Visualization Blitz Bombs on map of London - Fig. 5.2 of Modern Statistics for Modern Biology. By Andrzej Oles. A rendered version is shown here: http://rpubs.com/WolfgangHuber/401178

Blitz-19400907-latlng.RData

Blitz-19400907.Rmd

---
title: "London Blitz Bombs"
author: "Andrzej K. Oleś (main), Wolfgang Huber (contributions)"
output: BiocStyle::html_document
---

<font style="color:red"> Is there a good way to list the dependencies (packages) upfront?</font>

This gist produces a visualization of the location of the bombs that were dropped on London in the
night of September, 7th, 1940 based on data provided by the Guardian Data Store
& London Fire Brigade Records referenced by the website of the British National
Archives http://bombsight.org.

This is an alternative method to arrive at something like Figure 5.2 of the book [*Modern Statistics for Modern Biology*](http://www.huber.embl.de) by Susan Holmes and Wolfgang Huber.

Instead of running Steps 1 and 2, you can also use the precomputed dataframe `coordinates` and charge ahead to Step 3 directly.

# Download the street addresses

```{r data_source, message = FALSE, warning = FALSE, eval = !file.exists("Blitz-19400907-latlng.RData")}
library("googlesheets")

addresses_url <- "https://docs.google.com/spreadsheets/d/1rL68hnF9bHHg3p72ti_reSvwK12VEdABN5Q7YADXm3I"
addresses <- gs_read(gs_url(addresses_url))
```

# Query Google Maps to convert addresses into (latitude, longitude) coordinates

In order to query Google Maps, you need to set up the API key first, see `?googleway::set_key`. As an alternative, if you don't want to do this, you can use the dataframe `coordinates` that we provide along with this gist, see next step.

```{r geocode, eval = !file.exists("Blitz-19400907-latlng.RData")}
coordinates <- lapply(addresses$Location, function(loc) {
  x <- googleway::google_geocode(loc)
  x$results$geometry$location
}) %>% dpylr::bind_rows

save(coordinates, file = "Blitz-19400907-latlng.RData")
```

# Alternatively, load a precomputed dataframe

This step is alternative to Steps 1 and 2.

```{r load_data}
load("Blitz-19400907-latlng.RData")
```

# Display the locations on an online map and make a screenshot

We use the PhantomJS browser in order to do this. If it is not yet installed on your system, run the following chunk to do so:

```{r install_phantomjs, eval = FALSE}
## can we automaticallh check whether this needs to be run?
webshot::install_phantomjs()
```

Now we are ready to go:

```{r leaflet, message = FALSE}
library("leaflet")

bombsmap <- leaflet(coordinates) %>%
  addProviderTiles(providers$OpenStreetMap) %>%
  addCircleMarkers(color = "red",
                   radius = 5,
                   stroke = FALSE,
                   fillOpacity = 0.3) %>%
  setView(-0.1278, 51.5074, 12)

mapview::mapshot(bombsmap, file = "Blitz-19400907-latlng.png")
```

And here is the image:

![Blitz-19400907-latlng.png](Blitz-19400907-latlng.png)