# Web Source: SayRegExp blank separator
**URL:** https://community.mp3tag.de/t/web-source-sayregexp-blank-separator/13945
**Category:** Web Sources Discussion
**Created:** [September 30, 2012, 8:58am UTC](https://community.mp3tag.de/t/web-source-sayregexp-blank-separator/13945 "2012-09-30T08:58:40Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@vkostas](https://community.mp3tag.de/u/vkostas)
#### Post date: [September 30, 2012, 8:58am UTC](https://community.mp3tag.de/t/web-source-sayregexp-blank-separator/13945/1 "2012-09-30T08:58:40Z")
I have found out that SayRegExp separates the searches by space when blank separator has been chosen.
If I use a non-blank separator then result is correct.
In the following example I use the same source to extract artists.
Source is "Bryan Adams" ... "; Tina Turner" (I have joined correct separator to artist name)
result 1: "Bryan Adams ; Tina Turner" \<---- incorrect
result 2: "Bryan Adams; Tina Turner" \<---- CORRECT
If I set SayRegExp separator to comma (,) then I get
"Bryan Adams,; Tina Turner"
You can use following code to test it.
```
[Name]=discogs.com
[BasedOn]=http://api.discogs.com
[PreviewUrl]=http://www.discogs.com/release/%s
[AlbumUrl]=http://api.discogs.com/releases/%s
[SearchBy]=1284880
[ParserScriptIndex]=...
#
[ParserScriptAlbum]=...
#set stream to "artists": [{"join": "", "name joined": "Bryan Adams", "anv": ""}, {"join": "; ", "name joined": "; Tina Turner", "anv": ""}]
regexpreplace ".+" "\"artists\"\: \[\{\"join\"\: \"\", \"name joined\"\: \"Bryan Adams\", \"anv\"\: \"\"\}, \{\"join\"\: \"; \", \"name joined\"\: \"; Tina Turner\", \"anv\"\: \"\"\}\]"
outputto "artist"
sayregexp "(?<=\"name joined\": \").*?(?=\")" "" "}]"
outputto "albumArtist"
sayregexp "(?<=\"name joined\": \").*?(?=\")" "," "}]"
```
---
### Author:  [@vkostas](https://community.mp3tag.de/u/vkostas)
#### Post date: [October 21, 2012, 9:41pm UTC](https://community.mp3tag.de/t/web-source-sayregexp-blank-separator/13945/2 "2012-10-21T21:41:03Z")
No one has faced this issue?