In joebrew/vilaweb: Reproducible data analysis for www.vilaweb.cat

# Basic knitr options
library(knitr)
opts_chunk$set(comment = NA, 
               echo = FALSE, 
               warning = FALSE, 
               message = FALSE, 
               error = TRUE, 
               cache = FALSE,
               fig.path = 'figures/')

# Libraries
library(vilaweb)
library(rtweet)
library(tidyverse)
library(databrew)
library(translateR)
library(sentimentr) # https://github.com/trinker/sentimentr
require(RPostgreSQL)
require(readr)  
require(DBI)

if(file.exists('tl.RData')){
  load('tl.RData')
} else {
  # Connect to the db
  pg = DBI::dbDriver("PostgreSQL")
  con = DBI::dbConnect(pg, dbname="twitter")
  tl <- RPostgreSQL::dbGetQuery(
    con,
    paste0("SELECT * FROM twitter")
  )
  save(tl, file = 'tl.RData')  
  dbDisconnect(con)
  }


# Search for the words wordiation
find_word <- function(x){
  # grepl('lazos amarillos|llaços grocs', tolower(x))
  grepl('chiringuito', tolower(x))
}
tl <- tl %>%
  mutate(word = find_word(tweet))

pd <- tl %>%
  # filter(tolower(username) %in% tolower(people)) %>%
  filter(word) %>%
  filter(date >= '2018-01-01',
         date <= '2018-12-31') %>%
  group_by(person = username) %>%
  tally %>%
  ungroup 
left <- expand.grid(# word = sort(unique(tl$word)),
                    # date = seq(min(pd$date),
                    #            max(pd$date),
                    #            by = 'month'),
                    person = sort(unique(tl$username)))
pd <- left_join(left, pd)

pd$n[is.na(pd$n)] <- 0


make_simple_plot <- function(p = sort(unique(pd$person))){
  ppd <- pd %>% filter(person %in% p) %>%
    filter(person %in% c('ciudadanoscs',
                         'ciutadanscs',
                         'inesarrimadas',
                         'carrizosacarlos',
                         'oscar_puente',
                         'albert_rivera',
                         'vox_es',
                         'santi_abascal',
                         'espciudadana',
                         'societatcc',
                          'sanchezcastejon',
               'rogertorrent',
               'carmencalvo_',
               'inesarrimadas',
               'albert_rivera',
               'pablo_iglesias_',
               # 'alevysoler',
               'j_zaragoza_',
               'carrizosacarlos',
               'eva_granados',
               'ppcatalunya',
               'alejandrotgn',
               'albiol_xg',
               'societatcc',
               'joantarda',
               'santi_abascal',
               'pdemocratacat',
               'pablo_casado_',
               'ciudadanoscs',
               'socialistes_cat',
               'espciudadana',
               'pablocasado',
               'esquerra_erc', 
               'miqueliceta', 
               'perearagones',
               'albanodante76',
               'quimtorraipla',
               # 'cupnacional',
               'krls',
               'gabrielrufian',
               'junqueras','rogertorrent',
               'carlescampuzano',
               'pablo_iglesias_'
                         )) %>%
    mutate(person = paste0('@', person))
  ppd <- ppd %>% arrange(n) %>%
    filter(!is.na(n),
           n > -1,
           !person %in% congress$user_name)
  ppd$person <- factor(ppd$person, levels = ppd$person)
  ggplot(data = ppd,
       aes(x = person,
           y = n,
           # group = person,
           fill = person)) +
  geom_bar(stat = 'identity',
           color = 'black',
           lwd = 0.1) +
  theme_databrew() +
  labs(x = '',
       y = 'Tweets',
       title = 'Tweets con la palabra "chiringuito(s)"',
       # subtitle = '*Incluye también "golpista", "golpistas", etc.',
       subtitle = 'Año 2018') +
  theme(legend.position = 'none') +
  theme(#axis.text.x = element_text(angle = 90,
         #                          size = 7),
        plot.title = element_text(size = 24),
        plot.subtitle =  element_text(size = 24),
        strip.text = element_text(size = 25)) +
    scale_fill_manual(name = '',
                       values = rev(databrew::make_colors(n = length(unique(ppd$person)), r2g=TRUE, categorical = FALSE))) +
    geom_text(aes(label = n),
              alpha = 0.6,
              size = 3.5,
              nudge_y = 1) +
    coord_flip() +
    theme(axis.text.y = element_text(size = 12))

}
make_simple_plot()
# 
# ggsave('~/Desktop/x.png')