problems, rev 12354

[Mark Johnson]

Josh Sled wrote:

>On Sat, 2006-01-14 at 20:02 -0700, Mark Johnson wrote:
>  
>
>>I selected a simple checkbook, and made all top-level accounts 
>>placeholders (also Current Assets).  No opening balances.  Then gnucash 
>>unexpectedly quit.
>>
>>Here is the resulting backtrace:
>>Program received signal SIGSEGV, Segmentation fault.
>>[Switching to Thread -1224492832 (LWP 14760)]
>>0xb732a179 in g_type_check_instance_cast () from /usr/lib/libgobject-2.0.so.0
>>(gdb) bt
>>#0  0xb732a179 in g_type_check_instance_cast () from /usr/lib/libgobject-2.0.so.0
>>#1  0xb6ea9359 in gnc_tree_model_account_do_deletions (unused=0x0)
>>    at gnc-tree-model-account.c:1445
>>#2  0xb72aec03 in g_child_watch_add () from /usr/lib/libglib-2.0.so.0
>>#3  0xb72abc02 in g_main_depth () from /usr/lib/libglib-2.0.so.0
>>    
>>
>
>Yup ...  this is consistent with what I was seeing earlier today.
>There's some logic handing synchronizing account-deletions with the
>account-tree model that's not quite right, particularly when tearing
>down the new-account-hierarchy druid.  The way it manifests itself, I
>didn't run into it *before* making my changes, but started to as I was
>in the process of changing related code; frankly, I'm glad to see you
>run into it as well. :)
>
>In any case, I'm hoping to debug this in the balance of this weekend.
>
>  
>
Finally managed to get rev. 12369 compiled.  (Computer troubles and work 
delayed it; nothing to do with gnucash.)

The new-account-hierarchy-druid worked nicely.  The new accounts showed 
up as expected.  It was a nice touch that the top-level accounts are 
already marked as placeholders.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.gnucash.org/pipermail/gnucash-devel/attachments/20060117/5de8572a/attachment.html