebook table of contents overruns, cutting text and adding "..." (#186)
+- Codebook generation now supports a filepath (#185)
+- Added vertical space before append_text (#182)
+- Bugfix: append_text that is multiple lines du0lpicated vertical space. Collapsing. (#191)
+- Remove requirement for dev version of kableExtra (#184)
+
 ## crunchtabs 1.2.8
 
+- Documentation for generating codebooks (#180)
+- Tests and fixes for broken tests (#176)
+- Spacing and font type for codebooks (#178)
 - Additional refinements for codebooks such as title, subtitle, sample description (#152)
 - Fixes for warnings in existing code (#158)
 - Bugfix for logo assignment in excel (#155)
-- Added pagebreak_in_banner which stops multi-banner crosstabs from being spread across pages using "to be continued" (#161)
+- Added pagebreak_in_banner which stops multi-banner crosstabs from being spread across pages using "to be continued" (#161, #169)
+- Major rework of codebooks (#165, #172, #170, #160, #159)
 
 ## crunchtabs 1.2.7
 

R/codeBookSummary.R

@@ -71,19 +71,28 @@ codeBookSummary.CategoricalVariable <- function(x, multiple = FALSE, ...) {
   smry <- data.frame(crunch::table(x, useNA = "ifany"))
   names(smry) <- c("name", "n")
 
-  res <- merge(l, smry)
+  res <- merge(l, smry, sort = FALSE)
 
   if (multiple) {
-    res[with(res, order(id)),]
+    res = res[with(res, order(id)),]
   } else {
-    res = matrix(c(
+    res = as.data.frame(
+      matrix(c(
       res$id,
       res$name,
       res$n
-    ), ncol = 3)
-    res
+    ), ncol = 3),
+    stringsAsFactors = FALSE
+  )
+    names(res) <- c("id","name", "n")
   }
 
+  if (getOption("crunchtabs.codebook.suppress.zeros", default = FALSE)) {
+    warning('Zero count categoricals are supressed by options')
+    res <- res[res$n != "0",]
+  }
+
+  res
 }
 
 #' @describeIn codeBookSummary Prepares a codeBookSummary data.frame for a MultipleResponseVariable
@@ -129,24 +138,27 @@ codeBookSummary.CategoricalArrayVariable <- function(x, ...) codeBookSummary.Mul
 #' @describeIn codeBookSummary Prepares a codeBookSummary data.frame for a NumericVariable
 #' @export
 codeBookSummary.NumericVariable <- function(x, ...) {
+  mu <- round(mean(x, na.rm = T), 2)
+  std <- round(sd(x, na.rm = TRUE))
   minima <- round(min(x, na.rm = T), 2)
   maxima <- round(max(x, na.rm = T), 2)
   missings <- sum(is.na(as.vector(x)))
-
-  type_row <- c("Type", "Numeric")
-  range_row <- c("Range", paste0("[", minima,", ", maxima,"]"))
-
-  if (missings > 0) {
-    missings_row <- c("Missing", missings)
-    r <- rbind(
-      type_row, missings_row, range_row
-    )
-  } else {
-    r <- rbind(
-      type_row, range_row
-    )
-  }
-
+  n <- length(is.na(as.vector(x)))
+
+  #type_row <- c("Type", "Numeric")
+  #range_row <- c("Range", paste0("[", minima,", ", maxima,"]"))
+
+  r <- as.data.frame(
+    matrix(c(
+      mu,
+      std,
+      minima,
+      maxima,
+      n - missings,
+      missings), nrow = 1)
+  )
+
+  names(r) <- c("Mean", "SD", "Min", "Max","n", "Missing")
   rownames(r) <- NULL
   r
 
@@ -156,12 +168,26 @@ codeBookSummary.NumericVariable <- function(x, ...) {
 #' @export
 codeBookSummary.TextVariable <- function(x, ...) {
 
-  filled <- sum(as.vector(x) != "" | !is.na(as.vector(x)), na.rm = TRUE)
-  type_row <- c("Type", "Text")
-  missing <- c("Blank", length(as.vector(x)) - filled)
-  filled <- c("Filled", filled)
-
-  r <- rbind(type_row, missing, filled)
+  filled_verbs <- as.vector(x)
+  filled_verbs <- filled_verbs[!is.na(filled_verbs)]
+  filled_verbs <- filled_verbs[!filled_verbs %in% c("", "__NA__")]
+  #verbsamp <- sample(filled_verbs, 5, replace = TRUE)
+  #verbsamp <- paste0(substr(verbsamp, start = 1, stop = 27),
+  #       "...",
+  #       collapse = ", "
+  #)
+  filled <- length(filled_verbs)
+  missing <- length(as.vector(x)) - filled
+
+  r <- as.data.frame(
+    matrix(c(
+      filled,
+      missing,
+      ifelse(filled == 0, 0, max(nchar(filled_verbs)))
+      ), nrow = 1, ncol = 3), stringsAsFactors = FALSE
+  )
+
+  names(r) <- c("Filled","Missing", "Max Length")
   rownames(r) <- NULL
   r
 }
@@ -173,19 +199,16 @@ codeBookSummary.DatetimeVariable <- function(x, ...) {
   maxima <- max(x, na.rm = T)
   missings <- sum(is.na(as.vector(x)))
 
-  type_row <- c("Type", "Datetime")
-  range_row <- c("Range", paste0("[", minima,", ", maxima,"]"))
+  filled <- sum(!is.na(as.vector(x)))
+  range_row <- c(paste0("[", minima,", ", maxima,"]"))
+
+  r = data.frame(
+    Filled = filled,
+    Missing = missings,
+    Range = range_row,
+    stringsAsFactors = FALSE
+  )
 
-  if (missings > 0) {
-    missings_row <- c("Missing", missings)
-    r <- rbind(
-      type_row, missings_row, range_row
-    )
-  } else {
-    r <- rbind(
-      type_row, range_row
-    )
-  }
 
   rownames(r) <- NULL
   r