I noticed this before while doing my find account dialogue, at the start of
pull request 83 there is a patch for this and the account hidden setting.
On 22 September 2016 at 16:38, John Ralls <jra...@ceridwen.us> wrote:
>
> > On Sep 21, 2016, at 3:35 PM, Carsten Rinke <carsten.ri...@gmx.de> wrote:
> >
> > Hi,
> >
> > I try to make use of the "TaxRelated" flag of the accounts by calling
> xaccAccountSetTaxRelated (Account.c).
> >
> > I notice that on master this does not lead to the same result as on
> maint.
> >
> > To figure out what is wrong/different on master I simply added some
> printouts:
> >
> > ========= snip start ===========================
> >
> > gboolean
> > xaccAccountGetTaxRelated (const Account *acc)
> > {
> > GValue v = G_VALUE_INIT;
> > g_return_val_if_fail(GNC_IS_ACCOUNT(acc), FALSE);
> > qof_instance_get_kvp (QOF_INSTANCE(acc), "tax-related", &v);
> >
> > if (G_VALUE_HOLDS_BOOLEAN (&v))
> > printf("G_VALUE_HOLDS_BOOLEAN returns TRUE\n"); <=============
> > else
> > printf("G_VALUE_HOLDS_BOOLEAN returns FALSE, actually it is
> %s\n",G_VALUE_TYPE_NAME(&v)); <=============
> >
> > if (g_value_get_boolean (&v))
> > printf("g_value_get_boolean returns TRUE\n"); <=============
> > else
> > printf("g_value_get_boolean returns FALSE\n"); <=============
> >
> > return G_VALUE_HOLDS_BOOLEAN (&v) ? g_value_get_boolean (&v) : FALSE;
> > }
> >
> > void
> > xaccAccountSetTaxRelated (Account *acc, gboolean tax_related)
> > {
> > if (tax_related)
> > printf("Request to set TaxRelated TRUE\n"); <=============
> > else
> > printf("Request to set TaxRelated FALSE\n"); <=============
> >
> > GValue v = G_VALUE_INIT;
> > g_return_if_fail(GNC_IS_ACCOUNT(acc));
> >
> > g_value_init (&v, G_TYPE_BOOLEAN);
> > g_value_set_boolean (&v, tax_related);
> >
> > xaccAccountBeginEdit(acc);
> > qof_instance_set_kvp (QOF_INSTANCE (acc), "tax-related", &v);
> > mark_account (acc);
> > xaccAccountCommitEdit(acc);
> >
> > if (xaccAccountGetTaxRelated(acc))
> > printf("TaxRelated is now TRUE\n"); <=============
> > else
> > printf("TaxRelated is now FALSE\n"); <=============
> > }
> >
> > =========== snip end =========================
> >
> > The printout shows
> >
> > Request to set TaxRelated TRUE
> > G_VALUE_HOLDS_BOOLEAN returns FALSE, actually it is gchararray
> > g_value_get_boolean returns FALSE
> > TaxRelated is now FALSE
> >
> > My suspicion:
> >
> > in kvp_frame.cpp in function
> >
> > KvpValue*
> > kvp_value_from_gvalue (const GValue *gval)
> >
> > ============= snip start ===========
> > else if (type == G_TYPE_BOOLEAN)
> > {
> > auto bval = g_value_get_boolean(gval);
> > if (bval)
> > val = new KvpValue(g_strdup("true"));
> > ======= snip end =================
> >
> > the information is lost that the origin is boolean. After this point I
> get lost.
> >
> > What to do? Open a bug report?
> >
>
> Carsten,
>
> Sorry, I meant to pick this up yesterday, but I was focussed on Geert's PR
> and then forgot about it.
>
> There isn't a KVP_TYPE_BOOLEAN in maint or master. This is handled in
> maint by testing for the existence of the slot; if it exists the value is
> true, if it doesn't the value is false. The actual stored value is either
> an int, as in the case of Account's tax-related key, or a string "true".
>
> When I made KVP private I tried to standardize on the string version as
> that seemed to me to be more obvious when looking at a text XML file. It
> looks like I screwed up at lease xaccAccountGetTaxRelated and perhaps
> others. I'll look at fixing that right away.
>
> Regards,
> John Ralls
>
>
> _______________________________________________
> 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
