Squeak-dev November 2000

squeak-dev@lists.squeakfoundation.org

106 participants
325 discussions

[FIX] OrderedCollection>>atAll:
by Stephan Rudlof 29 Nov '00

29 Nov '00

"Change Set: Fix-OrderedCollection>>atAll: Date: 29 November 2000 Author: Stephan Rudlof Overrides SequenceableCollection>>atAll:. Evaluation (OrderedCollection with: 1 with: 2 with: 3) atAll: #(2) works with this fix." -- Stephan Rudlof (sr(a)evolgo.de) "Genius doesn't work on an assembly line basis. You can't simply say, 'Today I will be brilliant.'" -- Kirk, "The Ultimate Computer", stardate 4731.3

1 0

Re: Squeak and Children
by Mark Guzdial 29 Nov '00

29 Nov '00

At 3:12 PM -0800 11/27/00, Ken Kahn wrote: >Hi all. > >The reason I was particularly interested in the discussion of ToonTalk by >the Squeak community is because I'd like to understand better the relative >merits of the two systems. Here's my current understanding (note that I have >yet to check out the "Tiles" stuff in Squeak). The "Tiles" stuff (also called EToys or Viewers) is important here -- it's an easily-approachable level that can lead to programming textually in Squeak. I've got a walk-through of building one demo with EToys in my textbook. A draft chapter on the Morphic stuff including EToys is available in the Squeak Swiki at http://minnow.cc.gatech.edu/squeak/uploads/programmingmorphic.pdf >Squeak's special strengths are its mallebility, its consistency/elegance, >and its implementation/portability. > >ToonTalk's special strengths are its game-like program construction and >debugging environment, its support for programming by example, and its model >of concurrency. EToys also has a model of concurrency that's pretty powerful. I'll let others speak to the other points. > >Besides hearing whether the above summaries are accurate, I'd like to >understand the Squeak community's stance on children constructing programs. >For example, is the view that Squeak is well-suited for children above a >certain age? And if so what age? Is there anyone out there working with >young children and Squeak? If so how can one find out more? Various folk here are trying Squeak out with kids, including Kim Rose of Squeak Central and Stephane Ducasse. Mark -------------------------- Mark Guzdial : Georgia Tech : College of Computing : Atlanta, GA 30332-0280 Associate Professor - Learning Sciences & Technologies. Collaborative Software Lab - http://coweb.cc.gatech.edu/csl/ (404) 894-5618 : Fax (404) 894-0673 : guzdial(a)cc.gatech.edu http://www.cc.gatech.edu/gvu/people/Faculty/Mark.Guzdial.html

3 4

[FIX][2.9alpha] Install LargeIntegers v1.2
by Stephan Rudlof 29 Nov '00

29 Nov '00

Attention: Use this changeset only if you are sure to work with LargeIntegers v1.2! Greetings, Stephan "Change Set: InstallLargeIntegersV1.2-sr Date: 29 November 2000 Author: Stephan Rudlof Update: This changesets updates Integer>>anyBitOfMagnitudeFrom:to: to work with LargeIntegers v1.2." -- Stephan Rudlof (sr(a)evolgo.de) "Genius doesn't work on an assembly line basis. You can't simply say, 'Today I will be brilliant.'" -- Kirk, "The Ultimate Computer", stardate 4731.3

1 0

[FIX][plugin] LargeIntegers v1.2
by Stephan Rudlof 29 Nov '00

29 Nov '00

This fix correponds to the fix with subject [BUG][FIX] anyBitOfMagnitudeFrom:to:Fix-sr . It fixes the corresponding method inside the LargeIntegers plugin. Greetings, Stephan "Change Set: LargeIntegersV1.2-sr Date: 29 November 2000 Author: Stephan Rudlof Fix: LargeIntegersPlugin>>anyBitOfBytes: aBytesOop from: start to: stopArg Minor Fix/Improvement: >>testTwoComplementRightShift: moved it from LargePositiveInteger to Integer and corrected user messages. ToDo: (If commented out) you have to uncomment the primitive call in Integer>>anyBitOfMagnitudeFrom: start to: stopArg if you have updated the LargeIntegers module to v1.2 (or a VM with it). This is also made by fileIn the InstallLargeIntegersV1.2-sr changeset. " -- Stephan Rudlof (sr(a)evolgo.de) "Genius doesn't work on an assembly line basis. You can't simply say, 'Today I will be brilliant.'" -- Kirk, "The Ultimate Computer", stardate 4731.3

1 0

[BUG?]New Event Architecture Queue Sequencing?
by Andrew C. Greenberg 29 Nov '00

29 Nov '00

Running the following script in Morphic with a transcript open, I get some unexpected behavior. If you type a few keys, the transcript seems to correctly and timely capture all the key events. However, if I type a long sentence, along the lines of "Now is the winter of our Discontent, made glorious Summer by this son of York....!", all the key events appear to be captured, but after half of the events click by, there is a long pause (about a second or so) between each event. Hitting the shift key a few times seems to accelerate things somewhat, but, something seems wrongful. I started scanning the code, but I couldn't immediately identify the clear source of the problem. Any thoughts? "Exercise Keyboard stuff" showKeyEvent _ [:evt | ((evt at: 4) caseOf: { [0]->['character $', (evt at: 3) asCharacter asString, '(', (evt at: 3) asString,')']. [1]->['keyDown ', (evt at: 3) asString]. [2]->['keyUp ', (evt at: 3) asString]} otherwise: ['error']), ' [', ((evt at: 5) radix: 2), ']']. [((evt _ Sensor nextEvent) isNil not and: [(evt at: 1) = 2]) ifTrue: [Transcript show: Sensor eventQueue size asString, ': '. Transcript show: (showKeyEvent value: evt); cr]] repeat

1 0

[BUG] Win32 VM SystemDictionary getSystemAttribute: 1004
by mdr＠scn.org 29 Nov '00

29 Nov '00

Smalltalk getSystemAttribute: 1004 should return a string describing the "vm version" Instead it returns a string that describes the image version This is a problem with the most recent 2.8 Win32 VM. (I am putting together a general bug reporting template based on Daniel's work, and would like to include the basic VM information that can be seen when pressing F2 and then selecting "Display version information". As far as I can see, there are no alternative ways to get VM information) Mike NB: I have added this to the SourceForge VM bug database at http://sourceforge.net/bugs/?group_id=5506

1 0

Updates to SirenMIDIPlugin from Jay Hardesty
by Stephen T. Pope 29 Nov '00

29 Nov '00

Hello all, Jay Hardesty made a few extensions and fixes to the SirenMIDIPlugin; they are attached to this letter. Is anyone looking into porting the plugin to Linux or Windows? stp Jay Hardesty wrote: > > [ ... ] > > Here's a slightly altered version of sqMacMIDI.OMS.c > which seems to fix the problem for us - I'm not sure > if this was the right way to go about it, but so far > so good. Since input is already working fine for > you these changes may not be very interesting, but > sending them back just in case they might be of some > use (or maybe some amusement) somehow. > > The main changes were: > Uncommenting out a block of your code in sqOpenMIDI which > sets up OMS to read from all interfaces. This change made > me kind of nervous because I assume you're getting input > without running through this code... > I also added a condition in that same block of code > so that only devices checked 'Is a controller' in OMS Setup > are actually used for input: > (*nodeInfoList)->info[i].flags & kAnyIsController > > Added curly braces for a couple of nested if-else > statements in sqMIDIPlaySizeOn. [ ... ] > > Also here's a mindless little smalltalk method for > setting up the MIDICommands shared pool in MIDIPort > that adds a few more status messages. -- stp Stephen Travis Pope http://www.create.ucsb.edu/~stp 'From Squeak2.8 of 13 June 2000 [latest update: #2359] on 28 November 2000 at 10:29:55 pm'! !MIDIPort class methodsFor: 'as yet unclassified' stamp: 'jh 11/28/2000 22:28'! initializeMIDICommands " MIDIPort initializeMIDICommands " MIDICommands ifNil: [MIDICommands _ Dictionary new] ifNotNil: [MIDICommands keysAndValuesRemove: [:a :b | true]]. MIDICommands at: #noteOff put: 128; at: #noteOn put: 144; at: #polyTouch put: 160; at: #ctrlChange put: 176; at: #progChange put: 192; at: #chanTouch put: 208; at: #pitchWheel put: 224; at: #sysEx put: 240; at: #mtc put: 241; at: #spp put: 242; at: #midiClock put: 248; at: #midiStart put: 250; at: #midiContinue put: 251; at: #midiStop put: 252. (MIDIPort sharedPools includes: MIDICommands) ifFalse: [MIDIPort addSharedPool: MIDICommands]. MIDIPort compileAll ! ! /*************************************************************** * Squeak/Siren Interface to the Opcode MIDI System (OMS) * Stephen T. Pope -- stp(a)create.ucsb.edu * * Version History: see sqMIDI.h * * N.B.: This driver requires OMS version 2.5.3 or greater. * See http://www.opcode.com to get the latest version of OMS for free. * * Because of the (Mac/DOS) portability of OMS, this code should be * (more or less) portable to the DOS PC as well (please let me know). */ /*************************************************************** * Includes & Macros */ //#include "sq.h" // General Squeak definitions, includes sqMIDI.h // MIDI Includes -- STP -- these used to be in sq.h #include "OMS.h" // OMS definitions and structs #include <MIDI.h> // Apple MIDI Libraries #include "sqMIDI.h" // Squeak MIDI Structs and Prims #include "sqVirtualMachine.h" // The virtual machine proxy definition #include "Strings.h" #define sqSignature 'SQEK' // Creator ID. #define InputPortID 'in ' // I/O port IDs. #define OutputPortID 'out ' //#define DEBUG_MODE // Enable/Disable printfs (see macros below) // Debugging macros #ifdef DEBUG_MODE #define dprint1(str) printf(str) #define dprint2(str, val) printf(str, val) #define dprint3(str, v1, v2) printf(str, v1, v2) Boolean debug_mode = true; #else #define dprint1(str) #define dprint2(str, val) #define dprint3(str, v1, v2) Boolean debug_mode = false; #endif DEBUG_MODE extern struct VirtualMachine *interpreterProxy; // A few macros for error checking #define S_FAIL { interpreterProxy->success(false); return(-1); } #define S_ERR { interpreterProxy->success(false); return((int) err); } #define S_CHECK if(err != omsNoErr) { return(0 - (int) err); } #define S_CHECK_FAIL if(err != omsNoErr) { interpreterProxy->success(false); \ return(0 - (int) err); } #define S_CHECK_SIGNIN if(!gSignedInToMIDI) { return(-1024); } // memory access macros from interp.c #define longAtput(i, val) (*((int *) (i)) = val) #define instVarAtPutInt(obj, slot, val) \ longAtput(((char *)obj + (slot << 2)), (((int)val << 1) | 1)) /*************************************************************** * Global state variabless */ Boolean gSignedInToMIDI; // are we signed into MIDI driver? Boolean gEchoEnabled; // toggle to echo event input Boolean gInputEnabled; // toggle to enable event input Boolean gCacheControllers; // toggle to cache controller values /*************************************************************** * Controller value caches -- This is the minimum complement * A "larger" driver would cache 128*16 controllers and 128*16 key pressures */ unsigned char sqControllers[128]; // Controller value table unsigned char sqKeyPressures[128]; // Polyphonic key pressure table unsigned char sqChanPressures[16]; // Channel pressure value table int sqPitchBend[16]; // The value of the pitch wheel /*************************************************************** * OMS Driver Internals */ short gInPortRef; // refNum of the OMS input port short gOutPortRef; // refNum of the OMS output port short sqCallback; // semaphore for read callback short gCompatMode; // OMS compatibility mode #define NumPorts 8 // Max number of I/O ports #define MaxNameLen 64 // Max length of device names unsigned short gOutNodeRefs[NumPorts] = {0, 0, 0, 0, 0, 0, 0, 0}; sqMIDIioPort gNodeNames[NumPorts]; // The IO port table #define MaxInInQ 64 // The size of the input Q sqMIDIEvent sqMIDIInQ[MaxInInQ]; // the input event Q (an array) int itemsInInQ; // Index to write in the input Q int itemsInInQ2; // Pointer to input to read (<= iIIQ) OMSMIDIPacket thePacket; // Small (4-byte) MIDI packet OMSMIDIPacket255 theBigPacket; // Large packet for SysEx messages /*************************************************************** * Function Prototypes for I/O call-back utilities */ #define USESROUTINEDESCRIPTORS pascal void sqMIDIAppHook(OMSAppHookMsg *pkt, long myRefCon); pascal void sqMIDIReadHook(OMSPacket *pkt, long myRefCon); /*************************************************************** *************************************************************** * sqOpenMIDI -- Sign on to OMS, register callbacks, create ports * and connections, etc. */ int sqOpenMIDI(int callbackSemaphore, int inputMRSocket) { OSErr err; OMSAppHookUPP appHook; // pointer to call-back function OMSReadHookUPP readHook; // pointer to call-back function OMSAPI (long) versionID; // OMS driver version // OMSAPI(OMSIDListH) sqNodes; // list of I/O node IDs OMSConnectionParams sqConn; // connection parameter structure JH - uncommented this line OMSNodeInfoListH nodeInfoList; int i, j; // JH added tmp var j long LMGetCurrentA5(void); int interfaces; char *nname; Boolean found = false; interpreterProxy->success(true); //// It's not an error if we're already signed in. if (gSignedInToMIDI) return (0); dprint3("\n\tMIDI open -- sem = %d sock = %d\n", callbackSemaphore, inputMRSocket); sqCallback = 0; //// Make sure that OMS V2 is installed. versionID = OMSVersion(); if (versionID == 0) { interpreterProxy->success(false); return (-128); } if ((versionID & 0xff000000) != 0x02000000) { interpreterProxy->success(false); return (-129); } //// Set input event pointers itemsInInQ = 0; itemsInInQ2 = 0; //// Clear controller value tables for (i=0; i<128; i++) { sqControllers[i] = 64; sqKeyPressures[i] = 0; } for (i=0; i<16; i++) { sqChanPressures[i] = 0; sqPitchBend[i] = 0; } //// Set up call-back routine pointers. appHook = NewOMSAppHook(sqMIDIAppHook); readHook = NewOMSReadHook(sqMIDIReadHook); //// Sign in to OMS. err = OMSSignIn(sqSignature, (long)LMGetCurrentA5(), (unsigned char *) "Squeak/Siren", appHook, &gCompatMode); S_CHECK_FAIL gSignedInToMIDI = true; //// Add an input port. err = OMSAddPort(sqSignature, InputPortID, omsPortTypeInput, readHook, (long)LMGetCurrentA5(), &gInPortRef); if (err) goto errexit; //// Add an output port. err = OMSAddPort(sqSignature, OutputPortID, omsPortTypeOutput, NULL, 0L, &gOutPortRef); // no read hook for output if (err) goto errexit; //// Set up OMS to play to all output ports. nodeInfoList = OMSGetNodeInfo(omsIncludeOutputs + omsIncludeReal + omsIncludeVirtual); if (nodeInfoList != NULL) { interfaces = (*nodeInfoList)->numNodes; for (i=0; i<(*nodeInfoList)->numNodes; i++) { nname = (char *)(*nodeInfoList)->info[i].name; dprint2("\tOpen interface %s\n", nname); strcpy(gNodeNames[i].name, nname); gNodeNames[i].direction = 2; gOutNodeRefs[i] = (*nodeInfoList)->info[i].ioRefNum; } OMSDisposeHandle(nodeInfoList); } else goto errexit; // JH - uncommented following block of code //// Read from all input ports. sqConn.appRefCon = 0; nodeInfoList = OMSGetNodeInfo(omsIncludeInputs + omsIncludeReal + omsIncludeVirtual); if (nodeInfoList != NULL) { for (i=0; i<(*nodeInfoList)->numNodes; i++) { nname = (char *)(*nodeInfoList)->info[i].name; // JH - Actually only read from controllers if ((*nodeInfoList)->info[i].flags & kAnyIsController) { sqConn.nodeUniqueID = (*nodeInfoList)->info[i].uniqueID; err = OMSOpenConnections(sqSignature, InputPortID, 1, &sqConn, false); if (err) goto errexit; for (j=0; j < interfaces; j++) if (strcmp(nname, gNodeNames[j].name) == 0) { gNodeNames[j].direction += 1; found = true; } if (!found) { strcpy(gNodeNames[interfaces].name, nname); gNodeNames[interfaces].direction = 1; interfaces ++; } dprint2("\tConnect input %s\n", nname); } } OMSDisposeHandle(nodeInfoList); } //// Success gInputEnabled = false; // off by default gCacheControllers = false; // off by default gEchoEnabled = false; // off by default sqCallback = callbackSemaphore; interpreterProxy->success(true); return(interfaces); // Answer the number of interfaces found. errexit: OMSSignOut(sqSignature); gSignedInToMIDI = false; interpreterProxy->success(false); return ((int) err); } /*************************************************************** * sqCloseMIDI -- Close up shop and sign out from OMS */ int sqCloseMIDI() { interpreterProxy->success(true); S_CHECK_SIGNIN sqCallback = 0; // Perhaps I should do allNotesOff here... OMSSignOut(sqSignature); gSignedInToMIDI = false; gInputEnabled = false; gCacheControllers = false; gEchoEnabled = false; return(0); } /*************************************************************** * sqMIDIPlaySize -- Send a MIDI message out the output ports */ int sqMIDIPlaySizeOn (int message, int length, int interface) { int i; unsigned char *ptr = (unsigned char *) message; // It's an error if we're not signed in S_CHECK_SIGNIN // Play it now (i.e., no output buffer) if (length < 4) { // Handle a "normal" packet of 3 bytes thePacket.flags = 0; thePacket.len = length; for (i=0; i<length; i++) // copy output data into packet thePacket.data[i] = *(ptr + i); if (interface == 0) { // Interface 0 means play out all outputs for (i=0; i<NumPorts; i++) if (gOutNodeRefs[i] != 0) OMSWritePacket2(&thePacket, gOutNodeRefs[i], gOutPortRef); } // JH - added curly braces - CW seemed to disambiguate the if-if-else other than as intended(?) else OMSWritePacket2(&thePacket, gOutNodeRefs[interface - 1], gOutPortRef); } else { // Handle a big packet theBigPacket.flags = 0; theBigPacket.len = length; for (i=0; i<length; i++) // copy output data into packet theBigPacket.data[i] = *(ptr + i); if (interface == 0) { // Interface 0 means play out all outputs for (i=0; i<NumPorts; i++) if (gOutNodeRefs[i] != 0) OMSWritePacket2((struct OMSMIDIPacket *)&theBigPacket, gOutNodeRefs[i], gOutPortRef); // OMSWritePacket255(&theBigPacket, gOutNodeRefs[i], gOutPortRef); } // JH - added curly braces - CW seemed to disambiguate the if-if-else other than as intended(?) else OMSWritePacket2((struct OMSMIDIPacket *)&theBigPacket, gOutNodeRefs[interface - 1], gOutPortRef); // OMSWritePacket255(&theBigPacket, gOutNodeRefs[interface - 1], gOutPortRef); } interpreterProxy->success(true); return(length); } unsigned char lastCommand; int lastLength; /*************************************************************** * sqMIDIReadHook -- Interrupt routine called on in-coming MIDI messages */ pascal void sqMIDIReadHook(OMSPacket *pkt, long myRefCon) { sqMIDIEvent *inPtr; unsigned char *idata, *odata; int len, i, cmd, chan, value; long olda5 = SetA5(myRefCon); len = pkt->len; if (len == 0) // return if empty packet goto errexit; if (itemsInInQ == MaxInInQ) // return if input Q full goto errexit; // echo input to the output if (gEchoEnabled) for (i=0; i<NumPorts; i++) if (gOutNodeRefs[i] != 0) OMSWritePacket(pkt, gOutNodeRefs[i], gOutPortRef); cmd = (pkt->data[0]) & 0xF0; // Get "command" from 1st byte of packet if (gCacheControllers) { // Cache controller values in the driver if (cmd == ControlCmd) { // Read a control command chan = pkt->data[1]; value = pkt->data[2]; sqControllers[chan] = value; goto errexit; } else if (cmd == PchWheelCmd) { // Read a pitch wheel change chan = (pkt->data[0]) & 0x0F; i = pkt->data[1]; value = pkt->data[2]; sqPitchBend[chan] = (value << 7) + i; goto errexit; } else if (cmd == PolyPressCmd) { // Read polyphonic key pressure chan = pkt->data[1]; value = pkt->data[2]; sqKeyPressures[chan] = value; goto errexit; } else if (cmd == ChanPressCmd) { // Read channel key pressure chan = (pkt->data[0]) & 0x0F; value = pkt->data[1]; sqKeyPressures[chan] = value; goto errexit; } } if (!gInputEnabled) goto errexit; // Get the input Q record to use inPtr = &sqMIDIInQ[itemsInInQ++]; // Catch OMS packet length bug if ((cmd == PgmChngCmd) || (cmd == ChanPressCmd)) len = 2; else if (cmd != SysExCmd) len = 3; inPtr->len = len; // Copy the packet to the input Q inPtr->flags = (int)pkt->flags; inPtr->timeStamp = clock(); // = (int)pkt->beatTimeStamp; for OMS 2 input idata = &(inPtr->data[0]); odata = &(pkt->data[0]); if (len <= PKT_LEN) // Copy the data bytes for (i = 0; i < len; i++) *idata++ = *odata++; if (sqCallback != 0) { // Signal the read semaphore i = interpreterProxy->signalSemaphoreWithIndex(sqCallback);} errexit: SetA5(olda5); } /*************************************************************** * sqReadMIDIPacket -- Sent in response to the input semaphore * This is to up-load MIDI messages into the arguments (a MIDIPacket, * its max length, and its data ByteArray). */ int sqReadMIDIPacket(int ipacket, int maxSize, int idata) { sqMIDIEvent *outPtr; unsigned char *cdata; int len, i; unsigned char *pdata = (unsigned char *)idata; // The packet object is defined as: // Object subclass: #MIDIPacket // instanceVariableNames: 'length time flags data ' interpreterProxy->success(true); if (itemsInInQ == 0) return (0); if (sqCallback == 0) return (-1); // Set up pointers outPtr = &sqMIDIInQ[itemsInInQ2++]; dprint2("Reading %d bytes -- ", outPtr->len); if(debug_mode) printf("%x %x %x\n", outPtr->data[0], outPtr->data[1], outPtr->data[2]); len = outPtr->len; // copy the response fields if (len > maxSize) // packet longer than normal--answer length return (len); // copy the driver data into the packet object instVarAtPutInt(ipacket, 1, len); // length, time, flags instVarAtPutInt(ipacket, 2, (int)(outPtr->timeStamp)); instVarAtPutInt(ipacket, 3, (int)(outPtr->flags)); cdata = &(outPtr->data[0]); // copy MIDI message bytes into the packet for (i=0; i<len; i++) *pdata++ = *cdata++; if (itemsInInQ2 == itemsInInQ) // clear counters if Q is empty itemsInInQ2 = itemsInInQ = 0; return (len); // Answer len } /*************************************************************** * sqReadMIDIControl -- Answer one or more controller values from the cache */ int sqReadMIDIControl(int fromC, int toC, int intoArray) { int i; unsigned char *pdata = (unsigned char *)intoArray; // check controller range if ((fromC < 0) || (toC > 287) || (fromC > toC)) { interpreterProxy->success(false); return (-1); } pdata += 4; // Skip object header for (i = fromC; i <= toC; i++) { // Loop to read controllers if (i < 128) // 0-127 = controllers *pdata++ = ((int)sqControllers[i]); else if (i < 256) // 128-255 = key pressures *pdata++ = ((int)sqKeyPressures[i - 128]); else if (i < 272) // 256-271 = channel pressures *pdata++ = ((int)sqChanPressures[i - 256]); else // 272-287 = pitch bend (16-bit!) // Write single int into "Array" longAtput(intoArray, (sqPitchBend[i - 272])); } interpreterProxy->success(true); return(toC - fromC + 1); } /*************************************************************** * getDeviceName -- write the name of the selected device into the 2nd arg * Answer the ports directionality. */ int sqGetMIDIDeviceName(int index, int name) { if ((index < 0) || (index > NumPorts)) return(-1); strcpy((char *)name, gNodeNames[index].name); return(gNodeNames[index].direction); } /*************************************************************** * sqMIDIioctl -- General driver queries and status */ int sqMIDIioctl(int which, int onOff) { int answer = 0; interpreterProxy->success(true); switch (which) { case (sqMIDIInstalled): // Is a MIDI driver installed? answer = (int)OMSVersion(); // Can I communicate with the "real" driver? if (answer == 0) answer = false; answer = true; break; case (sqMIDIVersion): // What driver version is this? answer = (int)OMSVersion(); break; case (sqMIDIHasBuffer): // Is there a time-stamped output buffer? answer = false; // Answer false in the "default" driver break; case (sqMIDIHasDurs): // Is there a 1-call note command? answer = false; // Answer false in the "default" driver break; case (sqMIDIHasClock): // Does the driver have its own clock? answer = false; // Answer false in the "default" driver break; case (sqMIDIUseSemaphore): // Should the driver signal a semaphore on input? switch (onOff) { case sqMIDIQuery: // Answer value of flag answer = gInputEnabled; break; case sqMIDITurnOn: gInputEnabled = true; // Set flag break; case sqMIDITurnOff: gInputEnabled = false; // Clear flag break; } break; case (sqMIDIEcho): // Should we echo in-coming events in the driver? switch (onOff) { case sqMIDIQuery: answer = gEchoEnabled; // Answer value of flag break; case sqMIDITurnOn: gEchoEnabled = true; // Set flag break; case sqMIDITurnOff: gEchoEnabled = false; // Clear flag break; } break; case (sqMIDIControllerCache): // Should we cache controller values for polling? switch (onOff) { case sqMIDIQuery: answer = gCacheControllers; // Answer value of flag break; case sqMIDITurnOn: gCacheControllers = true; // Set flag break; case sqMIDITurnOff: gCacheControllers = false; // Clear flag break; } break; case (sqMIDIEventsAvailable): // How many events are in the input Q? answer = itemsInInQ - itemsInInQ2; break; case (sqMIDIFlushDriver): // Special flag to flush the driver's I/O buffers, itemsInInQ = itemsInInQ2 = 0; OMSAllNotesOff(); // and send all-notes-off break; default: // Default = FAIL interpreterProxy->success(false); } return(answer); } /*************************************************************** * sqMIDIAppHook -- Sent on OMS configuration changes; NO-OPs at present */ pascal void sqMIDIAppHook(OMSAppHookMsg *pkt, long myRefCon) { long olda5 = SetA5(myRefCon); switch (pkt->msgType) { // What to do when compatibility mode changes? case omsMsgModeChanged: // What to do? break; // What to do if a receiver is deleted? case omsMsgDestDeleted: // What to do? break; // What to do if the studio layout changes? case omsMsgNodesChanged: // What to do? break; } SetA5(olda5); } /*************************************************************** * E N D */

1 0

[BUG] Windows VM problem typing @
by mdr＠scn.org 29 Nov '00

29 Nov '00

I can not type (using a German keyboard -- Alt Gr Q) the "@" character when using the new Win32 VM 2.8 Build 4 November 4 at http://isgwww.cs.uni-magdeburg.de/~raab/squeak/SqueakExe.zip Nothing shows up with my 2.9alpha image with updates to 2774 (before the morphic changes). Nothing shows up with 2.9alpha image with all the recent public updates. Interestingly, if I run a vanilla 2.8 image, I do get the character Mike

2 1

Re: ChitChat, a C-like dialect (was Re: jpython anyone?)
by Andrew P. Black 29 Nov '00

29 Nov '00

>Okay, I did it. Check out the attachment. Bit of advice: it >wasn't designed to be "elegant" as in "ooooh!" :-) It was >designed to "feel right" for someone coming from C or (to a >lesser extent) Python. This is interesting. I never cease to be amazed how hard it is for some people to grok Smalltalk syntax, once they have been exposed to C. I find this with professional engineers, and with high school students. I have one suggestion for the declaration syntax. The colon is not particularly C-like. C programmers would write "int x". Of course, you ChitChat doesn't have types, but you might adopt the idea of using a keyword. So, instead of to:do (1, 8, {:(x) y: y = nextInt (myrandom, x); plotBetween:and (self, x, y); }); you might try to:do (1, 8, {param x; var y; y = nextInt (myrandom, x); plotBetween:and (self, x, y); }); This means that binary messages param: or var: can't be sent, but if one needs to do that, one can always use perform:with: as a workaround. Incidentally, C uses double quotes for strings, and single quotes for characters, so it probably makes sense to adopt that convention. Andrew

2 1

Re: Ascii Equivalents for KeyUp/KeyDown events
by Lex Spoon 28 Nov '00

28 Nov '00

I'm sure someone will post a more thorough response, but note that not all keys *have* an ASCII code. What is ASCII for F1, or up-arrow? -Lex "Andrew C. Greenberg" <werdna(a)mucow.com> wrote: > As I began messing around with the new event paradigm, I noted that > KeyUp and KeyDown events pass individual Key Codes, rather than their > ASCII equivalents. > > Is there any reason that the ASCII equivalent, when available, is not > carried somewhere in the keyUp/Down event objects? It would be > convenient and, particularly for games, faster, to have that > information be readily available without an additional lookup in some > applications without having to process two separate events. > > In the Macintosh implementation, this information is available > beneath the scenes -- in fact, it is separately parsed out of a > single MacOS event into two Squeak events. I'm not asking for a > change in the semantics -- just the convenience of having both > keycode and ascii equivalent in the same keydown record. > > Or am I missing something?

1 0

Jump to page: