[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

RE: Change in SVN ADD behaviour between 1.6.5 and 1.7.5

From: Mark Bertenshaw <mark.bertenshaw_at_fidessa.com>
Date: Tue, 26 Jun 2012 14:50:33 +0000

Stefan -

Thanks verifying this point. As a related matter, the error code returned in my case just happened to be 1. Is there a list of standard error codes returned from SVN ?

Regards,

--
Mark Bertenshaw
-----Original Message-----
From: Stefan Sperling [mailto:stsp_at_elego.de] 
Sent: 26 June 2012 15:44
To: Mark Bertenshaw
Cc: users_at_subversion.apache.org
Subject: Re: Change in SVN ADD behaviour between 1.6.5 and 1.7.5
On Tue, Jun 26, 2012 at 01:58:56PM +0000, Mark Bertenshaw wrote:
> Hi -
> 
> I have just cleared up a build issue which resulted from upgrading from Subversion 1.6.5 and 1.7.5. There is a point in the process where we add any newly created component logs with:
> 
>     SVN ADD -q $APPROOT/LOGS/*
> 
> In both versions of Subversion, this raises an error saying that the log file is already under version control. That's fine - the NANT script doesn't care about that. However, it seems that whilst 1.6.5 simply raised a warning for this process, 1.7.5 actually sets a return code of 1. NANT interprets this as an error, and the build process fails.
> 
> Was this change by design? I would prefer not to have to ignore errors in NANT, because it is feasible that other errors could occur and we would be none the wiser. On the other hand, I can't find an option for SVN ADD which says "ignore versioned files". I could write a script filtering by SVN STATUS, but this would be inelegant. At the moment, I am just ignoring errors for the NANT entry that calls SVN ADD.
> 
The exit code was deliberately changed for 1.7 so that scripts can detect whether they're trying to add files that don't exist or are already under version control.
In some scripting situations this is a good thing. In your situation it's probably a change for the worse... cannot make everyone happy :(
There was also some inconsistency in exit codes that svn subcommands returned in cases such as this, and with the new behaviour it is at least consistent across all subcommands.
********************************************************************************************************************************************************************************************
This message is intended only for the stated addressee(s) and may be confidential.  Access to this email by anyone else is unauthorised. Any opinions expressed in this email do not necessarily reflect the opinions of Fidessa. Any unauthorised disclosure, use or dissemination, either whole or in part is prohibited. If you are not the intended recipient of this message, please notify the sender immediately.
Fidessa plc - Registered office:
Dukes Court, Duke Street, Woking, Surrey, GU21 5BH, United Kingdom
Registered in England no. 3781700 VAT registration no. 688 9008 78
Fidessa LatentZero Limited  - Registered office:
Dukes Court, Duke Street, Woking, Surrey, GU21 5BH, United Kingdom
Registered in England no. 3656437 VAT registration no. 688 9008 78
Fidessa group plc - Registered Office:
Dukes Court, Duke Street, Woking, Surrey, GU21 5BH, United Kingdom
Registered in England no. 3234176 VAT registration no. 688 9008 78
Received on 2012-06-26 16:51:09 CEST

This is an archived mail posted to the Subversion Users mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.