Bluetooth PBAP specification expects for call history listing,
that parameter N: shall be empty when we cannot retrieve personal data
from PSE phone book.
Some devices(by example BH-903) are having problems when after N: parameter
unnecessary characters occurs. List of dialed/incoming calls on carkit
then is useless - carkit shows only blank lines and it's impossible to
determine who made call.
In previous version unnecessary semicolons were added after N:("N:;;;;")
to represent empty name.
Now if none of the contact fields is available, then adding real empty "N:"
parameter (without semicolons).
---
plugins/vcard.c | 35 +++++++++++++++++++++++++++++++++++
1 files changed, 35 insertions(+), 0 deletions(-)
diff --git a/plugins/vcard.c b/plugins/vcard.c
index 5948a4a..108f6bd 100644
--- a/plugins/vcard.c
+++ b/plugins/vcard.c
@@ -133,9 +133,44 @@ static void vcard_printf_begin(GString *vcards, uint8_t format)
vcard_printf(vcards, "VERSION:2.1");
}
+/* checks if there is at least one present contact field with personal data */
+static gboolean contact_fields_present(struct phonebook_contact * contact)
+{
+ if (contact->family && strlen(contact->family) > 0)
+ return TRUE;
+
+ if (contact->given && strlen(contact->given) > 0)
+ return TRUE;
+
+ if (contact->additional && strlen(contact->additional) > 0)
+ return TRUE;
+
+ if (contact->prefix && strlen(contact->prefix) > 0)
+ return TRUE;
+
+ if (contact->suffix && strlen(contact->suffix) > 0)
+ return TRUE;
+
+ /* none of the personal data fields is present*/
+ return FALSE;
+}
+
static void vcard_printf_name(GString *vcards,
struct phonebook_contact *contact)
{
+ if (contact_fields_present(contact) == FALSE) {
+ /*
+ * If fields are empty, then add only 'N:' as parameter.
+ * This is crucial for some devices ( Nokia BH-903) which
+ * have problems with history listings - can't determine
+ * if parameter is really empty if there are unnecessary
+ * characters after'N:' (like 'N:;;;;').
+ * We need to add only'N:' param - without semicolons.
+ */
+ vcard_printf(vcards, "N:");
+ return;
+ }
+
vcard_printf(vcards, "N:%s;%s;%s;%s;%s", contact->family,
contact->given, contact->additional,
contact->prefix, contact->suffix);
--
1.6.0.4
Hi Marcel,
On Fri, Jul 09, 2010, Marcel Holtmann wrote:
> > Bluetooth PBAP specification expects for call history listing,
> > that parameter N: shall be empty when we cannot retrieve personal data
> > from PSE phone book.
> > Some devices(by example BH-903) are having problems when after N: parameter
> > unnecessary characters occurs. List of dialed/incoming calls on carkit
> > then is useless - carkit shows only blank lines and it's impossible to
> > determine who made call.
>
> please be a bit consistent with your spaces here.
>
> > In previous version unnecessary semicolons were added after N:("N:;;;;")
> > to represent empty name.
> > Now if none of the contact fields is available, then adding real empty "N:"
> > parameter (without semicolons).
> > ---
> > plugins/vcard.c | 35 +++++++++++++++++++++++++++++++++++
> > 1 files changed, 35 insertions(+), 0 deletions(-)
> >
> > diff --git a/plugins/vcard.c b/plugins/vcard.c
> > index 5948a4a..108f6bd 100644
> > --- a/plugins/vcard.c
> > +++ b/plugins/vcard.c
> > @@ -133,9 +133,44 @@ static void vcard_printf_begin(GString *vcards, uint8_t format)
> > vcard_printf(vcards, "VERSION:2.1");
> > }
> >
> > +/* checks if there is at least one present contact field with personal data */
>
> "check if there is at least one contact field with personal data
> present"
>
> > +static gboolean contact_fields_present(struct phonebook_contact * contact)
> > +{
> > + if (contact->family && strlen(contact->family) > 0)
> > + return TRUE;
> > +
> > + if (contact->given && strlen(contact->given) > 0)
> > + return TRUE;
> > +
> > + if (contact->additional && strlen(contact->additional) > 0)
> > + return TRUE;
> > +
> > + if (contact->prefix && strlen(contact->prefix) > 0)
> > + return TRUE;
> > +
> > + if (contact->suffix && strlen(contact->suffix) > 0)
> > + return TRUE;
> > +
> > + /* none of the personal data fields is present*/
>
> "none of the personal data fields are present"
>
> > + return FALSE;
> > +}
> > +
> > static void vcard_printf_name(GString *vcards,
> > struct phonebook_contact *contact)
> > {
> > + if (contact_fields_present(contact) == FALSE) {
> > + /*
> > + * If fields are empty, then add only 'N:' as parameter.
> > + * This is crucial for some devices ( Nokia BH-903) which
>
> Please be consistent with spaces.
>
> > + * have problems with history listings - can't determine
> > + * if parameter is really empty if there are unnecessary
> > + * characters after'N:' (like 'N:;;;;').
> > + * We need to add only'N:' param - without semicolons.
>
> Same here.
I went ahead and fixed these issues myself so we can get another obexd
release out. The fixed patch is now upstream.
Johan
Hi Radek,
> Bluetooth PBAP specification expects for call history listing,
> that parameter N: shall be empty when we cannot retrieve personal data
> from PSE phone book.
> Some devices(by example BH-903) are having problems when after N: parameter
> unnecessary characters occurs. List of dialed/incoming calls on carkit
> then is useless - carkit shows only blank lines and it's impossible to
> determine who made call.
please be a bit consistent with your spaces here.
> In previous version unnecessary semicolons were added after N:("N:;;;;")
> to represent empty name.
> Now if none of the contact fields is available, then adding real empty "N:"
> parameter (without semicolons).
> ---
> plugins/vcard.c | 35 +++++++++++++++++++++++++++++++++++
> 1 files changed, 35 insertions(+), 0 deletions(-)
>
> diff --git a/plugins/vcard.c b/plugins/vcard.c
> index 5948a4a..108f6bd 100644
> --- a/plugins/vcard.c
> +++ b/plugins/vcard.c
> @@ -133,9 +133,44 @@ static void vcard_printf_begin(GString *vcards, uint8_t format)
> vcard_printf(vcards, "VERSION:2.1");
> }
>
> +/* checks if there is at least one present contact field with personal data */
"check if there is at least one contact field with personal data
present"
> +static gboolean contact_fields_present(struct phonebook_contact * contact)
> +{
> + if (contact->family && strlen(contact->family) > 0)
> + return TRUE;
> +
> + if (contact->given && strlen(contact->given) > 0)
> + return TRUE;
> +
> + if (contact->additional && strlen(contact->additional) > 0)
> + return TRUE;
> +
> + if (contact->prefix && strlen(contact->prefix) > 0)
> + return TRUE;
> +
> + if (contact->suffix && strlen(contact->suffix) > 0)
> + return TRUE;
> +
> + /* none of the personal data fields is present*/
"none of the personal data fields are present"
> + return FALSE;
> +}
> +
> static void vcard_printf_name(GString *vcards,
> struct phonebook_contact *contact)
> {
> + if (contact_fields_present(contact) == FALSE) {
> + /*
> + * If fields are empty, then add only 'N:' as parameter.
> + * This is crucial for some devices ( Nokia BH-903) which
Please be consistent with spaces.
> + * have problems with history listings - can't determine
> + * if parameter is really empty if there are unnecessary
> + * characters after'N:' (like 'N:;;;;').
> + * We need to add only'N:' param - without semicolons.
Same here.
> + */
> + vcard_printf(vcards, "N:");
> + return;
> + }
> +
> vcard_printf(vcards, "N:%s;%s;%s;%s;%s", contact->family,
> contact->given, contact->additional,
> contact->prefix, contact->suffix);
Regards
Marcel
Hi Radek,
> Bluetooth PBAP specification expects for call history listing,
> that parameter N: shall be empty when we cannot retrieve personal data
> from PSE phone book.
> Some devices(by example BH-903) are having problems when after N: parameter
> unnecessary characters occurs. List of dialed/incoming calls on carkit
> then is useless - carkit shows only blank lines and it's impossible to
> determine who made call.
>
> In previous version unnecessary semicolons were added after N:("N:;;;;")
> to represent empty name.
> Now if none of the contact fields is available, then adding real empty "N:"
> parameter (without semicolons).
> ---
> plugins/vcard.c | 27 +++++++++++++++++++++++++++
> 1 files changed, 27 insertions(+), 0 deletions(-)
>
> diff --git a/plugins/vcard.c b/plugins/vcard.c
> index 5948a4a..8c7147b 100644
> --- a/plugins/vcard.c
> +++ b/plugins/vcard.c
> @@ -123,6 +123,27 @@ static void add_slash(char *dest, const char *src, int len_max, int len)
> return;
> }
>
> +/* checks if there is at least one present contact field with personal data */
> +static gboolean contact_fields_present(struct phonebook_contact * contact)
> +{
> + if (contact->family && strlen(contact->family))
> + return TRUE;
I asked to use strlen(...) > 0 as test here. Please fix this.
> +
> + if (contact->given && strlen(contact->given))
> + return TRUE;
> + if (contact->additional && strlen(contact->additional))
> + return TRUE;
Why are these two tight together. Use an extra empty line.
> + if (contact->prefix && strlen(contact->prefix))
> + return TRUE;
> +
> + if (contact->suffix && strlen(contact->suffix))
> + return TRUE;
> +
> + /* none of the personal data fields is present*/
> + return FALSE;
> +}
> +
> static void vcard_printf_begin(GString *vcards, uint8_t format)
> {
> vcard_printf(vcards, "BEGIN:VCARD");
> @@ -136,6 +157,12 @@ static void vcard_printf_begin(GString *vcards, uint8_t format)
> static void vcard_printf_name(GString *vcards,
> struct phonebook_contact *contact)
> {
> + if (contact_fields_present(contact) == FALSE) {
> + /* if all fields are empty we need to have empty N: parameter */
> + vcard_printf(vcards, "N:");
I would be good to add a bit more extra text here why we are doing this.
You should mention the broken headset and what is expect outcome.
Also please sort the contact_fields_resent() function about
vcard_printf_name to keep them close together.
Regards
Marcel
Hi Radek,
> I've created patch problems with history listing with BH-903..
> After fix everything works as described in PBAP spec and problem no longer occurs.
>
> Patch is available on my gitorius master branch clone for obexd-packaging:
> git://gitorious.org/~jradosla/maemo-bluetooth/jradoslas-obexd-packaging.git
> It is located in: debian/patches/006-Fixed-empty-N-parameter-handling-for-VCARDs.patch
yeah, that is not how this works. Take the patch and send it to the
mailing list for review. I am not going to extract a patch from your
packaging repos nor is anybody else.
Regards
Marcel