(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
[11:06 AM PT] Andrew Hellershanks: Hello, everyone. Had a late lunch today so I didn't get here on time.
[11:06 AM PT] Ubit Umarov: well the code to do pbr is a lot heavier
[11:06 AM PT] Joe Magarac: Nothing has happened yet.
[11:06 AM PT] Ubit Umarov: wonder how good is that code..
[11:06 AM PT] Ubit Umarov: seem games with a lot more content using less gpu
[11:07 AM PT] Ubit Umarov: see..
[11:07 AM PT] Andrew Hellershanks: FS 7? There is a new major release coming soon?
[11:07 AM PT] Joe Magarac: FS 7 is out.
[11:07 AM PT] Ubit Umarov: fs it current release
[11:07 AM PT] Ubit Umarov: fs 7
[11:07 AM PT] Ubit Umarov: 7.1.9
[11:07 AM PT] Joe Magarac: Full PBR, rendering mostly in the GPU, users with low-end PCs are screaming.
[11:08 AM PT] Vincent.Sylvester @hg.zetaworlds.com: The slow performance is one thing, the crashes another
[11:08 AM PT] Joe Magarac: Works OK on an NVidia 640 from 2012, and really well on an NVidia 3070.
[11:08 AM PT] Ubit Umarov: terrain rendering is also done on gpu now
[11:08 AM PT] Andrew Hellershanks: Hm... In that case I will have to be sure to install it along side my 6.6 copy of FS.
[11:08 AM PT] Ubit Umarov: at least pbr terrain ??
[11:08 AM PT] Joe Magarac: Zero FS 7 crashes on Linux. Can't speak for Windows.
[11:09 AM PT] Ubit Umarov: release does crash a lot on opensim
[11:09 AM PT] Ubit Umarov: a bit less at sl
[11:09 AM PT] Andrew Hellershanks: Vincent, I only have a GTX 1050. I hope that is still good enough.
[11:09 AM PT] Ubit Umarov: they did release a beta that seems to crash less
[11:10 AM PT] Ubit Umarov: possible they will wait for webrtc before next release
[11:10 AM PT] Joe Magarac: It's mostly space. 2GB GPU will work on low settings. 4GB is probably OK. 8 GB is good.
[11:10 AM PT] Ubit Umarov: think ll wanted it bf august ?
[11:10 AM PT] Joe Magarac: It's "Integrated graphics" machines that suffer, I think.
[11:10 AM PT] Ubit Umarov: bay nothing is good :p
[11:11 AM PT] Joe Magarac: But I don't have one, so not sure.
[11:11 AM PT] Ubit Umarov: i will not disable my main gpu to test the cpu one
[11:11 AM PT] Ubit Umarov: your cpu does not have one?
[11:12 AM PT] Lyr Lobo: Joe, I have the RTX3080 at 10GB VRAM and it was not stable for me
[11:12 AM PT] Lyr Lobo: kept crashing
[11:12 AM PT] Lyr Lobo: on a clean install
[11:12 AM PT] Ubit Umarov: the cpu price of the gpu is like 20 usd
[11:12 AM PT] Andrew Hellershanks: The situation reminds me of XPlane 11 that also has a requirement for a high end video card (and CPU?)
[11:12 AM PT] Ubit Umarov: i saw no point on buying the cpu without it
[11:12 AM PT] Joe Magarac: Huh.
[11:13 AM PT] Ubit Umarov: well yeap fs 7 also uses more cpu
[11:13 AM PT] Joe Magarac: Yes. I've seen it keep 4 CPUs at 66%.
[11:13 AM PT] Andrew Hellershanks: fs6 uses enough CPU as it is.
[11:13 AM PT] Ubit Umarov: seems ll tried do use more threads...
[11:14 AM PT] Joe Magarac: Yes.
[11:14 AM PT] Ubit Umarov: well about opensim code changes during last week..
[11:14 AM PT] Ubit Umarov: hmmm none.. :)
[11:15 AM PT] Ubit Umarov: why change perfect bugs?
[11:15 AM PT] Ubit Umarov: :)
[11:15 AM PT] Andrew Hellershanks: :)
[11:16 AM PT] Gavin.Hird @grid.xmir.org:8002: bugs perfected over years
[11:16 AM PT] Vincent.Sylvester @hg.zetaworlds.com: You could almost call them features
[11:16 AM PT] Gavin.Hird @grid.xmir.org:8002: exactly - a win for all!
[11:16 AM PT] Ubit Umarov: yeap they are features
[11:17 AM PT] Vincent.Sylvester @hg.zetaworlds.com: That reminds me. So I been wondering about how the script syntax gets parsed and if there is a way to add branches to that
[11:18 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Specifically I been thinking about adding a strict spec; type keyword that places at the top of a script to change the way some functions work
[11:18 AM PT] Ubit Umarov: bad karma
[11:18 AM PT] Ubit Umarov: scripts should have a clean spec
[11:18 AM PT] Vincent.Sylvester @hg.zetaworlds.com: A few of the time functions and one or two others are not strictly to lsl spec and instead of changing the functions breaking all manners of scripts this would be simpler for those that need different functionality
[11:18 AM PT] Ubit Umarov: we already have a big mess
[11:19 AM PT] Ubit Umarov: and i did remove several things xmr had
[11:19 AM PT] Vincent.Sylvester @hg.zetaworlds.com: yoptions already has something for changing eval directions so I thought it would relatively simple, but I can't make heads or tails of it yet
[11:19 AM PT] Ubit Umarov: ( xmr is antecessor of yengine )
[11:20 AM PT] Ubit Umarov: yeap . but there where more
[11:20 AM PT] Ubit Umarov: and that is a bad idea
[11:20 AM PT] Ubit Umarov: i kept the those few cases
[11:20 AM PT] Ubit Umarov: switch.. etc
[11:21 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Why you think it'd be bad? It's same as if there was a ossl version of a function working to spec
[11:21 AM PT] Ubit Umarov: the spec should be simple and clean
[11:21 AM PT] Andrew Hellershanks: Is the differences in the implementation of some time related functions such that it would really would impact a large number of scripts?
[11:21 AM PT] Ubit Umarov: as i said we already have bad things
[11:22 AM PT] Ubit Umarov: will all the ossl etc
[11:22 AM PT] Ubit Umarov: ( http://opensimulator.org/wiki/YEngine for those that don't remember Y things )
[11:22 AM PT] Vincent.Sylvester @hg.zetaworlds.com: One or more of the timestamp functions report a different time as spec would suggest, but it's a fairly common function so changing to spec would break a lot of scripts
[11:23 AM PT] Vincent.Sylvester @hg.zetaworlds.com: It's not a complex change, but changing it now would mean trouble
[11:23 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Adding an ossl for it though, to me, seems counter to what ossl usually is
[11:23 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Hence the idea of introducing strict keyword
[11:24 AM PT] Ubit Umarov: well i only rmemeber the wall clock thing
[11:24 AM PT] Ubit Umarov: that os decided it should be the machine local time ???
[11:24 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Something like that
[11:24 AM PT] Ubit Umarov: bc most do not care about pacific time at all
[11:25 AM PT] Ubit Umarov: well many
[11:25 AM PT] Vincent.Sylvester @hg.zetaworlds.com: LL has some weird ideas on their spec which I don't agree with either, but I think for those that need it for some reason such a switch would be useful
[11:25 AM PT] Ubit Umarov: why should a grid in siberia use pst/pdt ??
[11:26 AM PT] Gavin.Hird @grid.xmir.org:8002: a common time reference is a good thing
[11:26 AM PT] Vincent.Sylvester @hg.zetaworlds.com: It used to be that some viewers had the timezone hardcoded so I think everyone stayed on it
[11:26 AM PT] Ubit Umarov: irritating how tpvs removed the utc option
[11:26 AM PT] Ubit Umarov: not if your grid is only for siberians..
[11:26 AM PT] Ubit Umarov: :)
[11:26 AM PT] Gavin.Hird @grid.xmir.org:8002: not sure PST should be the one, but...
[11:26 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Events search to this day reports the event times in local time zones which is not grid time, so that's lovely confusing
[11:27 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Not sure why there is no switch to force that to pst
[11:27 AM PT] Ubit Umarov: tpvs used to have at least the option to display UTC
[11:27 AM PT] Ubit Umarov: they removed it i thing
[11:27 AM PT] Gavin.Hird @grid.xmir.org:8002: LL requirement
[11:27 AM PT] Vincent.Sylvester @hg.zetaworlds.com: It's not just about time though in terms of spec differences
[11:27 AM PT] Ubit Umarov: who cares about ll on opensim..
[11:27 AM PT] Vincent.Sylvester @hg.zetaworlds.com: There are also things like floats not working quite the same due to how the code internally handles them
[11:28 AM PT] Vincent.Sylvester @hg.zetaworlds.com: We use more precision which rounds differently
[11:28 AM PT] Gavin.Hird @grid.xmir.org:8002: realizing the TPV code is 97% LL...
[11:28 AM PT] Ubit Umarov: shh vincent.Sylvester it is not that :P
[11:28 AM PT] Ubit Umarov: well there are little issues on lsl bc opensim had the bad decison of using doubles for them
[11:29 AM PT] Ubit Umarov: no idea the crap they where thinking.. well whatever
[11:29 AM PT] Ubit Umarov: i C# world there was the idea that doubles where even faster..
[11:29 AM PT] Ubit Umarov: in..
[11:30 AM PT] Ubit Umarov: in fact fpus did doubles as fast as floats
[11:30 AM PT] Ubit Umarov: and still do, but there are now somethings they do faster if float
[11:31 AM PT] Ubit Umarov: ofc no one did thing about memory cost
[11:31 AM PT] Ubit Umarov: and that memory access takes a LOT of time
[11:32 AM PT] Ubit Umarov: but well they made the decision and naow i can't change back ( yes i considered it a few times ;) )
[11:32 AM PT] Vincent.Sylvester @hg.zetaworlds.com: It would be nice to have for porting scripts or just to test different behavior. Who knows what nonsense LL come up with looking at the wiki entries for new LSL stuff
[11:32 AM PT] Ubit Umarov: well terrain height where stored in doubles
[11:33 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Another option would be something like a config var to change which function should be used, but that's not exactly user friendly
[11:33 AM PT] Ubit Umarov: something that at SL as like 10 valid bits or so
[11:33 AM PT] Ubit Umarov: bahh
[11:33 AM PT] Ubit Umarov: those i did change to floats
[11:33 AM PT] Ubit Umarov: no that would make scripts even more incompatible across regions
[11:33 AM PT] Ubit Umarov: NO :)
[11:34 AM PT] Jamie.Jordan @grid.kitely.com:8002: Hi everybody
[11:34 AM PT] Ubit Umarov: hi
[11:34 AM PT] Gavin.Hird @grid.xmir.org:8002: hi
[11:34 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Exactly so least problematic would be an in-script switch for it
[11:34 AM PT] Ubit Umarov: ossl permissions are a bad pain uff
[11:37 AM PT] Ubit Umarov: xmr/yengine was also overriding several funtions defined at LSL.cs
[11:37 AM PT] Ubit Umarov: sin() for example
[11:37 AM PT] Ubit Umarov: i removed that so it uses the normal LSLAPI
[11:38 AM PT] Ubit Umarov: in fact that was needed in a few cases for .net 8
[11:39 AM PT] Andrew Hellershanks: lslap :)
[11:39 AM PT] Ubit Umarov: bc the fact that IL has a native math function, does not mean it is the full thing
[11:40 AM PT] Ubit Umarov: ( .net also does ignore IL and replaces full funtions by their full name itno direct native code
[11:41 AM PT] Ubit Umarov: like MathF.Sin()
[11:41 AM PT] Ubit Umarov: well details :)
[11:42 AM PT] Ubit Umarov: btw ms also learned that floats matter
[11:42 AM PT] Ubit Umarov: and made the "new" mathF to support them
[11:42 AM PT] Ubit Umarov: Math.x only does ( mostly ) doubles
[11:43 AM PT] Ubit Umarov: so yeah back in 2007 .net ppl also did thing ppl should only use doubles...
[11:44 AM PT] Ubit Umarov: that does raise the old question of why ppl did chose crap c# for opensim
[11:44 AM PT] Motoko.Karu @login.digiworldz.com:8002: always wondered about that.. =)
[11:45 AM PT] Vincent.Sylvester @hg.zetaworlds.com: I don't mind the greater precision, it does make sense for var regions to have more space in variables when it comes to positioning and such and there is a way to cast in code to get the lsl spec rounding back quite easily
[11:45 AM PT] Vincent.Sylvester @hg.zetaworlds.com: I think there is a mantis about this already
[11:45 AM PT] Ubit Umarov: no vicent.. it is waste
[11:45 AM PT] Ubit Umarov: physices engines use float
[11:46 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Of course it is technically wasteful. No one needs sub-millimeter precision for stuff.... yet
[11:46 AM PT] Ubit Umarov: use of doubles every where like opensim did elsewhere was pure waste
[11:47 AM PT] Ubit Umarov: alos viewers interface uses floats
[11:48 AM PT] Vincent.Sylvester @hg.zetaworlds.com: In the grand scheme of things it's not so bad. Wasteful, but not catastrophically so and it's better to have a bit more than not enough in some cases
[11:48 AM PT] Ubit Umarov: ( some use of doubles is needed here and there where the math would otherwise blow up .. but special localized cases )
[11:48 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Silly as it is regarding physics, but that's a different beast altogether anyways
[11:50 AM PT] Ubit Umarov: well it is all abotu postions and rotations :)
[11:51 AM PT] Ubit Umarov: scene and main opensim does store those in floats
[11:52 AM PT] Ubit Umarov: what makes it harder to understand why doubles on scripts
[11:52 AM PT] Ubit Umarov: but well... things from 2007.. ghosts hunting us now :)
[11:52 AM PT] Vincent.Sylvester @hg.zetaworlds.com: More space to divide by zero xD
[11:54 AM PT] Ubit Umarov: well mb one day i change all that to 32bit float :)
[11:54 AM PT] Vincent.Sylvester @hg.zetaworlds.com: In other news. Appears that ossearch does mostly work under php8.3 minus some small things it blows up on which are just bugs in the code
[11:54 AM PT] Vincent.Sylvester @hg.zetaworlds.com: That was my weekend banging my head against a wall trying to work out what was faulty
[11:55 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Errors on unrelated line numbers are always fun
[11:55 AM PT] Ubit Umarov: man walls are expensive to repair..
[11:55 AM PT] Andrew Hellershanks: Vincent, which version osssearch?
[11:55 AM PT] Vincent.Sylvester @hg.zetaworlds.com: The webroot part, module code is unchanged
[11:56 AM PT] Vincent.Sylvester @hg.zetaworlds.com: I rewrote search in a framework remember, have been backporting fixes to the plain php one
[11:56 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Not sure I have caught all the bugs yet
[11:56 AM PT] Gavin.Hird @grid.xmir.org:8002: not to mention the work needed to remove old wallpaper...
[11:56 AM PT] Vincent.Sylvester @hg.zetaworlds.com: The other side of datasnapshot is also... well old
[11:57 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Need to look at that as well some day
[11:57 AM PT] Vincent.Sylvester @hg.zetaworlds.com: There was some defined and unused vars that caused the pdo prepare to fail
[11:57 AM PT] Vincent.Sylvester @hg.zetaworlds.com: Then there are some other vars used wrong in one of the search queries
[11:57 AM PT] Vincent.Sylvester @hg.zetaworlds.com: A missing category for places I think it was
[11:57 AM PT] Vincent.Sylvester @hg.zetaworlds.com: And a typo if memory serves right
[11:58 AM PT] Andrew Hellershanks: I'll have to take another look at the module. It has been a while since I last checkedit.
[11:59 AM PT] Vincent.Sylvester @hg.zetaworlds.com: It's still not clear what will become of search. I keep hearing that it is legacy, but it has yet to be removed
[12:00 PM PT] Andrew Hellershanks: Web search is also not currently supported.
[12:01 PM PT] Vincent.Sylvester @hg.zetaworlds.com: The url for it can be set in robust and the viewer will use it. If you write a page for it it can display it
[12:01 PM PT] Vincent.Sylvester @hg.zetaworlds.com: I have both working on my sandbox now
[12:02 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Writing a search engine is a pain in the rear though
[12:02 PM PT] Andrew Hellershanks: yes, that is true but I don't know of anyone who has written a page for it. I did start on one.
[12:02 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Can't use prebuilt stuff either as the data in search tables doesn't translate all that well
[12:02 PM PT] Joe Magarac: Gotta go, next meeting.
[12:04 PM PT] Andrew Hellershanks: ok, Joe. Thanks for dropping by. See you next week.
[12:04 PM PT] Ubit Umarov: 2 late :)
[12:04 PM PT] Vincent.Sylvester @hg.zetaworlds.com: I have to confirm that the rewrite is bug free and then I should be able to backport it to plain php again
[12:04 PM PT] Lyr Lobo: *smiles*
[12:04 PM PT] Andrew Hellershanks: :) yea. He doesn't stick around long enough.
[12:04 PM PT] orbert tatham: Ok, got here late, but gotta go anyway
[12:04 PM PT] orbert tatham: See you all later
[12:04 PM PT] Ubit Umarov: cya orbert
[12:04 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Or you can go through the framework version and seek out what changes were made
[12:04 PM PT] Andrew Hellershanks: Bye, orbert.
[12:06 PM PT] Gavin.Hird @grid.xmir.org:8002: in postgres you probably could build materialized views that had data suitable for search
[12:06 PM PT] Gavin.Hird @grid.xmir.org:8002: don't thin mysql supports that
[12:06 PM PT] Gavin.Hird @grid.xmir.org:8002: think*
[12:06 PM PT] Vincent.Sylvester @hg.zetaworlds.com: A compat layer would work too I suppose
[12:07 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Also not a big fan of like elastic or whatever bloated stuff for the queries you actually need to run
[12:07 PM PT] Gavin.Hird @grid.xmir.org:8002: materialized views are faster than accessing the tables directly
[12:07 PM PT] Gavin.Hird @grid.xmir.org:8002: but the content must be refreshed periodically, but that is not time critical for search
[12:08 PM PT] Andrew Hellershanks: It is tricky to do a search across multiple tables and combine the search results in to on stream.
[12:08 PM PT] Vincent.Sylvester @hg.zetaworlds.com: inner joins everywhere
[12:09 PM PT] Vincent.Sylvester @hg.zetaworlds.com: It's made more complicated by the fact the tables are using different charsets so the inner joins are awfully slow unless you make them all uniform
[12:09 PM PT] Vincent.Sylvester @hg.zetaworlds.com: When I first combined things like griduser table and profiles, people search took 20 seconds
[12:10 PM PT] Andrew Hellershanks: That's going to get rather ugly. Various queries of the groups database are bad enough.
[12:10 PM PT] Andrew Hellershanks: Vincent, ouch.
[12:10 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Oh reminds me. datasnapshot provides a region picture, but register never writes it to any table, the uuid for it that is
[12:11 PM PT] Vincent.Sylvester @hg.zetaworlds.com: It checks whether the image is there and sets has_pic but then just discards the uuid for the image itself
[12:11 PM PT] Gavin.Hird @grid.xmir.org:8002: yeah but a materialized view extracts data to what is essentially specialized tables, which is then accessed for the search. The last part is fasdt
[12:11 PM PT] Gavin.Hird @grid.xmir.org:8002: the extract takes time, but is run as a background refresh
[12:11 PM PT] Vincent.Sylvester @hg.zetaworlds.com: I stuffed that into allparcels, probably will delete the has_pic from it and just use if not null key or something
[12:12 PM PT] Vincent.Sylvester @hg.zetaworlds.com: On mysql you could probably use redis... oh wait nvm xD
[12:14 PM PT] Gavin.Hird @grid.xmir.org:8002: I guess I should get going
[12:14 PM PT] Gavin.Hird @grid.xmir.org:8002: have a great week all
[12:14 PM PT] Andrew Hellershanks: Vincent, what UUID? The data just comes from parsng some xml.
[12:14 PM PT] Vincent.Sylvester @hg.zetaworlds.com: cya
[12:14 PM PT] Ubit Umarov: cya
[12:15 PM PT] Vincent.Sylvester @hg.zetaworlds.com: snapshotuuid
[12:15 PM PT] Andrew Hellershanks: ok, Gavin. See you next week.
[12:16 PM PT] Vincent.Sylvester @hg.zetaworlds.com: In register php in the parser $has_pic etc. Fetches the image node and even writes the uuid to a var, but never writes it to any table, just whether a parcel has a picture or not
[12:17 PM PT] Vincent.Sylvester @hg.zetaworlds.com: So there is no way to actually fetch it short of using the hostregister to then ask the region for it, which is extra work the viewer does, but websearch would need some other way
[12:17 PM PT] Vincent.Sylvester @hg.zetaworlds.com: I just added it to allparcels table so I can fetch it from there to load the image
[12:18 PM PT] Andrew Hellershanks: I haven't seen it do anything with any picture info that may exist for the region or parcel. Might be useful for a web search. The search results aren't stored in any database unless you are suggesting they could be to save some time on subsequent searches.
[12:18 PM PT] Andrew Hellershanks: It has been a while since I've looked that the module code.
[12:18 PM PT] Vincent.Sylvester @hg.zetaworlds.com: The image tag will only exist if the parcel has a snapshot image
$has_pic = 0;
$image_node = $value->getElementsByTagName("image");
$image = "00000000-0000-0000-0000-000000000000";
if ($image_node->length > 0)
{
$image = $image_node->item(0)->nodeValue;
if ($image != "00000000-0000-0000-0000-000000000000")
$has_pic = 1;
}
[12:19 PM PT] Vincent.Sylvester @hg.zetaworlds.com: $image is never used past that, not written to tables
[12:19 PM PT] Vincent.Sylvester @hg.zetaworlds.com: That's really only change, putting it in allparcels
[12:19 PM PT] Vincent.Sylvester @hg.zetaworlds.com: $allparcelarray = array(
'regionUUID' => $regionuuid,
'parcelname' => $parcelname,
'ownerUUID' => $owneruuid,
'groupUUID' => $groupuuid,
'landingpoint' => $parcellanding,
'parcelUUID' => $parceluuid,
'has_picture' => $has_pic,
'snapshotUUID' => $image,
'infoUUID' => $infouuid,
'parcelarea' => $parcelarea
);
[12:20 PM PT] Vincent.Sylvester @hg.zetaworlds.com: $has_pic is only needed for parcels query for the has picture switch
[12:20 PM PT] Vincent.Sylvester @hg.zetaworlds.com: So yeah minor things
[12:21 PM PT] Andrew Hellershanks: Why write just that one piece of info in to a table when the rest of the information isn't being written to a table?
[12:22 PM PT] Vincent.Sylvester @hg.zetaworlds.com: It's technically not needed for the queries search does
[12:22 PM PT] Andrew Hellershanks: Does the viewer accept the actual uuid of the picture (if there is one)?
[12:22 PM PT] Vincent.Sylvester @hg.zetaworlds.com: I think the viewer does a query some other way to fetch it
[12:22 PM PT] Andrew Hellershanks nods
[12:23 PM PT] Vincent.Sylvester @hg.zetaworlds.com: It doesn't get sent to the viewer search, I just wanted to show the image like the viewer does so I needed to know the uuid of it
[12:23 PM PT] Andrew Hellershanks: That is probably why it doesn't pass it along although it would be easy enough to include it if there is a name for that parameter in the data packet.
[12:23 PM PT] Ubit Umarov: outch
[12:24 PM PT] Vincent.Sylvester @hg.zetaworlds.com: I mean why we need 3 tables for the parcel data I don't get either, allparcels, popular and parcels
[12:24 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Some of that stuff is a bit redundant and $has_pic you would not need to store if you just store the image uuid != null key means has pic
[12:24 PM PT] Vincent.Sylvester @hg.zetaworlds.com: So some cleanup and rethinking of code
[12:25 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Like I said, still tinkering with it and plan to perhaps expand it a bit more in preparation for websearch
[12:26 PM PT] Andrew Hellershanks: allparcels and parcels does sound redundant on the surface. Popular is interesting as there needs to be a way to determine what one considers something as popular.
[12:27 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Has something to do with "show in search" I think. I have not looked at cleaning that up yet, but yeah it's a bit redundant
[12:27 PM PT] Ubit Umarov: oops rl calls
[12:28 PM PT] Vincent.Sylvester @hg.zetaworlds.com: Might have been used to avoid doing inner joins
[12:28 PM PT] Ubit Umarov: cya ppl
[12:28 PM PT] Andrew Hellershanks: That could be. I would need to have another look at the code to see how thedata is handled.
[12:29 PM PT] Andrew Hellershanks: We are now a half hour past the hour mark. Any last minutes questions or comments before I wrap up todays meeting?
[12:30 PM PT] Jamie.Jordan @grid.kitely.com:8002: nope thanks yall
[12:31 PM PT] Andrew Hellershanks: I will see if I can find time to look at the search module again before next week. Been busy on some work related things so may not have the time.
[12:32 PM PT] Andrew Hellershanks: On that note, I will call todays gethering to a close. Thank you all for coming. See you again next week.
