update readme for package usage

DESCRIPTION

@@ -0,0 +1,26 @@
+Package: hateimparlament
+Title: Protocolanalysis of German Bundestag
+Version: 0.0.0.9000
+Authors@R: 
+    person(given = "First",
+           family = "Last",
+           role = c("aut", "cre"),
+           email = "first.last@example.com",
+           comment = c(ORCID = "YOUR-ORCID-ID"))
+Description: Downloads, parses and analyses protocols of the current German parliament (Bundestag).
+License: `use_mit_license()`, `use_gpl3_license()` or friends to pick a
+    license
+Encoding: UTF-8
+LazyData: true
+Roxygen: list(markdown = TRUE)
+RoxygenNote: 7.1.1
+Imports: 
+    dplyr,
+    pbapply,
+    rvest,
+    stringr,
+    xml2
+Suggests: 
+    rmarkdown,
+    knitr
+VignetteBuilder: knitr

NAMESPACE

@@ -0,0 +1,7 @@
+# Generated by roxygen2: do not edit by hand
+
+import(dplyr)
+import(pbapply)
+import(stringr)
+import(tibble)
+import(xml2)

R/fetch.R

@@ -0,0 +1,46 @@
+mk_absolute_url <- function(path) paste0("https://www.bundestag.de", path)
+
+mk_url <- function(offset) {
+    mk_absolute_url %$% sprintf("/ajax/filterlist/de/services/opendata/543410-543410?offset=%d",
+                                offset)
+}
+
+download_protocol <- function(path, name, download_dir) {
+    fp <- paste0(download_dir, name)
+    try %$% download.file(mk_absolute_url(path), fp, quiet=T)
+    progress <<- progress + 1
+    setTimerProgressBar(pb, progress)
+}
+
+fetch_batch <- function(offset, download_dir) {
+    stopifnot("Offset must be numeric" = is.numeric(offset))
+    mk_url(offset) %>%
+        rvest::read_html() %>%
+        as.character() %>%
+        str_match_all("/resource/blob/.*?/([0-9]*-data\\.xml)") %>%
+        `[[`(1) ->
+        paths
+    mapply(download_protocol,
+           paths[,1],
+           paths[,2],
+           MoreArgs=list(download_dir = download_dir))
+    return(length(paths) > 0)
+}
+
+# TODO: error handling
+# - what if: page not reachable
+# - wrong format, etc.
+fetch_all <- function(download_dir="records/") {
+    cat("Fetching all available records from bundestag.de. This may take a while ...\n")
+    
+    # create progress bar
+    pb <<- timerProgressBar(min=0, max=250, width=40, char="+")
+    progress <<- 0
+    # close progress bar on exit (also on error)
+    on.exit({close(pb); cat("Done.\n")})
+    # fetch batch by batch
+    offset <- 0
+    while(fetch_batch(offset, download_dir)) offset <- offset + 10
+    # if successful, set progressbar to 100%
+    setTimerProgressBar(pb, 250)
+}

R/hateimparlament-package.R

@@ -0,0 +1,15 @@
+#' @details
+#' hateimparlament ist ein großartiges Paket!
+#' @import tibble
+#' @import dplyr
+#' @import pbapply
+#' @import stringr
+#' @import xml2
+#' @keywords internal
+"_PACKAGE"
+
+# The following block is used by usethis to automatically manage
+# roxygen namespace tags. Modify with care!
+## usethis namespace: start
+## usethis namespace: end
+NULL

R/helpers.R

@@ -1,2 +1,4 @@
 `%$%` <- function(f, x) f(x)
 `%.%` <- function(f, g) function(...) f(g(...))
+
+clear_na <- function(xs) xs[!is.na(xs)]

R/parse.R

@@ -1,14 +1,33 @@
-source("config.R")
-source("../utils/helpers.R")
-library("xml2")
-library(tibble)
-library(magrittr)
-
 # for usage see the example at the end
 
+read_all <- function(path="records/") {
+    cat("Reading all records from", path, "\n")
+    available_protocols <- list.files(path)
+    res <- pblapply(available_protocols, read_one, path=path)
+
+    lapply(res, `[[`, "redner") %>%
+        bind_rows() %>%
+        distinct() ->
+        redner
+    
+    lapply(res, `[[`, "reden") %>%
+        bind_rows() %>%
+        distinct() ->
+        reden
+
+    lapply(res, `[[`, "talks") %>%
+        bind_rows() %>%
+        distinct() ->
+        talks
+
+    list(redner = redner, reden = reden, talks = talks)
+}
+
 # this reads all currently parseable data from one xml
-read_one <- function(name) {
+read_one <- function(name, path) {
-    x <- read_xml(paste0(DOWNLOAD_DIR, name))
+    x <- tryCatch(read_xml(paste0(path, name)),
+                  error = function(c) NULL)
+    if (is.null(x)) return(NULL)
     cs <- xml_children(x)
 
     verlauf <- xml_find_first(x, "sitzungsverlauf")
@@ -22,6 +41,7 @@ read_one <- function(name) {
         xml_find_all("rede") %>%
         parse_redenliste() ->
         res
+
     list(redner = redner, reden = res$reden, talks = res$talks)
 }
 
@@ -60,7 +80,8 @@ parse_rede <- function(rede_xml) {
     reden <- list()
     for (node in cs) {
         if (xml_name(node) == "p") {
-            if (xml_attr(node, "klasse") == "redner") {
+            klasse <- xml_attr(node, "klasse")
+            if (!is.na(klasse) && klasse == "redner") {
                 if (!is.na(cur_redner)) {
                     rede <- c(rede_id = rede_id,
                               redner = cur_redner,
@@ -111,10 +132,10 @@ parse_rednerliste <- function(rednerliste_xml) {
 # EXAMPLE USE
 
 # make sure data ist downloaded via fetch.R
-res <- read_one("19038-data.xml")
+# res <- read_one("19126-data.xml")
-
+# 
-res$redner
+# res$redner
-res$reden
+# res$reden
-res$talks
+# res$talks
 
 # -------------------------------

R/repair.R

@@ -0,0 +1,49 @@
+fraktionen <- c("AFD" = "AfD",
+                "BÜNDNIS90/" = "BÜNDNIS 90 / DIE GRÜNEN",
+                "BÜNDNIS90/DIEGRÜNEN" = "BÜNDNIS 90 / DIE GRÜNEN",
+                "FRAKTIONSLOS" = "Fraktionslos",
+                "DIELINKE" = "DIE LINKE",
+                "SPD" = "SPD",
+                "CDU/CSU" = "CDU/CSU",
+                "FDP" = "FDP")
+
+repair_fraktion <- function(fraktion) {
+    cleaned <- str_to_upper %$% str_replace_all(fraktion, "\\s", "")
+    fraktionen[cleaned]
+}
+
+# takes vector of titel and keeps longest
+longest_titel <- function(titel) {
+    if (all(is.na(titel))) NA_character_
+    else titel[which.max %$% str_length(titel)]
+}
+
+# takes character vector, removes duplicates and collapses
+collect_unique <- function(xs) xs %>% clear_na() %>% unique() %>% str_c(collapse="&") %>% na_if("")
+
+# expects a tibble of redner and repairs
+repair_redner <- function(redner) {
+    redner %>% mutate(fraktion = Vectorize(repair_fraktion)(fraktion)) %>% # fix fraktion
+        group_by(id, vorname, nachname) %>%
+        summarize(fraktion = collect_unique(fraktion),
+                  titel = longest_titel(titel),
+                  rolle_kurz = collect_unique(str_squish(rolle_kurz)),
+                  rolle_lang = collect_unique(str_squish(rolle_lang)))
+}
+
+repair_reden <- function(reden) {
+    # TODO: fill with content
+    reden
+}
+
+repair_talks <- function(talks) {
+    # TODO: fill with content
+    talks
+}
+
+# repairs all tables
+repair <- function(parse_output) {
+    list(redner = repair_redner(parse_output$redner),
+         reden = repair_reden(parse_output$reden),
+         talks = repair_talks(parse_output$talks))
+}

README.md

@@ -1,6 +1,26 @@
-#Test
+# How to develop
 
-Hallo ich teste das gerade
+Wie kann man entwickeln?
+
+```r
+# alles geht mit devtools (laedt auch noch ein paar andere pakete)
+library(devtools)
+
+# neu laden aller paket funktionen
+load_all()
+```
+Wir verwenden NIEMALS source, etc.! Außerdem NIEMALD library(...) verwenden, sondern
+um neue pakete hinzuzufuegen (als dependency), verwende:
+```r
+use_package("my-good-old-package")
+```
+Um paket imports verfuegbar zu machen, muss man diese in `R/hateimparlament-package.R`
+als `@import <package>` hinzufuegen.
+
+Um dokumentationen neu zu laden / zu erstellen (ruft roxgen auf)
+```r
+document()
+```
 
 # Herunterladen
 

man/hateimparlament-package.Rd

@@ -0,0 +1,18 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/hateimparlament-package.R
+\docType{package}
+\name{hateimparlament-package}
+\alias{hateimparlament}
+\alias{hateimparlament-package}
+\title{hateimparlament: Protocolanalysis of German Bundestag}
+\description{
+Downloads, parses and analyses protocols of the current German parliament (Bundestag).
+}
+\details{
+hateimparlament ist ein großartiges Paket!
+}
+\author{
+\strong{Maintainer}: First Last \email{first.last@example.com} (\href{https://orcid.org/YOUR-ORCID-ID}{ORCID})
+
+}
+\keyword{internal}

scraping/config.R

@@ -1,2 +0,0 @@
-DOWNLOAD_DIR = "../data/" # warning: this is not created (should maybe)
-

scraping/fetch.R

@@ -1,36 +0,0 @@
-source("../utils/helpers.R")
-source("config.R")
-library(rvest)
-library(stringr)
-
-mk_absolute_url <- function(path) paste0("https://www.bundestag.de", path)
-
-mk_url <- function(offset) {
-    mk_absolute_url %$% sprintf("/ajax/filterlist/de/services/opendata/543410-543410?offset=%d",
-                                offset)
-}
-
-download_protocol <- function(path, name) {
-    fp <- paste0(DOWNLOAD_DIR, name)
-    try %$% download.file(mk_absolute_url(path), fp, quiet=T)
-}
-
-fetch_batch <- function(offset) {
-    stopifnot("Offset must be numeric" = is.numeric(offset))
-    mk_url(offset) %>%
-        read_html() %>%
-        as.character() %>%
-        str_match_all("/resource/blob/.*?/([0-9]*-data\\.xml)") %>%
-        `[[`(1) ->
-        paths
-    mapply(download_protocol, paths[,1], paths[,2])
-    return(length(paths) > 0)
-}
-
-# TODO: error handling
-# - what if: page not reachable
-# - wrong format, etc.
-fetch_all <- function() {
-    offset <- 0
-    while(fetch_batch(offset)) offset <- offset + 10
-}

vignettes/.gitignore

@@ -0,0 +1,2 @@
+*.html
+*.R

vignettes/funwithdata.Rmd

@@ -0,0 +1,36 @@
+---
+title: "funwithdata"
+output: rmarkdown::html_vignette
+vignette: >
+  %\VignetteIndexEntry{funwithdata}
+  %\VignetteEngine{knitr::rmarkdown}
+  %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+  collapse = TRUE,
+  comment = "#>"
+)
+```
+
+```r
+read_all() %>% repair() -> res
+
+reden <- res$reden
+redner <- res$redner
+talks <- res$talks
+
+# first tries
+
+left_join(reden, redner, by=c("redner" = "id")) %>%
+    group_by(fraktion) %>%
+    summarize(n = n()) %>%
+    ggplot(aes(x = fraktion, y = n)) +
+    geom_bar(stat = "identity")
+```
+
+
+```{r setup}
+library(hateimparlament)
+```