Re: Windows Nightly Builds

[John Ralls]

> On Apr 25, 2016, at 7:12 AM, John Ralls  wrote:
> 
> 
>> On Apr 25, 2016, at 2:26 AM, Robert Fewell <14ubo...@gmail.com> wrote:
>> 
>> I see from the build logs that a new build was created on the 24th, ending
>> f4f-setup.exe but is not in the list of downloads, I wonder if the disk is
>> FULL ?
>> 
> 
> No, that was from me running the build by hand on the build VM to make sure 
> that the changes to the build scripts for CMake worked.
> 

And I've now copied it to code so you can download and test it.

Regards,
John Ralls



___
gnucash-devel mailing list
gnucash-devel@gnucash.org
https://lists.gnucash.org/mailman/listinfo/gnucash-devel

Re: segfault on customer report

[John Ralls]

> On Apr 25, 2016, at 3:15 AM, will Snow  wrote:
> 
>   Hi All,
> 
>   After running Reports -> business -> customer report gnucash has a
>   segfault (git master branch c739f4fae442e583389dd8c47100a38a06aac4c4)
> 
>   I'm interested in helping out with gnucash so looked a bit further. It
>   looks as though the segfault occurs when the get_slot method is run
>   because the book variable is empty.
> 
>   src/libqof/qof/qofbook.cpp
>   1194 KvpValue*
>   1195 qof_book_get_option (QofBook *book, GSList *path)
>   1196 {
>   1197 KvpFrame *root = qof_instance_get_slots(QOF_INSTANCE (book));
>   1198 Path path_v {KVP_OPTION_PATH};
>   1199 for (auto item = path; item != nullptr; item =
>   g_slist_next(item))
>   1200 path_v.push_back(static_cast(item->data));
>   1201 return root->get_slot(path_v);
>   1202 }
>   After looking a bit deeper it seems that SWIG_MustGetPtr (see below)
>   returns null to arg1 (book) which is then passed to qof_book_get_option
>   resulting in qof_instance_get_slots returning null instead of an object
>   and then it segfaults when trying to run get_slots.
> 
>   I thought adding some sort of check on the root variable but that's
>   probably wrong and not solving the underlying problem? I'm not sure
>   where to go from here with regards to fixing the issue. Any pointers
>   would be helpful. Thanks.
> 
>   ./src/engine/swig-engine.c
>   static SCM
>   _wrap_qof_book_get_option (SCM s_0, SCM s_1)
>   {
>   #define FUNC_NAME "qof-book-get-option"
> QofBook *arg1 = (QofBook *) 0 ;
> GSList *arg2 = (GSList *) 0 ;
> SCM gswig_result;
> SWIGUNUSED int gswig_list_p = 0;
> KvpValue *result = 0 ;
> {
>   arg1 = (QofBook *)SWIG_MustGetPtr(s_0, SWIGTYPE_p_QofBook, 1, 0);
> }
> arg2 = gnc_scm_to_gslist_string(s_1);
> result = (KvpValue *)qof_book_get_option(arg1,arg2);
> gswig_result = gnc_kvp_value_ptr_to_scm(result);
> 
> return gswig_result;
>   #undef FUNC_NAME

The problem is probably upstream of _wrap_qof_book_get_option(): Whatever is 
calling it from Scheme is likely passing a nullptr in s_0.  If Scheme isn't one 
of your languages please file a bug.

Regards,
John Ralls


___
gnucash-devel mailing list
gnucash-devel@gnucash.org
https://lists.gnucash.org/mailman/listinfo/gnucash-devel

Re: Windows Nightly Builds

[John Ralls]

> On Apr 25, 2016, at 2:26 AM, Robert Fewell <14ubo...@gmail.com> wrote:
> 
> I see from the build logs that a new build was created on the 24th, ending
> f4f-setup.exe but is not in the list of downloads, I wonder if the disk is
> FULL ?
> 

No, that was from me running the build by hand on the build VM to make sure 
that the changes to the build scripts for CMake worked.

Regards,
John Ralls
___
gnucash-devel mailing list
gnucash-devel@gnucash.org
https://lists.gnucash.org/mailman/listinfo/gnucash-devel

Re: How the date populates when transactions are duplicated

[David T.]

Glad you found a solution that works for you, although I’m scratching my head 
about your comment regarding “tabbing through” autofill. Once autofill supplies 
you a transaction, the entire transaction is duplicated (including notes and 
comments), and pressing enter commits it, which is a pretty quick process, all 
told. There is no need to tab through all fields.

Regardless, glad you have it sorted.
David

> On Apr 25, 2016, at 5:45 AM, will Snow  wrote:
> 
>   Thanks David,
> 
>   I found autofill takes longer, it's quicker to duplicate the record and
>   just enter a date rather than tabbing though and autofilling. However,
>   it looks as though the scheduled transactions feature solves the issue
>   I was having.
> 
>   Thanks again,
>   Will
> 
>   Sent: Saturday, April 23, 2016 at 2:13 PM
>   From: David 
>   To: "will Snow" , gnucash-devel@gnucash.org
>   Subject: Re: How the date populates when transactions are duplicated
> 
>   Will,
> 
>   Not that I'm a programmer or anything, but I can't see why you'd do
>   this. It seems to me that a user could set the date to their starting
>   point and use autofill to create their transactions. Having yet another
>   preference setting will only add to general user confusion, I think.
> 
>   David
> __
> 
>   From: will Snow 
>   Sent: Sat Apr 23 06:35:46 EST 2016
>   To: gnucash-devel@gnucash.org
>   Subject: How the date populates when transactions are duplicated
> 
> Hi All,
> 
> I noticed that when transactions are duplicated the date field uses the 
> latest d
> ate rather than the date of the original transaction. I tend to do my 
> accounts a
> ll at once towards the end of the year so this seems to slow me down quite a 
> lot
> as I have to enter an entire date all the time rather than just bumping up the
> day in the date field. So I was thinking about adding an option somewhere 
> where
> the user can specify how the date field populates on duplication (or perhaps 
> thi
> s already exists?).
> 
> It looks as though the relevant code is inside 
> src/gnome-utils/dialog-dup-trans.
> c
> and the flow goes as: gnc_dup_trans_dialog -> gnc_dup_trans_dialog_internal 
> -> g
> nc_dup_trans_dialog_create
> 
> Moreover, it looks as though date and date_p are key variables but I'm not 
> sure
> on where to pull the date stored in the transaction in the parent window. Or 
> if
> I should be doing it in some other way?
> 
> Any advice is greatly appreciated.
> 
> Thanks,
> Will
> __
> 
>   gnucash-devel mailing list
>   gnucash-devel@gnucash.org
>   [1]https://lists.gnucash.org/mailman/listinfo/gnucash-devel
> 
> References
> 
>   1. https://lists.gnucash.org/mailman/listinfo/gnucash-devel
> ___
> gnucash-devel mailing list
> gnucash-devel@gnucash.org
> https://lists.gnucash.org/mailman/listinfo/gnucash-devel


___
gnucash-devel mailing list
gnucash-devel@gnucash.org
https://lists.gnucash.org/mailman/listinfo/gnucash-devel

segfault on customer report

[will Snow]

   Hi All,

   After running Reports -> business -> customer report gnucash has a
   segfault (git master branch c739f4fae442e583389dd8c47100a38a06aac4c4)

   I'm interested in helping out with gnucash so looked a bit further. It
   looks as though the segfault occurs when the get_slot method is run
   because the book variable is empty.

   src/libqof/qof/qofbook.cpp
   1194 KvpValue*
   1195 qof_book_get_option (QofBook *book, GSList *path)
   1196 {
   1197 KvpFrame *root = qof_instance_get_slots(QOF_INSTANCE (book));
   1198 Path path_v {KVP_OPTION_PATH};
   1199 for (auto item = path; item != nullptr; item =
   g_slist_next(item))
   1200 path_v.push_back(static_cast(item->data));
   1201 return root->get_slot(path_v);
   1202 }
   After looking a bit deeper it seems that SWIG_MustGetPtr (see below)
   returns null to arg1 (book) which is then passed to qof_book_get_option
   resulting in qof_instance_get_slots returning null instead of an object
   and then it segfaults when trying to run get_slots.

   I thought adding some sort of check on the root variable but that's
   probably wrong and not solving the underlying problem? I'm not sure
   where to go from here with regards to fixing the issue. Any pointers
   would be helpful. Thanks.

   ./src/engine/swig-engine.c
   static SCM
   _wrap_qof_book_get_option (SCM s_0, SCM s_1)
   {
   #define FUNC_NAME "qof-book-get-option"
 QofBook *arg1 = (QofBook *) 0 ;
 GSList *arg2 = (GSList *) 0 ;
 SCM gswig_result;
 SWIGUNUSED int gswig_list_p = 0;
 KvpValue *result = 0 ;
 {
   arg1 = (QofBook *)SWIG_MustGetPtr(s_0, SWIGTYPE_p_QofBook, 1, 0);
 }
 arg2 = gnc_scm_to_gslist_string(s_1);
 result = (KvpValue *)qof_book_get_option(arg1,arg2);
 gswig_result = gnc_kvp_value_ptr_to_scm(result);

 return gswig_result;
   #undef FUNC_NAME
___
gnucash-devel mailing list
gnucash-devel@gnucash.org
https://lists.gnucash.org/mailman/listinfo/gnucash-devel

Re: How the date populates when transactions are duplicated

[will Snow]

   Thanks David,

   I found autofill takes longer, it's quicker to duplicate the record and
   just enter a date rather than tabbing though and autofilling. However,
   it looks as though the scheduled transactions feature solves the issue
   I was having.

   Thanks again,
   Will

   Sent: Saturday, April 23, 2016 at 2:13 PM
   From: David 
   To: "will Snow" , gnucash-devel@gnucash.org
   Subject: Re: How the date populates when transactions are duplicated

   Will,

   Not that I'm a programmer or anything, but I can't see why you'd do
   this. It seems to me that a user could set the date to their starting
   point and use autofill to create their transactions. Having yet another
   preference setting will only add to general user confusion, I think.

   David
 __

   From: will Snow 
   Sent: Sat Apr 23 06:35:46 EST 2016
   To: gnucash-devel@gnucash.org
   Subject: How the date populates when transactions are duplicated

Hi All,

I noticed that when transactions are duplicated the date field uses the latest d
ate rather than the date of the original transaction. I tend to do my accounts a
ll at once towards the end of the year so this seems to slow me down quite a lot
 as I have to enter an entire date all the time rather than just bumping up the
day in the date field. So I was thinking about adding an option somewhere where
the user can specify how the date field populates on duplication (or perhaps thi
s already exists?).

It looks as though the relevant code is inside src/gnome-utils/dialog-dup-trans.
c
and the flow goes as: gnc_dup_trans_dialog -> gnc_dup_trans_dialog_internal -> g
nc_dup_trans_dialog_create

Moreover, it looks as though date and date_p are key variables but I'm not sure
on where to pull the date stored in the transaction in the parent window. Or if
I should be doing it in some other way?

Any advice is greatly appreciated.

Thanks,
Will
 __

   gnucash-devel mailing list
   gnucash-devel@gnucash.org
   [1]https://lists.gnucash.org/mailman/listinfo/gnucash-devel

References

   1. https://lists.gnucash.org/mailman/listinfo/gnucash-devel
___
gnucash-devel mailing list
gnucash-devel@gnucash.org
https://lists.gnucash.org/mailman/listinfo/gnucash-devel

Re: Windows Nightly Builds

[Robert Fewell]

I have previously built successfully on my XP VM so that was why I tried to
load a nightly build on it, seemed to be the simplest way to prove as it
takes so long to build on it.

I see from the build logs that a new build was created on the 24th, ending
f4f-setup.exe but is not in the list of downloads, I wonder if the disk is
FULL ?

Once I have a successful build on XP, I can install on windows 10 to prove
/ debug on their,

Robert

On 24 April 2016 at 23:31, Wm  wrote:

> In article  9pytszkpnf...@mail.gmail.com>
> Robert Fewell <14ubo...@gmail.com> wrote:
> >
> > OK, the build was created last night and I have downloaded and installed
> on
> > my XP VM but fails to start properly, console output below...
>
> I built on an XP system (real not VM) yesterday and got a working
> system.  I'm building again at the moment in an XP VM but it takes
> ages.
>
> Given that it will build under XP but not Win 8.1 here I wonder if
> your XP VM isn't missing something that the build relies on ?
>
> I'm using a modern.IE XP VM running under VirtualBox on Win 8.1 and
> will let you know if that works when it finishes.
>
> --
> Wm
>
>
___
gnucash-devel mailing list
gnucash-devel@gnucash.org
https://lists.gnucash.org/mailman/listinfo/gnucash-devel

Re: Windows Nightly Builds

[Wm]

In article 
Robert Fewell <14ubo...@gmail.com> wrote:
>
> OK, the build was created last night and I have downloaded and installed on
> my XP VM but fails to start properly, console output below...

I built on an XP system (real not VM) yesterday and got a working 
system.  I'm building again at the moment in an XP VM but it takes 
ages.

Given that it will build under XP but not Win 8.1 here I wonder if 
your XP VM isn't missing something that the build relies on ?

I'm using a modern.IE XP VM running under VirtualBox on Win 8.1 and 
will let you know if that works when it finishes.

-- 
Wm

___
gnucash-devel mailing list
gnucash-devel@gnucash.org
https://lists.gnucash.org/mailman/listinfo/gnucash-devel