Attention is currently required from: daniel, laforge, pespin.

neels has posted comments on this change by pespin. ( 
https://gerrit.osmocom.org/c/osmo-mgw/+/39224?usp=email )

Change subject: mgw: CRCX: Split mgcp header pars parsing into a previous step
......................................................................


Patch Set 5:

(5 comments)

File include/osmocom/mgcp/mgcp_protocol.h:

https://gerrit.osmocom.org/c/osmo-mgw/+/39224/comment/219b8149_7ba9106c?usp=email
 :
PS4, Line 7: #define MGCP_PARSE_HDR_PARS_OSMUX_CID_WILDCARD (-1)
> Yeah in general, but in this case it's not a closed-set of options, but 2 
> specific options within in […]
i disagree with that position. i feel that lists of separate #defines for any 
number of related ints, nevermind the used range, is a hack from times where 
our early ancestors didn't know about curly braces yet and had barely 
discovered fire ... enums are an actual part of the C language proper that can 
be used as return type or argument type to clarify things, #defines are a 
preprocessor shim, much like a flea circus...

i will not block this, but i challenge you to admit that mentioning the int 
range is just a lame excuse.


File src/libosmo-mgcp/mgcp_msg.c:

https://gerrit.osmocom.org/c/osmo-mgw/+/39224/comment/df8aab49_10bc8aff?usp=email
 :
PS4, Line 210:                  goto mgcp_header_done;
> I prefer handling it below as it is now, and as it used to be too. […]
It is a label with a single caller, that simply does "return 0": there is no 
logic to preserve.
It is the textbook example of meaningless bloat.

it is a non-functional cosmetic issue, so i am dialing back a bit, but i find 
this beyond weird.


https://gerrit.osmocom.org/c/osmo-mgw/+/39224/comment/e66fc108_5abbf072?usp=email
 :
PS4, Line 218:  return 0;
> To me this makes it easier to understand that you can finish properly based 
> on whether you arrived t […]
empty words.

(resolving this one so we don't say the same things twice)


File src/libosmo-mgcp/mgcp_protocol.c:

https://gerrit.osmocom.org/c/osmo-mgw/+/39224/comment/57f448d3_6904a35a?usp=email
 :
PS4, Line 873:          hpars->remote_osmux_cid = 
MGCP_PARSE_HDR_PARS_OSMUX_CID_UNSET;
> Yeah well a lot of stuff is quite fucked up, but we cannot fix it all in one 
> commit. […]
This patch does exactly the right thing: it introduces proper layering 
boundaries. So I would like the new code to observe the boundaries that it 
introduces. This is not about a fix of earlier code.


https://gerrit.osmocom.org/c/osmo-mgw/+/39224/comment/90590321_9523a011?usp=email
 :
PS4, Line 899:  endp->x_osmo_ign |= hpars->x_osmo_ign;
> Yes |= because it's in  the endp, it's shared by several conns. […]
ok!



--
To view, visit https://gerrit.osmocom.org/c/osmo-mgw/+/39224?usp=email
To unsubscribe, or for help writing mail filters, visit 
https://gerrit.osmocom.org/settings?usp=email

Gerrit-MessageType: comment
Gerrit-Project: osmo-mgw
Gerrit-Branch: master
Gerrit-Change-Id: I3ee5158c254213203830fe9c38de11c15b4b19c1
Gerrit-Change-Number: 39224
Gerrit-PatchSet: 5
Gerrit-Owner: pespin <[email protected]>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: daniel <[email protected]>
Gerrit-Reviewer: laforge <[email protected]>
Gerrit-Reviewer: neels <[email protected]>
Gerrit-Attention: laforge <[email protected]>
Gerrit-Attention: pespin <[email protected]>
Gerrit-Attention: daniel <[email protected]>
Gerrit-Comment-Date: Thu, 16 Jan 2025 11:55:23 +0000
Gerrit-HasComments: Yes
Gerrit-Has-Labels: No
Comment-In-Reply-To: neels <[email protected]>
Comment-In-Reply-To: pespin <[email protected]>

Reply via email to