| *** tpb has joined #tp | 00:00 | |
| *** ChanServ sets mode: +o tpb | 00:00 | |
| *** Greywhind has joined #tp | 00:02 | |
| tansell | it would be good if you could prepare a weekly summary before each meeting and a "this is what i plan for next week" | 00:02 |
|---|---|---|
| Greywhind | tansell: ok. the blog posts are too detailed for the summary? | 00:02 |
| tansell | Greywhind, the blog posts are good - but I'm more after "this is what I actually achieved" | 00:03 |
| tansell | only once a week | 00:03 |
| Greywhind | tansell: ok | 00:03 |
| Greywhind | i'll write a summary next time. | 00:03 |
| tansell | yeah | 00:03 |
| tansell | it'll become more important when I get busier | 00:03 |
| Greywhind | for next week, i plan to start by continuing to fix the bugs that appear whenever i start the client | 00:03 |
| tansell | and can't look over your shoulder as much | 00:04 |
| Greywhind | i plan to finalize drawing of objects with multiple coordinates as much as possible at this point | 00:04 |
| tansell | how does the undefined problem effect that? | 00:05 |
| tansell | A number of types of objects you should consider include | 00:05 |
| tansell | * Wormholes (IE an object which is connected between to locations) | 00:05 |
| Greywhind | well, it might make the drawing of some objects impossible to perform accurately, if they have a reference object with multiple coordinates | 00:06 |
| tansell | * Mine fields which can be an arbitrary shape | 00:06 |
| tansell | * Scanner ranges? | 00:06 |
| tansell | anything else you can come up with | 00:06 |
| tansell | Lee was suggesting that wormholes might need more work because they could have properties like | 00:07 |
| tansell | single directional | 00:07 |
| tansell | might not no where they pop-out | 00:07 |
| tansell | etc | 00:07 |
| Greywhind | well, i'm thinking i'll draw anything with multiple coordinates like a starsystem with a line between each coordinate | 00:07 |
| Greywhind | does that sound reasonable? | 00:08 |
| tansell | Greywhind, possibly | 00:08 |
| tansell | Greywhind, before you get to far along, it might be worth drawing some stuff on paper | 00:09 |
| Greywhind | then things like scanner ranges can be specified simply by a circular set of coordinates | 00:09 |
| tansell | what happens if the coordinates are very very close together | 00:09 |
| Greywhind | minefields by a polygon of coordinate | 00:09 |
| Greywhind | *coordinates | 00:09 |
| tansell | they should probably be drawn as icons | 00:09 |
| Greywhind | etc. | 00:09 |
| Greywhind | i'll also try to start using media | 00:09 |
| Greywhind | although that may need to wait for next week | 00:09 |
| Greywhind | but instead of trying to figure out the type of object first, i'll just use whatever media it points me to if it has any | 00:10 |
| Greywhind | that way, future rulesets can be drawn as they like rather than as the client attempts to auto-generate | 00:10 |
| tansell | Greywhind, okay your going to have to figure out things like "icon" verses full image, etc | 00:10 |
| tansell | nobody has hit that problem before | 00:11 |
| Greywhind | tansell: yeah, i'll try to look into that kind of thing | 00:11 |
| tansell | so I think people have been just ignoring it :) | 00:11 |
| Greywhind | might have to have specific requirements for the media if it's to be drawn on the starmap | 00:11 |
| Greywhind | like a specific name or tag or something | 00:11 |
| Greywhind | we'll see | 00:11 |
| tansell | so my suggestion would be to use file endings | 00:13 |
| Greywhind | ok | 00:13 |
| tansell | have you seen | 00:13 |
| tansell | http://www.thousandparsec.net/wiki/Media.gz | 00:13 |
| tpb | Title: Media.gz - Thousand Parsec Wiki (at www.thousandparsec.net) | 00:13 |
| Greywhind | no | 00:13 |
| tansell | what is missing there is that it requires a client to download the complete media to find out things like size | 00:14 |
| Greywhind | ok | 00:14 |
| Greywhind | so by file endings, you mean what exactly? | 00:14 |
| tansell | .png | 00:15 |
| tansell | Thousand Servers will not give full URLs for the media to be used, instead it will give a URL leaving out the media extension. It is up to the client to choose the file type that is requires. | 00:15 |
| tansell | For example, | 00:15 |
| tansell | Game's base url: http://media.thousandparsec.net/media/ | 00:15 |
| tansell | Object's media url: planets/planet1 | 00:15 |
| tansell | If the client supported mng and the file is present in the Media.gz file, it could download | 00:15 |
| tansell | http://media.thousandparsec.net/media/planets/planet1.mng | 00:15 |
| tansell | The different types of media will be present in the Media.gz file, so the client can choose from them. | 00:16 |
| tansell | icon verse full size is the same problem as 2d art verses 3d art | 00:16 |
| Greywhind | ok, gotcha | 00:16 |
| tansell | the media downloader stuff needs a bit of work | 00:18 |
| Greywhind | i'll look at it when i get to that part... which file is it? | 00:19 |
| Landon | tansell: poke | 00:22 |
| llnz | bbl | 00:31 |
| *** llnz has quit IRC | 00:31 | |
| tansell | Greywhind, it's the Media thread in libtpclient-py | 00:43 |
| tansell | Landon, how goes things? | 00:44 |
| Greywhind | tansell: ok. | 00:44 |
| Greywhind | now i have to set up git-cl on this machine | 00:44 |
| Landon | nothing tangible yet, mainly reading the ogre manual and sketching out the battle spec so I have some quick references to it | 00:46 |
| Landon | going to try getting python-ogre on my netbook from my desktops installation btw | 00:47 |
| Landon | and see if that works | 00:47 |
| *** jnengland77 has quit IRC | 00:53 | |
| Greywhind | tansell: http://codereview.mithis.com/2 | 00:59 |
| tpb | Title: Issue 2: Fixed panelPicture to figure out types of objects TP04 style. - Code Review (at codereview.mithis.com) | 00:59 |
| tansell | Greywhind, done | 01:02 |
| Greywhind | tansell: ok | 01:02 |
| tansell | I wonder why cherez's change ended up at 1001 | 01:02 |
| cherez | I hope Greywhind doesn't put up 999 more, or we might have problems. | 01:03 |
| Greywhind | heh | 01:04 |
| tansell | cherez, you saw my comments on your patch right? | 01:04 |
| cherez | Yeah. | 01:05 |
| tansell | okay great | 01:05 |
| cherez | I'm doing the pack/unpack testing now. | 01:05 |
| CIA-46 | Greywhind tpclient-pywx-tp04 * r49060472af5b / (extra/objectutils.py windows/main/panelPicture.py): | 01:06 |
| CIA-46 | Fixed panelPicture to figure out types of objects TP04 style. Note: will need to look | 01:06 |
| CIA-46 | for self-specified media in each object in the future. | 01:06 |
| Greywhind | ok, time to write a blog post and then sleep | 01:07 |
| cherez | tansell: Should negative timestamps be possible? | 01:23 |
| mithro_ | cherez: dunno - how are they represented | 01:37 |
| cherez | The packed value is derived as: t = long(time.mktime(t.timetuple())) | 01:39 |
| cherez | But time.mktime doesn't work with negative times and throws an exception. | 01:40 |
| cherez | I could write a replacement function that would work around that, but I'm not sure how much pre-1970s timestamps will matter. | 01:40 |
| tansell | but can they even be stored? | 01:40 |
| tansell | is it stored in an unsigned int? | 01:43 |
| cherez | It's presently unsigned, but there's no reason for it to be. | 01:44 |
| cherez | It looks like Python's datetime objects internally use a signed int. | 01:44 |
| tansell | cherez, it has to be compatible with cpp-server | 01:46 |
| cherez | And there will only be inconstencies before January 1st 1970 and after January 19th 2038. | 01:46 |
| tansell | is it a 32bit number or 64bit? | 01:46 |
| cherez | There are versions for both in xstruct. | 01:46 |
| tansell | test boundaries and wrong input | 01:47 |
| cherez | They both have the same boundaries. | 01:47 |
| cherez | The ones I just mentioned. | 01:47 |
| cherez | So between times 0 and 2^31-1 | 01:48 |
| tansell | well a 64bit number has a bigger boundry then that | 01:48 |
| cherez | Yes, but we convert to and from datetimes. | 01:48 |
| cherez | So even if our int64 is okay with it, it will crash in the conversion. | 01:49 |
| tansell | cherez, will it? does a datetime only support 32bit numbers? | 01:49 |
| cherez | Signed 32-bit integers. | 01:50 |
| cherez | So 1970 +- 68 years. | 01:51 |
| cherez | I don't think we can circumvent that without either dropping datetime or reimplementing it. | 01:52 |
| tansell | cherez, so document the limitations :) | 01:52 |
| cherez | Where should I document them? | 01:52 |
| cherez | Just at the header of the file? | 01:53 |
| tansell | Possibly | 01:57 |
| tansell | where ever seems the best | 01:57 |
| tansell | maybe a "limitations" section at the top | 01:57 |
| cherez | tansell: http://codereview.mithis.com/1001 | 02:10 |
| tpb | Title: Issue 1001: Fixed floating point numbers and added float and timestamp tests. - Code Review (at codereview.mithis.com) | 02:10 |
| tansell | cherez, I think you're better of making a light weight class for this | 02:11 |
| cherez | For datetime objects? | 02:12 |
| tansell | I'll put in the comments | 02:12 |
| cherez | Does anything use xstruct for timestamps now? | 02:12 |
| tansell | cherez, I think they do | 02:18 |
| tansell | check libtpproto-py | 02:18 |
| cherez | Definitely seems so. | 02:21 |
| *** peres has joined #tp | 02:24 | |
| tansell | cherez, comments | 02:24 |
| cherez | Reading. | 02:25 |
| mithro_ | cherez: a good theory to follow is that you are up at 2:00am in the morning trying to debug this code | 02:26 |
| mithro_ | try and make them as clear and easy to debug as possible | 02:26 |
| cherez | That will probably be happening in half an hour. :P | 02:27 |
| cherez | I'll be implementing your suggestions now. | 02:27 |
| *** tansell_laptop has joined #tp | 02:50 | |
| *** greenlion has joined #tp | 03:02 | |
| *** llnz has joined #tp | 03:11 | |
| cherez | tansell: Is this more readable? http://codereview.mithis.com/1001 | 03:22 |
| tpb | Title: Issue 1001: Fixed floating point numbers and added float and timestamp tests. - Code Review (at codereview.mithis.com) | 03:22 |
| tansell | cherez, almost | 04:09 |
| tansell | but repeat after me, *never*, never, never use backticks :P | 04:09 |
| cherez | Alright, I'll turn them into repr()'s | 04:10 |
| tansell | otherwise it's much better | 04:10 |
| tansell | cherez, there is %r | 04:10 |
| tansell | which does what you want | 04:10 |
| cherez | Really? | 04:10 |
| tansell | yes | 04:11 |
| cherez | Woohoo. | 04:11 |
| cherez | Are strings null terminated? | 04:11 |
| tansell | cherez, no | 04:11 |
| tansell | python uses a length character | 04:11 |
| cherez | I mean in xstruct's packing. | 04:11 |
| tansell | cherez, no | 04:12 |
| tansell | we use a len packing scheme | 04:12 |
| tansell | see the protocol document | 04:12 |
| tansell | http://www.thousandparsec.net/tp/dev/documents/protocol3.php#Basics | 04:14 |
| tpb | <http://ln-s.net/3EkT> (at www.thousandparsec.net) | 04:14 |
| tansell | know and love that protocol documents :P | 04:14 |
| *** greenlion has quit IRC | 04:17 | |
| *** JLP_ has joined #tp | 04:17 | |
| *** JLP has quit IRC | 04:17 | |
| cherez | tansell: The protocol says UTF-8, but xstruct looks like it only supports ASCII right now. | 04:19 |
| tansell | why do you think that? | 04:20 |
| tansell | check unpack_string | 04:20 |
| cherez | I might not remember UTF-8 well, but I believe that will just give us the ASCII string that happens to have the same representation as the UTF-8 string. | 04:23 |
| cherez | Doesn't str only support 8-bit characters? | 04:23 |
| tansell | yes | 04:23 |
| tansell | but Python has a unicode object | 04:24 |
| tansell | hence why | 04:24 |
| tansell | pack_string does | 04:24 |
| tansell | unicode(s).encode('utf-8') | 04:24 |
| tansell | which basically encoded the unicode string into utf-8 format | 04:24 |
| cherez | My pack_string does "s = str(s)" | 04:25 |
| tansell | unicode(u'') does nothing | 04:25 |
| tansell | cherez, hrm | 04:25 |
| cherez | Are you sure your version is what's on the repository? | 04:25 |
| tansell | you might want to check it against libtpproto-py's xstruct.py as that is the most up-to-date | 04:25 |
| cherez | Hey, it's the code I was going to write! | 04:26 |
| cherez | I'll forward port that into libtpproto2-py. | 04:26 |
| tansell | yeah | 04:28 |
| tansell | you should also backport your changes and test to libtpproto-py | 04:28 |
| *** greenlion has joined #tp | 04:39 | |
| cherez | I don't think I'm understanding how to pack lists. | 04:43 |
| tansell | cherez, oh? | 04:44 |
| cherez | Can lists contain compound data? | 04:44 |
| tansell | what do you mean? | 04:44 |
| cherez | Like [bb] ? | 04:45 |
| tansell | sure | 04:46 |
| tansell | we have things a lot more complicated then that | 04:46 |
| tansell | [jSI[II]] | 04:46 |
| *** cherez has left #tp | 04:53 | |
| *** cherez has joined #tp | 04:53 | |
| mithro_ | [18:46] <tansell> sure | 05:00 |
| mithro_ | [18:46] <tansell> we have things a lot more complicated then that | 05:00 |
| mithro_ | [18:46] <tansell> [jSI[II]] | 05:00 |
| mithro_ | going to dinner now | 05:01 |
| mithro_ | bblr | 05:01 |
| *** Landon has quit IRC | 05:01 | |
| *** Landon has joined #tp | 05:03 | |
| *** jmtan has quit IRC | 05:19 | |
| cherez | mithro_: New version for you to look at. | 05:22 |
| tansell | cherez, lgtm | 05:33 |
| cherez | Alright, I'll push it then. | 05:33 |
| tansell | cherez, still not checking error cases however | 05:34 |
| cherez | That's next. | 05:34 |
| tansell | okay great | 05:34 |
| cherez | I'll be checking to make sure exceptions get thrown whien they should be? | 05:34 |
| tansell | yeah | 05:34 |
| tansell | IE when you try and encode an float into an int | 05:34 |
| tansell | or you can invalid unicode | 05:35 |
| tansell | etc | 05:35 |
| CIA-46 | cherez libtpproto2-py * r080ecf71e8a8 / (testxstruct.py tp/netlib/xstruct.py): Fixed floating point numbers and added float and timestamp tests. | 05:39 |
| CIA-46 | cherez libtpproto2-py * r5d573789afb0 /testxstruct.py: Implemented separate pack/unpack testing. | 05:39 |
| CIA-46 | cherez libtpproto2-py * r16a1e196cc0e / (testxstruct.py tp/netlib/xstruct.py): Testing for timestamp objects and documenting timestamp limitations. | 05:39 |
| CIA-46 | cherez libtpproto2-py * r5bfd73ae9b2f /testxstruct.py: Cleaned up the tests. | 05:39 |
| CIA-46 | cherez libtpproto2-py * re370e08a3006 / (testxstruct.py tp/netlib/xstruct.py): Unicode support, tests for lists and strings. | 05:39 |
| *** llnz2 has joined #tp | 05:41 | |
| *** llnz has quit IRC | 05:41 | |
| *** llnz2 is now known as llnz | 05:41 | |
| cherez | I'm writing a list of things that could go wrong. | 06:05 |
| cherez | I just realized how futile that sounds. | 06:06 |
| llnz | hehe | 06:11 |
| * llnz ponders how to set up the media in minisec | 06:15 | |
| cherez | tansell: I have failure tests up for review. | 06:19 |
| tansell | cherez, hrm - you forgot to squash before pushing | 06:20 |
| cherez | Doh. :/ | 06:20 |
| tansell | I can reset the remote repository if you want to try again | 06:20 |
| tansell | cherez, btw your better of using mithro: as it flashes both my work and home laptop | 06:21 |
| cherez | Ahh, alright. | 06:21 |
| cherez | Yikes, it's 5:30 AM. | 06:33 |
| cherez | Good night. | 06:33 |
| * llnz wanders off | 07:55 | |
| llnz | later all | 07:55 |
| *** llnz has quit IRC | 07:55 | |
| *** tuna-fish has quit IRC | 08:09 | |
| *** mithro_ has quit IRC | 08:58 | |
| *** reac has joined #tp | 09:02 | |
| *** bddebian has joined #tp | 10:09 | |
| bddebian | Heya | 10:10 |
| *** jmtan has joined #tp | 10:27 | |
| *** tuna-fish has joined #tp | 10:32 | |
| *** Noughmad has joined #tp | 11:14 | |
| *** Noughmad_ has joined #tp | 11:23 | |
| *** DTRemenak has quit IRC | 11:57 | |
| *** DTRemenak has joined #tp | 12:05 | |
| *** Noughmad has quit IRC | 12:09 | |
| *** Noughmad_ has quit IRC | 12:23 | |
| *** alanp has joined #tp | 12:37 | |
| *** peres has quit IRC | 12:39 | |
| *** alanp has quit IRC | 12:48 | |
| *** nash has quit IRC | 12:57 | |
| *** Erroneous has joined #tp | 13:44 | |
| *** reac has quit IRC | 14:00 | |
| *** jmtan has quit IRC | 14:31 | |
| *** Greywhind has quit IRC | 15:27 | |
| *** tuna-fish has quit IRC | 15:36 | |
| *** bcarlyon|laptop has joined #tp | 15:37 | |
| *** bcarlyon|laptop has quit IRC | 16:03 | |
| *** bcarlyon|laptop has joined #tp | 16:04 | |
| *** llnz has joined #tp | 16:18 | |
| llnz | monring all | 16:19 |
| *** Greywhind has joined #tp | 16:32 | |
| *** jnengland77 has joined #tp | 16:47 | |
| *** tuna-fish has joined #tp | 16:51 | |
| Landon | morning | 17:02 |
| *** mhilmi has joined #tp | 17:16 | |
| alanp-home | hey | 17:18 |
| alanp-home | what do you guys use to generate your xml? | 17:18 |
| alanp-home | should I just use vim? | 17:18 |
| alanp-home | or some gui app? | 17:18 |
| llnz | alanp-home: i typically use quanta (html/xml editing tool for kde) | 17:22 |
| llnz | vim would be ok | 17:22 |
| alanp-home | kde eh | 17:28 |
| alanp-home | works for me | 17:28 |
| *** bddebian has quit IRC | 17:29 | |
| * alanp-home trying to figure out how to use quanta | 17:31 | |
| llnz | it probably is not the best tool | 17:33 |
| alanp-home | ah | 17:34 |
| llnz | every time I open the protocol.xml file it asks to turn it into an HTML file | 17:35 |
| * alanp-home hasn't done much cpp+xml, checking for recommended procedures online | 17:36 | |
| llnz | We have tiny xml in the sources already | 17:37 |
| alanp-home | yeah i see it in the risk ruleset | 17:42 |
| llnz | i'll move it to the core (tpserver) if you think it would be useful | 17:46 |
| llnz | actually, i might anyway | 17:46 |
| alanp-home | we can leave it out of core for now i think | 17:46 |
| alanp-home | but it might be beneficial to add it | 17:46 |
| alanp-home | since we're now using it in two places, and i am sure it will be used in more | 17:47 |
| alanp-home | llnz: do you use an ide? | 17:52 |
| llnz | yes, kdevelop3 | 17:52 |
| * llnz uses kde as his desktop | 17:53 | |
| llnz | (when not at work) | 17:53 |
| * alanp-home also | 17:53 | |
| alanp-home | which dist? | 17:53 |
| alanp-home | of linux i mean | 17:53 |
| llnz | debian testing currently | 17:54 |
| *** jnengland77 has quit IRC | 17:55 | |
| *** jnengland77 has joined #tp | 17:55 | |
| *** jnengland77 has joined #tp | 17:56 | |
| *** jnengland77 has left #tp | 17:57 | |
| *** jnengland77 has joined #tp | 17:57 | |
| *** jnengland77 has joined #tp | 17:58 | |
| *** jnengland77 has quit IRC | 17:59 | |
| *** Greywhin1 has joined #tp | 18:06 | |
| *** jnengland77 has joined #tp | 18:08 | |
| *** tuna-fish has quit IRC | 18:11 | |
| *** Greywhind has quit IRC | 18:18 | |
| *** greenlion has quit IRC | 18:24 | |
| *** tuna-fish has joined #tp | 18:24 | |
| alanp-home | llnz: ping | 19:12 |
| llnz | alanp-home: pong | 19:12 |
| alanp-home | http://pastebin.com/m4664f1d1 does that look like an alright xml layout for comps? | 19:12 |
| tpb | Title: pastebin - collaborative debugging tool (at pastebin.com) | 19:12 |
| alanp-home | my xmlfu is not advanced, so I want to run it by you | 19:13 |
| llnz | as a micro format it looks ok, the top level tag could probably be "Components" | 19:14 |
| *** nash has joined #tp | 19:15 | |
| llnz | are you going to use the tag names inside the property list to find the propertyid? | 19:15 |
| llnz | hi nash | 19:15 |
| nash | heyo | 19:15 |
| alanp-home | llnz: what do you mean "find"? | 19:16 |
| llnz | iirc, the property list in Component is a propertyid to string map, so will you use the tag name to lookup the property (and thefore the id) before adding it to the map? | 19:18 |
| alanp-home | oh | 19:18 |
| alanp-home | the index right? | 19:18 |
| alanp-home | propertylist[propertyIndex["Speed"]] = "(lambda (design) 0)"; | 19:18 |
| llnz | yeah | 19:21 |
| alanp-home | yes, for sure | 19:21 |
| alanp-home | i had to capatalize the property to match because of that | 19:22 |
| tansell | morning | 20:10 |
| llnz | hi tansell | 20:14 |
| tansell | Greywhin1, ping? | 20:15 |
| Greywhin1 | tansell: pong | 20:17 |
| *** Greywhin1 is now known as Greywhind | 20:17 | |
| tansell | Greywhind, how goes things today? | 20:18 |
| tansell | I have a meeting at 11:00 here - so just wanted to check there is nothing we need to chat about first | 20:19 |
| Greywhind | tansell: a little bit slowly, since i'm trying to figure out how best to add media support to the starmap | 20:19 |
| Greywhind | i submitted one patch for review | 20:19 |
| Greywhind | http://codereview.mithis.com/3 | 20:19 |
| tpb | Title: Issue 3: Fixed the code that determines whether an object can move in panelStarMap.py to work with tp04. - Code Review (at codereview.mithis.com) | 20:19 |
| tansell | Greywhind, I would go for something easy to start with :P | 20:24 |
| tansell | Greywhind, the left pane box and the icons in the system tree for example | 20:24 |
| Greywhind | the picturePanel? | 20:25 |
| tansell | yeah | 20:27 |
| tansell | Greywhind, done! | 20:27 |
| *** mithro_ has joined #tp | 20:30 | |
| tansell | Greywhind, the starmap problem is actually really hard | 20:47 |
| Greywhind | tansell: hmm... i was thinking adding another icon type, that draws an image | 20:48 |
| Greywhind | and using that if the correct media reference exists for an object | 20:48 |
| tansell | Greywhind, how does that work for things like the orbit icons for systems | 20:52 |
| Greywhind | tansell: i was thinking that a system would either choose to use the media reference, and just draw the star, or use the standard system icon style as it does now | 20:53 |
| Greywhind | maybe in the future it could be extended to draw the child images around it too | 20:53 |
| tansell | Greywhind, there isn't a huge amount of space on the starmap | 20:55 |
| Greywhind | well, obviously the images would need to be fairly small | 20:55 |
| Greywhind | the size of current icons | 20:55 |
| tansell | which is pretty tiny | 20:58 |
| Greywhind | i think it would make the starmap look more game-like to have some images, especially for fleets | 21:01 |
| tansell | Greywhind, those little triangle icons don't take up much space | 21:02 |
| tansell | and when you have 500 of them | 21:02 |
| tansell | that is a good thing :) | 21:02 |
| Greywhind | true | 21:02 |
| Greywhind | but custom objects like minefields will probably need their own images, because we'll never draw everything exactly as it wants to be drawn by trying to auto-detect | 21:02 |
| llnz | I wonder if we need a type of object type, being point, area, "line/link", etc | 21:04 |
| *** tansell_laptop has quit IRC | 21:07 | |
| *** tansell_laptop has joined #tp | 21:07 | |
| tansell_laptop | even the small 16x16 icons in the star system pane are too big for the starmap | 21:24 |
| Greywhind | well, i'll start with the picture panel | 21:24 |
| Greywhind | and we'll deal with the starmap another time, i guess | 21:24 |
| Greywhind | is it important that i remove that single deleted blank line from the commit i submitted for feedback? | 21:25 |
| Greywhind | if so, how would i do so? | 21:25 |
| Greywhind | just make another commit with the line returned and then squash it? | 21:25 |
| tansell_laptop | Greywhind, do a --amend commit would probably be best | 21:27 |
| Greywhind | i'll take a look | 21:27 |
| Greywhind | tansell_laptop: i'm having trouble coming up with any other way to figure out if an object can move. all objects have velocity... | 21:46 |
| Greywhind | having the ability to be given a move order seems like a reasonable way to determine it | 21:46 |
| tansell_laptop | has an order which as a position? | 21:47 |
| Greywhind | oh - orders can have a "pos," i see... | 21:47 |
| Greywhind | that would probably be better than using the name, yes | 21:47 |
| tansell_laptop | but if you imagine a "packet thrower" might have a "throw packet here" order | 21:48 |
| tansell_laptop | which would be caught by that | 21:51 |
| Greywhind | true | 21:51 |
| tansell_laptop | maybe we can draw different orders types in different colors | 21:52 |
| Greywhind | does that have anything to do with this solution, or is it a separate thought? | 21:52 |
| tansell_laptop | well - I mean the lines | 21:53 |
| tansell_laptop | then you could see if it's an intercept verses a throw-packet | 21:54 |
| tansell_laptop | brb | 21:54 |
| tansell | then you could go down the route of something like the resources | 21:55 |
| tansell | where you say "Display only paths which are of this type" | 21:55 |
| tansell | a fairly flexible solution? | 21:56 |
| Greywhind | well, this is just to show or hide the waypoint button | 21:57 |
| Greywhind | but i'm not sure what lines you're talking about | 21:57 |
| tansell | okay | 21:57 |
| tansell | you know how we loop over the objects orders | 21:57 |
| tansell | and find a "path" the object is going to take | 21:58 |
| Greywhind | no, actually... not sure i do | 21:58 |
| Greywhind | i haven't looked much at the order panel and such | 21:58 |
| tansell | Greywhind, tell a fleet to move somewhere, and then somewhere again | 21:59 |
| tansell | http://git.thousandparsec.net/gitweb/gitweb.cgi?p=tpclient-pywx.git;a=blob;f=windows/main/overlays/Path.py;h=0911bd5cc93ce36ef4719651336db1dc1dfc93e2;hb=tp04 | 22:01 |
| tpb | <http://ln-s.net/3EzP> (at git.thousandparsec.net) | 22:01 |
| *** bddebian has joined #tp | 22:04 | |
| Greywhind | tansell: well, in this case, the fleet won't necessarily have any orders | 22:05 |
| Greywhind | tansell: we need to find out if it supports move orders, not if it already has some | 22:05 |
| tansell | they are a related problem | 22:05 |
| Greywhind | will this subtype attribute help? if subtype == constants.ARG_ABS_COORD: | 22:06 |
| Greywhind | or does that have the same packet thrower problem? | 22:06 |
| tansell | well that has the same thrower problem | 22:07 |
| Greywhind | i figured | 22:07 |
| tansell | so basically what I'm suggesting is a UI rework | 22:07 |
| tansell | where we ask the user to select which 'paths' to show on the starmap | 22:08 |
| tansell | (and thus which paths can be added using the waypoint system) | 22:08 |
| Greywhind | hm | 22:09 |
| tansell | does that make more sense? | 22:09 |
| Greywhind | yeah, that makes more sense | 22:09 |
| tansell | so basically they can go | 22:09 |
| tansell | "Oh a green path is a packet thrower" | 22:09 |
| tansell | "and a blue path is a ship moving" | 22:09 |
| Greywhind | yeah, that makes sense | 22:10 |
| tansell | the idea still need a bit of work | 22:10 |
| tansell | (what about ship mounted packet throwers :P) | 22:10 |
| Greywhind | i don't know how soon i can do a change like that... i might have to leave it as it is for now | 22:11 |
| tansell | multiple order queue might be another solution | 22:11 |
| tansell | Greywhind, well I would alteast change it to be consistent with the path stuff for now | 22:11 |
| Greywhind | i'll try to do that | 22:12 |
| tansell | and that path code should probably end up in your objectutils stuff to | 22:12 |
| Greywhind | well, i'm actually not sure that path code will work with tp04, since i haven't looked into how orders work now | 22:12 |
| tansell | Greywhind, the only real change is the fact that objects can have multiple order queues | 22:13 |
| tansell | so instead of using the object id - you use the order queue id | 22:13 |
| tansell | (and an object will reference a bunch of order queues) | 22:13 |
| Greywhind | so cache.orders[obj.id] won't work anymore, but cache.orders[orderqueue.id] will? | 22:13 |
| tansell | more like | 22:14 |
| Greywhind | where orderqueue is one of the object's order queues, that is | 22:14 |
| tansell | cache.orders[obj.orderqueue1.id] | 22:14 |
| Greywhind | ah. | 22:14 |
| Greywhind | yeah | 22:14 |
| tansell | where 'orderqueue1' is dynamic | 22:14 |
| tansell | (the name) | 22:15 |
| Greywhind | right | 22:15 |
| Greywhind | i'll try to figure out a method for that at some point | 22:15 |
| Greywhind | maybe "get order queue by name" or something | 22:15 |
| tansell | IE It could be an attribute called "Orders" or "Build Queue" or | 22:15 |
| tansell | "Attack Orders" | 22:15 |
| Greywhind | and a "get all order queues" | 22:15 |
| Greywhind | i have to go eat | 22:15 |
| tansell | fixing the orders panel is probably a good short term goal too | 22:16 |
| * tansell is heading to lunch himself | 22:16 | |
| tansell | see ya in a bit! | 22:16 |
| Greywhind | i was thinking of doing that after fixing drawing, but i might move it up | 22:16 |
| llnz | Greywhind: just fyi, i'm intending to add media info to the objects in minisec this long weekend | 23:16 |
| Greywhind | llnz: ok | 23:17 |
| tansell | back | 23:17 |
| Greywhind | tansell: so, two things | 23:23 |
| tansell | I have to run off in 5 minutes again (busy friday :() | 23:23 |
| Greywhind | tansell: well, the old way that the path manager calculated stuff | 23:23 |
| Greywhind | doesn't work with dynamic orders | 23:24 |
| tansell | Greywhind, orders have always been dynamic | 23:24 |
| Greywhind | hmm | 23:25 |
| Greywhind | when i did: | 23:25 |
| Greywhind | argument_name, subtype = orderdesc.names[0] | 23:25 |
| Greywhind | it said the orderdesc didn't have a names attribute | 23:25 |
| Greywhind | so something must have changed | 23:25 |
| tansell | It might be a change to the general structure of dynamic objects | 23:26 |
| tansell | s/objects/frames/ | 23:26 |
| tansell | because the on-wire format for orders has not changed | 23:26 |
| Greywhind | well, the other thing is that hasattr(orderdesc, 'pos') doesn't ever return true | 23:27 |
| Greywhind | even though dir(orderdesc) shows a 'pos' | 23:27 |
| Greywhind | tansell: i can get around that with: | 23:31 |
| Greywhind | for property in order.properties: | 23:31 |
| Greywhind | if property.name == 'pos': | 23:31 |
| Greywhind | canmove = True | 23:31 |
| Greywhind | but i'm confused as to why it doesn't have a .pos attribute | 23:31 |
| *** mhilmi has quit IRC | 23:36 | |
| Greywhind | i think i'm actually going to check if it has any parameters of type OrderParamAbsSpaceCoords or OrderParamRelSpaceCoords | 23:36 |
| Greywhind | that'll be more robust than "pos" | 23:36 |
| *** jmtan has joined #tp | 23:38 | |
| Landon | hey jmtan | 23:47 |
| jmtan | hi Landon, did you resolve the pythonogre issues? | 23:47 |
| Landon | maybe, I'm missing a few libraries on my laptop | 23:47 |
| Landon | so i've been tracking those down | 23:47 |
| jmtan | oh, which libraries are those? | 23:51 |
| Landon | not sure | 23:52 |
| Landon | the last time I tried I got a message that GEM was missing or something like that | 23:52 |
| Landon | (soon after I got a kernel panic) | 23:52 |
| Landon | :P | 23:52 |
| Landon | oh just remembered there was a prereq script, I'll try that | 23:54 |
| jmtan | the prereq script in pythonogre should be for compiling only (i think) | 23:56 |
| Landon | oh good point, its only dev stuff being installed | 23:56 |
| jmtan | is your kernel unmodified? some googling shows that gem is a kernel feature | 23:58 |
| jmtan | http://kernelnewbies.org/Linux_2_6_28#head-b957b19f6139b6bbbfabaf790bf643b1746985d6 | 23:58 |
| tpb | <http://ln-s.net/2i-9> (at kernelnewbies.org) | 23:58 |
| Landon | yeah it's unmodified | 23:59 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!