World: r3wp
[All] except covered in other channels
older newer | first last |
Maxim 15-Jan-2007 [2363] | it seems as if R3 will allow us to link to rebol itself... so this might be possible... as you have noticed... too in some of Carl's R3 posts. |
Pekr 15-Jan-2007 [2364] | I am not sure it is possible. Is Python just some kind of direct C extension or what? |
Maxim 15-Jan-2007 [2365x6] | internally it using the same type structure (immutable strings, etc) |
and in loads C stubs easily. so you don't have to try and make it work... its python's main feature... its so easy, everyone is able to port their legacy C stuff and then use it within a scripted env. | |
and most C libs are within python as-is, so things like sockets, regexp, threads and such use the same mechanics and nomenclature... so there is very little adapting a part from the peculiar indent grouping. | |
obviously, python has other merits, but making C integration that easy made many developpers "take the risk" of trying it out, since it did not means wasting years of investment. | |
there is nothing like proving an easy, lazy path to migration. | |
proving= providing | |
Pekr 15-Jan-2007 [2371] | that sounds good. I tried to look into Ruby and Python, and if I would learn another new language, I would learn python - the syntax is pretty readable ... |
Maarten 15-Jan-2007 [2372] | Yes, and the language is as capable as rebol, except for symbolic manipulation. But with a good parser generator you can do a lot |
Pekr 16-Jan-2007 [2373x2] | guys, don't you know any application, which would be able to generate traffic? I want to install two Mikrotik routers in Dual Nstreme mode, and I would like to test real-life thoughput. Copying one big file amongst two ends is interesting test for line stability, but there is a difference if you run one connection, or few hundreds ones ... |
IIRC there was some perl tool to test Apache, but it required Apache (webserver) on the other end .... I need something like "point-to-point" testing tool ... | |
Volker 16-Jan-2007 [2375] | rebol would notwork for that? |
Pekr 16-Jan-2007 [2376x3] | Volker - did not think about that :-) |
I would have to implement some kind of tasking ... not sure 1 rebol process is able to generat the throughput ... | |
and - how would I measure real packet size? I am not interested in data-only size .... Hmm, but ethernet plus ip plus tcp header have some constant overhead I think :-) | |
Volker 16-Jan-2007 [2379x2] | with async? |
if one of the gurus steps in to write the script^^ | |
Pekr 16-Jan-2007 [2381x2] | I have my own, very short but working - multiserver .... |
I just should generate 1000 opened connections and try to send random data .... | |
Volker 16-Jan-2007 [2383] | without async there are little hidden waits for handshaking AFAIK. |
Pekr 16-Jan-2007 [2384] | yes, probably connection times ... hmm, interesting idea, I might try it anyway .... |
Volker 16-Jan-2007 [2385] | or a script to launch some processes, each using sync io. |
Pekr 16-Jan-2007 [2386] | I can start multiple Rebol instances with the same script. I think that multiplexing to some 100 ports per instance should be ok, no? I can open connections first, then start sending data .... |
Gabriele 16-Jan-2007 [2387x2] | iperf |
http://dast.nlanr.net/Projects/Iperf/ | |
Maarten 16-Jan-2007 [2389] | Petr, iperf is the de facto tool for this. We use it in the Detective as well |
Pekr 16-Jan-2007 [2390x2] | ok, thanks guys, looks good. I am just studying command-line options ... |
I have my rebol multi-server, hopefully it would work, but I would need to think of how to do calculations :-) | |
Luis 24-Jan-2007 [2392] | .\ |
Ladislav 26-Jan-2007 [2393x2] | I posted two REBOL puzzles recently not receiving any response. Just interested to know why, since I don't think they are really hard. |
(are they?) | |
Anton 27-Jan-2007 [2395x2] | I thought the questions were a little unclear. |
but now that I reread I understand. | |
Gregg 15-Feb-2007 [2397] | Just busy here (haven't even been on this world for a while). I think the abstract/guru-level stuff attracts a much smaller crowd than the "write the fastest code" kind of challenge. Maybe it's the competetive nature versus deep thinking and problem solving. |
Maxim 16-Feb-2007 [2398] | or just that must of us dont have the proper knowledge to compete ;-) |
Henrik 23-Feb-2007 [2399] | I made a code snippet checklist. I thought it might be handy, rather than pasting miles of code inside the groups. |
Maxim 23-Feb-2007 [2400] | good idea! |
[unknown: 5] 25-Feb-2007 [2401] | And |
Maxim 27-Feb-2007 [2402] | maybe you should add the author in the title? |
[unknown: 9] 20-Apr-2007 [2403x4] | I finally got Opera to download and run on my PC. I find it very nice, but SLOW. |
I like the thumbnail view of all sites by default on the "new tab." Since I'm new to it, and a few of you around here already know - what are teh top best features about it? | |
If I were to do the same for FF it would: | |
- Fast - Same interface on all the platforms I use - Firebug (killer debugging tool) - Getting better every day | |
Maxim 20-Apr-2007 [2407] | firebug IS awesome. |
[unknown: 9] 21-Apr-2007 [2408] | I like Opera's MAgic Wand (auto fills in Password). I need to feel it in more examples. But it seems cool. |
Alek_K 22-Apr-2007 [2409] | Opera is slow?! From my experience Way faster on older computers (1 GHz, Windows XP) than Firefox. If Win98 - older version is better. |
Sunanda 22-Apr-2007 [2410] | I have intermittent problems with Opera. Sometimes it takes a very long time to display a page. When it works, it is fine; when it is sulking like that, it is a total pain. There is some inconclusive discussion in their support forum, and they never replied to my feedback/bug notification. |
[unknown: 9] 22-Apr-2007 [2411x2] | It seems to be connected to JavaScript heavy sites. |
But ther is no quesiton, FF is faster than Opera on everything I'm testing, including Yahoo. | |
older newer | first last |