but some things in the E2 code fail to compile when C++ exceptions are enabled.
I don't think that would be hard to fix for anyone, I wonder why it hasn't been fixed yet.
Posted 7 August 2014 - 19:05
I have a question regarding some recent httpstream changes, I see a fix for chunked transfer for the code 200, then it was reverted back with a comment saying that it breaks some other streams, could I get some more details, I see no problem doing a chunked transfer if response code is 200 and there is a chunked transfer header.
probably pieterg can answer this ?
Posted 7 August 2014 - 21:41
I found where is the problem
this wont work anyway due to the incorrect string comparison expected value:
if (statuscode == 206 && strncasecmp(linebuf, "transfer-encoding: chunked", strlen("transfer-encoding: chunked")))
it needs to be changed to:
if (statuscode == 206 && strncasecmp(linebuf, "transfer-encoding: chunked", strlen("transfer-encoding: chunked")) == 0)
so to make it work for 200 and 206, we should simply do:
if (strncasecmp(linebuf, "transfer-encoding: chunked", strlen("transfer-encoding: chunked")) == 0) { isChunked = true; }
Edited by zakalibit, 7 August 2014 - 21:45.
Posted 8 August 2014 - 09:39
a bit better impl would be:
if (!isChunked && strncasecmp(linebuf, "transfer-encoding: chunked", strlen("transfer-encoding: chunked")) == 0) { isChunked = true; }to avoid excessive string comparisons
This sounds very reasonable, thanks!
* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.
Posted 8 August 2014 - 09:57
Zakalibit, comitted. I forgot to mention your credit, excuse!
* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.
Posted 8 August 2014 - 09:58
do not know where it shows exactly, but when I saw into its qtdemux.c under PLi, there f.eg. is
demux->upstream_size = seekable ? stop : -1;and then it cannot be boolean, imho
AFAIK Littlesat has committed this. Littlesat can you please acknowledge patches you have committed?
* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.
Posted 8 August 2014 - 10:22
Zakalibit, comitted. I forgot to mention your credit, excuse!
I think there is no need to check for status code, I think the code should be as I posted above, it will work as is now as well, but code checking does not give anything useful there.
And could you please bring all strcmp() to the same form, i.e. !strcmp(), as we have strcmp() == 0 variation as well
Edited by zakalibit, 8 August 2014 - 10:24.
Posted 8 August 2014 - 14:29
I mean if you apply/commit/push something, that you leave a message in the relevant thread that it's done, so others (from the team AND the contributor) can see in a blink of an eye whether it's done or not.
* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.
0 members, 15 guests, 0 anonymous users