In courtsbr/scraperTJRS:

Para fazer o download dos DJE utilize o seguinte código:

meses_a_coletar <- readRDS('/home/storage/platipus/projects/scrapers/scraperTJRS/meses_a_coletar.rds')

local_pdf <- '/mnt/storage/platipus/raw/estadual/TJRS/dje/'

library(dplyr)

resultado <- meses_a_coletar  %>% 
  filter(!is.na(data)) %>%
#  head(100) %>% 
  select(-mesano) %>% 
  plyr::mdply(download_diario_oficial_tjrs, pasta = local_pdf)

baixados <- data_frame(arq = list.files(local_pdf)) %>%
  mutate(nome_caderno = gsub('.pdf','',substr(arq,21,nchar(arq)), fixed = T),
         data = substr(arq, 10, 19))

Rpdf <- readPDF(control = list(text = "-layout"))


djes <- list.files(local_pdf, full.names = T)

arqs <- Corpus(URISource(djes[1:10]), 
                   readerControl = list(reader = Rpdf)) %>% 
  tm_map(content_transformer(tolower))

dtm <- DocumentTermMatrix(arqs)

meses_a_coletar <- readRDS('meses_a_coletar.rds')

local_page_matchs <- '/home/storage/platipus/raw/estadual/TJRS/pages/'

dir.create(local_page_matchs, recursive = T)

edicoes <- meses_a_coletar %>% 
  filter(!is.na(ed)) %>% 
  with(unique(ed))

ed_test <- edicoes[1]

links = character()

for(e in edicoes){
  links <- c(links,unlist(matches_busca_livre_tjrs('bamerindus',e)))
}

links_hsbc = character()

for(e in edicoes){
  links_hsbc <- c(links,unlist(matches_busca_livre_tjrs('hsbc',e)))
}

links_sistema = character()

for(e in edicoes){
  links_sistema <- c(links,unlist(matches_busca_livre_tjrs('banco sistema',e)))
}

links <- bind_rows(data_frame(link = links, banco = 'bamerindus'),
         data_frame(link = links_hsbc, banco = 'hsbc'),
         data_frame(link = links_sistema, banco = 'banco sistema'))

links <- links %>% 
         mutate(id = 1:n(),
                path = paste0(local_page_matchs,'page_',banco,'_',id,'.pdf'))

saveRDS(links, 'paginas_bam_hsbc_sist.rds')

links %>% 
  select(link, path) %>% 
  plyr::d_ply(c('link','path'),function(x){
    final <- x$link %>% 
      rvest::html_session() %>% 
      rvest::html_nodes('iframe') %>% 
      rvest::html_attr('src')

    u <- sprintf('http://www3.tjrs.jus.br/servicos/diario_justica/%s',final)

    download_pdf_tjrs(u, x$path)
    })

local_page_matchs_txt <- '/home/storage/platipus/raw/estadual/TJRS/pages/txt'

dir.create(local_page_matchs_txt)


links %>% 
  mutate(path_txt = gsub('pages','txt',path)) %>% 
  plyr::d_ply(.c('link','path'), function(x){

  })

courtsbr/scraperTJRS documentation built on May 29, 2019, 12:37 p.m.