00:00:00 --- log: started forth/05.09.15 00:00:22 In the MAF document I referenced a while back, there's an implementation. 00:00:43 maf? 00:01:25 Minimal ANS Forth, a Forth written in Forth. 00:01:37 Ah, yes. The one I downloaded 00:02:42 Can I ask how you started getting into your compoany/developing Quartus Forth etc? 00:03:47 I heard about the Pilot handheld; found out it could be programmed for with relative ease and low barrier-to-entry, and got excited about the idea of programming directly on it. 00:04:20 When was that? 00:04:31 Late '96, I think. 00:04:37 * amca nods 00:04:54 Then found out there wasnt a Forth implementation for Palm and decided to write one? 00:05:24 Right. 00:05:45 Or were there Forth implementations, but not quality ones? 00:05:55 There were none when I began. 00:07:14 Is there any competition now? 00:08:34 There've been a couple of other efforts, only one of which had any backing, but I believe they're both abandoned at this point. Quartus Forth is the only actively-supported Forth for the Palm OS, as far as I'm aware. 00:08:55 * amca nods 00:10:36 --- quit: JasonWoof ("off to bed") 00:10:39 So did you start Quartus in 97? 00:10:56 That sounds about right. 00:15:48 What did you start coding it in? Did you bootstrap it? 00:16:14 I started it in a 68000 cross-assembler. 00:16:21 --- quit: madwork (Read error: 110 (Connection timed out)) 00:16:25 Ah. 00:16:30 My notes say I started it on 1997.7.18. 00:16:51 Hmm. Almost a year after you got your palm 00:16:52 I undoubtedly began earlier than that, research, gathering tools, etc. 00:16:58 I see 00:17:06 How old were you around then? 00:17:16 I'm going from reconstructed memory; I may not have got the device until early-mid '97. 00:17:23 ok 00:17:48 I suppose that would have made me nearly 30, at the time. 00:18:06 Can I ask what you had been doing up till then? 00:18:21 Reading back-issues of the Reader's Digest. 00:18:55 Ah. Sounds like Quartus really improved your life then 00:19:31 I wouldn't go that far. I mean, you've got Quotable Quotes; Word Power; Humor in Uniform; I Was Joe's Lung; that sort of thing. 00:20:07 lol. So Quatus *really* improved your life then ;) 00:20:19 Although I must admit I like reading them too... 00:20:56 When I was a kid I'd pick up back-issues cheap. Always something to read. 00:21:05 * amca nods 00:21:16 Full of info for an info whore like me 00:21:19 and laughs 00:29:35 It's well into the wee hours here; I'm off. 00:30:09 Enjoy your sleep then 00:30:15 Thanks! 00:30:30 :) np 00:46:43 --- quit: thinfu_ (Remote closed the connection) 00:46:43 --- join: thinfu (i=thin@bespin.org) joined #forth 00:46:43 --- mode: ChanServ set +o thinfu 01:19:28 --- quit: virsys (Read error: 110 (Connection timed out)) 01:20:29 --- join: virsys (n=virsys@or-65-40-181-238.dyn.sprint-hsd.net) joined #forth 02:29:52 --- join: amca_ (n=plump@as-bri-4-1-151.ozonline.com.au) joined #forth 02:50:10 --- quit: amca (Read error: 113 (No route to host)) 02:50:22 --- nick: amca_ -> amca 03:16:45 --- quit: amca ("d34d") 04:11:58 --- join: tathi (n=josh@pdpc/supporter/bronze/tathi) joined #forth 04:47:05 --- join: amca (n=plump@as-bri-4-1-175.ozonline.com.au) joined #forth 04:48:43 Hello 04:49:02 Raystm2: are you at work or home? 05:08:11 hi amca 05:08:29 How are ye tathi ? 05:08:36 The thing in asm that I was working on yesterday is a Forth, BTW 05:08:48 I'm still fighting with the assembler, but I think I've got it on the run... :) 05:08:50 How are you? 05:08:58 hehe :) 05:09:02 Not too bad 05:09:07 Got some sleep today :) 05:09:53 What chip you coding it for? 05:10:04 PPC 05:10:21 Ah, yes 05:10:28 currently runs under Linux, or boots under Open Firmware, though no drivers yet. 05:10:32 Is there any code for it online yet? 05:17:59 not really... 05:17:59 Is it something you would like to email? 05:17:59 actually, it's in my SVN repository, so I guess it is online. 05:17:59 Ah. Open to the public? 05:17:59 read access... http://josh.qualdan.com:3/svn/forth/ should work 05:17:59 cool thanks 05:17:59 I wanna have a look at what PPC Forth implementation looks like :) 05:17:59 How is it going? 05:18:54 it works ok. Still pretty minimal though. 05:19:02 I keep tinkering with the internals rather than adding functionality. 05:19:33 Changing how it is structured/refactoring it? 05:19:42 yeah. 05:20:16 This has been my most "normal" forth -- I generally work on it for a bit, then drop it for a while to go experiment with crazier ideas. 05:20:19 --- join: PoppaVic (n=pete@0-1pool46-87.nas30.chicago4.il.us.da.qwest.net) joined #forth 05:20:27 But I keep coming back to it every 6 months or so. 05:20:38 Well that is probably a better thing to be fiddling with internals rather than adding features to a buggy thing 05:20:46 Howdy 05:20:58 Helo again! :) 05:21:00 Oh, I don't know that it was buggy, just that I'm working towards cleaner code. 05:21:00 watt I miss? 05:21:02 Hi PoppaVic 05:21:24 Hi tathi,amca 05:21:35 it's getting to be fairly good (or at least it fits the inside of my head). I can come back to it after 6 months and it still makes sense :) 05:21:43 tathi: Sorry, that's what I meant by "buggy". To me code that is inelegant is buggy >.> 05:21:51 ah. :) 05:22:08 no. Elegance is not all that thrilling 05:22:19 PoppaVic: Maybe not to you 05:22:29 PoppaVic: We are chatting about tathi's PPC asm Forth 05:22:45 Readability, easy and comments where apropos are a lot more useful ;-) 05:23:23 PoppaVic: To me that *is* elegance 05:23:27 Ahh 05:23:55 Moreso than fancy hacks, although I can appreciate the good one's like Duff's Device too 05:24:00 lol 05:24:03 Some folks seem to believe that "elegance" is using the perfect, tightest, fastest routine. 05:24:13 LOL 05:24:22 That is raw power, not elegance 05:24:32 of course.. you can't then port or read it, but: it's "elegant" ;-) 05:25:13 Different things are different elegance to different pplk I suppose 05:25:24 I sometimes see folks advocating recursion that way, too 05:26:02 Recursion can be elegant as long as it is appropriate and adds to readability IMO 05:26:36 yeah. There are places where it fits beautifully; there are other places where a simple loop is just tidier. 05:27:02 yep 05:27:02 otoh, I've never been one to suggest someone recurse right outta' stack. 05:27:17 tathi: I like your air hocky idea :) 05:28:39 amca: yeah, I actually have a table. 05:28:49 but I didn't use self-leveling epoxy, so... 05:28:56 oy 05:28:57 it's still waiting for me to finish sanding it level. 05:29:02 ow 05:29:23 Is self levelling epoxy less viscous than normal epoxy? 05:29:37 I gather the chemistry is rather complicated. 05:29:37 what in the world needed epoxy? 05:30:10 PoppaVic: I built an air-hockey table (mostly out of wood), and I finished the top surface in epoxy 05:30:17 ahh 05:30:24 I see. 05:30:35 amca: in high school I worked as a lab assistant at a place that did epoxies 05:30:42 PoppaVic: He wants a nice laquered finish on the front of his monitor ;) 05:30:48 I still have the 2 cans of nice Hobby-Poxy pain ;-) 05:30:58 tathi: Ah. So you have an idea how complicated it is? 05:31:02 pain/t/ 05:31:06 in the troubleshooting lab. Seemed like half our problems were that some big plant was having trouble with there finishes not leveling quite right. 05:31:33 I think that the problem is that it hardens by forming long chains, so if you don't get it just right it affects the surface tension and you get dimples. 05:31:35 Ah 05:31:42 amca: now-now... Poxy ain't laqueer ;-) 05:31:44 yep 05:32:04 What *is* the difference between the two. Im pretty bad when it comes to paints 05:32:36 I can't recall ever using a lacquer. Enamels, yeah. 05:32:57 I used to recall the diffs.. Been too many years now. 05:33:10 Laquers are transparent, arent they? 05:33:14 yeah...I should finish up the air-hockey table and publish some plans for it. 05:33:21 they can be; don't have to be 05:33:54 Can enamels and epoxies be transparent? 05:34:05 sure, afaik 05:34:10 yeah. 05:34:22 color is always a function of pigments 05:34:38 epoxies tend to be a bit on the yellowish side, but basically transparent. 05:34:49 although, I can recall some epoxies that "clear" was a bit inaccurate on. 05:34:57 what tathi said 05:35:40 tathi: Do you know what the difference between laquer, epoxies and enamels are? 05:35:59 not really. 05:36:13 One of these days I'll have two projects: making gramps gun-cabinet about 2-3" deeper; and building a decent reloading-bench. 05:36:31 I mean, epoxy is a two-part thing (have to mix resin/hardener), the others aren't 05:37:22 * amca nods 05:37:27 Like epoxy glue 05:37:52 tathi: Why did you pick epoxy for the surface instead of ;laquer or enamal? 05:37:53 lacquer: any of various clear or colored synthetic coatings made by dissolving nitrocellulose or other cellulose derivatives together with plasticizers and pigments in a mixture of volitale solvents used to impart a high gloss to a surface. 05:38:26 Oooh! I ought to look at wikip 05:38:43 and...sounds like enamel is a catch-all term for latex or oil-based paint. 05:39:13 yeah, I was afraid they'd do that 05:39:15 but that it originally referred to a finish of fired glass. 05:39:17 So it basically the type of chemicals used as opposed to the paint function? 05:39:40 absolutely 05:40:04 some cultures did better with one or the other; both have slightly diverse uses 05:40:18 Ah 05:40:34 "baked enamel" is a fairly tough coat 05:40:49 ..also used on jewelry and such 05:41:12 I knew about the origin of "enamel" so I wondered why it applied to oil/latex based paints 05:41:41 I would guess they threw up their hands and appropriated it as a descriptive-term. 05:42:26 certainly a high-gloss white latex-enamel looks nice on doorframes. 05:42:57 ..and, perhaps they have other additives that make it act more like a real enamel as it dries? 05:43:38 Is one of the three more weather resistant? I would guess that lacquers would be the most durable 05:43:38 (btw, never EVER buy and use oil-based paint on homes - ewwwww!) 05:43:54 heh 05:43:58 iirc, there are indoor and outdoor latex-enamels 05:44:17 --- join: aardvarx (n=folajimi@shell2.sea5.speakeasy.net) joined #forth 05:44:19 oh crap 05:44:27 Good day. 05:44:41 tathi: the house I once owned: the idiotic family had stolen industrial yellow-oilbase and painted the kitchen and LR. It was a WHORE to repaint. 05:44:51 hi aardvarx ;-) 05:44:59 gday aardvarx 05:44:59 hi PoppaVic. 05:45:06 gday amca 05:45:16 PoppaVic: How come it was a whore to repaint? 05:45:29 PoppaVic: Couldnt just paint over it with water based paints? 05:45:43 amca: you have to use etchants and sanding to get enough 'tooth' for any new paint to adhere. 05:46:10 ah. Nice and glossy then 05:46:17 What's with the 403 on the TWiki? 05:46:41 ..and for all that... Those walls were easier than the one they "papered"... There were like 5 layers.. Some were flocked, some were contact-paper. 05:47:02 aardvarx: haven't looked this AM. Lemme' load up. 05:47:28 lol 05:47:32 shit 05:47:41 hmmm 05:47:47 forbidden? Weird 05:47:56 I guess it's not just me. 05:48:11 ok.. I guess we irked someone. Somehow. 05:48:30 Quartus: yo! you awake? 05:50:46 Maybe his system went down/oral? No clue 05:51:57 --- join: ccfg (i=ccfg@dsl-roigw3de0.dial.inet.fi) joined #forth 05:55:04 Anyway, one of the questions that came up yesterday was whether or not you plan to use an alternative to Make for Metabuilder. Is that true? 05:56:20 To build itself, I've been using GNU make, yeah. However, once built, I'd like to step away from the assorted-makes if possible. the shell-ish thingie or another should act for makes 05:56:39 Specifics, man. Specifics. 05:56:46 Will http://freshmeat.net/articles/view/1715 help? 05:57:28 There are a variety of Make alternatives listed there. 05:57:37 You keep looking for hard-code and a Five-Year Plan, and I'm trying to get things lined up to build tools to build tools ;-) 05:58:59 Decent url - reading 05:59:27 Well, if you're dissatisfied with something, now will be the time to address it. Nothing is written in stone. However, some of the discussions here illustrated why projects start off with Make, but regret the decision later. 05:59:48 Yeah. And, I try to keep makefiles dead-simple. 06:00:50 "An enamel paint is a paint that dries to an especially hard, usually glossy, finish.This is a commonly used, yet fanciful term, implying that an ordinary latex or oil-based paint has the same properties as true, fired vitreous enamel." 06:01:04 amca: that sounds right 06:02:07 I've doesnloaded and seen some of those tools 06:02:14 doWn 06:03:23 This is not a contract; but it is helpful to have this written down so that in the future if questions regarding the chosen approach(es) arise... 06:03:36 you will have something to look back to. 06:03:53 I agree. And, should we ever get the wiki back, we can address these issues 06:05:35 I've already code in place that sweeps thru /usr/include and the "standard headers" list, and checks on all defines. So, some "tools" have had some experiments done. BUT, mind you: not in forth(ish) - just in C. 06:06:15 That's another issue I do not understand. 06:06:27 Where does C end and Forth begin? 06:06:37 After seeing all the insanity that relies on sh, and make, and sed and m4 and - well, I've pretty well concluded we've language rather than speed issues 06:07:03 aardvarx: It depends where you want it to 06:08:05 If you mean "Where do you end C to start the forthish code?" then, when I get a system (lib/api) in place that lets us interact, extend, save and load. At that point, we can begin forthish code-dev. 06:09:18 * aardvarx checks Wikipedia for what an API is... 06:09:54 think headers(structs, types, proto) 06:10:13 then the lib has code/funcs, structs & data 06:10:43 aardvarx: Application Programming Interface. The interface you use to use a code library so you dont have to know about the internals 06:10:46 You have to have the "critical mass" to leave the C for the forthish 06:10:56 eg Windows API for creating windows etc 06:11:51 amca, I *MUST* understand API to be a good programmer!!! 06:11:54 :P 06:12:13 aardvarx: wtf? 06:12:25 he's right ;-) The term is prevalent 06:12:38 --- join: madwork (n=foo@derby.metrics.com) joined #forth 06:12:57 hi mad 06:13:22 --- quit: tathi ("leaving") 06:14:14 amca, I have always heard of APIs, but did not think much of them. From what I am reading now, it seems it behooves me to comprehend the concept behind APIs. 06:14:24 absotively 06:14:50 C-fiends can talk of headers and libs and plugins: but they need to know of "API" to work with them. 06:14:55 Ah, yes. 06:15:04 some other languages may NOT have an equivalent. 06:15:06 aardvarx: I thought you might be being sarcastic or something 06:15:38 nah, he was being sincere ;-) 06:15:41 aardvarx: In face when you code your programs, it is a good idea to create a clean API yourself between the model and the user iinterface 06:15:52 PoppaVic: ;P 06:16:10 exactly - regardless if the "user" is another programmer or a app-user that CAN'T program. 06:17:36 aardvarx: The advantage of having a clean API is that you can use the library for multiple uses and if you want to change how your library is implemented, you can keep the same API. 06:17:58 yep. changes to guts don't need to change the "interface" 06:18:22 aardvarx: Do you know much about different implementations of stacks? 06:19:17 Nope 06:19:40 I just started reading about stack traces two days ago... 06:19:42 Do you know what a linked list is? 06:20:13 amca, I am only slightly familiar with it; I'm willing to learn, though. :) 06:20:40 amca, I *CAN* create an API? 06:20:47 Yes 06:20:52 I thought only Corporations can do that. 06:20:56 no 06:21:14 aardvarx: What langs do you know? 06:21:25 C and QBasic. 06:21:38 I'm learning C++ now. 06:21:38 cool 06:21:46 http://rafb.net/paste/results/exQpP677.html 06:22:26 that's some older code of mine 06:22:30 ..an "API" 06:22:59 Say you have a program that has a number that you want to manipulate in an opque way, so that if you decide to change how the number is stored, you still have the same interface 06:23:17 yep like a CELL or DCELL or QCELL 06:23:39 okay... 06:24:27 that old API I posted would allow any program to "interface" with stacks and all the gymnastics - w/o having to write a pile of support themselves. 06:24:43 What??? 06:24:46 So your interface to the data structure could be "void set_number(int num); int get_number(void);" 06:24:55 That's less than 100 lines! 06:25:03 PoppaVic, are you serious? 06:25:10 the .c file is never seen, it'd be in a lib 06:25:30 aardvarx: That is just the interface, not the implementation. The idea is that the implementation is hidden from the library user 06:25:35 all they do is #include and remember to -labstrack 06:25:37 Okay. 06:26:12 These API's are why I develop in the order which I do. 06:26:28 tools-to-write-tools 06:26:45 amca, are you saying that get_number() is the data structure? 06:26:51 nope 06:26:55 the interface 06:27:05 Okay. 06:27:12 No. get_number is part of the the API to the data structure. 06:27:14 amca - use the term 'CELL' instead of 'int' - it's more abstracted 06:27:32 PoppaVic: Im just giving a very basic example 06:28:06 CELL get(Something *context, offset o); 06:28:20 PoppaVic: Too complicated an example 06:28:23 CELL put(Something *context, offset o); 06:28:28 hehe - yeah, sorry 06:28:39 amca, I see what you're saying. The data structure is contacted by the API, not directly by the user. 06:28:39 Good form, but :) 06:28:43 Right? 06:28:44 yes 06:28:49 should we throw in vtables now? ;-) 06:28:55 lol No! 06:29:18 aardvarx: Ill rafb it it will be easier to understand 06:29:31 aardvarx: the datastruct can be opaque, or abstract, or literal (dunno the proper term there) 06:30:01 amca, okay. 06:30:02 aardvarx: have you used FILE* and handles? 06:30:21 PoppaVic, is FILE a data type? 06:30:28 If not, no. 06:30:41 a FILE* is intended to be opaque. A handle is abstract 06:31:12 I guess this has nothing to do with fopen() or fclose() 06:31:18 C programmers are NOT supposed to dick around within what either represent, w/o going trhu the provided API 06:31:27 aardvarx: sure it does 06:32:47 In fact, folks that DO get inside the FILE struct (via the *) and tinker around, soon discover that some other machine will blow up - because only the system is supposed to KNOW about the guts. And they can vary across platforms. 06:33:58 In fact, the only reason folks always use to pick one system over the other (open vs fopen), is the API's for either are not reflective. 06:34:33 PoppaVic: fopen is higherlavel IO than open as well 06:34:43 yep, and I use it almost exclusively 06:34:54 the api's are just not mirrored 06:35:06 Say for example you had this code. number.h is the interface, or API - what users of the library know about, and number.c is theimplementation - what users of the library dont (or shouldnt) know. http://rafb.net/paste/results/vrWMRc15.html 06:35:20 aardvarx: That last one was for you btw 06:36:16 should do "static int _number=0;" and a leading _ is reserved 06:36:45 is it? I thought it was 2 leading _'s? Doh 06:36:54 What should I use instead to indicate privateness? 06:36:55 the 'static' tells the compiler to not add the term to a symbol-table 06:37:21 soo, no outside code can 'find' the var 06:37:37 aardvarx: Can you undertand that code? 06:38:12 PoppaVic: What is a better prefix to use then? 06:38:17 amca, just a sec. 06:38:34 I'd use some 'namespace' suggestion... like example_ 06:38:59 Ah, yes 06:39:04 (and recall that some prefixes like that are ALSO "reserved", the bastards) 06:39:22 O_O 06:39:30 Who reserved them? GNU? 06:39:40 This is another fine reason for our vm/engine ;-) 06:39:55 amca: GNU maybe, but the committees/stds anyway 06:40:02 PoppaVic, I need to write that down. 06:40:10 Could you please clarify it? 06:40:16 It took me weeks to run them all down. 06:40:27 amca, number.c is hidden, right? 06:40:31 Well it isnt in the 89 std I dont htink. Is it in the 99 one? 06:40:43 stuff like 'E' prefix is reserved for errors; "mem" and "str" are reserved, etc. 06:40:59 aardvarx: The user will see number.c as a compiled library, like number.o or number.a 06:41:04 amca: could be - all I know is I ran into an abundance of irritations 06:41:23 PoppaVic: With -Wall under gcc? 06:41:39 aardvarx: And the .h file too of course 06:41:47 I went thru the web and wrote tests and such, I'd have to look around for the code and notes 06:44:03 amca, so would I do '#include ' in my C-code? 06:44:17 Im pasting an example program now 06:44:24 right 06:44:34 http://rafb.net/paste/results/zMaZq384.html 06:44:34 and, you'd link the object/lib to gcc 06:45:25 PoppaVic: Any suggestions about the coding style? What order should I have had the includes in? 06:45:39 for your example? 06:45:46 yep 06:46:07 aardvarx: Does that look clear? Any Qs about it? 06:46:16 I always include the systems-stuff first, then my own stuff (possibly installed), and the "local.h" stuff 06:46:21 Yeah, that was very helpful :) 06:46:25 cool 06:46:32 Now, the important part 06:47:01 the exception to my above would be a possible LEADING #include "config.h" 06:47:11 Say you have discovered a bug in your code (like PoppaVic pointed out) and you also want to change the implementation of your library 06:47:30 * PoppaVic is in the corner and watching 06:47:40 With a well written API, you could use it the same and not worry about the implementation details. 06:47:49 eg http://rafb.net/paste/results/hwaJmp90.html 06:48:11 You would still use the API functions and number.h exactly the same way 06:48:54 PoppaVic: you would name the library libnumber.a wouldnt you? 06:49:32 ahh, usually. For a static-lib .a - the "lib" prefix is always used, and never mentioned at -l 06:49:49 * aardvarx compares the header files for differences... 06:50:03 * amca nods 06:50:34 So you would use gcc -l number myprogram.c -o myprogram 06:50:40 aardvarx: ahhh - the "interface" (API) remained the same ;-) 06:51:50 Two things: 06:52:02 But the internals are different - the float being used instead of an int. Of course this is a trivial example, but it shows the basis behind APIs 06:52:04 go on 06:52:12 well, I'm sorta' adamant that it should always be akin to " [-o target] [] [-llibraries] 06:52:45 (naturally above MIGHT change a bit because order can be significant) 06:53:00 First, is "number.c" the data structure? I'm used to having to use 'struct' to define a structure... 06:53:24 ahhh 06:53:29 Second, are you just casting int to float and back? 06:53:35 aardvarx: No. 'int _number'/'float p_number' is the data structure in this case 06:53:35 number.c is the source-code. 06:54:11 number.c might be considered the never-seen ABI 06:54:17 I see. 06:54:51 But in this case there can be only one data structure of this type in the program. It would be possible to have it set up so that you could have multiple ones in your api, usually by using handles etc. Do you want to see an example of that? 06:55:05 * aardvarx nods enthusiastically. 06:55:19 Yes! 06:55:23 Okay. PoppaVic will have to debug it, cause I hadnt thought of this yet :) 06:55:37 To your other Q 06:55:38 I have to interfere for a moment... Please remember C++ is not remotely C - it's a language of it's own. And, the whole 'classes' and inheritance/methods mess is a real mess to begin from. 06:56:38 now, while I am trying to beat the fovm_ struct: what did you want me to write/post? 06:56:49 In the first implementation , the library stores the data as an integer, and in the second one it stores the number represented at a floating point number. I use casts (int) and (float) to convert them back and forth 06:57:13 PoppaVic: Just want you to double check my new implementation 06:57:19 and API 06:57:20 and in both cases amca mentions: the users and API haven't a clue HOW the data is maintained internally. 06:57:48 I did check - looked ok. Maybe a notch too trivial for him? Maybe a struct & union example? 06:58:02 PoppaVic: No, I havent written it yet 06:58:24 So, number.h allows access to the set_number() and get_number() functions? 06:58:35 oh. OK. lemme know when ready, I'll go back to my corner so aardvarx doesn't get confused. 06:58:47 aardvarx: specifies HOW 06:59:28 PoppaVic, but if I don't include the header file, I can't use the functions, right? 06:59:48 welllll...... Let's say you cannot, just to make you write good code 07:00:03 Okay. 07:00:37 as a very good rule: NEVER use funcs/vars that w/o including the proper headers 07:01:06 Fair enough. 07:01:11 using "gcc -W -Wall ..." is a very good way to do CYA. 07:01:22 (cover yer ass) 07:01:29 I'm already in the habit of "gcc -Wall -pedantic -ansi ..." 07:01:50 Wait a minute... you're using -W and -Wall concurrently... isn't that redundant? 07:02:21 PoppaVic: What header is malloc in? 07:03:08 stdlib.h 07:03:32 always "man foo" and heed the includes if unsure 07:03:52 aardvarx: no, it just triggers some extra noise 07:04:20 yeah, some folks like -pedantic, I use it rarely - as a self-check. 07:04:53 "-ansi" specifies defines affecting includes and code and extra screams. 07:06:03 You need to understand: I've been at this for morethan 20 years, but never paid or got a degree in coding. So, some of my habits and suggestions are not "purist". 07:06:36 PoppaVic: Shock! Horro! ;) 07:06:42 (Horror) 07:09:50 Is it safe to say the code is "clean" if it passes all of those flags? 07:10:21 syntax error free, yes. Not bugfree 07:10:31 I see. 07:22:38 aardvarx: still there? 07:22:38 Yes. 07:22:38 --- join: virl (n=hmpf@chello062178085149.1.12.vie.surfer.at) joined #forth 07:22:38 aardvarx: We are still working on the new API - havent coded in C for a while. :) 07:22:38 Should be ready soon 07:22:38 TYT 07:28:37 he's about ready to unveil ;-) 07:28:56 http://rafb.net/paste/results/9qarde90.html 07:29:06 oops wrong window 07:29:14 loading 07:29:16 that was for PoppaVic 07:29:29