From ba8af03b0fec489ea6ad589155159fce6c99f8e9 Mon Sep 17 00:00:00 2001
From: Lukas Gehrke <lukasgehrke@Lukass-MacBook-Pro.local>
Date: Wed, 2 Jan 2019 09:04:29 +0100
Subject: [PATCH] =?UTF-8?q?Fixt=20getter=20f=C3=BCr=20ersten=20paragraph.?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
---
r/GetBirthdate.R | 17 ++++++++++-------
1 file changed, 10 insertions(+), 7 deletions(-)
diff --git a/r/GetBirthdate.R b/r/GetBirthdate.R
index a00060c..dbe0cdd 100644
--- a/r/GetBirthdate.R
+++ b/r/GetBirthdate.R
@@ -13,7 +13,7 @@ library(data.table)
### Try to extract birthdate from infobox
### If there is no infobox, try to extract from introduction text
getBirthdate <- function(article) {
- # check if vcard exists
+ # check
if(!grepl("vcard", article)) {
# check first paragraph
introduction <- getIntroduction(article)
@@ -22,8 +22,6 @@ getBirthdate <- function(article) {
birthdate <- str_extract_all(introduction, "\\([^()]+\\)")[[1]]
# remove parentheses
birthdate <- substring(birthdate, 2, nchar(birthdate)-1)
-
- return(birthdate)
} else {
# retrun Null if there is no birthdate
return(0)
@@ -87,11 +85,16 @@ getIntroduction <- function(article) {
xml_find_all(page, ".//br") %>%
xml_remove
- # Get first paragraph
- introduction <- page %>%
+ # Get all paragraphs
+ paragraphs <- page %>%
html_nodes("p") %>%
- html_text() %>%
- .[[1]]
+ html_text()
+
+ # there will be some leading paragraphs containing only "\n"
+ # remove those leading paragraphs
+ remove <- c("\n")
+ cleaned <- setdiff(paragraphs, remove)
+ introduction <- cleaned[1]
# Return first paragraph
return(introduction)
--
GitLab