Squeak-dev March 1999

squeak-dev@lists.squeakfoundation.org

153 participants
338 discussions

RE: VM cleanup (was Re: [VM] NCM performance report)
by Raab, Andreas 27 Mar '99

27 Mar '99

Andrew, > If this is already in place and I missed it, mea culpa. > The 'mea culpa' is accepted - everything you were talking about is in there ;-) The addresses are *of course* cached - the OS dependent lookup gets only invoked if the function has not been loaded before. > But without > it, let's leave critical innerloop-style primitives inside the VM > until we have polished and refined the primitive plugin process > somewhat; or else Tim's project (one that ought to be undertaken) may > lead to a substantial degradation of performance. > The overhead for invoking a (cached) named primitive can be substantial if we're talking about stuff like Object>>at: or Object>>at:put: but given that the primitive takes more than say 50 micro seconds the overhead shouldn't be noticable at all. What happens when invoking a named primitive is: * Fetch the first literal of the method * Make sure it is an array of the right size * Load the session id and the function address from the array * If function address ~~ 0 and session id == current session: Call the primitive * Otherwise: do the slow lookup and everything else [Just as a side note: The entire Balloon stuff uses named primitives and I don't think that you would notice a measurable difference when switching to indexed prims] > Some thought ought to go to plugin library naming conventions and > version-testing conventions as well. > I'm not sure what you understand by plugin naming conventions, but there is a sort-of versioning system for the plugins: Whenever the VM attempts to load the plugin it calls the plugins 'setInterpreter' function with a version number and a vector of VM functions. The major version number of the VM must match the major version with which the plugin was compiled and the minor version must be at least of the minor version with which the plugin was compiled. There is a unique list of exported functions from the VM (btw, contained in 'sqVirtualMachine.h') and the plugin is linked against these settings. Note, however, that plugins can also be used to simply structure the VM. A Squeak plugin can be added when the Interpreter sources are generated (see Interpreter>>translate:doInlining:) and I actually understand Tim more like he want's to have a simple way of getting rid of parts of the VM by just excluding these guys from the list of plugins. Andreas -- PLEASE NOTE NEW EMAIL ADRESS! +===== Andreas Raab ========= (andreasr(a)wdi.disney.com) ==+ | Walt Disney Imagineering Phone: +1 818 544 5016 I I Glendale, CA Fax: +1 818 544 4544 I +======< http://isgwww.cs.uni-magdeburg.de/~raab >========+ > ---------- > From: Andrew C. Greenberg > Reply To: squeak(a)cs.uiuc.edu > Sent: Friday, March 26, 1999 5:06 PM > To: squeak(a)cs.uiuc.edu > Cc: recipient list not shown > Subject: Re: VM cleanup (was Re: [VM] NCM performance report) > > > Dan's worries about image size in the recent NCM thread reminded me to > ask a > > question I had almost let slip out of my mind:- > > when/who/how is the VM going to get cleaned out; by which I mean to ask > if > > anyone is actively pursuing moving many of the 'extra' prims out into > Plugin- > > land thus making a minimum VM much more easy to produce? > > . . . > > > Along with minimum images, headlessness and loadable > > partial-images, a small VM > > might make a practical scripting utility possible. It would > > certainly load and > > startup more rapidly. > > > > So, is anyone already doing this? > > Before assigning that task, let's see if we can't tune the primitive > plugins somewhat better first. I haven't studied the code for > awhile, but as I recall, the present primitive plugin (at least for > the Mac) performs a table lookup of the plugin name in the module > tables with EACH AND EVERY CALL to the primitive. Accordingly, > primitives deep in an inner loop can face a substantial overhead just > from the primitive calls. > > So some sort of caching arrangement needs to be accomplished to assure: > > (1) that modules are loaded exactly once for each session in > which they are used (but the caches must be cleared on image startup > shutdowns; > > (2) that each primitive is looked up exactly once for each > session in which it is used (likewise, with caches cleared on image > startup/shutdowns; and > > (3) that the overhead of this protocol isn't itself large in > comparison with the overhead of present-day primitive calls once the > module is loaded and the primitive has been called at least once. > > If this is already in place and I missed it, mea culpa. But without > it, let's leave critical innerloop-style primitives inside the VM > until we have polished and refined the primitive plugin process > somewhat; or else Tim's project (one that ought to be undertaken) may > lead to a substantial degradation of performance. > > Some thought ought to go to plugin library naming conventions and > version-testing conventions as well. >

2 1

Re: PDAgui
by Edward P Luwish 27 Mar '99

27 Mar '99

Tim - I was the guy who sent you the question about psion5/linux & Squeak the other night - I was inspired to correspond with you because of recent activity on the psion/linux mailing list from Alexander and others regarding a GUI. I have just joined the squeak list, so I guess there is no return now. I am well aware that Squeak is not a GUI in the same way that X or Windows or the Macintosh desktop is, having used Squeak and Smalltalk for some time now. It did appeal to me as a substitute, on a PDA, for the functions of the graphic interface, windowing system and PDA applications all rolled into one. In that respect, the size of the system is not as relevant, since it includes far more functionality. I am committed to experimenting with Squeak on the Psion 5 regardless of the direction of mainstream development of Linux on the Psion. As I mentioned in my private email to you, I would be interested in an EPOC32 version as well. I do have all the development tools from Symbian (and, ecch, Microsoft) to do this, but until now only lacked a worthy project. The only major issues in adapting existing C code to the EPOC32 environment is that it must be jacketed in C++ and that ALL memory must be managed through the kernel, and all code must be ROMable (meaning, for example, that no static or toplevel variables are allowed anywhere, unless they are constants initialized at compile time). The interpreter would have to be hand-edited to deal with this restriction, and later I would want to modify the interpreter generator in Squeak so as not to have to do this for every release. Sorry to disappoint you, but I don't think the latest SDK from Symbian is divorced from Windows or Microsoft Visual C++. The recommended method of development is still to compile your application under VC++ and run/debug it on the included EPOC32 simulator under Windows. Then you recompile for the ARM processor using the GNU compiler suite included on the SDK and download it to the Psion, or Geofox, or Ericsson GSM telephone, whatever. Because of this methodology, Symbian did not develop native debugging tools like gdb that would allow for development without Windows directly on the EPOC32 target system. The SDK does include source code for many, but not all, the ARM development tools, which run in the Cygwin (UNIX on Windows) environment, so they can be built to run on Linux. Because I can swallow my rising gorge and use Windows if I absolutely have to, I can take some of the unpleasant work off your hands - I will probably bother you with some hopefully intelligent questions from time to time. My goal is to have a Smalltalk system on a PDA complete with classes replacing the PDA apps, and hooks to interface with existing apps that can not be immediately replaced. The final result is a PDA that can be instructed to adapt to the user rather than the other way around. Ultimately, it would be nice to replace the ROM in my Psion with a direct-to-the-hardware Smalltalk system. Ed

2 1

Re: Steps to Modularity - Incremental Snapshots
by jecel＠lsi.usp.br 26 Mar '99

26 Mar '99

Congratulations Dan and Ted!! I have always felt that the monolithic image has been the number one reason why Smalltalk has failed to reach the masses (and Java just reinforced this idea). I would just like to comment that "single root" segments worry me somewhat. It is nearly impossible to avoid the "leaking of internal pointers" unless you limit very strictly how segments may be used. If you have a whole bunch of objects that are pointed to from the rest of the image via a single local root object, all it takes is one message to that root that causes it to return a pointer to one of its children and having that pointer stored in an instance variable of a non-segment object to ruin the single rooted structure: myInstVariable := singleRoot children first. If a little care is taken in synchronizing the GC, then storing that internal pointer in a temporary variable won't cause the same problem. The write barrier in the GC (ObjectMemory>>storePointer:intoObject: withValue:) could be patched to detect this internal pointer leak, but doing so effeciently wouldn't be easy at all. -- Jecel

2 1

Re: Multilingual Squeak
by Russell Allen 26 Mar '99

26 Mar '99

Hi > >It wouldn't be hard to add a translator to translate everything from one >language to another with some control/help by the user. > >Ivan Tomek, > If you have a look at the babelfish translator that altavista uses (http://babelfish.alstavista.com) you will see the limits and possibilities of this approach... In Squeak the strict "Object message: Object" grammer certainly simplifies the parsing of Squeak sentences. However, a translating browser would still have to cope with translating all of the symbols in the system (class names, messages, temp variables etc), and would be better if it could make at least a credible stab at translating comments by the user. You'd have to (at a minimum) cope with noun phrases such as anOrderedCollection FormInspectView ShortIntegerArray as well as verbs in various forms: scanFor: addSubclass: isVariable atPutAll: You probably know more about how to approach this then I do; but it would be (as far as I can see) quite difficult to do, requiring a fairly complex NLP engine. One of the problems would be that many of the symbols are not "gramatically correct", but are paraphrases/simplifications of longer phrases... However, it would be a great solution to the problem of a multi-language image - browse and make changes in whatever language you choose! :) Russell ---------------------------------------- Russell Allen russell.allen(a)firebirdmedia.com ----------------------------------------

3 3

Re: Multilingual Squeak
by jecel＠lsi.usp.br 26 Mar '99

26 Mar '99

oachim Durchholz wrote: > > Mark Wai wrote: > > Pre-mature conclusion and always follow the given can only lead to > > missed opportunities to make this a better world. > > I don't think my conclusions are premature. I have seen several tries at > this one. Apple Computer translated the keywords of a scripting > language, as did Microsoft. Both failed miserably; the experiment was > cancelled almost instantly once it hit the street (i.e. within a year's > notice). They failed exactly because they gave up too soon! Back in the Fortran/Basic/Clipper days there were lots of people here in Brazil who programmed even though they didn't know much English. That is no longer possible with VisualBasic, C++/MFC and much less in Squeak. So programmers aren't very representative of the general populations, just as this list's audience is obviously composed of only English speaking people. Given that, it should be no surprise that non-English scripting languages weren't well received. I hate it when a friend asks me to see something in his machine and he is running a Portuguese version of Windows or of the applications. Not that I don't know Portuguese :-), but I am used to the English versions and trying to find things in a different version is a major waste of time. So if you were to ask me, I would say I prefer Win98 in English. If they were to listen to me they would lose a lot of sales since I am not representative of most potential customers. To be a programmer today you have to know English - of course if you ask those people they won't be interested in a multilingual version of anything. And if you do introduce the product anyway (fixing the other problems Joachim Durchholz mentioned) it will take a long time for people interested in programming but not doing it due to their lack of English to notice that things have changed and to take advantage of it. So being stunned by the initial reaction and giving up in less than a year doesn't prove at all that the idea has no merit. -- Jecel

1 0

released: correspondents networking framework 1a2
by Craig Latta 26 Mar '99

26 Mar '99

Hi-- A new release of the correspondents networking framework is available. This one is one alpha 2. The socket support actually works! It's missing complete serial port and MIDI support (I'll integrate from existing sources), a few situations are undefined, a few non-critical things are undeclared, and I'm writing more examples. As you may recall, the correspondents framework handles much of the busywork in writing new clients and servers. It includes a NetStream class whose instances can stream over several different external resources, including TCP and UDP sockets, MIDI ports, and serial ports. It uses a Berkeley-style primitive interface, and synchronization based on (Smalltalk) semaphores, instead of the current polling approach. Timeouts are supported, via those semaphores. Servers can handle multiple clients. This release uses new networking primitives. The source for them, and a new virtual machine, are included for the Win95 platform. It's at http://www.netjam.org/self/projects/smalltalk/correspondents.html. Please try it out and let me know what you think! thanks, -C -- Craig Latta composer and computer scientist craig.latta(a)netjam.org www.netjam.org latta(a)interval.com Smalltalkers do: [:it | All with: Class, (And love: it)]

3 2

Newbie Q: Geographic Transformation Package Availability?
by John-Reed Maffeo 26 Mar '99

26 Mar '99

Does anyone have any Smalltalk programs that you could share which address any cartographic related calculations, transformations, or display? I am going to try writing a Spatial Data Browser that I can use to view cartographic data (available on the web from the USGS). This project is part of my ongoing effort to learn how to do OO<foo>. I am a newbie to everything and any help would be appreciated. Thank You, John-Reed Maffeo Mesa, Arizona, USA

2 1

Re: Multilingual Squeak
by Michael S. Klein 26 Mar '99

26 Mar '99

Just changing the user-visible string in an app is the usual way to go for multi-lingual applications. I was once teaching Scheme to some Japanese-speaking clients (with the help of a translator -- aargh) I showed them VisualWorks... particularly that they could use kanji for methods, variables, etc, without the compiler barfing. They seemed surprised. Still, even within an English speaking community of programmers, it seems difficult (boring/not done) to actually get a *lot* of consistency amongst Smalltalk developers on interfaces/message-semantics. Huzzah to X3J20 for even trying. Perhaps this is the price we pay for dynamic typing. Like VW, dynamic typing is expensive, but worth it :-) -- Mike mklein(a)alumni.caltech.edu

1 0

[BUG] Sockets on 68k MacOS 8.1 standalone
by O'NEEL Bruce 26 Mar '99

26 Mar '99

Hi, First, the combo of Squeak, PWS, and Swiki produce one of the coolest applications I've ever seen. Fantastic. Alas, there is one glitch. On quitting Squeak I get a hard hang (no mouse response, etc), or, in MacsBug I get an illegal instruction. The config: PowerBook 520, 33mhz LC040 MacOS 8.1 OpenTransport 1.3 Squeak 2.3 as released. I set up the TCP/IP panel to connect via: AppleTak(MacIP), Configure: Using MacIP Manually, with a hard coded IP address of 1.2.3.4. Then I configure the AppleTalk to use RemoteOnly. RemoteOnly is from the Apple OpenTransport SDK and is version 2.1. Why all this silliness? So that I can run TCP/IP applications without a net connection, ie, the server and client on the same system. This is documented a touch clearer in http://developer.apple.com/qa/nw/nw59.html. Works great, PWS serves pages to Scamper with no problem. A PB540 with the Apple PPC upgrade, but otherwise identically configured works fine in this same config and doesn't hang on exit. To be honest this is a small problem and probably not worth much of anyone's time. First, I'm probably one of the few people who this bothers. Second because this is an bizzare config and because I can just reboot the machine with macsbug when I want to exit :-). I guess exiting Squeak is at some level evil, right? Finally, if I find it really annoying I'll try to track it down myself. I've attached the StdLog log below. Thanks! cheers bruce Content-type: text/plain; NAME=stdlog; CHARSET=US-ASCII Content-description: StdLog from Macsbug Content-disposition: inline; filename=stdlog MacsBug 6.5.4a4, Copyright Apple Computer, Inc. 1981-98 Illegal Instruction at 014216B0 26-Mar-1999 2:25:54 PM (since boot = 3 minutes) Current application is "Squeak VM 2.3 68K" Machine = 72 (PowerBook500), System $0810, sysu = $01008000 ROM version $067C, $30F1, $0000 (ROMBase $40000000) VM is off NIL^ = $40010000 Stack space used = +628 Address 014216B0 is in the "Squeak VM 2.3 68K" heap at 013BFCC0 It is 00022ED0 bytes into this heap block: Start Length Tag Mstr Ptr Lock Prg Type ID File Name * 013FE7E0 000234DC+08 N 68040 Registers D0 = 00000000 A0 = 014216B0 USP = 00000000 D1 = 00000000 A1 = 01F1EFAC MSP = 00000000 D2 = 00000000 A2 = 015192B1 ISP = 01FBA2F4 D3 = 00000000 A3 = 01846F64 VBR = 00000000 D4 = 00000018 A4 = 01846F34 CACR = 80008000 SFC = 0 D5 = 01438FC4 A5 = 01FBA568 DFC = 5 D6 = 000003BC A6 = 01FBA314 PC = 014216B0 D7 = 00000000 A7 = 01FBA2F4 SR = SmxnZvc Int = 0 Disassembling from 01421692 No procedure name 01421692 EOR.W D6,(A2)+ | BD5A 01421694 BCHG D0,D2 | 0142 01421696 MOVE.B $013E,-(A2) | 1538 013E 0142169A CMPA.W $0142(A4),A6 | BCEC 0142 0142169E MOVE.B $013E,-(A2) | 1538 013E 014216A2 CMP.L $0142,D6 | BCB8 0142 014216A6 MOVE.B $013E,-(A2) | 1538 013E 014216AA CMP.W #$0142,D6 | BC7C 0142 014216AE MOVE.B $013E,-(A2) | 1538 013E 014216B2 EOR.L D5,-(A0) | BBA0 014216B4 BCHG D0,D2 | 0142 014216B6 MOVE.B $013E,-(A2) | 1538 013E 014216BA CMP.L (A2)+,D2 | B49A 014216BC BCHG D0,D2 | 0142 014216BE MOVE.B $013E,-(A2) | 1538 013E 014216C2 SUB.W (A4),D6 | 9C54 014216C4 BCHG D0,D2 | 0142 014216C6 MOVE.B $013E,-(A2) | 1538 013E 014216CA DIVU.W $0142(A0),D7 | 8EE8 0142 014216CE MOVE.B $013E,-(A2) | 1538 013E Heap zones #1 32 3727K 00002000 to 003A5E2F SysZone^ #2 32 4K 00002800 to 00003813 ! #3 32 47K 000F43D0 to 001003C3 #4 32 31877K 003A5E30 to 022C73D3 #5 32 172K 01394750 to 013BF9F3 #6 32 12106K 013BFCC0 to 01F92563 "Squeak VM 2.3 68K" ApplZone^ TheZone^ Targ etZone #7 32 47K 01FCE020 to 01FD9EE3 #8 32 903K 021D9430 to 022BB2B3 "Finder" WARNING: One or more heaps may be corrupt. Use HC ALL (Heap Check) for a thorough check. Checking all heaps The System heap at 00002000 is ok The heap at 00002800 is bad Nonrelocatable block: Pointer to zone is bad Block header 00002854 4000 000C 0000 04D0 4007 EBC0 DB6D B6DB @******-@**¿*m* The heap at 000F43D0 is ok The Process Manager heap at 003A5E30 is ok The heap at 01394750 is ok The "Squeak VM 2.3 68K" heap at 013BFCC0 is ok The heap at 01FCE020 is ok The "Finder" heap at 021D9430 is ok Totaling the "Squeak VM 2.3 68K" heap at 013BFCC0 Total Blocks Total of Block Sizes Free 0013 #19 00079FC0 #499648 Nonrelocatable 0014 #20 00B57010 #11890704 Relocatable 0049 #73 00001890 #6288 Locked 0000 #0 00000000 #0 Purgeable and not locked 0000 #0 00000000 #0 Heap size 0070 #112 00BD2860 #12396640 The target heap is the System heap at 00002000 Totaling the System heap at 00002000 Total Blocks Total of Block Sizes Free 0040 #64 0000534C #21324 Nonrelocatable 01AD #429 00187650 #1603152 Relocatable 04DB #1243 00217450 #2192464 Locked 00D3 #211 00187DB0 #1605040 Purgeable and not locked 0052 #82 00024F90 #151440 Heap size 06C8 #1736 003A3DEC #3816940 The target heap is the "Squeak VM 2.3 68K" heap at 013BFCC0 Displaying File Control Blocks fRef File Vol Type Fl Fork LEof 0002 System Bruce 68k zsys dW rsrc #5956722 0060 Bruce 68k **** dw data #1047552 00BE Bruce 68k **** dw data #1047552 04C8 OpenTpt Serial Arbitr Bruce 68k libr dw rsrc #7638 0526 Open Transport Library Bruce 68k libr dw rsrc #587356 0584 Shared Library Manager Bruce 68k INIT dw rsrc #168325 05E2 Open Transport Library Bruce 68k libr dw rsrc #587356 0640 Open Tpt AppleTalk Li Bruce 68k libr dw rsrc #541758 069E Open Tpt Internet Lib Bruce 68k libr dw rsrc #485159 06FC OpenTpt Remote Access Bruce 68k libr dw rsrc #355731 075A OpenTpt Modem 68K Bruce 68k libr dw rsrc #56729 07B8 OpenTpt Remote Access Bruce 68k libr dw rsrc #355731 0816 Remote Access Log Bruce 68k lzlg dW data #12288 0874 OpenTpt Serial Arbitr Bruce 68k libr dw rsrc #7638 08D2 Appearance Extension Bruce 68k INIT dw rsrc #633281 0930 Appearance Extension Bruce 68k INIT dw rsrc #633281 098E Users & Groups Data F Bruce 68k BTFL dW data #12288 09EC Finder Bruce 68k FNDR dW rsrc #502012 0A4A Finder Preferences Bruce 68k pref dW rsrc #510 0AA8 Desktop DB Bruce 68k BTFL dW data #86016 0B06 Desktop DF Bruce 68k DTFL dW data #656082 0B64 Squeak VM 2.3 68K Bruce 68k APPL dW rsrc #1481 0BC2 StdLog Bruce 68k TEXT dW data #6427 #71 FCBs, #33 in use (including #10 fonts not listed), #38 free Displaying resource information: > Map $013BFDF8, flags $0000, file $0B64 = Squeak VM 2.3 68K + Map $00151EA4, flags $001E, file $08D2 = Appearance Extension S Map $00002064, flags $000D, file $0002 = System [Skipped $000A maps belonging to font files] Calling chain using A6 links A6 Frame Caller 01FBA560 01421CD8 01FBA440 001012D2 CallUniversalProc+00DD2 01FBA400 013F2B7C setInterpreter+0D600 01FBA3E0 013EF3AA setInterpreter+09E2E 01FBA3B4 013CE184 gePrimitiveSetOffset+02B48 01FBA368 013DE8D2 primitiveFloatArraySubScalar+031F2 01FBA344 013EE616 setInterpreter+0909A 01FBA330 002C8BA0 ExitToShell+00018 Return addresses on the stack Stack Addr Frame Addr ISA Caller 01FBA54C 68K 002AEFEE 'scod BFAF 0002'+0F17E 01FBA548 01FBA544 68K 0029626A 01FBA540 68K 0029626A 01FBA53C 68K 01FBA4BE 01FBA528 68K 002961C6 01FBA524 68K 002961C6 01FBA506 68K 001239FE 01FBA4DC 68K 01FBA4EA 01FBA4D0 01FBA4CC 68K 00296968 01FBA4C0 01FBA4BC 68K 00296FC8 01FBA4AC 01FBA4A8 68K 002970F2 01FBA4A4 68K 00296DD0 01FBA490 68K 01FBA4BE 01FBA474 68K 00296DD0 01FBA448 68K 002A4932 'scod BFAF 0002'+04AC2 01FBA444 01FBA440 68K 01421CD8 01FBA404 01FBA400 68K 001012D2 CallUniversalProc+00DD2 01FBA3FC 68K 0029779E 01FBA3E4 01FBA3E0 68K 013F2B7C setInterpreter+0D600 01FBA3B8 01FBA3B4 68K 013EF3AA setInterpreter+09E2E 01FBA36C 01FBA368 68K 013CE184 gePrimitiveSetOffset+02B48 01FBA368 68K 01FBA3B2 01FBA35C 68K 013CE0C0 gePrimitiveSetOffset+02A84 01FBA34C 68K 013DF650 primitiveFloatArraySubScalar+03F70 01FBA348 01FBA344 68K 013DE8D2 primitiveFloatArraySubScalar+031F2 01FBA334 01FBA330 68K 013EE616 setInterpreter+0909A 01FBA318 01FBA314 68K 002C8BA0 ExitToShell+00018 01FBA314 68K 01FBA32E 01FBA310 01FBA30C 68K 013EB502 setInterpreter+05F86 01FBA2F4 68K 001006B2 CallUniversalProc+001B2 Displaying memory from 0 00000000 4001 0000 4001 0000 0029 FEC6 0029 FEC8 @***@****)**)*» 00000010 0029 FECA 0029 FECC 4000 26F8 4000 26FA *)* *)*Ã@*&*@*&* Closing log

1 0

Re: [VM] NCM performance report
by Alan Lovejoy 26 Mar '99

26 Mar '99

Ian Piumarta wrote: > Helge, > > > I wonder what a J2-NCM VM could do... > [snip] > What does cost (a *lot*) is block activation, since I have to go > through the maps to find the initial nPC based on the intial vPC > stored in the closure. (For compiled methods this cost is only paid > when flushing a context out of the context cache, which happens > relatively infrequently.) My #1 request would therefore be to have > block bodies compiled into their own CompiledMethods, separate from > the home method. (Or does NCM do this already??? I've not been > following the thread very closely.) I haven't looked at Tim's NCM implementation yet (Hi, Tim!), butI know that the VisualWorks implementation has something called a CompiledBlock, which shares a common superclass with CompiledMethod (namely, CompiledCode). I think this is essentially what you requested. Tim and I worked together on a project (that shall remain nameless) where we modified the Squeak VM in a way very similar to Tim's current NCM implementation (well, actually Tim and Don Charnley reworked the VM, and I reworked the CompiledMethod class). It **is** cleaner and more elegant, and makes CompiledMethods more easily extensible. --Alan Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Alan Lovejoy Content-Disposition: attachment; filename="vcard.vcf" Attachment converted: Anon:vcard.vcf 12 (TEXT/ttxt) (00007A78)

1 0

← Newer
1
2
3
4
5
6
7
8
...
34
Older →

Jump to page: