IRC logs for #opendune on OFTC at 2009-12-16
⏴ go to previous day
08:57:46 <Xaroth> TrueBrain: I don't get why most of the php svn browsers (websvn and the like) don't use pecl svn
08:59:01 <Xaroth> it gives access to most (if not all) svn commands, without having to parse the output (websvn, for instance, lets it output as xml, and parses that... resulting in increased memory usage due to sucky xml parser)
08:59:46 <Xaroth> with pecl it's just svn_ls("svn://svn.opendune.org/") and it'll fetch it.. same with file:// and other handlers that normal svn client understands
09:32:30 *** ChanServ sets mode: +v Yexo
10:40:12 <Xaroth> going to make this svn display thing 1 repo per site
10:50:14 <Xaroth> not going to bother with support for multi-repo display, especially as it requires me to do all kinds of wierd stuff to my concept :P
11:31:01 <TrueBrain> Xaroth: just make sure to monitor memory usage of this pecl svn
11:31:21 <TrueBrain> I wrote a few svn handlers for python, they were as memory efficient as SVN can get, still .. it consumes a lot :p
11:37:03 <Xaroth> can't really test it on this machine tho, only 67 revisions :P
11:37:11 <Xaroth> note to self, writing custom pagination SUCKS ARSE
11:37:22 <Xaroth> btw, going to Evo after lunch, want something from it? :P
11:37:35 <Xaroth> like, turn off some wikipedia servers or so :P
12:33:34 <Xaroth> [msg] => -Add: named 2 more GUI functions
12:34:35 <Xaroth> svn_log(rev, SVN_REVISION_HEAD, SVN_REVISION_HEAD)
12:34:58 <Xaroth> trying to reproduce websvn one page at a time
12:36:14 <Xaroth> that's not much work tbh :P
12:36:19 <Xaroth> made my own config class and stuff
12:36:28 <Xaroth> updated smarty (by lack of better template engine)
12:36:37 <Xaroth> and started fiddling with stuff
12:36:55 <TrueBrain> in a moment, after scrubs, I am going to start with GUI shit :)
12:38:52 <Xaroth> <b>Fatal error</b>: Allowed memory size of 16777216 bytes exhausted (tried to allocate 31 bytes) in <b>/var/www/svndune/classes/svn.class.php</b> on line <b>25</b><br />
12:39:01 <TrueBrain> and there it starts :p
12:39:08 <Xaroth> you don't want to know what I tried :P
12:39:24 <Xaroth> svn_ls(repo, SVN_REVISION_HEAD, SVN_LS_RECURSE)
12:39:33 <TrueBrain> I am just happy to read that you are limited to 16MB memory
12:40:03 <Xaroth> just shows that recursive is a bad thing :)
12:40:25 <Xaroth> also means that I will be disabling recursive :P
12:43:59 <Xaroth> not getting the 'created_rev' entity
12:44:06 <Xaroth> as clearly src was created -before- 771
12:44:10 <Xaroth> it should be called modified_rev
12:44:16 <Xaroth> or just plain 'rev' ...
12:46:09 <Xaroth> nothing that'll kill me tho :P
13:03:41 <TrueBrain> walking GUI functions .. hitting dead ends :p
13:05:47 <glx> better than "infinite" loops ;)
13:19:58 <TrueBrain> okay, back to square one .. approach two ..
13:23:49 <TrueBrain> I think I just need to convert a GUI function, and see where that leads us ...
13:25:16 <glx> I'll try to convert 0AEC functions
13:26:09 <glx> I noticed something strange in emu_GUI_Viewport() around l__013A
13:27:10 <glx> emu_bx can't be > 3, but there are 5 values in switch
13:27:18 <glx> maybe it can be negative then
13:27:42 <TrueBrain> I doubt that, but we will see :)
13:29:10 <glx> 0163 015c 0155 016a 0082 <-- I'd say jump to 0082 is impossible :)
13:30:43 <TrueBrain> jump to 0082 is also weird
13:30:49 <TrueBrain> so I guess it is a slip of the decompiler
13:30:58 <TrueBrain> we have many of them ;)
13:31:12 <TrueBrain> what is the word after 0082?
13:31:15 <glx> indeed this section handles the little arrows to move the map
13:31:32 <TrueBrain> it is the other switch
13:31:37 <TrueBrain> so yes, the decompiler fucked up
13:32:47 <glx> this function handles everything :)
13:32:56 <glx> move map, set destination, set target
13:33:08 <TrueBrain> all clicks in the viewport
13:33:13 <TrueBrain> I didn't expect anything else ;)
13:33:47 <glx> yes and mouse move of course (for structure place)
13:34:28 <glx> hmm too bad there's a "not easy" unresolved jump
13:34:56 <TrueBrain> more the question: is it important?
13:35:55 <glx> well there's space between 046E and 04BE
13:36:13 <glx> so maybe something important
13:36:50 <TrueBrain> so you need to figure out what 6D8F does ;)
13:37:48 <glx> currently all tests on 6D8F are for 0 and 4
13:43:04 <TrueBrain> in the middle of a conversion, so I will do that after that :)
13:43:22 <glx> I'm just reading the functions for now
13:44:01 <TrueBrain> hmm .. some labels on buttons are centered
13:44:07 <TrueBrain> (by reading the code)
13:44:25 <glx> to find where I need to create crash.bin and where not :)
13:44:26 <TrueBrain> and for some reason X-positions are stored divided by 8
13:47:33 <TrueBrain> @base 10 16 [calc 0x4192 - 0x413a]
13:47:36 <glx> I remember >> 3 for 93E5 and 95E5 arrays
13:48:02 <TrueBrain> @base 10 16 [calc 0xb6 - 0x9a]
13:51:24 <TrueBrain> somewhere I made a conversion error .. I hate it when that happens :p
13:58:03 <TrueBrain> Cancel, Previous, Yes and No are centered
13:58:06 <TrueBrain> rest is left aligned
13:59:39 <TrueBrain> f__B4F2_1303_0017_56DB <- some special entries for 'text'
14:00:13 <TrueBrain> @calc 0x10000 - 0xFFF2
14:00:37 <TrueBrain> owh, wait .. 0xFFF2 .. 0xFFFF
14:55:30 <DorpsGek> SVN: truebrain (r772) -Add: figured out a few more variables
15:03:38 <DorpsGek> SVN: truebrain (r773) -Add: C-ified GUI_Widget_DrawButton()
15:04:00 <TrueBrain> took just 1+ hour :p
15:17:13 <DorpsGek> SVN: truebrain (r774) -Add: forgot to name 3 variables which were pretty obvious
15:26:23 <TrueBrain> Netbeans sometimes does not correct show the call graph :(
15:29:38 <TrueBrain> @base 10 16 [calc 0x4192 - 0x4062]
15:32:01 <TrueBrain> @base 10 16 [calc 0x4062 + 2 * 20 * 8]
15:32:09 <TrueBrain> @base 10 16 [calc 0x4062 + 2 * 19 * 8]
15:34:48 <DorpsGek> SVN: truebrain (r775) -Add: more naming around Widgets
15:43:24 <TrueBrain> I understand why it does a divide for the X position (320 doesn't fit in a byte), but I don't get why by 8 ...
15:43:38 <TrueBrain> but okay .. I also can't make sense about the button hover code :p
15:44:14 <DorpsGek> SVN: truebrain (r776) -Fix (r775): forgot to change one 0x13 to 19
16:12:47 <DorpsGek> SVN: truebrain (r777) [JIT] -Add: mapped another 13 functions (French and German language)
16:13:35 <DorpsGek> SVN: truebrain (r778) -Update (r777): update decompiled code
16:14:16 <TrueBrain> glx: I can't resume your crash bin, it indicates it is already decompiled
16:16:12 <TrueBrain> (which is not possible of course, but still)
16:27:31 <TrueBrain> ha, my first working memory patch :)
16:27:48 <TrueBrain> Could not find file 'ROCKET.VOC'
16:27:55 <TrueBrain> dunno what it does, but it breaks things :p
16:29:10 <TrueBrain> can't get to the function you were looking for
16:29:43 <TrueBrain> so I don't know what it does :p
16:30:25 <TrueBrain> yes, it has to do which (digital) driver is loaded
16:30:28 <TrueBrain> so lets ignore it for now :)
16:31:02 <TrueBrain> I guess someone should fix the sound shit ...
16:45:37 <TrueBrain> sound driver causes a lot of code to be executed in 0xAB000
16:45:47 <TrueBrain> don't know what to think about that
16:47:28 <TrueBrain> the code looks fine, so I don't see any real harm ...
16:47:42 <TrueBrain> but I guess it depends which audio driver you load
16:49:21 <TrueBrain> and welcome back Xaroth ;)
16:49:45 <DorpsGek> SVN: truebrain (r779) [JIT] -Add: mapped another 292 functions (audio driver 'support')
16:51:13 <DorpsGek> SVN: truebrain (r780) -Update (r779): update decompiled code
16:51:19 <TrueBrain> glx: there, that should get you on the way a bit
16:51:26 <TrueBrain> there is still an unresolved, but not one I could fix easily
16:52:04 <Xaroth> gah, my pagination function fails
16:52:51 <TrueBrain> bad for stats .. 50 more real functions :(
16:54:21 <TrueBrain> now time for some food!!
17:11:30 <Xaroth> AH, much better, pagination works
18:28:58 <Xaroth> TrueBrain/glx: any direct stuff you want different in the 'new' svn browser?
18:29:07 <TrueBrain> Xaroth: mostly, easy browsing the diff
18:29:16 <TrueBrain> Prev / Next are essential
18:29:24 <Xaroth> TrueBrain goes for the kill :P
18:29:43 <TrueBrain> Xaroth: for the rest, clear overview of revisions
18:29:54 <TrueBrain> where sanity is there on where you can click
18:30:02 <TrueBrain> also like trac, or redmine for that matter
18:30:48 <TrueBrain> but walking diffs in a clear and visible way is the most important of all
18:33:38 <Xaroth> svn_diff is harder than i'd thought
18:34:12 <TrueBrain> there ar elibs who take care of most things
18:41:19 <Xaroth> it's more a matter on how to get a proper diff
18:41:36 <Xaroth> as you need to specify a start revision
18:41:43 <Xaroth> which can't be < revision_file_added
18:41:49 <Xaroth> or > revision_file_removed
18:44:05 <TrueBrain> good luck .. enjoy subversion
18:45:17 <glx> so emu_GUI_Viewport() should be emu_GUI_Widget_Viewport_Handler() I think
18:45:52 <TrueBrain> lets keep those renames to C-ifying :)
18:45:58 <TrueBrain> I renamed the whole shit 3 times today :p
18:46:31 <TrueBrain> then when you are done, you have a very good overview what it does exactly :)
18:46:37 <TrueBrain> then I guess we can find a really good name ;)
18:46:57 <TrueBrain> as I wonder: does it only handle mouse, or also keyboard?
18:50:24 <Xaroth> websvn actually does those checks as well :o
19:00:22 <TrueBrain> glx: btw, I am working on GUI_Widget_Draw atm
19:22:05 *** Alberth has joined #openDune
19:24:56 *** ChanServ sets mode: +v Yexo_
19:26:41 <TrueBrain> okay, this is weird ... static array with jump labels ..
19:27:41 <TrueBrain> - emu_bx = emu_get_memory16(emu_ss, emu_bp, -0xE);
19:27:42 <TrueBrain> - emu_decw(&emu_bx);
19:27:47 <TrueBrain> - if (emu_bx <= 0x5) goto l__08A7;
19:27:59 <TrueBrain> so for bp-0xE valid values are 1 .. 5, right?
19:28:52 <TrueBrain> 949 is never called
19:29:08 <Alberth> I should learn to read such code :)
19:29:25 <TrueBrain> you should, and help us out ;)
19:30:19 <Alberth> a 80386 cpu manual would be the first stop, right?
19:30:26 <TrueBrain> that, or just the code
19:30:39 <TrueBrain> assembly is very simplified code
19:30:47 <TrueBrain> (even more the 16bit part of the 386 instruction set)
19:30:59 <TrueBrain> you need a general knowledge about stack
19:31:02 <Alberth> i probably pays off to have a semantics for ss, bp, bx, and such
19:31:17 <TrueBrain> and about how registers are used
19:31:33 <TrueBrain> (ss+sp is stackpointer, ss+bp is framepointer, ...)
19:31:35 <Alberth> I did do 6502, 68000, and ARM2 assembly language programming
19:31:45 <TrueBrain> so you should know how a stack works :)
19:32:04 <TrueBrain> so look at decompiled/cs__1A34.c, emu_Unit_SetDestination
19:32:12 <TrueBrain> you should be able to figure out what that does without too much effort I guess
19:33:49 <Alberth> 80x86 always scared me with its weird overlapping segments ;)
19:34:15 <TrueBrain> still there, but you won't notice it that much :)
19:35:17 <Alberth> first a bit of reading about registers, I think
19:36:02 <TrueBrain> [abcd]x general registers, di/si mostly too, but needs to be preserved over function calls, bp = framepointer, sp = stackpointer, cs:ip is current instruction, and I believe that are all of them :p
19:37:11 * Alberth prefers a nice flat 32b memory space :)
19:40:07 <Xaroth> Alberth: if I can do it, you can too :P
19:44:33 <Alberth> about 'can' I am pretty much convinced :p
19:46:31 <Xaroth> If I can understand parts of it, you can understand all of it
19:51:21 * Alberth considers the idea that 'part' may be equal to the empty set, and discards that idea.
19:51:59 <Alberth> sorry, doing quite mathematical algorithm coding at work atm :)
19:55:01 <Xaroth> svn_log is quite.. useful
19:55:18 <TrueBrain> wasn't my question, but tnx :p
19:55:29 <Xaroth> Hover is floating, Hoover is .. a dam?
19:56:19 <Alberth> "Only Hoover brand vacuum cleaners deliver superior cleaning performance every time." :)
19:58:10 <Alberth> Xaroth: I am getting more and more annoyed with svn to be honest.
19:58:46 <TrueBrain> wrt = With Respect To
19:58:53 <TrueBrain> for the people who do not sms that much (like me)
19:59:28 <Xaroth> Regard/Respect/Reference
19:59:36 <Alberth> crappy merging, ignoring my changes without telling me, sub-dir behavior
19:59:51 <Xaroth> never tried merging :P
20:00:27 <TrueBrain> Subversion is not the best, no :)
20:02:41 <Alberth> it is still a leap forwards from CVS :)
20:02:56 <TrueBrain> git is not the next step
20:02:59 <TrueBrain> Mercurial is still so-so
20:03:04 <TrueBrain> and I still need to test Bazaar
20:03:13 <TrueBrain> it is nice, but it has its problems
20:03:21 <TrueBrain> (mostly related to his age)
20:05:06 <Alberth> git seems just way too complicated (hear-say though, never used it up to now)
20:05:21 <TrueBrain> git is overcomplicated, yes, but also very very kernel orientated
20:05:26 <TrueBrain> with other words: you need A LOT of knowledge to use it
20:05:59 <Alberth> bazaar is simple in idea, just a repo with a working dir
20:06:13 <Alberth> for branching, you make a clone eg
20:07:01 <TrueBrain> uint16 positionRight = positionLeft + w->width - 1;
20:07:07 <Alberth> and then they started hacking around the problems with light-weight clones etc. At that point I switched to hg, and stopped with bzr
20:07:09 <TrueBrain> emu_push(positionLeft + w->width - 1);
20:08:10 <TrueBrain> wow .... my C-ification is correct!! :o :o :o
20:08:22 <Alberth> it was compiled code from C right?
20:08:32 <TrueBrain> ignore 01F7, and yes
20:08:41 <TrueBrain> those are 'libs', given by Borland C compiler
20:08:51 <TrueBrain> so they violate the framepointer from time to time, and shit like that
20:09:31 <Alberth> they had a very bad optimizer then :p
20:11:23 <glx> and we can thanks that ;)
20:17:50 <TrueBrain> Alberth: feel free to say: no optimization
20:17:54 <TrueBrain> and yes, we are very happy with that ;)
20:37:33 <Xaroth> woop, pagination seems to work.. ish :o
20:37:46 <Xaroth> and even, by default, configurable amount of commits per page
20:37:49 <TrueBrain> and I made the next function 'static' :) I love doing that ;)
20:38:03 <Xaroth> this thing of mine is already smarter than websvn :P
20:38:36 <TrueBrain> we already did 8352 lines in C ...
20:38:45 <TrueBrain> just 101000 more to go :p :p
20:39:32 <TrueBrain> haha: 101010 more to go even :p
20:39:39 <TrueBrain> Total Physical Source Lines of Code (SLOC) = 101,010
20:39:41 <TrueBrain> what are the odds ;)
20:42:35 <TrueBrain> that is: it doesn't decompile one function now, but it does one it shouldn't :p
20:47:04 <DorpsGek> SVN: truebrain (r781) -Add: C-ified GUI_Widget_Draw()
20:47:11 <TrueBrain> and there is another blob :)
20:47:25 <TrueBrain> and now I just need to know what is in type '5' widgets
20:50:19 <TrueBrain> I hope not too bad?
20:51:05 <glx> no your function and the function I work on
20:51:20 <TrueBrain> which function were you working on?
20:51:37 <glx> void GUI_Widget_Viewport(Widget *w)
20:51:50 <TrueBrain> how did that conflict?
20:52:02 <glx> both at the end of the file
20:54:24 <DorpsGek> SVN: truebrain (r782) [JIT] -Add: mapped another 4 functions (missing widget type)
20:56:17 <Xaroth> You almost confused me there, TrueBrain :P
20:56:24 <Xaroth> I was testing the svn log thing i made
20:56:31 <Xaroth> and suddenly the revision changed
20:56:45 <Xaroth> then I realised the message changed as well in the debug window :P
20:58:37 <Xaroth> what amazes me more is that it's working quite fast even though the repo is not on my server
20:59:44 <Xaroth> I'll refrain from doing too much work on the design for now
21:01:24 <TrueBrain> k ... code-deduplicated a lot
21:02:36 <Xaroth> do you use the actual repo browser a lot (as in, browsing the directory structure on websvn)
21:03:24 <Xaroth> k, then I'll work on that last :)
21:07:39 <glx> more important is diff and prev/next :)
21:08:27 <Xaroth> or for pages, in general
21:08:28 <TrueBrain> yes, from rev to rev+1
21:08:34 <Xaroth> cuz pagination got a prev/next button :P
21:08:36 <TrueBrain> that is what you do most
21:08:47 <TrueBrain> when you wake up, you go to svn.opendune.org, select the latest rev you have seen
21:08:51 <TrueBrain> and then want to walk all the revs
21:10:10 <TrueBrain> I think that is _the_ primary function how we use it
21:11:00 <DorpsGek> SVN: truebrain (r783) -Add (r782): missing widget type, and deduplicate code (which Dune2 implemented via slightly different ways)
21:11:20 <TrueBrain> now it ismuch more clear to me what the difference between those types is
21:11:28 <TrueBrain> image, rectangle, text, custom proc, ...
21:11:33 <Xaroth> right, that should actually be doable.
21:12:07 <TrueBrain> only widget type 3 is weird
21:12:13 <TrueBrain> as it only pushes the 'cs' of the drawProc
21:12:18 <TrueBrain> feels like it is abused ...
21:12:19 <Xaroth> where's the trac page of ottd?
21:12:44 <TrueBrain> drawProc can be a pointer to text, to a proc, to something else, and the .cs field is abused
21:12:50 <TrueBrain> so it is just a general 'userdata' field
21:13:00 <TrueBrain> (to translate nicely to C)
21:13:06 <Xaroth> oh, that should definetely be doable.. I think...
21:14:02 <Xaroth> it's doable by just svn_diff .. but that'll be messy .. or with multiple svn_diff's combined with a svn_log
21:14:55 <TrueBrain> just as long as you show 1 rev per page ;)
21:17:16 <Xaroth> I definitely need to borrow the bug-linking-code at some point
21:17:19 <Xaroth> at some later point, probably
21:18:04 <DorpsGek> SVN: truebrain (r784) -Fix: too much copy/paste
21:20:50 *** Xaroth sets mode: +vvvv tneo SmatZ proyvind planetmaker
21:20:51 *** Xaroth sets mode: +v Alberth
21:23:07 <DorpsGek> SVN: truebrain (r785) -Add: named the draw modes via an enum
21:23:21 <TrueBrain> just 3 unknown modes
21:25:09 <DorpsGek> SVN: truebrain (r786) -Add: named one more mode
21:32:51 <glx> hmm 2DCE seems to be a big data segment
21:40:53 <Alberth> ok, enough random modifications for tonight
21:49:40 <Xaroth> TrueBrain: any preference to changelog ordering of files?
21:50:15 <TrueBrain> Xaroth: how do you mean?
21:50:44 <Xaroth> grouped per action (Add/Delete/Modify)
21:52:49 <Xaroth> svn log shows dirnames as added as well :o
21:53:55 <Xaroth> fuck, crashed firefox :P
21:56:16 <TrueBrain> hmm .. I am so tempted to buy a 150 euro xbox, to jtag hack him, so I can run custom software on it
21:56:18 <TrueBrain> just because I can ...
21:57:35 <TrueBrain> but that means I can play all the XBox Live Arcade games!!! :) :)
21:58:43 *** proyvind_ has joined #openDune
22:00:32 <TrueBrain> no need; don't worry about that
22:01:35 <Xaroth> definetely need to restrict that
22:06:37 <Xaroth> TrueBrain: Deleted files yes or no content?
22:06:45 <Xaroth> new files quite obvious it does
22:07:19 <Xaroth> (glx's oppinion is also welcome ofcourse)
22:08:11 <glx> deleted file should be full of -
22:08:26 <Xaroth> yes, but does those need to be displayed :P
22:08:57 <glx> nice to compare C-ified function when the source file is deleted
22:13:06 <Xaroth> what I'm worried about is that when lots of files have been modified the thing will go apeshit due to too much data
22:14:28 <TrueBrain> it should display deleted files as '-', yes
22:15:01 <TrueBrain> okay .. I like what I have done with the place (Widget_Draw)
22:15:06 <TrueBrain> clear, to the point, and spot-on :)
22:23:19 <TrueBrain> (or is that bragging?)
22:23:57 <Xaroth> it's not bragging if it's the truth
22:39:28 <TrueBrain> I really wonder what f__2B6C_0197_00CE_4D32 and f__2B6C_0292_0028_3AD7 do .. it has to do with clickable regions ..
23:06:39 <Xaroth> instead of parsing what comes after @@ index index @@
23:06:43 <Xaroth> it parsed what came before it
23:06:47 <Xaroth> which is.. not what I wanted
23:16:59 <Xaroth> just means I can parse the generated diff
23:17:43 <Xaroth> it's being a bit slow tho, but that might be because it's trying to load the entire diff of r773
23:18:04 <TrueBrain> rsync the svn to your local disk
23:18:19 <TrueBrain> (just make sure to rsync in the right direction :p)
23:18:40 <Xaroth> first time i actually have a use for rsync :P
23:19:32 <Xaroth> ugh, i'll do it the old fashion tar way :P
23:19:42 <Xaroth> cba installing certificates on this machine
23:19:43 <TrueBrain> rsync -av truebrain@secure.opendune.org:/var/repos/svn svn
23:22:03 <Xaroth> this... will take a while
23:23:46 <Xaroth> (Couldn't open a repository) Unable to open an ra_local session to URL 180001 (Couldn't open a repository) Unable to open repository 'file:///var/repos/svn/opendune'
23:24:00 <Xaroth> (yes, i was actually lazy enough to keep the directory structure :P )
23:24:43 <TrueBrain> make sure the dir really exists
23:24:49 <TrueBrain> and that it isn't /var/repos/svn/svn/opendune
23:25:00 <Xaroth> it's opendune/opendune
23:25:10 <TrueBrain> I am a freaking mind reader!
23:25:22 <TrueBrain> (I just love it when I am spot-on ;))
23:26:21 <Xaroth> ... my girlfriend now 'owns' the repo on my server
23:27:00 <TrueBrain> but nothing a chmod can't solve :p
23:27:14 <Xaroth> on a 'slowish' machine
23:27:36 <TrueBrain> you know ... I have nothing to do tomorrow
23:27:39 <TrueBrain> what a lovely feeling
23:27:50 <Xaroth> last day before the weekend for me tomorrow
23:28:01 <Xaroth> 1.2ghz athlon with 512 ram :o
23:28:04 <Xaroth> actually faster than I thought
23:28:32 <TrueBrain> why wouldn't it be?
23:28:53 <Xaroth> might need a bit optimization tho
23:29:37 <Xaroth> 3.12 by removing a substr :O
23:30:17 <TrueBrain> owh, displaying the page took 3.5 seconds
23:30:36 <Xaroth> well, parsing the entire thing
23:30:56 <TrueBrain> then 3 seconds is A LOT
23:31:04 <TrueBrain> as it is a relative small commit
23:31:10 <Xaroth> websvn takes a second or two as well
23:31:28 <TrueBrain> takes 0.7s here to fetch the data
23:31:33 <TrueBrain> so I would expect 1 second
23:31:54 <TrueBrain> 900 yes, including headers
23:32:05 <Xaroth> it sais quite a bit more here :P
23:32:18 <TrueBrain> then you are not diffing r773 against r772
23:32:22 <Xaroth> so something is going wrong
23:32:53 <TrueBrain> the diff between 700 and 773 is smaller
23:33:49 <TrueBrain> I have no clue what you are doing :p
23:35:02 <Xaroth> something's going wrong in the calling
23:35:04 <TrueBrain> that should contain MUCH more lines
23:35:15 <TrueBrain> '0' btw is no revision
23:35:29 <TrueBrain> 149797 lines, it should give
23:35:35 <TrueBrain> well, that is including headers, but still
23:36:23 <TrueBrain> 700:HEAD gives about that amount of lines
23:37:07 <TrueBrain> patric@smallbrain:/prog/opendune/trunk$ svn diff -r 704:HEAD | wc -l
23:37:10 <TrueBrain> patric@smallbrain:/prog/opendune/trunk$ svn diff -r 703:HEAD | wc -l
23:37:20 <Xaroth> 1.8 seconds now, with 10000 lines.. still going wrong
23:38:04 <Xaroth> This page was created in 0.181144952774 seconds
23:38:20 <TrueBrain> that is more like it
23:38:27 <TrueBrain> 181ms .. for PHP this is 'normal' I guess
23:38:36 <TrueBrain> but it is in the range of acceptable
23:38:40 <TrueBrain> and nothing a cache can't solve
23:38:54 <Xaroth> I wonder if smarty's cache can cope with it
23:39:02 <TrueBrain> if your URL is good enough, it will
23:39:15 <TrueBrain> I have run 1M hits a day website on a 600MHz, so ...
23:39:42 <Xaroth> 0.2seconds with a var_dump on the returned array
23:39:43 <TrueBrain> you need a good URL scheme, and Smarty can take care of the rest :)
23:40:40 <TrueBrain> now convert to 'easy' readable ;)
23:41:07 <Xaroth> mind, the 'log' template is still wacky as fook
23:41:27 <Xaroth> shows 4 lines, a header, revision+nr, author+name message+message
23:41:36 <Xaroth> then a list of the last 25 commits, in a table
23:41:42 <Xaroth> then a 'pagination' UL
23:41:58 <Xaroth> no formatting what-so-ever :P
23:42:46 <Xaroth> i've not noticed any leaks yet with the pecl svn plugin
23:42:52 <Xaroth> apache running on 1.8% mem
23:43:08 <TrueBrain> and I can't find where widget type 6 is used :p
23:43:27 <Xaroth> maybe in the score screen?
23:44:19 <Xaroth> the 'map' display to select new mission
23:44:39 <Xaroth> er, the secret question textbox?
23:44:49 <TrueBrain> checked checked and checked
23:45:00 <Xaroth> this is like the puzzle
23:45:04 <TrueBrain> wait, maybe the weird box with an error
23:45:07 <Xaroth> there's 1 missing hole, find the missing piece :P
23:45:09 <TrueBrain> how to creat that ...
23:45:24 <Xaroth> the one you get in M1?
23:45:39 <Xaroth> missing1, turn on the help thingies
23:45:42 <TrueBrain> no, when there is a file missing :p
23:46:00 <Xaroth> doubt it'll hit, but still
23:46:10 <Xaroth> it's a frame, image and text
23:46:19 <TrueBrain> I think it is a closed rectangle
23:50:44 <TrueBrain> when I force the call to that function, it seems to do .. nothing :p
23:55:10 <TrueBrain> can't find 3, nor 6 ...
23:55:16 <TrueBrain> hard to figure what they do then :p
23:57:28 <glx> I start to know some global variable "address" by heart
23:57:44 <TrueBrain> haha, I know that problem :p
23:59:54 <TrueBrain> still no widget 3 or 6 ...
23:59:57 <glx> I don't have it in global.h ;)
continue to next day ⏵