iis log file reader, closes ##23

Ironholds · Nov 30, 2017 · 041cd35 · 041cd35
1 parent df43836
commit 041cd35
Show file tree

Hide file tree

Showing 6 changed files with 39 additions and 5 deletions.
diff --git a/.gitignore b/.gitignore
@@ -2,3 +2,4 @@
 .Rhistory
 .RData
 inst/doc
+*.DS_Store
diff --git a/R/RcppExports.R b/R/RcppExports.R
@@ -2,10 +2,10 @@
 # Generator token: 10BE3573-1514-4C36-9D1C-5A225CD40393
 
 internal_split_clf <- function(requests) {
-    .Call(webreadr_internal_split_clf, requests)
+    .Call(`_webreadr_internal_split_clf`, requests)
 }
 
 internal_split_squid <- function(requests) {
-    .Call(webreadr_internal_split_squid, requests)
+    .Call(`_webreadr_internal_split_squid`, requests)
 }
 
diff --git a/R/readers.R b/R/readers.R
@@ -309,4 +309,25 @@ read_s3 <- function(file){
   data <- readr::read_log(file = file, col_types = types, col_names = names)
   data$request_time <- readr::parse_datetime(data$request_time, format = "%d/%b/%Y:%H:%M:%S %z")
   return(data)
-}
+}
+
+#'@title Read Microsoft IIS Logs
+#'@description \code{read_iis} provides a reader for Microsoft IIS access logs.
+#'
+#'@param file the full path to the IIS log file you want to read.
+#'
+#'@examples
+#'# Using the inbuilt testing dataset
+#'iis_data <- read_iis(system.file(file.path("extdata", "iis.log"), package = "webreadr"))
+#'
+#'@export
+
+read_iis <- function(file){
+  names <- c("date", "time", "s-ip", "cs-method", "cs-uri-stem", "cs-uri-query", "s-port", 
+             "cs-username", "c-ip", "cs(User-Agent)", "sc-status", "sc-substatus", 
+             "sc-win32-status", "time-taken")
+  types <- "cccccciccciiii"
+  data <- readr::read_log(file = file, col_types = types, col_names = names, skip = 4)
+  data$date <- as.POSIXct(paste(data$date, data$time), format = "%Y-%m-%dT%H:%M:%S %z", tz = "UTC")
+  return(data[,!names(data) == "time"])
+}
diff --git a/inst/extdata/iis.log b/inst/extdata/iis.log
@@ -0,0 +1,5 @@
+#Software: Microsoft Internet Information Services 8.0
+#Version: 1.0
+#Date: 2016-02-15 23:44:42
+#Fields: date time s-ip cs-method cs-uri-stem cs-uri-query s-port cs-username c-ip cs(User-Agent) cs(Referer) sc-status sc-substatus sc-win32-status time-taken
+2016-02-15 23:44:42 10.10.2.18 GET / - 443 - 198.20.69.74 - - 200 0 995 29761
diff --git a/src/RcppExports.cpp b/src/RcppExports.cpp
@@ -7,7 +7,7 @@ using namespace Rcpp;
 
 // internal_split_clf
 DataFrame internal_split_clf(CharacterVector requests);
-RcppExport SEXP webreadr_internal_split_clf(SEXP requestsSEXP) {
+RcppExport SEXP _webreadr_internal_split_clf(SEXP requestsSEXP) {
 BEGIN_RCPP
     Rcpp::RObject rcpp_result_gen;
     Rcpp::RNGScope rcpp_rngScope_gen;
@@ -18,7 +18,7 @@ END_RCPP
 }
 // internal_split_squid
 DataFrame internal_split_squid(CharacterVector requests);
-RcppExport SEXP webreadr_internal_split_squid(SEXP requestsSEXP) {
+RcppExport SEXP _webreadr_internal_split_squid(SEXP requestsSEXP) {
 BEGIN_RCPP
     Rcpp::RObject rcpp_result_gen;
     Rcpp::RNGScope rcpp_rngScope_gen;

diff --git a/tests/testthat/test_readers.R b/tests/testthat/test_readers.R
@@ -45,4 +45,11 @@ test_that("S3 files can be read",{
   expect_equal(ncol(data), 18)
   expect_equal(nrow(data), 6)
   expect_equal(class(data$request_time), c("POSIXct","POSIXt"))
+})
+
+test_that("iis files can be read",{
+  data <- read_iis(system.file(file.path("extdata", "iis.log"), package = "webreadr"))
+  expect_equal(ncol(data), 14)
+  expect_equal(nrow(data), 1)
+  expect_equal(class(data$date), c("POSIXct","POSIXt"))
 })