I changed it. It does look tidier.

 

== Rene

 

 

From: [email protected] [mailto:[email protected]] On Behalf Of 
Konstantin Vayner
Sent: vrijdag 18 juni 2010 9:04
To: Rene Kluwen
Cc: Kannel Devel
Subject: Re: [PATCH] SMPP box multipart messages

 

Rene,

 

Looks good. +1

 

Though i dont like the "for (;;) {" approach in catenate_msg very much... it 
looks stressing before going through the code in detail ;)

I'd rather have a do { .. } while(++max > total) -> and then instead of 
unreachable return do a failure code...

But then again it seems like a common practice in gwlib and throughout kannel 
code so it shouldnt be an issue

 

$ grep -rn 'for[ ]*(;;)' gw*

gw/urltrans.c:346:    for (;;) {

gw/smsc/smsc.c:361:    for (;;) {

gw/smsc/smsc_cimd.c:125:    for (;;) {

gw/smsc/smsc_cimd.c:297:    for (;;) {

gw/smsc/smsc_cimd.c:580:    for (;;) {

gw/smsc/smsc_cimd.c:597:    for (;;) {

gw/smsc/smsc_cimd.c:688:    for (;;) {

gw/smsc/smsc_oisd.c:1248:    for (;;) {

gw/smsc/smsc_cimd2.c:2053:    for (;;) {

gw/smsc/smsc_emi.c:1289:    for (;;) {

gw/smsc/smsc_smasi.c:945:        for (;;) {

gw/smsc/smsc_sema.c:724:    for (;;) {

gw/smsbox.c:1121:    for (;;) {

gw/smsbox.c:3179:    for (;;) {

gwlib/http.c:138:    for (;;) {

gwlib/http.c:2126:    for (;;) {

gwlib/mime.c:138:    for (;;) {

gwlib/fdset.c:328:    for (;;) {

gwlib/utils.c:244:    for (;;) {

gwlib/utils.c:403:    for (;;) {

 

Regards,

  Konstantin

 

On Thu, Jun 17, 2010 at 9:25 PM, Rene Kluwen <[email protected]> wrote:

Okay, to comply with Kannel's common practices, hereby I send in a patch for
open smppbox that allows for all parts of multipart messages to be sent via
the same smsc (letting bearerbox split the message up).

The patch applies to current svn trunk of smppbox.

Known issue: This patch submits a deliver_sm message for all parts of the
original message. Kannel complains about it, because it can only find the
first part in the dlr table.
Since this seems to be a Kannel issue and not smppbox, I am letting it this
way. This doesn't affect any practical, operational issues.

Special thanks to Tomasz Konopka for letting me test in his production
environment.

Can I have any votes on this?

== Rene

 

Reply via email to