 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : CMM2: V5.05.06RC19: Scotty more power
  Page 2 of 4   |
|||||
| Author | Message | ||||
thwill Guru  Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
Here you go: With RC19: Colour Maximite 2 MMBasic Version 5.05.06RC19 Copyright 2011-2020 Geoff Graham Copyright 2016-2020 Peter Mather > list "new_bug.bas" Option Explicit On Option Default None test_exists() End Sub test_exists() Local f$ = Mm.Info$(Current) Print fil.exists%(f$) Print fil.exists%(fil.get_parent$(f$) + "/" + fil.get_name$(f$)) End Sub Function fil.exists%(f$) ' On Error Skip fil.exists% = LCase$(Dir$(f$, ALL)) = LCase$(fil.get_name$(f$)) End Function ' Gets the name of file/directory 'f$' minus any path information. Function fil.get_name$(f$) Local i% For i% = Len(f$) To 1 Step -1 If InStr("/\", Mid$(f$, i%, 1)) > 0 Then Exit For Next i% fil.get_name$ = Mid$(f$, i% + 1) End Function ' Gets the parent directory of 'f$', or the empty string if it does not have one. Function fil.get_parent$(f$) Local i% For i% = Len(f$) To 1 Step -1 If InStr("/\", Mid$(f$, i%, 1)) > 0 Then Exit For Next i% If i% > 0 Then fil.get_parent$ = Left$(f$, i% - 1) End Function > run "new_bug.bas" 1 Error in line 15: FIL.EXISTS is not declared And with RC15: > Colour Maximite 2 MMBasic Version 5.05.06RC15 Copyright 2011-2020 Geoff Graham Copyright 2016-2020 Peter Mather > run "new_bug.bas" 1 1 Please note you can't use Mm.Info(FileSize f$) as a general mechanism for determining the existence of a file/directory because it returns -1 for both existing directories and non-existent files. Best wishes, Tom Edited 2020-11-16 01:56 by thwill Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 8592 |
Sorry but I'm not prepared to look at it while you use constructs like fil.exists% = LCase$(Dir$(f$, ALL)) = LCase$(fil.get_name$(f$)) This is not valid MMBasic syntax If it happened to work in the past that was a complete fluke and the only bug was that it should have give an error then Edited 2020-11-16 03:33 by matherp |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
You've tried to sell me this before and I'm still not buying  Are you trying to tell me that it is invalid to set a variable equal to the result of an expression ... that sounds bananas, as well as being incorrect. The following 3 formulations for the test all produce the same error: With brackets: fil.exists% = (LCase$(Dir$(f$, ALL)) = LCase$(fil.get_name$(f$))) With NOT and <>: fil.exists% = Not (LCase$(Dir$(f$, ALL)) <> LCase$(fil.get_name$(f$))) With IF: If LCase$(Dir$(f$, ALL)) = LCase$(fil.get_name$(f$)) Then fil.exists% = 1 Else fil.exists% = 0 EndIf Best wishes, Tom Edited 2020-11-16 04:11 by thwill Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
If '=' is really not intended to behave like the other dyadic logical operators then this exception needs to be documented in big/bold letters in the manual. Generally accepted BASIC syntax for assignment is: LET var = expr "Evaluates the expression and assigns the value to the variable. The LET keyword is optional." Since: a = b is a valid expression that evaluates to 1 (or for some BASICs -1) if 'a' equals 'b' and 0 if not then: LET var = a = b should be valid syntax, and given LET is optional so should: var = a = b Which despite your protests to the contrary does in general appear to be how MMBasic behaves, IMO this bug has nothing to do with that. Sorry, Tom Edited 2020-11-16 04:27 by thwill Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 8592 |
Coukld take some time to find and fix - workround is local s$=fil.get_parent$(f$) + "/" + fil.get_name$(f$) print fil.exists%(s$) |
||||
| William Leue Guru Joined: 03/07/2020 Location: United StatesPosts: 383 |
Of course, in C and its derivatives like C++, the single-equals operator is transitive and does chain in exactly this way, and this is an explicitly legal usage. I have to confess that I have never seen assignment chaining in any BASIC and I am surprised it works here. I wonder if this is an inadvertent side effect and might be dependent on some specific implementation detail that could be changed any time. I have never seen a formal language definition for any BASIC including MMBASIC, so it makes me wonder. -Bill |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
Thanks Peter. I'd appreciate it if you could shake out the validity or not of var = a = b with Geoff. Does anyone have access to other "Microsoft BASIC" (not VB) "standard" BASICs to see whether it is widely accepted or not ? Kind regards, Tom Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| William Leue Guru Joined: 03/07/2020 Location: United StatesPosts: 383 |
Here is a link to a formal BNF language syntax definition for Dartmouth BASIC. It of course does not tell us anything about MMBasic. https://watermark.silverchair.com/15-1-37.pdf?token=AQECAHi208BE49Ooan9kkhW_Ercy7Dm3ZL_9Cf3qfKAc485ysgAAAqUwggKhBgkqhkiG9w0BBwagggKSMIICjgIBADCCAocGCSqGSIb3DQEHATAeBglghkgBZQMEAS4wEQQMDQvxDHIB-0DwSNELAgEQgIICWLRfcQv7L16kQwbrznaR_DTsm4T3yAvS8SA37s4CZAm55Y7IiADGJI_wJCRE4bET17t0SR5VMPmZxaWKp9HvkeNy4UAq2gwpNz9Zol9tKsBY0OS0_09-ZDYPQ6v2Au7yuGBIl7CF3jwJZG4gf-o5PcnGXyJdO9iKbTg3SjMjpaQExraB0Rp3YydZ9BqHmgqH-_J1yMK24lvvBRvGSv_n8i_WoifkgkmEYaK-jMMJql8PVGi-kwi1cygeOgtuWmRcYWK9VZA-YuTgTdIfICe_hcaOOFHd_8tBVbFnYhdYf3qKQm3g8N-pJnwoRkQHRWBo9u66GxBiL4JZ65iGU1U2CKbaWk1ayVGJpsFobtgRsHRLmbYTnSkLC1qhbqn5CVTWvtLsofc2tPWDO6eI9bUIYYj9Mxf5nBm1VNyzsV5NVCvdEs1DfaQpPFB0Du-D4QXkNI1siA_jvyt7HrI-8BDbUNBkOkx6InaeG1KUe5mhTSEb6WypTeCk_caZRTjtuDZBPjr5u3Vk8wsRJPRLw0_HdWS_vMfPrHhruWP6z7iRMHQmZjuRlHJ0ZOsGs1rVYGLHM9t0LHcfLGFewG1077FvkILFM591UCeBMn3FeVaz8aU-LYhb5VJut79ZJ9KdgVZrcARaYpOwsQrJtvz0x3VMlNRDAiJCZiiLLjUpfhwxbOCF1I1yOuLepWd2QtzKy9tHRaD57fY9_OkGq-fa5ufmq_1xQufJMXpzjnnw4kDJ6sHlPbyrEz6-1TwN2-OY-n2oUdjBUl-qJbfgS5ZlP3S_n2VbQuUwrMlbFg |
||||
CircuitGizmos Guru Joined: 08/09/2011 Location: United StatesPosts: 1421 |
Does: var = (a = b) change things? Micromites and Maximites! - Beginning Maximite |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
I don't believe it to be syntactically ambiguous so I would be surprised if it is not allowed. A quick check shows it to be valid syntax for BBC BASIC, C64 BASIC and Sinclair (ZX-81) BASIC. Note that in the first two cases the value of true is -1 whilst for Sinclair, like MMBasic it is 1. Feeling confident, Tom Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
Nope, as indicated in my earlier post this has the same behaviour with regards this bug, I believe is unrelated to this issue of syntax since it is reproducible using an IF THEN ELSE ENDIF formulation. I'm going to go out on a limb here and say that MMMBasic is supposed to, and in fact does support var = a = b, I've been using it enough, and that for once Peter might be mistaken. Tom Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 8592 |
But it is ambiguous What does c=7; a = b= c; do in C? |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
Hang on Bill, this is not "assignment chaining", the equivalent "C" expression for this construct is: var = a == b not: var = a = b Best wishes, Tom Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
LeoNicolas Guru Joined: 07/10/2020 Location: CanadaPosts: 445 |
in MSX Basic works fine 10 DIM X,Y,Z 20 X=1 30 Y=2 40 Z=X=Y 50 PRINT Z The result is 0 For Y=1 (equals X) the result is -1 |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
This is an example of assignment chaining and it assigns the value of 'c', i.e. 7 to 'a' and 'b'  But BASIC is not C and has different rules.By not syntactically ambiguous I meant that a BASIC parser could unambiguously determine the meaning of = from its context. If a statement starts "LET|DIM|CONST|STATIC|LOCAL var =" or "var =" (where var might be an array variable) then "=" is the assignment operator, otherwise it is the dyadic equality operator. I've cited 3 (4 if you count MMBasic) versions of BASIC where the = operator behaves as I have described and used. Do you have a counter example ? This is "fun"  ,Tom Edited 2020-11-16 05:25 by thwill Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
And MSX Basic is definitely a Microsoft BASIC such as that with which MMBasic claims compatibility ... let's not start to argue whether 1 or -1 should be the true value of true  Tom Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 5735 |
I've never seen var = a = b as valid BASIC. I'm rather surprised that you can get a result out of it at all . I'd have expected a syntax error. I must admit that I've never actually tried it. :) I can imagine it working in Sinclair BASIC (if it will let you enter it) as that has a lot of quirks (some that I love, like string slicing being valid on the LHS of the equation). var = (a = b) is perfectly valid, of course. Surely var = a = b could mean any of: 1. set var to a then set a to b 2. set both a and b equal to var 3. set var to the true/false result of (a=b) It seems very ambiguous to me. Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| thwill Guru Joined: 16/09/2019 Location: United KingdomPosts: 3848 |
And a+b could mean push 'a' onto the stack, pop the top two values of the stack and add them, push the result onto the stack and then push 'b' onto the stack, but it doesn't because we are using the rules/syntax of BASIC not RPN. When I used the word "ambiguous" I meant to the parser, not to a user unfamiliar with the language's idiosyncrasies. Can anyone cite a (non-toy) BASIC that does not support var = a = b ? Best wishes, Tom Edited 2020-11-16 06:32 by thwill Game*Mite, CMM2 Welcome Tape, Creaky old text adventures |
||||
RetroJoe Senior Member  Joined: 06/08/2020 Location: CanadaPosts: 290 |
Just checked on my Apple //e. AppleSoft BASIC also supports this syntax. ] a=7=7 ] print a 1 I don’t think it is ambiguous at all if you know that BASIC does not have an explicit assignment operator (I’ve always been partial to Pascal’s := myself ...). Whatever semantic ambiguity exists is caused by the optional use of the keyword LET. If LET is used consistently, then the code reads like English I.e. Let (this variable) take on the value of (everything that appears on the RHS of the = sign). Edited 2020-11-16 06:51 by RetroJoe Enjoy Every Sandwich / Joe P. |
||||
TassyJim Guru Joined: 07/08/2011 Location: AustraliaPosts: 5911 |
var = a = b The LET is only implied for the first equals sign. LET var = a = b The second equals sign is going to be an equality test. To make it clear and to make sure that you don't get caught by order of precedence, use brackets. var = (a = b) This all had nothing to do with the bug you found and only served as a distraction. would have been a better example to show the bug. Jim VK7JH MMedit  MMBasic Help |
||||
| Page 2 of 4 |
|||||
 Print this page |
