From aec52e7683412ba86d3a525cb2b0ae2a0400a5e6 Mon Sep 17 00:00:00 2001
From: nwickel <n.wickelmaier@iwm-tuebingen.de>
Date: Thu, 31 Aug 2023 16:03:21 +0200
Subject: [PATCH] Removed code for merge and join; deleted some TODOs

---
 code/02_preprocessing.R | 36 ------------------------------------
 1 file changed, 36 deletions(-)

diff --git a/code/02_preprocessing.R b/code/02_preprocessing.R
index a6438b6..e66394d 100644
--- a/code/02_preprocessing.R
+++ b/code/02_preprocessing.R
@@ -186,7 +186,6 @@ for (art in aws) {              # select artwork
 
 head(dat2[, c("artwork", "event", "trace")], 50)
 tail(dat2[, c("artwork", "event", "trace")], 50)
-# TODO: How to handle popups from glossar???
 
 rm(aws, i, j, last_event, art)
 
@@ -246,9 +245,6 @@ table(is.na(dat2[dat2$glossar == 1, "trace"]))
 # How many glossar_files are only associated with one artwork?
 lut[sapply(lut$artwork, length) == 1, "glossar_file"]
 
-# TODO: Fill in the ones that are associated with one artwork
-# --> Can't come up with something -- maybe ask AK???
-
 single <- lut[sapply(lut$artwork, length) == 1, "glossar_file"]
 tmp <- subset(dat2, is.na(dat2$trace))$popup
 inside <- unique(tmp[tmp %in% lut[sapply(lut$artwork, length) == 1, "glossar_file"]])
@@ -284,8 +280,6 @@ dat2[14110:14130, ]
 
 # TODO: Integrate for loop into for loop above
 
-# TODO: "glossar" entry should be changed to the corresponding artwork
-
 # TODO: For now: Exclude not matched glossar entries
 
 df <- subset(dat2, !is.na(dat2$trace))
@@ -434,38 +428,8 @@ dat_openPopup <- openPopup_wide[, c("fileid.start", "fileid.stop", "event",
 rm(num_start, openPopup_wide)
 
 
-# Merge all
-# system.time({
-# dat_all <- merge(dat_trans, dat_flipCard, all = TRUE)
-# dat_all <- merge(dat_all, dat_openTopic, all = TRUE)
-# dat_all <- merge(dat_all, dat_openPopup, all = TRUE)
-# })
-#
-# # check
-# nrow(dat_all) == (nrow(dat_trans) + nrow(dat_flipCard) +
-#                   nrow(dat_openTopic) + nrow(dat_openPopup))
-#
-# dat_all <- dat_all[order(dat_all$date.start), ]
-# rownames(dat_all) <- NULL
-#
-
-# TODO: from here on NA... WHY??
-# dat_all[19426:19435, ]
-
 # TODO: Should card maybe also be filled in for "openPopup"?
 
-# dat_all2 <- dplyr::full_join(dat_trans, dat_flipCard)
-# dat_all2 <- dplyr::full_join(dat_all, dat_openTopic)
-# dat_all2 <- dplyr::full_join(dat_all, dat_openPopup)
-#
-# nrow(dat_all2) == (nrow(dat_trans) + nrow(dat_flipCard) +
-#                   nrow(dat_openTopic) + nrow(dat_openPopup))
-#
-# dat_all2 <- dat_all2[order(dat_all2$date.start), ]
-# rownames(dat_all2) <- NULL
-# TODO: --> same result - but faster. Need it?
-# --> Would hate to depend on dplyr...
-
 #' ## Use `rbind()` instead...
 # --> unbeatable in terms of time!