Merge pull request #40 from MicheleNuijten/hotfix

Hotfix

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: statcheck
 Title: Extract Statistics from Articles and Recompute P-Values
-Version: 1.4.0
-Date: 2020-04-30
+Version: 1.4.0-beta.2
+Date: 2021-03-17
 Authors@R: c(
     person("Michele B.", "Nuijten", email = "m.b.nuijten@uvt.nl", 
            role = c("aut", "cre")),

R/regex.R

@@ -31,8 +31,10 @@ RGX_TEST_DF <- paste0("(", RGX_T_DF, "|", RGX_R_DF, "|", RGX_Q_DF, "|", RGX_F_DF
 
 # test value
 # this is the same for every type of test
+# the part "[^a-zA-Z\\d\\.]{0,3}" is to extract punctuation marks that could 
+# signal a weirdly encoded minus sign
 # note that impossible values such as r > 1 are excluded at a later stage
-RGX_TEST_VALUE <- "[<>=]\\s?[^a-z\\d]{0,3}\\s?\\d*,?\\d*\\.?\\d+\\s?,"
+RGX_TEST_VALUE <- "[<>=]\\s?[^a-zA-Z\\d\\.]{0,3}\\s?\\d*,?\\d*\\.?\\d+\\s?,"
 
 # p-values
 # this is the same for every type of test

tests/testthat/test-extract-t-tests.R

@@ -46,6 +46,7 @@ test_that("variations in the t-statistic are retrieved from text", {
   txt2 <- "t(28) = 2,000.20, p = .03"
   txt3 <- "t(28) < 2.20, p = .03"
   txt4 <- "t(28) > 2.20, p = .03"
+  txt5 <- "t(28) = %^&2.20, p = .03" # read as -2.20
 
   result <- statcheck(c(txt1, txt2, txt3, txt4), messages = FALSE)
 
@@ -81,6 +82,7 @@ test_that("corrected degrees of freedom in t-tests are retrieved from text", {
 test_that("incorrect punctuation in t-tests are not retrieved from text", {
   txt1 <- "t(28) = 2.20; p = .03"
   txt2 <- "t[28] = 2.20, p = .03"
+  txt3 <- "t(28) = .2.20, p = .03"
 
   expect_output(statcheck(c(txt1, txt2), messages = FALSE), "did not find any results")
 })

tests/testthat/test-file-to-txt.R

@@ -5,7 +5,7 @@ context("Check if statistics from an article are correctly retrieved and parsed"
 # individual pdf files
 test_that("statistics from a pdf are correctly retrieved and parsed", {
 
-  pdf_file <- system.file("test_materials/NuijtenEtAl_2016_ReportingErrorsPsychology.pdf",
+  pdf_file <- system.file("test_materials/nuijten.pdf",
                           package = "statcheck")
 
   result <- checkPDF(pdf_file, messages = FALSE)
@@ -49,8 +49,8 @@ test_that("statistics from all pdfs in a folder are correctly retrieved and pars
 # individual html files
 test_that("statistics from a html are correctly retrieved and parsed", {
 
-  html_file <- system.file("test_materials/NuijtenEtAl_2016_ReportingErrorsPsychology.html",
-                          package = "statcheck")
+  html_file <- system.file("test_materials/nuijten.html",
+                           package = "statcheck")
 
   result <- checkHTML(html_file, messages = FALSE)
   result_1tailed <- checkHTML(html_file, messages = FALSE, OneTailedTxt = TRUE)
@@ -63,7 +63,7 @@ test_that("statistics from a html are correctly retrieved and parsed", {
   # check errors
   expect_equal(result[[VAR_ERROR]], c(FALSE, FALSE, FALSE, TRUE, FALSE, TRUE))
   expect_equal(result[[VAR_DEC_ERROR]], c(FALSE, FALSE, FALSE, TRUE, FALSE, TRUE))
-
+  
   # check errors with one-tailed test detection
   expect_equal(result_1tailed[[VAR_ERROR]], c(FALSE, FALSE, FALSE, TRUE, FALSE, FALSE))
   expect_equal(result_1tailed[[VAR_DEC_ERROR]], c(FALSE, FALSE, FALSE, TRUE, FALSE, FALSE))
@@ -101,7 +101,7 @@ test_that("statistics from all pdfs and htmls in a folder are correctly retrieve
   dir <- system.file("test_materials", package = "statcheck")
 
   result <- checkdir(dir, subdir = FALSE, messages = FALSE)
-
+  
   # extract 59 tests (6 from html and 53 from pdf)
   expect_equal(nrow(result), 59)
 })