buildings.Rmd

---
title: "Building Occupancy"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, message = FALSE, warning = FALSE)
```

```{r}
library(tidyverse)
library(readxl)
library(ggmap)
```

## Instructional use of Buildings

```{r}
days <- as.list(dir("data/classes"))
attend <- map(days, function(x) read_csv(file.path("data/classes", x)))
names(attend) <- str_remove(unlist(days), ".csv")
days <- days == "sunday.csv"
sunday <- attend[["sunday"]] %>%
  rename(building = "Building Name") %>%
  mutate(day = "sunday") %>%
  select(day, building, Student, Instructor)
attend <- bind_rows(attend[!days], .id = "day") %>%
  rename(ind = "Instructor Student Ind",
         ct = "Distinct count of Emplid",
         building = "Building Name") %>%
  pivot_wider(names_from = "ind", values_from = "ct") %>%
  select(day, building, Student, Instructor)
attend <- bind_rows(attend, sunday) %>%
  pivot_longer(Student:Instructor, names_to = "role", values_to = "count") %>%
  arrange(role, building)
```

```{r}
attend_wk <- attend %>%
  group_by(role, building) %>%
  summarize(count = sum(count)) %>%
  ungroup %>%
  pivot_wider(names_from = "role", values_from = "count")
```

```{r}
write_csv(attend_wk, "attend_week.csv")
```

## Research use of Buildings

```{r}
floor <- read_excel("data/ResearchRebootPhase2-2020-08-01.xls",
                         sheet = "PeopleBuildingsData") %>%
  rename(People = "Person Count") %>%
  mutate(Floor = ifelse(Floor == "No", "0", Floor),
         People = ifelse(People == "No", "0", People),
         Floor = as.numeric(Floor),
         People = as.numeric(People)) %>%
  group_by(Building, Floor) %>%
  summarize(People = sum(People, na.rm = TRUE)) %>%
  ungroup
```

```{r}
building <- floor %>%
  group_by(Building) %>%
  summarize(People = sum(People, na.rm = TRUE)) %>%
  ungroup %>%
  filter(People > 0 & !is.na(People)) %>%
  mutate(bldg_num = str_remove(Building, "[A-Z]* -.*"),
         bldg_name = str_remove(Building, ".*- "),
         bldg_title = tolower(str_remove_all(bldg_name, 
           " (Memorial|Building|Bldg|Laboratory|Laboratories|Hall|Center|Street|St)")),
         bldg_title = str_replace(tolower(bldg_title), "^(.*)-(.*)", "\\2 \\1"),
         bldg_title = str_remove(bldg_title, "deluca "))
```

```{r}
write.csv(building %>% select(Building, bldg_name, People), "buildings.csv", row.names = FALSE)
```

### Map of campus buildings

I can only carry this so far because names do not match well.

<https://accessibility.fpm.wisc.edu/campus-building-locations-entrances/>

<https://www.storybench.org/geocode-csv-addresses-r/>

The map is built on https://www.mapbox.com/, which offers a large amount of customization on how to display a map.  It seems like there’s some way to add data to this, but I only glanced through they docs: https://docs.mapbox.com/help/how-mapbox-works/.  I’ve never used this, and it looks pretty complicated, but I’d wager to guess that someone in the university has experience in this in order to create the map.wisc.edu site.  


```{r}
addr <- read_csv("data/campus_address.csv") %>%
  mutate(Name = ifelse(grepl("SEWELL", Name), "Sewell Social Sciences", Name),
         Name = ifelse(Name == "ANIMAL HEALTH & BIOMEDICAL SCIENCES BLDG", "Hansen Biometical Sciences", Name),
         Name = ifelse(Name == "MEAT SCIENCE AND MUSCLE BIOLOGY LAB", "Meat Science & Animal Biologics Discovery", Name),
         bldg_title = str_remove(tolower(Name), " (memorial|building|bldg|laboratory|laboratories|hall|center|street|st)"),
         bldg_title = str_replace(tolower(bldg_title), "^(.*), (.*)", "\\2 \\1"),
         bldg_title = str_replace(tolower(bldg_title), "^(.*)-(.*)", "\\2 \\1"),
         bldg_title = ifelse(grepl("zoology", bldg_title), 
                             str_replace(bldg_title, "zoology", "integrative biology"), bldg_title))
```

```{r}
m <- pmatch(building$bldg_title, addr$bldg_title)
building$bldg_name[is.na(m)]
```


```{r}
tmpfn2 <- function(x) {
  if(is.na(x[1]) | x[1] == "NA") {
    x[2]
  } else {
    b <- strsplit(x[2], ",")[[1]]
    m <- match(x[1], b)
    if(!is.na(m)) {
      x[1]
    } else {
      ""
    }
  }
}
tmpfn <- function(a, b) {
  m <- data.frame(a, b)
  apply(m, 1, tmpfn2)
}
m <- match(building$bldg_title, addr$bldg_title)
ma <- map(as.list(building$bldg_title), function(x) agrep(x, addr$bldg_title))
tmp <- cbind(m, sapply(ma, function(x) paste(x, collapse = ",")), building$bldg_name)
colnames(tmp) <- c("VCRGE","Access", "Name")
tmp <- as_tibble(tmp) %>% 
  mutate(Access = tmpfn(VCRGE, Access),
         VCRGE = as.numeric(VCRGE))
tmp %>%
  filter(is.na(VCRGE) & Access == "") %>% 
  select(Name)
```

```{r}
tmp
```