Re: [Ql-Users] Fw: query
On 13/04/2012 16:10, Norman Dunbar wrote: On 13/04/12 16:06, Rich Mellor wrote: You are correct - the full error is: Cannot open flp1_libguide_dbs as a database. -15 Press ENTER ... The database starts with DBAS As I say, the program runs fine under Minerva and I have opened it and resaved it. It still will not open under a JS ROM on q-emulator. The source code isn't complicated. Open the file as a qdos file first, then take the 32 bit channel number and use that to open the file as a database. If the database id returned is negative, we have an error, bale out after displaying the error. So the problem must be/is in the DBAS library or Q-Emulator I think, especially as it opens fine in other ROMs. Difficult one to solve I think. Cheers, Norm. Daniele has looked at this issue and to quote his comments: The bad parameter error is related to a file system TRAP#3 call with D0=$44. Do you know what this trap is supposed to do? In any case, this is probably not the primary issue, as on Minerva this TRAP is not called for some reason. Either the code really doesn't work on JS, or there is an incompatibility somewhere, perhaps in the Q-emuLator file drivers. -- Rich Mellor RWAP Services Specialist Enuuk Auction Programming Services www.rwapservices.co.uk ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] Fw: query
On 15 Apr 2012, at 08:59, Rich Mellor wrote: On 13/04/2012 16:10, Norman Dunbar wrote: On 13/04/12 16:06, Rich Mellor wrote: You are correct - the full error is: Cannot open flp1_libguide_dbs as a database. -15 Press ENTER ... The database starts with DBAS As I say, the program runs fine under Minerva and I have opened it and resaved it. It still will not open under a JS ROM on q-emulator. The source code isn't complicated. Open the file as a qdos file first, then take the 32 bit channel number and use that to open the file as a database. If the database id returned is negative, we have an error, bale out after displaying the error. So the problem must be/is in the DBAS library or Q-Emulator I think, especially as it opens fine in other ROMs. Difficult one to solve I think. Cheers, Norm. Daniele has looked at this issue and to quote his comments: The bad parameter error is related to a file system TRAP#3 call with D0=$44. Do you know what this trap is supposed to do? This Trap is not implemented on all the normal QDOS and SMSQ/E systems. In any case, this is probably not the primary issue, as on Minerva this TRAP is not called for some reason. Either the code really doesn't work on JS, or there is an incompatibility somewhere, perhaps in the Q-emuLator file drivers. George ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] Fw: query
Op Sun, 15 Apr 2012 09:59:18 +0200 schreef Rich Mellor r...@rwapservices.co.uk: On 13/04/2012 16:10, Norman Dunbar wrote: On 13/04/12 16:06, Rich Mellor wrote: You are correct - the full error is: Cannot open flp1_libguide_dbs as a database. -15 Press ENTER ... The database starts with DBAS This error is produced by Libguide_exe and not QDOS. I had exactly the same error when I forgot to load DATA_BIN first. After loading DATA_BIN first everything was OK (Q-Emulator 2.5, JS). ... Daniele has looked at this issue and to quote his comments: The bad parameter error is related to a file system TRAP#3 call with D0=$44. Do you know what this trap is supposed to do? In any case, this is probably not the primary issue, as on Minerva this TRAP is not called for some reason. Either the code really doesn't work on JS, or there is an incompatibility somewhere, perhaps in the Q-emuLator file drivers. Have you tried a more recent version of DATA_BIN? The supplied one is so old there is no version number, now 2.13 (21/2/98). Bob -- The BSJR QL software site at: http://members.upc.nl/b.spelten/ql/ ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] Fw: query
On 15/04/2012 11:12, Bob Spelten wrote: Op Sun, 15 Apr 2012 09:59:18 +0200 schreef Rich Mellor r...@rwapservices.co.uk: On 13/04/2012 16:10, Norman Dunbar wrote: On 13/04/12 16:06, Rich Mellor wrote: You are correct - the full error is: Cannot open flp1_libguide_dbs as a database. -15 Press ENTER ... The database starts with DBAS This error is produced by Libguide_exe and not QDOS. I had exactly the same error when I forgot to load DATA_BIN first. After loading DATA_BIN first everything was OK (Q-Emulator 2.5, JS). ... Daniele has looked at this issue and to quote his comments: The bad parameter error is related to a file system TRAP#3 call with D0=$44. Do you know what this trap is supposed to do? In any case, this is probably not the primary issue, as on Minerva this TRAP is not called for some reason. Either the code really doesn't work on JS, or there is an incompatibility somewhere, perhaps in the Q-emuLator file drivers. Have you tried a more recent version of DATA_BIN? The supplied one is so old there is no version number, now 2.13 (21/2/98). Bob I don't have a more recent version of DATA_BIN - could you maybe email me one - if it works, I will send it onto Quanta and ask that they replace the version used in the library guide. -- Rich Mellor RWAP Services Specialist Enuuk Auction Programming Services www.rwapservices.co.uk ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] Fw: query
Have you tried a more recent version of DATA_BIN? The supplied one is so old there is no version number, now 2.13 (21/2/98). Bob I don't have a more recent version of DATA_BIN - could you maybe email me one - if it works, I will send it onto Quanta and ask that they replace the version used in the library guide. There is a v1.21 string embedded in the Library version, although I'm not sure if that refers to the library guide database itself. The boot program on my qemulator loads v2.12 of the DATA_BIN. So we know the error persists with v2.12 at least. Disk SP07 in Quanta library has version 2.13 of DATA_BIN (which I’ll send to Rich privately to see if it helps), while disk SP08 has the sources for anyone who wants to investigate the problem. This is a list of changes, although the list stops at v2.10, after that there appears to only have been sporadic updates to individual files, possibly by Phil Borman. Dilwyn CHANGES TO DATABASE HANDLING V2.00 Names handling introduced into the vectors, Basic interface, and C interface. The FSD.ORDR (ORDER) algorithm was rewritten to be alot faster. Array parameters made available to Basic procedures. When the comparison function is called from the C fsd_search, fsd_incld, fsd_excld functions; the first field descriptor is at fld_list[1] not fld_list[0]. V2.01 The memory allocating routines now allocate memory with the owner as the job who opened/created the database. The user memory interface routines have the owner Job ID handed over in D2. Some minor bugs corrected. V2.02 DATA_BIN DBAS_BIN now display messages when called. DATA_BIN prints to QDOS channel $, and DBAS_BIN prints to channel #0 of the Basic in which it is called. DATA_BIN has a QJUMP config style configuration block for default order key length and compare case dependence, as well as for activating the database thing. All procedures in the superbasic interface now have a function version (preceded by FDB_) that returns the error code as a float. Reordering of records after appending updating now proceeds with the same algorithm as ordering, and so is alot faster. FSD.ORDR (ORDER) now preserves the select status of the records. Note, however, that unordering the database no longer releases the memory, but must be followed by a call to FSD.RES (RESET) to do so. The size of the order string keys can now be increased through use of FSD.SOKY (SORDKEY). FSD.SEL (INCLUDE EXCLUDE) now check the records in file order rather than any imposed order, and so go faster. FSD.SEL FSD.FIND-FSDF.FSE (INCLUDE, EXCLUDE, SEARCH) now can take a Contains string condition, and have a flag to NOT the result. Also AND linkages take priority over OR linkages. The record expansion routine (for dynamic records) has been rewritten, and so is now faster. Therefore FSD.FIND FSD.SEL (FIND, LOCATE, SEARCH, INCLUDE EXCLUDE) go faster, as well as record positioning exporting. OPEN_DDIR now loads the inbuilt field name table. A new database editor program has been added that requires the pointer enviroment to run. However, its source is very long, so its sources and the ALTER_BIN sources are to be submitted on another disk. V2.03 The v2.01 memory allocation improvement is now documented. The machine code test program TEST_MC2 now allocates memory with the supplied job ID. V2.10 The TRAP #4 flag is now tested cleared on SMS2 systems. A new set of vectors to manipulate subfields within string fields has been added documented (Basic commands begin SUBF_). A new pair of vectors to add and remove fields has been added (ADD_FIELD and REMOVE_FIELD in Basic). The DBADDR function has now been transfered to DBAS_BIN, instead of having its own files. The Basic function FETCH should now work properly. A new vector has been added to copy the selected records, in order, from an open database to a new database file. (COPY_DATA in Basic) ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] Fw: query
On 15/04/2012 12:27, Dilwyn Jones wrote: Have you tried a more recent version of DATA_BIN? The supplied one is so old there is no version number, now 2.13 (21/2/98). Bob I don't have a more recent version of DATA_BIN - could you maybe email me one - if it works, I will send it onto Quanta and ask that they replace the version used in the library guide. There is a v1.21 string embedded in the Library version, although I'm not sure if that refers to the library guide database itself. The boot program on my qemulator loads v2.12 of the DATA_BIN. So we know the error persists with v2.12 at least. Disk SP07 in Quanta library has version 2.13 of DATA_BIN (which I’ll send to Rich privately to see if it helps), while disk SP08 has the sources for anyone who wants to investigate the problem. This is a list of changes, although the list stops at v2.10, after that there appears to only have been sporadic updates to individual files, possibly by Phil Borman. Dilwyn CHANGES TO DATABASE HANDLING V2.00 Names handling introduced into the vectors, Basic interface, and C interface. The FSD.ORDR (ORDER) algorithm was rewritten to be alot faster. Array parameters made available to Basic procedures. When the comparison function is called from the C fsd_search, fsd_incld, fsd_excld functions; the first field descriptor is at fld_list[1] not fld_list[0]. V2.01 The memory allocating routines now allocate memory with the owner as the job who opened/created the database. The user memory interface routines have the owner Job ID handed over in D2. Some minor bugs corrected. V2.02 DATA_BIN DBAS_BIN now display messages when called. DATA_BIN prints to QDOS channel $, and DBAS_BIN prints to channel #0 of the Basic in which it is called. DATA_BIN has a QJUMP config style configuration block for default order key length and compare case dependence, as well as for activating the database thing. All procedures in the superbasic interface now have a function version (preceded by FDB_) that returns the error code as a float. Reordering of records after appending updating now proceeds with the same algorithm as ordering, and so is alot faster. FSD.ORDR (ORDER) now preserves the select status of the records. Note, however, that unordering the database no longer releases the memory, but must be followed by a call to FSD.RES (RESET) to do so. The size of the order string keys can now be increased through use of FSD.SOKY (SORDKEY). FSD.SEL (INCLUDE EXCLUDE) now check the records in file order rather than any imposed order, and so go faster. FSD.SEL FSD.FIND-FSDF.FSE (INCLUDE, EXCLUDE, SEARCH) now can take a Contains string condition, and have a flag to NOT the result. Also AND linkages take priority over OR linkages. The record expansion routine (for dynamic records) has been rewritten, and so is now faster. Therefore FSD.FIND FSD.SEL (FIND, LOCATE, SEARCH, INCLUDE EXCLUDE) go faster, as well as record positioning exporting. OPEN_DDIR now loads the inbuilt field name table. A new database editor program has been added that requires the pointer enviroment to run. However, its source is very long, so its sources and the ALTER_BIN sources are to be submitted on another disk. V2.03 The v2.01 memory allocation improvement is now documented. The machine code test program TEST_MC2 now allocates memory with the supplied job ID. V2.10 The TRAP #4 flag is now tested cleared on SMS2 systems. A new set of vectors to manipulate subfields within string fields has been added documented (Basic commands begin SUBF_). A new pair of vectors to add and remove fields has been added (ADD_FIELD and REMOVE_FIELD in Basic). The DBADDR function has now been transfered to DBAS_BIN, instead of having its own files. The Basic function FETCH should now work properly. A new vector has been added to copy the selected records, in order, from an open database to a new database file. (COPY_DATA in Basic) ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm I take it then, I am not the only one with this problem - Dilwyn also has seen it! The v2.13 of DATA_BIN does not help either unfortunately... -- Rich Mellor RWAP Services Specialist Enuuk Auction Programming Services www.rwapservices.co.uk ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] Quanta Library Guide
Have you tried a more recent version of DATA_BIN? The supplied one is so old there is no version number, now 2.13 (21/2/98). Bob Have tried v2.13 of DATA_BIN dbase handler on Q-emuLator, and got the following results: 1. works ok on version JM ROM 2. fails on version JS supplied with QemuLator 3. Fails on version JS-4M ROM (the 4MB RAM version) 4. fails on standard JS ROM from my website (just in case of outside possibility ROM supplied with QemuLator was corrupt) 5. fails also on jsu.rom 5. works OK on Minerva 1.97 Unfortunately my JS QL is boxed up in the attic and I can't get at it. Can anyone try the library guide on a JS ROM QL which then ought to help us try to ascertain if it is a JS ROM or q-emulator-with-JS-rom problem, and then pin it down to what in DBAS is happening. Dilwyn ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] Quanta Library Guide
Daniele has looked at this issue and to quote his comments: The bad parameter error is related to a file system TRAP#3 call with D0=$44. Do you know what this trap is supposed to do? Here's the Trap #3 d0=$44 notes for the DATAREF_DOC document from DBAS. Dilwyn 2) EUSING THE DATABASE TRAP To find out the database version number and the vector base address (pointer to list of vectors), a trap must be used. The trap is only available on DIRECTORY DEVICE DRIVERS (these are devices which can produce a directory, ie: respond to a DIR procedure from BASIC) that were linked in when the CALL procedure or the DATA_EXT procedure was issued. So make sure all the Device Drivers that you are going to use are installed before the package is CALLed or DATA_EXTed. DIRECTORY DEVICE DRIVERS include things like MDV, FLP, WIN, RAM, and N, but not NET, PIPE, CON, SCR, SER, and PAR. To use the package with the network file server, any machine that wishes to use it, must have the package in memory, or a crash will certainly occur. Eg: QL net no. 1 has DATA_BIN in memory, and a hard disk. QL net no. 2 has neither. If QL 2 accesses the trap (see below) on QL 1's hard disk, then the operation may be successful. If it then tries to access vectors at the address returned, a crash will almost certainly occur. However if QL 2 has DATA_BIN in memory, and QL 1 has or has not, then: All accesses by QL 2 to the package will not crash, provided no duff values (eg: Database ID) are supplied. The trap: FS.DBASE TRAP #3 D0=$44 Call parameters Return Parameters D1 D1.L Ascii version number D2 D2.L Preserved D3.W TimeoutD3.L Preserved A0.L Channel ID A0.L Preserved A1 A1.L Vector base address A2 A2.L Preserved Error return: ERR.BP Not implemented ERR.NO Channel not open ERR.NC Not complete ERR.NI Not implemented The vector base address should never change, so it should be safe to do this trap once only. ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
[Ql-Users] SMSQE java
Hi all, I've recently started thinking about doing an smsqe emulator in Java, and finally put my hands to keyboard. I use an MC68000 emulator by somebody called Tony Headford (emulator which unfortunately I had to debug quite extensively). Anyway, to cut a long story short, this is now getting to a point where I think I could show a very early pre-alpha version (note the qualifiers) to anybody interested. The state of the project as it is now: This is an adapted gold card version - mode 4 only, normal QL mode 4 screen 512x256), imperfect keyboard handling (probably no altkeys, ctrl not working correctly), no mdvs (obviously) or floppies, no win drive, but an NFA drive (Native File access) that allows you to read/write to a native drive, no mouse or sound. I've successfully run this basic prog: 10 a=RESPR(4) 20 LBYTES nfa2_C68dhrystone,a 30 SEXEC ram1_p,a,4,10 40 PRINTdone and then exec'd the program from ram1. Unfortunately, this prog wouldn't work on such a barebone machine (nor under a simple QPC machine with just these screen sizes) complaining about some res missing,I don't know why. The emulator is very slow, probably down to the speed of an original Ql. Some of that is certainly due to very imperfect way I handle screen emulation for the time being. It runs under an incorporated monitor program. If anybody is interested in alpha testing this, please drop me an email with SMSQUMULATOR as subject line (in upper case). You are expected to know how to compile SMSQE since no precompiled version comles with it. I'll email the sources as a zip file, so you will need a working SMSQE to compile the source (it involves running a simple basic prog...). the sources are about 7 MB, the emulator is about 1 MB on top of that. The thing is designed to run under Java 7 , any older version probably won't cut it. Don't expect an answer to any email today! Wolfgang ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] SMSQE java
Wolfgang Lenerz wrote: I've recently started thinking about doing an smsqe emulator in Java, and finally put my hands to keyboard. I use an MC68000 emulator by somebody called Tony Headford (emulator which unfortunately I had to debug quite extensively). Anyway, to cut a long story short, this is now getting to a point where I think I could show a very early pre-alpha version (note the qualifiers) to anybody interested. Super news! Well done, Wolfgang - my best wishes to you with this project. This tallies with something expressed by Chris Zeinstra from Groningen, Netherlands last year, who expressed an interest on QL Forum in doing a similar project in Java - see http://www.qlforum.co.uk/viewtopic.php?f=8t=170#p888 and http://www.qlforum.co.uk/viewtopic.php?f=19t=178 Interestingly he mentioned the Tony Headford emulator, plus looking into X68000, a Japanese emulator in Java, emulating a popular 68000 micro in Japan (to quote from his posting in September. Dilwyn Jones ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
[Ql-Users] FORMAT ram1_mdv2 Oddities
I have an issue 6 QL (JS ROM), and have plugged a Super Gold Card into it. It seems a good QL with good working microdrive units. However, although FORMAT ram1_mdv1 works to make an image of the cartridge in the left hand drive, FORMAT ram1_mdv2 just waits and then gives up - FORMAT failed. Does anyone have any suggestions as to what may be the problem ?? I have swapped the ROM over on the Super Gold Card, and tried a second Super Gold Card, but it still exhibits the sme issue. The only thing I have not done is removed the 68008 processor. The microdrive units respond to all standard QDOS commands without any problems.. -- Rich Mellor RWAP Services Specialist Enuuk Auction Programming Services www.rwapservices.co.uk ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] FORMAT ram1_mdv2 Oddities
As far as I know.It only works with mdv1_. - Original Message - From: Rich Mellor r...@rwapservices.co.uk To: ql-us...@q-v-d.com Sent: Sunday, April 15, 2012 6:09 PM Subject: [Ql-Users] FORMAT ram1_mdv2 Oddities I have an issue 6 QL (JS ROM), and have plugged a Super Gold Card into it. It seems a good QL with good working microdrive units. However, although FORMAT ram1_mdv1 works to make an image of the cartridge in the left hand drive, FORMAT ram1_mdv2 just waits and then gives up - FORMAT failed. Does anyone have any suggestions as to what may be the problem ?? I have swapped the ROM over on the Super Gold Card, and tried a second Super Gold Card, but it still exhibits the sme issue. The only thing I have not done is removed the 68008 processor. The microdrive units respond to all standard QDOS commands without any problems.. -- Rich Mellor RWAP Services Specialist Enuuk Auction Programming Services www.rwapservices.co.uk ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] FORMAT ram1_mdv2 Oddities
Ralf, Rich, the manual says different (From the Trump card manual): FORMAT ram1_mdv2 loads an image of mdv2 into RAM Disk 1 The RAM Disk can even load a Microdrive with a damaged directory. It cannot, however, load a Microdrive with a damaged map. Note the example - It's identical to what Rich tried to do, so it should work on drive 2, too. From what you write, Rich, I understand mdv2_ is working fine on other commands? Did you try on a newly reset machine? Cheers, Tobias Am 15.04.2012 um 20:52 schrieb Ralf Reköndt: As far as I know.It only works with mdv1_. - Original Message - From: Rich Mellor r...@rwapservices.co.uk To: ql-us...@q-v-d.com Sent: Sunday, April 15, 2012 6:09 PM Subject: [Ql-Users] FORMAT ram1_mdv2 Oddities I have an issue 6 QL (JS ROM), and have plugged a Super Gold Card into it. It seems a good QL with good working microdrive units. However, although FORMAT ram1_mdv1 works to make an image of the cartridge in the left hand drive, FORMAT ram1_mdv2 just waits and then gives up - FORMAT failed. Does anyone have any suggestions as to what may be the problem ?? I have swapped the ROM over on the Super Gold Card, and tried a second Super Gold Card, but it still exhibits the sme issue. The only thing I have not done is removed the 68008 processor. The microdrive units respond to all standard QDOS commands without any problems.. -- Rich Mellor RWAP Services Specialist Enuuk Auction Programming Services www.rwapservices.co.uk ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
Re: [Ql-Users] SMSQE java
This sounds exciting and worthwhile - is there anything the average joe can do to support/encourage your work? Dave On Sun, Apr 15, 2012 at 10:54 AM, Dilwyn Jones dil...@evans1511.fsnet.co.uk wrote: Wolfgang Lenerz wrote: I've recently started thinking about doing an smsqe emulator in Java, and finally put my hands to keyboard. I use an MC68000 emulator by somebody called Tony Headford (emulator which unfortunately I had to debug quite extensively). Anyway, to cut a long story short, this is now getting to a point where I think I could show a very early pre-alpha version (note the qualifiers) to anybody interested. Super news! Well done, Wolfgang - my best wishes to you with this project. This tallies with something expressed by Chris Zeinstra from Groningen, Netherlands last year, who expressed an interest on QL Forum in doing a similar project in Java - see http://www.qlforum.co.uk/viewtopic.php?f=8t=170#p888 and http://www.qlforum.co.uk/viewtopic.php?f=19t=178 Interestingly he mentioned the Tony Headford emulator, plus looking into X68000, a Japanese emulator in Java, emulating a popular 68000 micro in Japan (to quote from his posting in September. Dilwyn Jones ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm ___ QL-Users Mailing List http://www.q-v-d.demon.co.uk/smsqe.htm
