From e110780ad506ebe0f9bcc6cb8d29d295257a5a44 Mon Sep 17 00:00:00 2001
From: Erik de Vries <erik@devries.pm>
Date: Wed, 16 Jan 2019 19:21:20 +0100
Subject: [PATCH] merger: idiotic fix for a non-problem, see comment on line 32

---
 R/merger.R | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/R/merger.R b/R/merger.R
index 440c462..5c0c0c0 100644
--- a/R/merger.R
+++ b/R/merger.R
@@ -29,6 +29,10 @@ merger <- function(row, out, text, clean) {
       # Regex removes all words consisting of or containing numbers, @#$%
       # Punctuation is not taken into account, as it is already filtered out, see above
     {if(clean == T) str_replace_all(.,"\\S*?[0-9@#$%]+[^\\s]*", "")  else . }
+    # In the very rare but obviously occuring (CxqrOmMB4Bzg6Uhtzw0P) case that a document consists only of punctuation, return an empty string
+    if (length(lemmas) == 0 ){
+      lemmas <- ''
+    }
     return(lemmas)
   }
   # Replacing $-marked punctuation with their regular forms