---
title: "Charlotte Tweets"
output: 
  flexdashboard::flex_dashboard:
    social: menu
    source: embed
---


```{r setup, include=FALSE}
library(flexdashboard); library("readr"); library("dplyr"); library("lubridate"); library("streamgraph"); library("htmlwidgets");  library(xts); library(lubridate); library(dygraphs); library(reshape2)

#tweets <- read.csv("./CLT_beer_tweets.csv", stringsAsFactors = F)
tweets <- read.csv("./CharlotteTweets20Sample.csv", stringsAsFactors = F)

#tweets$text <- iconv(tweets$text, "latin1", "ASCII", sub="")

# Read my tweets
tweets_df <- tweets %>%
  select(postedTime, body) %>%
  mutate(text = tolower(body))

# Pick hashtags with regexp
hashtags_list <- regmatches(tweets_df$text, gregexpr("#[[:alnum:]]+", tweets_df$text))

# Create a new data_frame with (timestamp, hashtag) -pairs
hashtags_df <- data_frame()
for (i in which(sapply(hashtags_list, length) > 0)) {
  hashtags_df <-
    bind_rows(hashtags_df,
              data_frame(
                timestamp = paste0(substr(tweets_df$postedTime[i], 1, 10), " 00:00:00 EDT"),
                hashtag = hashtags_list[[i]]
              ))
}

# Process data for plotting
hashtags_df <- hashtags_df %>%
  # Pick top 10 hashtags
  filter(hashtag %in% names(sort(table(hashtag), decreasing=TRUE))[1:10]) %>%
  # Group by day
  mutate(day = as.Date(timestamp)) %>%
  group_by(day, hashtag) %>%
  summarise(value = n())
```

### Dygraphs: Line graphs for time series {data-height=500}

```{r}
tweets$estTime <- ymd_hms(tweets$postedTime, tz = "EDT")

agg.time <- tweets %>% group_by(estTime, geo.type) %>% summarise(Count = n())
agg.time$estTime <- strptime(agg.time$estTime, "%Y-%m-%d")

aqm <- melt(agg.time, id=c("estTime","geo.type"), measure.vars = c("Count"), na.rm=TRUE)
count <- dcast(aqm, estTime ~ geo.type, fun.aggregate = sum)
count$estTime <- with_tz(count$estTime, tz = "EDT")

tweet.time <- xts(
  x = count[,2:3],
  order.by = count$estTime,
  tz = "EDT"
)

 dygraph(tweet.time,  main = "Tweet Counts") %>%
    dyOptions(colors = RColorBrewer::brewer.pal(3, "Dark2"), includeZero = TRUE) %>%
    dyAxis("x", drawGrid = FALSE) %>%
    dyAxis("y", label = "Tweet Count") 
```


### Streamgraph: What are the top hashtags? {data-height=500}

```{r}
# Create streamgraph
sg <- streamgraph(data = hashtags_df, key = "hashtag", value = "value", date = "day",
                  offset = "silhouette", interpolate = "cardinal", scale = "date") %>%
        sg_legend(TRUE, "hashtag: ")

sg

```