data-ch.r

#
# scrape full MP listing
#

if (!file.exists("data/sponsors-ch.csv")) {
  
  deputes = data.frame()
  for (i in 48:54) {
    
    cat("Parsing MPs in legislature", i, "... ")
    
    h = htmlParse(paste0("http://www.lachambre.be/kvvcr/showpage.cfm?section=/depute&language=fr&rightmenu=right_depute&cfm=cvlist54.cfm?legis=", i, "&today=n"))
    h = xpathSApply(h, "//a[contains(@href, 'key=')]/@href")
    
    cat(length(h), "pages\n")
    
    for (j in h) {
      
      hh = htmlParse(paste0("http://www.lachambre.be/kvvcr/", j))
      hh = data.frame(legislature = i,
                      nom = str_clean(xpathSApply(hh, "//form[@id='myform']/center/h2", xmlValue)),
                      mandate = paste0(xpathSApply(hh, "//div[@id='section']/*/a[contains(@href, 'lactivity=')]", xmlValue), collapse = ";"),
                      photo = xpathSApply(hh, "//img[contains(@src, 'cv/')]/@src"),
                      url = j,
                      bio = xpathSApply(hh, "//table[2]/tr[@valign='top'][2]/td/p", xmlValue))
      deputes = rbind(deputes, hh)
      cat(".")
      
    }
    
    cat("\n")
    
  }
  
  write.csv(deputes, "data/sponsors-ch.csv", row.names = FALSE)
  
}

deputes = read.csv("data/sponsors-ch.csv", stringsAsFactors = FALSE)

# download photos

for (i in unique(deputes$photo)) {

  photo = gsub("/site/wwwroot/images/cv", "photos_ch", i)

  if (!file.exists(photo)) {
    
    h = GET(paste0("http://www.lachambre.be", i))
    
    if (h$status_code == 200)
      writeLines(content(h, "text"), photo)
    
  }
  
  if (file.exists(photo)) {
  
    deputes$photo[ deputes$photo == i ] = photo
  
  } else {
    
    deputes$photo[ deputes$photo == i ] = NA
  
  }

}

deputes$bio = str_clean(gsub("▀ ", "", deputes$bio))

# biographies that failed to scrape properly
deputes$bio[ deputes$url == "showpage.cfm?section=/depute&language=fr&rightmenu=right_depute&cfm=cvview54.cfm?key=06536&lactivity=54" ] = "Député socialiste (PS) de la circonscription électorale du Brabant wallon depuis le 30 juillet 2014, en remplacement de monsieur André Flahaut, élu ministre du gouvernement de la Communauté française. Membre du groupe PS. Né à Mons le 10 juillet 1967."
deputes$bio[ deputes$url == "showpage.cfm?section=/depute&language=fr&rightmenu=right_depute&cfm=cvview54.cfm?key=00951&lactivity=54" ] = "Député démocrate Humaniste (cdH) de la circonscription électorale de Bruxelles-Capitale depuis le 30 juillet 2014, en remplacement de madame Céline Fremault, élue membre du gouvernement de la Région de Bruxelles-Capitale. Député de la circonscription électorale de Bruxelles-Hal-Vilvorde du 20 mars 2008 au 13 juin 2010, du 7 décembre 2011 au 16 mai 2013 et du 16 mai 2013 au 25 mai 2014. Membre du groupe cdH. Né à Fataki (Congo belge) le 17 janvier 1958. Sénateur du 13 juin 1999 au 18 mai 2003 et du 12 juillet 2007 au 20 mars 2008. Docteur en médecine, chirurgie et accouchements. Spécialisation en médecine tropicale. Certificat d'épidémiologie et de biostatistique. Maîtrise en politique internationale. Maîtrise en management des institutions de soins de santé. Ancien directeur général de 'Handicap International'. Ancien directeur général adjoint de Médecins sans Frontières. Vice-président de l'Hôpital Brugman. Président du comité de gestion du Fonds des maladies professionnelles. Ancien conseiller communal de Bruxelles. Ancien échevin de Bruxelles. Conseiller communal de Woluwe-Saint-Pierre. Chevalier de l'Ordre de Léopold. Président de la commission des Naturalisations."
deputes$bio[ deputes$url == "showpage.cfm?section=/depute&language=fr&rightmenu=right_depute&cfm=cvview54.cfm?key=04491&lactivity=54" ] = "Député libéral (Open Vld) de la circonscription électorale d Anvers du 7 décembre 2011 au 25 mai 2014 et depuis le 30 juillet 2014, en remplacement de madame Annemie Turtelboom, élue ministre du gouvernement flamand le 25 juillet 2014. Député de la circonscription électorale d'Anvers du 7 décembre 2011 au 25 mai 2014. Membre du groupe Open Vld. Né à Turnhout le 25 avril 1965. Licencié en sciences économiques appliquées. Chef d'entreprise. Ancien conseiller provincial (Anvers) Bourgmestre de Merksplas."

# constituencies (simplified to province due to changing nature over elections)
y = gsub("(.*)\\d+?(.*)", "\\1", deputes$bio)
y = gsub("(.*)(arrondissement|arrondsissement|circonscription( électoral)?(e)?|kieskring)", "", y)
y = gsub("^\\s(d\\s|d'|d\u0092|de\\s|du\\s|de\\sla\\s)+", "", y)
y = gsub("\\s(depuis|sedert)(.*)|\\sdu\\s$|\\s\\(PS\\)", "", y)
y = str_trim(gsub("\\s-\\s", "-", y))

# fill in a few values that are not captured by the code above
y[ deputes$url == "showpage.cfm?section=/depute&language=fr&rightmenu=right_depute&cfm=cvview54.cfm?key=00632&lactivity=48" ] = "Termonde"
y[ deputes$url == "showpage.cfm?section=/depute&language=fr&rightmenu=right_depute&cfm=cvview54.cfm?key=00106&lactivity=51" ] = "Bruxelles-Hal-Vilvorde"

# ==============================================================================
# CHECK CONSTITUENCIES
# ==============================================================================

# Province_du_Brabant_flamand
y[ y %in% c("Brabant flamand") ] = "Province_du_Brabant_flamand"
y[ y %in% c("Bruxelles", "Bruxelles-Capitale") ] = "Province_du_Brabant_flamand" # "Bruxelles"
y[ y %in% c("Bruxelles-Hal-Vilvorde", "Bruxelles-Halle-Vilvorde") ] = "Province_du_Brabant_flamand" # "Bruxelles-Hal-Vilvorde"
y[ y %in% c("Louvain") ] = "Province_du_Brabant_flamand" # "Louvain"

# Province_du_Brabant_wallon
y[ y %in% c("Brabant wallon", "Brabant-wallon") ] = "Province_du_Brabant_wallon"
y[ y %in% c("Nivelles") ] = "Province_du_Brabant_wallon"

# Province_de_Flandre-Occidentale
y[ y %in% c("Flandre occidentale", "Flandre Occidentale") ] = "Province_de_Flandre-Occidentale"
y[ y %in% c("Courtrai", "Courtrai-Roulers-Tielt", "Roulers-Tielt") ] = "Province_de_Flandre-Occidentale"
y[ y %in% c("Furnes-Dixmude-Ostende", "Furnes-Dixmude-Ypres-Ostende", "Ypres") ] = "Province_de_Flandre-Occidentale"
y[ y %in% c("Bruges") ] = "Province_de_Flandre-Occidentale"

# Province_de_Flandre-Orientale
y[ y %in% c("Flandre orientale") ] = "Province_de_Flandre-Orientale"
y[ y %in% c("Gand-Eeklo") ] = "Province_de_Flandre-Orientale"
y[ y %in% c("Saint-Nicolas", "Saint-Nicolas-Termonde", "Termonde") ] = "Province_de_Flandre-Orientale"
y[ y %in% c("Alost", "Alost-Audenarde", "Audenarde") ] = "Province_de_Flandre-Orientale"

# Province_de_Hainaut
y[ y %in% c("Hainaut", "Henegouwen") ] = "Province_de_Hainaut"
y[ y %in% c("Charleroi", "Charleroi-Thuin", "Thuin") ] = "Province_de_Hainaut"
y[ y %in% c("Mons", "Mons-Soignies", "Soignies") ] = "Province_de_Hainaut"
y[ y %in% c("Tournai-Ath-Mouscron", "Ath-Tournai-Mouscron") ] = "Province_de_Hainaut"

# Province_d'Anvers
y[ y %in% c("Anvers") ] = "Province_d'Anvers"
y[ y %in% c("Malines", "Malines-Turnhout", "Turnhout") ] = "Province_d'Anvers"

# Province_de_Liège
y[ y %in% c("Liège") ] = "Province_de_Liège"
y[ y %in% c("Verviers") ] = "Province_de_Liège"
y[ y %in% c("Huy-Waremme") ] = "Province_de_Liège"

# Province_de_Limbourg
y[ y %in% c("Hasselt", "Hasselt-Tongres- Maaseik", "Hasselt-Tongres-Maaseik", "Tongres-Maaseik") ] = "Province_de_Limbourg"
y[ y %in% c("Limbourg", "Limburg") ] = "Province_de_Limbourg"

# Province_de_Luxembourg
y[ y %in% c("Arlon-Marche-en-Famenne-Bastogne", "Arlon-Marche-en-Famenne-Bastogne-Neufchâteau-Virton", "Neufchâteau-Virton") ] = "Province_de_Luxembourg"
y[ y %in% c("Luxembourg") ] = "Province_de_Luxembourg"

# Province_de_Namur
y[ y %in% c("Namur", "Namur-Dinant-Philippeville", "Dinant-Philippeville") ] = "Province_de_Namur"

table(y[ !grepl("_", y)], exclude = NULL)
deputes$constituency = y

cat("Checking constituencies,", sum(is.na(deputes$constituency)), "missing...\n")
for (i in na.omit(unique(deputes$constituency))) {
  
  g = GET(paste0("https://", meta[ "lang" ], ".wikipedia.org/wiki/", i))
  
  if (status_code(g) != 200)
    cat("Missing Wikipedia entry:", i, "\n")
  
  g = xpathSApply(htmlParse(g), "//title", xmlValue)
  g = gsub("(.*) — Wikipédia(.*)", "\\1", g)
  
  if (gsub("\\s", "_", g) != i)
    cat("Discrepancy:", g, "(WP) !=", i ,"(data)\n")
  
}

deputes$sexe = str_extract(deputes$bio, "Député(e)?")
deputes$sexe[ deputes$nom == "Juliette Boulet" ] = "Députée" # Flemish text
deputes$sexe[ deputes$nom == "Paul Meeus" ] = "Député" # typo
deputes$sexe[ deputes$sexe == "Députée" ] = "F"
deputes$sexe[ deputes$sexe == "Député" ] = "M"

# mandate years (rough approximation)
#deputes$from = as.numeric(sapply(str_extract_all(deputes$mandate, "[0-9]{4}"), min))
#deputes$to = as.numeric(sapply(str_extract_all(deputes$mandate, "[0-9]{4}"), max))
#deputes$nyears = deputes$to - deputes$from + 1

# mandate years (detailed)
deputes$mandate = sapply(deputes$mandate, function(x) {
  x = as.numeric(unlist(str_extract_all(x, "[0-9]{4}")))
  paste0(seq(min(x), max(x)), collapse = ";")
})

deputes$annee_naissance = str_extract(deputes$bio,
                                      "(N|n)é(e)? (a|à|te) (\\w|,|\\.|'|-|\\(|\\)|\\s)+ le \\d+(er)? \\w+ \\d{4}")
deputes$annee_naissance = as.numeric(str_extract(deputes$annee_naissance, "[0-9]{4}"))
deputes$annee_naissance[ deputes$nom == "Juliette Boulet" ] = 1981 # Flemish text
deputes$annee_naissance[ deputes$nom == "Magda Raemaekers" ] = 1947 # Wikipedia additions
deputes$annee_naissance[ deputes$nom == "Frédéric Daerden" ] = 1970
deputes$annee_naissance[ deputes$nom == "Georges Dallemagne" ] = 1958
deputes$annee_naissance[ deputes$nom == "Frank Wilrycx" ] = 1965
deputes$annee_naissance[ deputes$nom == "Koen Metsu" ] = 1981
deputes$annee_naissance = as.integer(deputes$annee_naissance)

# full URL
deputes$url = paste0("http://www.lachambre.be/kvvcr/", deputes$url)

# ============================================================================
# QUALITY CONTROL
# ============================================================================

# - might be missing: born (int of length 4), constituency (chr),
#   photo (chr, folder/file.ext)
# - never missing: sex (chr, F/M), nyears (int), url (chr, URL),
#   party (chr, mapped to colors)

cat("Missing", sum(is.na(deputes$annee_naissance)), "years of birth\n")
stopifnot(is.integer(deputes$annee_naissance) & 
            nchar(deputes$annee_naissance) == 4 | is.na(deputes$annee_naissance))

cat("Missing", sum(is.na(deputes$constituency)), "constituencies\n")
stopifnot(is.character(deputes$constituency))

cat("Missing", sum(is.na(deputes$photo)), "photos\n")
stopifnot(is.character(deputes$photo) &
            grepl("^photos(_\\w{2})?/(.*)\\.\\w{3}", deputes$photo) | is.na(deputes$photo))

stopifnot(!is.na(deputes$sexe) & deputes$sexe %in% c("F", "M"))
# stopifnot(!is.na(deputes$nyears) & is.integer(deputes$nyears)) ## computed on the fly
stopifnot(!is.na(deputes$url) & grepl("^http(s)?://(.*)", deputes$url))
stopifnot(deputes$party %in% names(colors))

#
# scrape dossiers in legislatures 47-54
#

root = "http://www.dekamer.be/kvvcr/showpage.cfm?section=/flwb&language=fr&cfm=Listdocument.cfm?legislat="
cat("Scraping raw lower chamber data...\n")

for (i in 47:54) {
  
  file = paste0("data/dossiers-ch", i, ".csv")
  if (!file.exists(file)) {
    
    cat("Scraping legislature", i, "... ")
    
    dossiers = data.frame()
    
    h = htmlParse(paste0(root, i))
    h = xpathSApply(h, "//a[@class='link']/@href")
    
    cat(length(h), "pages\n")
    for (j in h) {
      
      hh = htmlParse(paste0("http://www.dekamer.be/kvvcr/", j))
      hh = xpathSApply(hh, "//div[@class='linklist_0']/a/@href")
      hh = gsub("(.*)&dossierID=(.*)", "\\2", as.character(hh))
      
      cat("Page", sprintf("%3.0f", which(h == j)),
          "scraping", sprintf("%3.0f", length(hh)), "dossiers\n")
      
      for (k in hh) {
        
        t = paste0("http://www.dekamer.be/kvvcr/showpage.cfm?section=flwb&language=fr&leftmenu=none&cfm=/site/wwwcfm/search/fiche.cfm?ID=", i, "K", k, "&db=FLWB&legislat=", i)
        t = try(readHTMLTable(t, which = 1, header = FALSE, stringsAsFactors = FALSE))
        if (!"try-error" %in% class(t)) {
          
          names(t) = c("variable", "value")
          t$variable = str_clean(t$variable)
          t$value = str_clean(t$value)
          
          t = data.frame(legislature = i, dossier = k,
                         t[ t$variable != "" | !is.na(t$value), ])
          dossiers = rbind(dossiers, t)
          
        } else {
          
          cat("Error: dossier", k, "legislature", i, "\n")
          
        }
        
      }
      
    }
    
    write.csv(dossiers, file, row.names = FALSE)
    
    cat("Scraped", nrow(dossiers), "rows over",
        n_distinct(dossiers$dossier), "dossiers\n")
      
  }
  
}

# parse sponsors from raw dossiers

dir = dir("data", pattern = "dossiers-ch\\d{2}.csv", full.names = TRUE)
for (k in dir) {
  
  file = gsub("dossiers", "sponsors", k)
  
  if (!file.exists(file)) {
    
    cat("\nParsing Chamber, legislature", gsub("\\D", "", k), "...\n")
    cat("Building", file)
    
    d = read.csv(k, stringsAsFactors = FALSE)  
    d$uid = paste0(d$legislature, "K", sprintf("%04.0f", d$dossier))
    
    uids = unique(d$uid)
    authors = data.frame()
    
    cat(" :", length(uids), "dossiers\n")
    # sink(gsub("csv$", "log", file))
    for (i in uids) {
      
      dd = subset(d, uid == i & !grepl("^NL|NL$", d$variable))
      # cat("Parsing dossier", i)
      
      status = dd$value[which(grepl("Etat d'avancement", dd$variable)) + 1]
      if (!length(status))
        status = NA
      
      topic = unique(dd$value[which(grepl("Eurovoc", dd$variable)) + 2])
      topic = paste0(topic, collapse = ",")
      
      sd = which(dd$variable == "Document principal")
      dd = dd[ min(sd):nrow(dd), ]
      
      sd = c(1, which(dd$variable == "Document(s) suivant(s)"))
      if (length(sd) < 2)
        sd = c(sd, nrow(dd))
      
      for (n in 1:(length(sd) - 1)) {
        
        ddd = dd[ sd[ n ]:sd[ n + 1 ], ]
        type = ddd$value[ which(ddd$variable == "Type")[1] + 1 ]
        
        au = which(ddd$value == "AUTEUR")
        if (length(au)) {
          au = sapply(au, function(x)
            paste(ddd$value[ x + 2 ], ddd$value[ x + 1 ], "[", ddd$value[ x + 3 ], "]"))
          au = unique(au)
        } else {
          au = NA
        }
        
        cs = which(ddd$value == "SIGNATAIRE")
        if (length(cs)) {
          cs = sapply(cs, function(x)
            paste(ddd$value[ x + 2 ], ddd$value[ x + 1 ], "[", ddd$value[ x + 3 ], "]"))
          cs = unique(cs)
        } else {
          cs = NA
        }
        
        a = na.omit(c(au, cs))
        a = a[ !grepl("ZZZ|0", a) ]
        
        # subset to cosponsored legislation
        if (length(a) > 1) {
          
          # cat("\n -", type, ":", length(au), "author(s)",
          #     ifelse(!length(na.omit(cs)), "",
          #            paste(length(na.omit(cs)), "cosponsor(s)")))
          
          authors = rbind(authors,
                          cbind(uid = paste0(i, "-", n),
                                dossier = i, document = n,
                                type, topic, status,
                                n_au = length(a),
                                authors = paste0(na.omit(au[ !grepl("ZZZ|0", au) ]), collapse = ";"),
                                cosponsors = paste0(na.omit(cs[ !grepl("ZZZ|0", cs) ]), collapse = ";")))
          
        } else if (length(a) == 1) {
          
          # cat("\n -", type, ":", a)
          authors = rbind(authors,
                          cbind(uid = paste0(i, "-", n),
                                dossier = i, document = n,
                                type, topic, status, n_au = 1,
                                authors = a,
                                cosponsors = ""))
          
        } else {
          
          # cat("\n -", type, ":", gsub("\\d|\\[|\\]|ZZZ|\\s", "", au))
          
        }
        
      }
      
      # cat("\n\n")
      
    }
    # sink()
    
    write.csv(authors, file, row.names = FALSE)
    
  }
  
}

b = bind_rows(lapply(dir("data", pattern = "sponsors-ch", full.names = TRUE),
                     read.csv, stringsAsFactors = FALSE))

b$legislature = substr(b$uid, 1, 2)

b$n_a = b$n_au # total
b$n_au = 1 + str_count(b$authors, ";")
b$n_au[ is.na(b$authors) ] = NA

b$n_co = 1 + str_count(b$cosponsors, ";")
b$n_co[ b$cosponsors == "" ] = 0
b$n_co[ is.na(b$cosponsors) ] = 0

b = subset(b, authors != "") # removes only one row -- dossier 52K1939 [ 52 ]

b$type[ grepl("AMENDEMENT|DECISION DE NE PAS AMENDER", b$type) ] = "AMENDEMENT"
b$type[ grepl("ADDENDUM|AVIS|BUDGET|CORRIGE|CONCLUSIONS|DECISION|ERRATA|PROJET|RAPPORT|ANNEXE|TEXTE", b$type) ] = "AUTRE"
b$type[ grepl("PROPOSITION DE LOI", b$type) ] = "PROPOSITION DE LOI"
b$type[ grepl("PROPOSITION", b$type) & !grepl("PROPOSITION DE LOI", b$type) ] = "PROPOSITION"

# print(table(b$type, b$legislature, exclude = NULL))

b$status[ grepl("ADOPTE|TERMINE", b$status) ] = "ADOPTE"
b$status[ grepl("CADUQUE|REJETE", b$status) ] = "REJETE"
b$status[ grepl("SANS OBJET|RETIRE", b$status) ] = NA

# print(table(b$type, b$status, exclude = NULL))

# join authors and cosponsors

b$sponsors = paste(b$authors, b$cosponsors, sep = ";")
b$sponsors = gsub(";(NA)?$", "", b$sponsors)

#
# name and party fixes
#

# Sabien (Lahaye-)Battheu [ 51-54 ]
b$sponsors = gsub("Sabien Battheu", "Sabien Lahaye-Battheu", b$sponsors)
b$sponsors = gsub("Karine Jiroflée", "Karin Jiroflée", b$sponsors)

# Emile Coulonvaux, PLP [ 47 ]
b$sponsors = gsub("\\[ PLP \\]", "[ PVV ]", b$sponsors)

# Jan Decorte INDEP/ROSSEM [ 48 ]
b$sponsors = gsub("Jan Decorte \\[ INDEP \\]", "Jan Decorte [ ROSSEM ]", b$sponsors)

# Arthur Honoré Buysse, libéral flamand [ 49-50 ]
b$sponsors = gsub("\\[ LIB \\]", "[ PVV ]", b$sponsors)

# party code fixes [53]
b$sponsors = gsub("Els Van Hoof \\[ 0 \\]", "Els Van Hoof [ CD&V - N-VA ]", b$sponsors)
b$sponsors = gsub("Stefaan De Clercq \\[ 0 \\]", "Stefaan De Clercq [ CD&V ]", b$sponsors)
b$sponsors = gsub("Myriam Delacroix-Rolin \\[ 0 \\]", "Myriam Delacroix-Rolin [ cdH ]", b$sponsors)
b$sponsors = gsub("Fatma Pehlivan \\[ 0 \\]", "Fatma Pehlivan [ sp.a ]", b$sponsors)

#
# party simplifications
#

b$sponsors = gsub("\\[ Groen(!)? \\]", "[ ECOLO ]", b$sponsors)
b$sponsors = gsub("\\[ (Agalev-)?(ECOLO|Ecolo)(-Groen)?(!)? \\]", "[ ECOLO ]", b$sponsors)
# b$sponsors = gsub("\\[ (PS|SP|sp.a)(-spirit)?(\\+Vl\\.Pro)? \\]", "[ SOC ]", b$sponsors)
b$sponsors = gsub("\\[ PS \\]", "[ SOC-F ]", b$sponsors)
b$sponsors = gsub("\\[ (SP|sp.a)(-spirit)?(\\+Vl\\.Pro)? \\]", "[ SOC-V ]", b$sponsors)
b$sponsors = gsub("\\[ (PVV|Open Vld|VLD) \\]", "[ LIB-V ]", b$sponsors)
b$sponsors = gsub("\\[ (MR|FDF|PRL|FDFPPW|PRLFDF) \\]", "[ LIB-F ]", b$sponsors)
b$sponsors = gsub("\\[ (VU|VU-ID|N-VA) \\]", "[ VOLKS ]", b$sponsors)
b$sponsors = gsub("\\[ (CVP|CD&V) \\]", "[ CDEM-V ]", b$sponsors)
b$sponsors = gsub("\\[ (PSC|cdH) \\]", "[ CDEM-F ]", b$sponsors)
b$sponsors = gsub("\\[ CD&V - N-VA \\]", "[ CDEM-V/VOLKS ]", b$sponsors)
b$sponsors = gsub("\\[ VB \\]", "[ VLAAMS ]", b$sponsors)
b$sponsors = gsub("\\[ PTB-GO! \\]", "[ PT ]", b$sponsors)
b$sponsors = gsub("\\[ MLD \\]", "[ DLB ]", b$sponsors) # Laurent Louis, always alone
b$sponsors = gsub("\\[ (ONAFH|INDEP|INDEP-ONAFH) \\]", "[ IND ]", b$sponsors)

# last fixes on duplicates (leave at end)
b$sponsors = gsub("Lisette Nelis-Van Liedekerke \\[ LIB-V \\]",
                  "Lisette Nelis-Van Liedekerke [ CDEM-V ]", b$sponsors) # l. 48
b$sponsors = gsub("Philippe Dallons \\[ SOC-F \\]", "Philippe Dallons [ ECOLO ]", b$sponsors) # l. 49
b$sponsors = gsub("Pierrette Cahay-André \\[ LIB-F \\]", "Pierrette Cahay-André [ CDEM-F ]", b$sponsors) # l. 49
b$sponsors = gsub("Richard Fournaux \\[ CDEM-F \\]", "Richard Fournaux [ LIB-F ]", b$sponsors) # l. 51
b$sponsors = gsub("Vincent Decroly \\[ IND \\]", "Vincent Decroly [ ECOLO ]", b$sponsors) # l. 50

# note: fixes do not cover sponsors from legislature 47,
# for which there is no sponsor details (age, sex, etc.)

## all keywords
# kw = unlist(strsplit(b$topic, ","))
# table(kw)[ table(kw) > 600 ]

write.csv(b, "data/bills-ch.csv", row.names = FALSE)