How to pipe cypher-scripts (UTF-8) containing umlauts to neo4j using cypher-shell (Windows)?

Question

The Cypher-script "Umlauts.cypher" (UTF-8, created in Windows-environment) contains one transaction for importing an "umlaut-object" containing some umlauts:

:BEGIN
CREATE(:Umlautobj{text:"ÄäÖöÜü"});
:COMMIT

For piping the script (Windows environment) the following code is used in a cmd-shell

TYPE Umlauts.cypher|cypher-shell.bat -u neo4 - p secret

Now, I try a Cypher-Query(Firefox-Browser) to get the text-information from the new node:

MATCH (n:Umlautobj) RETURN n.text

I get:

"Ã„Ã¤Ã–Ã¶ÃœÃ¼"

But I expected to get:

"ÄäÖöÜü"

How can I import umlauts correctly?

[Windows 10, neo4j-community 3.5.15, neo4j-DB on the same machine]

And if you just do a `TYPE Umlauts.cypher`, do characters are well displayed ? — logisima, Mar 06 '20 at 13:13
Try [using the notepad](https://stackoverflow.com/a/13464816/974731) to confirm that the file is actually in UTF-8, or to change its encoding if needed. — cybersam, Mar 06 '20 at 19:28
@cybersam: I allready confirmed UTF-8-encoding using Notepad++ ... :-/ — CB_Dev, Mar 10 '20 at 08:23
I also tried 'chcp 65001' in cmd to ensure UTF-8 is used ... But it does not help. — CB_Dev, Mar 10 '20 at 08:35

score 0 · Answer 1 · edited Jun 08 '20 at 08:24

I was having the same issue and searching for the solution. Couldn't find anything so write a small python script to resolve this.

    from py2neo import Graph
    neo_url = 'http://neo4j:password@localhost:7474/db/data'
    graph = Graph(neo_url)
    file_name = "test.cypher"

    with open(file_name, encoding="utf8") as file_in:
        query = ""
        for line in file_in:        
            if line.strip().lower() == ":begin" or line.strip().lower() == ":commit":
                continue

            query += line
            if line.strip()[-1] == ";":
                graph.run(query)
                query = ""

How to pipe cypher-scripts (UTF-8) containing umlauts to neo4j using cypher-shell (Windows)?

1 Answers1