home *** CD-ROM | disk | FTP | other *** search
- all
- H#: 61555 S12/SPECTRA Publishing
- 20-Aug-90 12:02:00
- Sb: #61280-VERSION NUMBERS
- Fm: Mason Landstreet 71220,1547
- To: Bob Zale: PowerBASIC R&D 76304,1303 (X)
-
- Bob,
-
- Set up the code attheend to reflect a 3 digit number up to 512. This would
- allow you to bit map the ptches for up to 9 patches (a 5 digit number would
- take you up to 16 or so). This would show the patches and would allow a quick
- check for prerequisite patches.
-
- Mason
-
-
- Press <CR> for next or type CHOICES !
- H#: 61536 S12/SPECTRA Publishing
- 20-Aug-90 10:07:47
- Sb: #61503-EXTERNAL STATEMENT
- Fm: George Green 70565,276
- To: Bob Zale: PowerBASIC R&D 76304,1303 (X)
-
- <SIGH>...OK, THANKS!
- -George \ex es
- just washed my hands, can't do a thing with them...
-
- Press <CR> for next or type CHOICES !
- H#: 61630 S12/SPECTRA Publishing
- 20-Aug-90 20:03:51
- Sb: #61503-#EXTERNAL STATEMENT
- Fm: Dan Robins 70007,3264
- To: Bob Zale: PowerBASIC R&D 76304,1303 (X)
-
- Bob,
- One quick question for you....
- Why is it you can't name a .PBU file the same as the .PBU call itself:
-
- I call the .PBU file DAN.PBU
- I would like to call the SUB which has been made into a unit the same,
- ie... SUB DAN (blah,blah,blah,blah)
- .....
- END SUB
-
- You also get the same duplicate name error at the $LINK "DAN.PBU" in a
- program which attempts to link in the .PBU.
- Any thoughts or revelations about this?
-
- Dan
-
- There is 1 Reply.
-
- Press <CR> for next or type CHOICES !
- H#: 61684 S12/SPECTRA Publishing
- 21-Aug-90 07:38:08
- Sb: #61630-EXTERNAL STATEMENT
- Fm: Bob Zale: PowerBASIC R&D 76304,1303
- To: Dan Robins 70007,3264
-
- True, Dan, you can't name a unit file the same as a procedure. I'm afraid
- there's no "grand" explanation for this, just the way that the symbol table
- works, internal to the compiler. We keep track of each procedure by the module
- (unit) name, as well as the procedure name itself. This process requires that
- all names be unique, by design, rather than by some other hidden reason.
- Obviously, this could therefore be changed, although at the possible expense of
- some extra code, and maybe some efficiency. But if you folks need that change,
- well, let us know . . .
- Anyone care to "vote" on this issue ? ? ?
- Bob Zale (PowerBASIC R&D)
-
- Press <CR> for next or type CHOICES !
- H#: 61558 S12/SPECTRA Publishing
- 20-Aug-90 12:25:36
- Sb: #ARRAY confusion
- Fm: George Green 70565,276
- To: [F] Spectra Support 75300,214 (X)
-
- OK... Here's another problem I've come up with in my quest to convert over to
- PB units.
-
- I have one particular SUB that works fine as a simple call, but when put in the
- UNIT gives error 488..."Array is Dynamic".
-
- The array in question is DIMed using STATIC eg DIM STATIC PROBLEM (20). It's
- specified as PUBLIC in the main program and as EXTERNAL in the UNIT It's
- cleared often using ERASE in the main program.
-
- Other than the above, the only other times it appears in the program are to
- assign its elements values, nowhere is it re-dimentioned.
-
- Why does PB now think it's a DYNAMIC array?
-
- -George Green
-
-
-
- There is 1 Reply.
-
- Press <CR> for next or type CHOICES !
- H#: 61576 S12/SPECTRA Publishing
- 20-Aug-90 14:44:58
- Sb: #61558-#ARRAY confusion
- Fm: George Green 70565,276
- To: George Green 70565,276 (X)
-
- ..Answering my own question?
-
- It turns out that the arrays in question are assigned values via the UserEnt
- sub routine from the (fixed) TURBO BASIC DB TOOLKIT. Could it be that this sub
- is causing the problem? Anyway, I converted the arrays to DYNAMIC and the ERASE
- command to REDIM. This works, but doesn't tell me what the original problem
- was. BTW, putting the STATIC command in the SUB within the UNIT didn't help.
-
- -George Green \ex
-
- There is 1 Reply.
-
- Press <CR> for next or type CHOICES !
- H#: 61587 S12/SPECTRA Publishing
- 20-Aug-90 16:29:34
- Sb: #61576-ARRAY confusion
- Fm: Bob Zale: PowerBASIC R&D 76304,1303
- To: George Green 70565,276 (X)
-
- George--
- The answer is simply that all arrays which are declared PUBLIC and/or
- EXTERNAL must be dynamic by definition. The reason is simply this: when
- compiling other modules, the compiler would have no way of knowing whether the
- array was dynamic or static. In the example you gave, the unit file contains
- the ONLY reference anywhere as to whether the array is static or dynamic. When
- PB is compiling the main module, and encounters the ERASE statement, it has no
- way of knowing which type of array it is. Remember, ERASE is valid for both
- types of arrays, but the action taken is VERY different.
- Bob Zale (PowerBASIC R&D)
-
- Press <CR> for next or type CHOICES !
- H#: 61622 S12/SPECTRA Publishing
- 20-Aug-90 19:28:20
- Sb: #VERSION NUMBERS
- Fm: AL MUSELLA 76114,637
- To: Bob Zale: PowerBASIC R&D 76304,1303 (X)
-
- Bob,
- I can only reproduce the problem with a large program. The
- individual modules work correctly, but when I put them together, it
- won't compile. I am getting an error 462 message on the wrong line - and
- possibly it may be the wrong error message. The editor (and PBC) tells me
- that the error is on line 86 of a module that is only 17 lines long!. I didn't
- locate this bug yet, but I know from experiance that it is something simple,
- if PB tells me the correct line.
- The source code (including .pbus) is rather large (460K unzipped, or 150K
- after using PKZIP) and it would be easier if I could send you a
- 5 1/4" diskette. Just let me know where to send it.
- Maybe the problem is just the size of the files, and I should just be
- getting an out of memory error message instead of an error 462? - but PB and
- PBC both put me on the same line - and shouldn't run out of space on the same
- line, should they?
- Al
-
- BTW, why would anyone only use PATCH #3 without previous patches?
- You should set it up so that a patch would only work if all previous patches
- were applied. That way, everyone has the same version. The most likely reason
- for not applying a patch is that the person doesn't know about it, or forgets
- to apply the patch. Stop us from making this mistake!!
-
-
- There is 1 Reply.
-
- Press <CR> for next or type CHOICES !
- H#: 61694 S12/SPECTRA Publishing
- 21-Aug-90 07:50:45
- Sb: #61622-VERSION NUMBERS
- Fm: Bob Zale: PowerBASIC R&D 76304,1303
- To: AL MUSELLA 76114,637
-
- Al --
- Based upon the error messages, I agree it seems you are seeing a probable
- "bug" in PB. I'd love to take a look at the code to see if there is some
- explanation for it, or else to get it corrected asap. I'm sending you my
- mailing address by e-mail, so I'll look forward to getting your disk.
- I agree, I can't imagine any reason why a person would want to install ONLY
- Patch3, but I suppose it's possible. Perhaps they need PATCH3 at once for a
- programming project, but don't have the others readily available. Should we
- "put them out of business" by refusing to run? I don't know for sure. Also
- we've considered the possibility that we might have to issue some patches ONLY
- for specific hardware. If PB didn't run properly on an ACME computer, we might
- need to issue a patch only for ACME users. What then for the non-ACME folks?
- I understand your concerns, but it a difficult choice.
- Bob Zale (PowerBASIC R&D)
-
- Press <CR> for next or type CHOICES !
- H#: 61640 S12/SPECTRA Publishing
- 20-Aug-90 21:02:35
- Sb: UTILITY UPLOAD
- Fm: AL MUSELLA 76114,637
- To: ALL
-
- Hi,
- Did you ever have the problem of hundreds of files accumulating
- in your PB directory, not knowing which ones were important or current? I
- wrote and uploaded to the library a utility which, when supplied with the main
- filename, will use PKZIP (or you can substitute your favorite compression
- utility) to combine all of the files needed to compile the main file. It will
- also optionally print out a list of which $include and $link files are present
- in your program.
- You can ZIP the important files, then erase the rest(make sure you
- do a complete backup first, incase a mistake is made!!!!)
- The program also includes source code for the dos find first and find next
- functions.
- If you have any questions or comments, let me know.
- If anyone is interested, I am writing a MAKE utility, which will
- insure that all $link files in a program were compilied after the source was
- last changed, and if not current, will automatically compile them. If anyone
- wants it I will be glad to upload this also.
-
- Al
-