The message format corrupted?

The message format corrupted? SearchSearch
Author Message
Baumann
Unregistered guest
Posted on Monday, January 12, 2004 - 07:54 am:   

i read the mms conformance , wut's difference between "CONTENT ID" & "Content LOCATION"?
it semms that it's always CONTENT ID used when start parameter presented. when to use the Content Location? should the Content ID include '<' & '>' as the conformance specified?

the attached file below is wut i sent out, but the mmsc returns me

data: 12 80 01 04 20 71 61 70 .... qap
data: 70 6c 69 63 61 74 69 6f plicatio
data: 6e 2f 76 6e 64 2e 77 61 n/vnd.wa
data: 70 2e 6d 6d 73 2d 6d 65 p.mms-me
data: 73 73 61 67 65 00 a6 52 ssage..R
data: 65 73 69 6e 2f 32 2e 31 esin/2.1
data: 2e 39 00 92 04 40 02 46 .9...@.F
data: e8 85 80 ab 48 54 54 50 ....HTTP
data: 2f 31 2e 30 20 47 44 47 /1.0 GDG
data: 5a 2d 50 2d 47 57 30 38 Z-P-GW08
data: 2d 57 41 50 20 28 54 72 -WAP (Tr
data: 61 66 66 69 63 2d 53 65 affic-Se
data: 72 76 65 72 2f 35 2e 31 rver/5.1
data: 2e 31 2d 52 20 5b 63 20 .1-R [c
data: 73 53 66 20 5d 29 00 8c sSf ])..
data: 81 98 74 72 61 6e 73 69 ..transi
data: 64 31 30 37 33 39 31 39 d1073919
data: 36 36 30 00 8d 90 92 83 660.....
data: 93 45 72 72 6f 72 20 4d .Error M
data: 65 73 73 61 67 65 20 46 essage F
data: 6f 72 6d 61 74 20 43 6f ormat Co
data: 72 72 75 70 74 00 rrupt.

i wonder i am stumbled.
text/htmlmymms
mymms.txt (17.2 k)
Baumann
Unregistered guest
Posted on Monday, January 12, 2004 - 08:08 am:   

the subject is chinese
( i ignored the charset question, just using the Text-String encode rule.
in this example, becoz the first is 0x2a(* symbol), the 0x7f is useless)

and mms text is also chinese.

i once encode the mms with Content Location(without '<', '>') it's okay,
in this example i sent with both Content Location & Content ID, but failed.
any1 can tell me where i stumbled? thanks lots in advance.
Marty
New member
Username: Marty

Post Number: 34
Registered: 10-2003
Posted on Monday, January 12, 2004 - 09:09 am:   

Try to use the mmscomp.exe included in the demo NowMMS application and compare the output with what you've created.
Baumann
Unregistered guest
Posted on Monday, January 12, 2004 - 10:55 am:   

there having the differences is quite normal.
becoz mms header have some optional & the smil would be different
Bryce Norwood - NowSMS Support
Board Administrator
Username: Bryce

Post Number: 1584
Registered: 10-2002
Posted on Monday, January 12, 2004 - 09:03 pm:   


quote:

i read the mms conformance , wut's difference between "CONTENT ID" & "Content LOCATION"?
it semms that it's always CONTENT ID used when start parameter presented. when to use the Content Location? should the Content ID include '<' & '>' as the conformance specified?




Well, only a "Content-ID" field can be referenced by a "start=" parameter. So you should at least have a "Content-ID" field associated with your SMIL file.

Elsewhere, "Content-ID" and "Content-location" really only matter as a way of referencing the objects from a SMIL file. If you reference by "Content-id", then your "href" must start with "cid:".

To keep things simple, I usually just recommend that you encode both a "Content-ID" and "Content-location" and assign the same value to each.

As for why the MMSC is rejecting your message, well, I'm not sure if I am able to download the binary correctly from your attachment here.

But a couple of things I see ...

If the Chinese characters in your subject header are not encoded as UTF-8, then you should include a character set encoding within the subject header. (Maybe just try for testing, a subject that has simple English characters, so that you can determine first if this is the field that is causing problems.)

Also, the Content-type header does not look right. Check the WSP specification and look at the definition of Value-length encoding (which is used in the content-type header). If the length is > 30 (in this case it is 31), then you must use a length-quote.
Baumann
Unregistered guest
Posted on Tuesday, January 13, 2004 - 01:30 am:   

thanks u , Mr. Bryce
you said
If you reference by "Content-id", then your "href" must start with "cid:".

but i retrieved some mms,
it uses the content-id , content uri, and parameter name for the multipart entry's cotenttype.
in the received mms, there is no cid existed after the string src= . even the mms conformance example shows the cid, i dont know its right use,
maybe it's said in some rfc ?
Baumann
Unregistered guest
Posted on Tuesday, January 13, 2004 - 01:58 am:   

i encode subject field with the 0x7f prefix, so it will be right Text-String.so the mmsc should not report mms format error.
Bryce Norwood - NowSMS Support
Board Administrator
Username: Bryce

Post Number: 1604
Registered: 10-2002
Posted on Tuesday, January 13, 2004 - 06:26 pm:   

I'm not sure where to point you as a reference. I'm sure there is a spec (probably the SMiL spec) that talks about when to use "cid:" and when not to.

But from my experience, in a SMIL file, if you have a "src=" reference to a particular content-id named "xxxx", then you need to specify "src=cid:xxxx" in the reference.

If you reference by "Content-location:", then you do not include the "cid:", it is just "src=file".

The subject line may or may not be a problem. I'm just saying that if it is Chinese characters, you might need to specify the character set using Encoded-string-value instead of Text-string.

It is easy enough for you to change the subject line to determine if the problem you are experiencing is related to that.

But I still think the problem is your content-type header, as I mentioned previously.
Baumann
Unregistered guest
Posted on Wednesday, January 14, 2004 - 01:59 am:   

yes, the big problem is my content type header. thank u again