[JENKINS-1352] Special characters in job names causes 404 error

[jenkins-infra-bot]

• Go to the New job page.
• For the name of the job, enter 'Contrôleur' (French for 'controller').
• Hudson then attempts to open page
  http://localhost:8080/job/Contr%C3%B4leur/configure
  and I get the following error:
  <<
  Status Code: 404
  Exception:
  Stacktrace:

(none)

Generated by Winstone Servlet Engine v0.9.10 at Mon Mar 03 15:51:41 CET 2008
>>

However, URL http://localhost:8080/job/Contr%F4leur/configure seems to work OK.

My recommandation: prevent the user from entering those special characters. As
seen with issue #929, allowing those special characters might have other issues
down the road.

---

Originally reported by elefevre, imported from: Special characters in job names causes 404 error

• status: Closed
• priority: Minor
• component(s): core
• resolution: Fixed
• resolved: 2008-03-13T23:56:35+00:00
• votes: 0
• watchers: 0
• imported: 2025-11-24

Raw content of original issue

	
Go to the New job page.
	
For the name of the job, enter 'Contrôleur' (French for 'controller').
	
Hudson then attempts to open page
http://localhost:8080/job/Contr%C3%B4leur/configure
and I get the following error:
<<
Status Code: 404
Exception:
Stacktrace:



(none)


Generated by Winstone Servlet Engine v0.9.10 at Mon Mar 03 15:51:41 CET 2008
>>

However, URL http://localhost:8080/job/Contr%F4leur/configure seems to work OK.

My recommandation: prevent the user from entering those special characters. As
seen with issue #929, allowing those special characters might have other issues
down the road.

• environment: Platform: All, OS: All

[jenkins-infra-bot]

kohsuke:

• Original comment link
• Raw content of original comment:

  Limiting character set to ASCII is simply not acceptable in many parts of the
  world, particularly in East Asia countries.
  
  In this case, whoever decoding the URL is doing it incorrectly — URL escape is
  always UTF-8, and %F4 is iso-8859-1 encoding.
  

Limiting character set to ASCII is simply not acceptable in many parts of the
world, particularly in East Asia countries.

In this case, whoever decoding the URL is doing it incorrectly — URL escape is
always UTF-8, and %F4 is iso-8859-1 encoding.

[jenkins-infra-bot]

kohsuke:

• Original comment link
• Raw content of original comment:

  Note to myself when filing a bug against Winstone. See
  http://en.wikipedia.org/wiki/Percent-encoding and in particular RFC 3986. Also
  see Eliote Rusty Harold's comment in
  http://weblogs.java.net/blog/joconner/archive/2005/07/charset_traps.html
  

Note to myself when filing a bug against Winstone. See
http://en.wikipedia.org/wiki/Percent-encoding and in particular RFC 3986. Also
see Eliote Rusty Harold's comment in
http://weblogs.java.net/blog/joconner/archive/2005/07/charset_traps.html

[jenkins-infra-bot]

kohsuke:

• Original comment link
• Raw content of original comment:

  Filed in
  http://sourceforge.net/tracker/index.php?func=detail&aid=1909365&group_id=98922&atid=622497
  and patched in 0.9.10-hudson-3
  

Filed in
http://sourceforge.net/tracker/index.php?func=detail&aid=1909365&group_id=98922&atid=622497
and patched in 0.9.10-hudson-3

[jenkins-infra-bot]

kohsuke:

• Original comment link
• Raw content of original comment:

  This was fixed in 1.190.
  

This was fixed in 1.190.