TB_SpellChecking-Utility

1 Attachment(s)

I made a new spellchecker to check thinBasic-Files for bugs and typos - (the old one was pretty slow)
- but this time I avoided the use of String-methods as Len and Mid$ etc... :)

MinVersion 1.9.6.0 or 1.9.7.0 - when came the drag-in to TBGL?
It works the following:

Run the script from thinAir,
drag any .tBasic, .tBasicC, .tBasicU or .inc -files from windows-explorer or from desktop into the spellchecker-window. Check starts immediately and stops if some serious problem in script structure is found.

It checks for #IF+ If + Select + Do/Loop + While/Wend + Repeat/Until + For/Next + Type + Union + Sub/Function-nesting levels, also checks for even parenthesis & quotes, and for typos and a few things more. So if some expression is used just once per script it will inform about some dubious expression that you might to check back for.
It will inform you, if found anything - click on the information to display the codeline then for further examinations.

Only one problem to which I have no solution yet: It complains uneven paranthesis if the script uses line-continuation without underscore.

_____________________________________________________________________________________

Edit: @Eros
(who else could answer this question?)

Is there a way to add in some user-defined function or utility as this to thinAir? So I could instantly run the spellcheck from within thinAir on the current active scripts text without cumbersome actions as save script, change to spellchecker, start it, (search folder of script to check) dragging that there - checking here and then back to thinAir for corrections

- I imagine like some menuitem on Tools\Utilities\"User defined" which can be choosen through thinAir-options - set User-defined-utility-Path + Command$ (OS_GetCommands) there then thinAir "knows" if to pass script-filename & path or plain text-string of current script or whatever (list of all current opened scriptnames, crlf-delimited?) as command to the user-utility...:talk38:... ok, I would rewrite it, so you pass some tB-script as String to the function and get some $crlf-delimited error-report back?

Rene,

Do you have a link to an unofficial 1.9.7.0?

I wont be able to checkout your spell-checker
until later today. Plus, I want to study your
script to see how you did it without Len and
MID$.

Bill

No sorry Billbo, I don't have a link to 1.9.7. I don't even know if it exists :oops: - perhaps the latest TBGL-Package can help - but I don't know for sure...

Anyway- I made another file for demonstration-purpose which runs without TBGL - uses Console & File only - but not very comfortable since this is supposed to run only once on some filename that you type into the code before running it. This is intended to demonstrate

Code:

Function SpellCheck_Proceed(ByVal sScript As String) As String

where sScript is String of Script-content and result-string is the Error-report - small TBMain-function explains usage

Code:

Uses "FILE" ' needed to load keywords... Uses "CONSOLE" ' for demonstration-purpose '[] top of script Begin Const %await_NONE %await_EndFunction %await_EndSub %await_EndWith %await_EndType %await_EndUnion %await_DPlusENDIF ' = "#ENDIF" %await_EndIf %await_Case %await_EndSelect %await_Next %await_Wend %await_Loop %await_Until End Const Dim sKeyword() As String ' array of tB-keywords Dim sCodeLine() As String ' read-in codelines (change by check in memory) Dim sErrors As String Dim sExpression() As String ' store expressions Dim lExpression() As Long ' line of expression Dim nExpressions As Long ' # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # Function TBMain() ' change sFile to any you want... Local sFile As String = APP_Path + "SampleScripts\General\ByRef\Byref_Array_UDT.tBasic" Local myErrorReport As String = SpellCheck_Proceed(FILE_Load(sFile)) If StrPtrLen(StrPtr(myErrorReport)) Then Local sOut() As String Local nLines, i As Long nLines = Parse( myErrorReport, sOut, $CRLF) PrintL "found " + Str$(nLines - 1) + " messages:" PrintL For i = 1 To nLines - 1 PrintL "Line-Number " + Val("&H" + Peek$(StrPtr(sOut(i)), 8)) + " reports:" PrintL Peek$(StrPtr(sOut(i)) + 8, StrPtrLen(StrPtr(sOut(i))) - 8 ) PrintL Next If Val("&H" + Peek$(StrPtr(sOut(nLines)), 8)) = -1 Then PrintL "Check aborted because of script-structure-problem" Else PrintL "Check finished" EndIf Else PrintL "No problems found" EndIf WaitKey End Function ' ------------------------------------------------------------------------------ Function SpellCheck_Proceed(ByVal sScript As String) As String ' pass some tB-Script and receive Error-String delimited by $CRLF ' if no errors found returns empty string ' every single error-textline starts with 8 Bytes Hex$ of the scriptline-number ' (these bytes are represented here in comment-text as "§") ' somewhat like "§§§§§§§§ErrorTextHere" + $crlf ' at the end get 8 Bytes more appended -only if any error found at all- ' and mean ' -1 check has been stopped through script-structure-problem ' 0 check ran through until end ' two errors would look now like this ' "§§§§§§§§ErrorText1" + $crlf + "§§§§§§§§ErrorText2" + $crlf + Hex$(a 0 or -1 here, 8) ' first get keywords... If Not FILE_Exists(APP_Path + "thinAir\Syntax\thinBasic\thinBasic_Keywords.ini") Then MsgBox(0, "Can not find thinBasic_Keywords.ini", %MB_OK , "Unable to run") Return "" EndIf Local lKeywords As String = Ucase$(FILE_Load(APP_Path + "thinAir\Syntax\thinBasic\thinBasic_Keywords.ini")) If Not Parse(lKeywords, sKeyword, $CRLF) Then MsgBox(0, "thinBasic_Keywords.ini seems to be corrupted", %MB_OK , "Unable to run") Return "" Endif ' parse the script to lines, all ucase$ ReDim sCodeLine(1) sScript = Ucase$(sScript) Local nLines As Long = Parse(sScript, sCodeline, $CRLF) If nLines < 1 Then Return "" Local cl, lLen As Long ' go through all lines, remove Rem etc. sort numerals out ' combine them to actual lines if line-continuation etc. For cl = 1 To nLines Check_Codelines(cl) Next ' next step is checking "If/Elseif" followed by "Then" ' every "Sub" has an "End Sub" etc. ' looks quite confusing much here, but is not that wild... For cl = 1 To nLines lLen = StrPtrLen(StrPtr(sCodeline(cl))) If lLen > 1 Then If Peek$(StrPtr(sCodeline(cl)), 2) = "IF" Then Check_Then(cl, "IF") Else If lLen > 2 Then If Peek$(StrPtr(sCodeline(cl)), 3) = "SUB" Then Check_End(cl, "SUB") Else If lLen > 3 Then If Peek$(StrPtr(sCodeline(cl)), 4) = "WITH" Then Check_End(cl, "WITH") ElseIf Peek$(StrPtr(sCodeline(cl)), 4) = "TYPE" Then Check_End(cl, "TYPE") Else If lLen > 4 Then If Peek$(StrPtr(sCodeline(cl)), 5) = "UNION" Then Check_End(cl, "UNION") Else If lLen > 5 Then If Peek$(StrPtr(sCodeline(cl)), 6) = "ELSEIF" Then Check_Then(cl, "ELSEIF") ElseIf Peek$(StrPtr(sCodeline(cl)), 6) = "SELECT" Then If lLen >= 11 Then If Peek$(StrPtr(sCodeline(cl)), 11) <> "SELECT CASE" Then sErrors += Hex$(cl, 8) + "SELECT missing: CASE" + $CRLF Else sErrors += Hex$(cl, 8) + "SELECT missing: CASE" + $CRLF EndIf Else If lLen > 7 Then If Peek$(StrPtr(sCodeline(cl)), 8) = "FUNCTION" Then Check_End(cl, "FUNCTION") Else If lLen >= 11 Then If Peek$(StrPtr(sCodeline(cl)), 11) = "BEGIN CONST" Then If lLen = 11 Then sCodeLine(cl) += " " Check_End(cl, "CONST") sCodeLine(cl) = TrimFull$(sCodeline(cl)) Else If lLen >= 15 Then If Peek$(StrPtr(sCodeline(cl)), 15) = "BEGIN CONTROLID" Then If lLen = 15 Then sCodeLine(cl) += " " Check_End(cl, "CONTROLID") sCodeLine(cl) = TrimFull$(sCodeline(cl)) Else If lLen > 16 Then If Peek$(StrPtr(sCodeline(cl)), 17) = "CALLBACK FUNCTION" Then Check_End(cl, "FUNCTION") EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf EndIf Next ' now check nesting-levels, sort out known keywords, binary (&B) and hex (&H) For cl = 1 To nLines If StrPtrLen(StrPtr(sCodeline(cl))) > 1 Then If Not Check_Loops(cl) Then ' structural problem, makes no sense further checking sErrors += Hex$(-1, 8) Return sErrors EndIf EndIf Next ' now check all collected unknown expressions ' if the long lExpression(x) = 0 then this ' sExpression(x) has been found more than once already If nExpressions Then For cl = 1 To nExpressions If lExpression(cl) <> 0 Then sErrors += Hex$(lExpression(cl),8) + "dubious expression: " + sExpression(cl) + $CRLF Next EndIf If StrPtrLen(StrPtr(sErrors)) Then sErrors += Hex$(0, 8) Return sErrors End Function ' ------------------------------------------------------------------------------ Sub Check_Codelines(ByVal atLine As Long) Static i, nChars, lParenthesis, lStart As Long Static bChar, bSpace As Byte Static sBuffer As String Static inQuote, bContinueRemark, bContinueLine As Boolean sCodeLine(atLine) = TrimFull$(sCodeLine(atLine)) If StrPtrLen(StrPtr(sCodeLine(atLine))) < 1 Then bContinueLine = FALSE lStart = 0 Exit Sub EndIf bSpace = 1 inQuote = FALSE sBuffer = String$(StrPtrLen(sCodeLine(atLine)), 32) nChars = 0 If Not lStart Then lStart = atLine For i = 1 To StrPtrLen(StrPtr(sCodeline(atLine))) bChar = Peek(StrPtr(sCodeLine(atLine)) + i - 1) If bContinueRemark Then If bChar = 43 Then ' */ If i < StrPtrLen(StrPtr(sCodeline(atLine))) Then If Peek(StrPtr(sCodeline(atLine)) + i) = 47 Then bContinueRemark = FALSE EndIf EndIf Else If bChar = 34 Then inQuote = inQuote XOR TRUE If Not inQuote Then If Not bSpace Then bContinueLine = FALSE Select Case bChar Case 32 If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, 32) nChars += 1 bSpace = 1 EndIf Case 34 Case 38 ' & If i < StrPtrLen(StrPtr(sCodeLine(atLine))) Then Select Case Peek(StrPtr(sCodeline(atLine))+i) Case 66, 72 ' &B, &H Poke(StrPtr(sBuffer) + nChars, 38) nChars += 1 bSpace = 0 End Select EndIf Case 39 ' ' Exit For Case 40 ' ( lParenthesis += 1 If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, 32) nChars += 1 bSpace = 1 EndIf Case 41 ' ) lParenthesis -= 1 If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, 32) nChars += 1 bSpace = 1 EndIf Case 42 To 46 ' * + , - . If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, 32) nChars += 1 bSpace = 1 EndIf Case 47 ' /* If i < StrPtrLen(StrPtr(sCodeline(atLine))) Then If Peek(StrPtr(sCodeline(atLine)) + i) = 43 Then bContinueRemark = TRUE EndIf EndIf If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, 32) nChars + 1 bSpace = 1 EndIf Case 48 To 57 ' 0 to 9 If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, bChar) nChars += 1 bSpace = 0 EndIf Case 58 To 60 ' : ; < If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, 32) nChars += 1 bSpace = 1 EndIf Case 61 ' = If nChars >= 8 + bSpace Then If Peek$(StrPtr(sBuffer) + nChars - 8 - bSpace, 8) = "FUNCTION" Then ' remove "Function =" If nChars = 8 + bSpace Then sBuffer = "" nChars = 0 Else If Peek(StrPtr(sBuffer) + nChars - 9 - bSpace) = 32 Then nChars -= 8 + bSpace sBuffer = Peek$(StrPtr(sBuffer), nChars) EndIf EndIf Else If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, 32) nChars += 1 bSpace = 1 EndIf EndIf Else If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, 32) nChars += 1 bSpace = 1 EndIf EndIf Case 62, 92 ' > \ If bSpace = 0 Then Poke(StrPtr(sBuffer) + nChars, 32) nChars += 1 bSpace = 1 EndIf Case 73 ' I - combine "End" + "If" to "EndIf" If i < StrPtrLen(StrPtr(sCodeline(atLine))) And bSpace = 1 And nChars >= 4 And TRUE Then If Peek(StrPtr(sCodeline(atLine)) + i ) = 70 Then 'F If nChars = 4 Then If Peek$(StrPtr(sBuffer), 4) = "END " Then nChars -= 1 Else If Peek$(StrPtr(sBuffer) + nChars - 5, 5) = " END " Then nChars -= 1 EndIf EndIf EndIf EndIf Poke(StrPtr(sBuffer) + nChars, bChar) nChars += 1 bSpace = 0 Case 82 ' R (REM) if bSpace then If i + 2 <= StrPtrLen(StrPtr(sCodeline(atLine))) Then If Peek(StrPtr(sCodeline(atLine)) + i) = 69 Then If Peek(StrPtr(sCodeline(atLine))+ i + 1) = 77 Then If i + 2 = StrPtrLen(StrPtr(sCodeline(atLine))) Then Exit For If Peek(StrPtr(sCodeline(atLine))+ i + 2) = 32 Then Exit For EndIf EndIf EndIf Endif Poke(StrPtr(sBuffer) + nChars, bChar) nChars += 1 bSpace = 0 Case 95 ' _ If bSpace Then If i = StrPtrLen(StrPtr(sCodeline(atLine))) Then bContinueLine = TRUE Else If Peek(StrPtr(sCodeline(atLine)) + i) = 32 Then bContinueLine = TRUE Else Poke(StrPtr(sBuffer) + nChars, bChar) nChars += 1 bSpace = 0 EndIf EndIf Else Poke(StrPtr(sBuffer) + nChars, bChar) nChars += 1 bSpace = 0 EndIf Case Else Poke(StrPtr(sBuffer) + nChars, bChar) nChars += 1 bSpace = 0 End Select EndIf EndIf Next If nChars Then If lStart = atLine Then sCodeLine(lStart) = TrimFull$(Peek$(StrPtr(sBuffer), nChars)) Else sCodeLine(lStart) += " " + TrimFull$(Peek$(StrPtr(sBuffer), nChars)) sCodeline(atLine) = "" EndIf Else sCodeLine(atLine) = "" bContinueLine = FALSE EndIf If Not bContinueLine Then If inQuote Then sErrors += Hex$(lStart, 8) + "Missing quotes" + $CRLF If lParenthesis < 0 Then sErrors += Hex$(lStart, 8) + "Parenthesis - missing (" + $CRLF ElseIf lParenthesis > 0 Then sErrors += Hex$(lStart, 8) + "Parenthesis - missing )" + $CRLF EndIf lParenthesis = 0 lStart = 0 EndIf End Sub ' - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ' - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Sub Check_Then(ByVal Start As Long, ByVal Keyword As String) Local sArg() As String Static i, j, nArgs As Long If StrPtrLen(StrPtr(sCodeline(Start))) > StrPtrLen(StrPtr(Keyword)) Then ' not followed by a space is not the keyword we search for If Peek(StrPtr(sCodeline(Start)) + StrPtrLen(StrPtr(Keyword))) <> 32 Then Exit Sub EndIf If StrPtrLen(StrPtr(sCodeline(Start))) < StrPtrLen(StrPtr(Keyword)) + 5 Then sErrors += Hex$(Start, 8) + Keyword + " missing: THEN" + $CRLF Exit Sub EndIf nArgs = Parse(sCodeline(Start), sArg, " ") If nArgs < 2 Then sErrors += Hex$(lStart, 8) + Keyword + " missing: THEN" + $CRLF Exit Sub EndIf i = 0 : j = 0 Do i = Array Scan sArg(1+i), Collate Ucase, = "THEN" If i Then j += 1 Loop Until i = 0 Or i = nArgs If Keyword = "IF" And sArg(nArgs) <> "THEN" And TRUE Then 'remove one-line if sCodeline(Start) = "" For i = 2 To nArgs If sArg(i) <> "THEN" Then If sArg(i) <> "ELSE" Then If sArg(i) <> "ENDIF" Then sCodeline(Start) += sArg(i) If i < nArgs Then sCodeline(Start) += " " EndIf EndIf EndIf Next i EndIf If j < 1 Then sErrors += Hex$(Start, 8) + Keyword + " missing: THEN" + $CRLF ElseIf j > 1 Then sErrors += Hex$(Start, 8) + Keyword + " redundant: THEN" + $CRLF EndIf End Sub ' - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Sub Check_End(ByVal Start As Long, ByVal Keyword As String) Static i, lsLen, lcLen As Long Static sSearch As String If StrPtrLen(StrPtr(sCodeline(Start))) > StrPtrLen(StrPtr(Keyword)) Then ' not followed by a space is not the keyword we search for If Peek(StrPtr(sCodeline(Start)) + StrPtrLen(StrPtr(Keyword))) <> 32 Then Exit Sub EndIf If StrPtrLen(StrPtr(sCodeline(Start))) < StrPtrLen(StrPtr(Keyword)) + 2 Then sErrors += Hex$(Start, 8) + Keyword + " missing: identity" + $CRLF Exit Sub EndIf sSearch = "END " + Keyword lsLen = StrPtrLen(StrPtr(sSearch)) For i = Start To UBound(sCodeline) lcLen = StrPtrLen(StrPtr(sCodeline(i))) If lcLen >= lsLen Then If Peek$(StrPtr(sCodeline(i)) + lcLen - lsLen, lsLen) = sSearch Then Exit Sub EndIf If i > Start Then If lcLen > 3 Then If Peek$(StrPtr(sCodeline(i)), 4) = "SUB " Then Exit For If lcLen > 8 Then If Peek$(StrPtr(sCodeline(i)), 9) = "FUNCTION " Then Exit For If lcLen > 17 Then If Peek$(StrPtr(sCodeline(i)), 18) = "CALLBACK FUNCTION " Then Exit For EndIf EndIf EndIf EndIf Next sErrors += Hex$(Start, 8) + "missing: END " + Keyword + $CRLF End Sub ' - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ' - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Function Check_Loops(ByVal Start As Long) As Boolean Local sArg() As String Static i, j, nArgs, lAwait As Long Static lError, lFound As Long Static bFound As Boolean Static sAwait, sFound As String If sAwait = "" Then sAwait = MKL$(%await_NONE) ' returns false if structural problem - ' unable to find the right loop then, has to stop checking nArgs = Parse(sCodeline(Start), sArg, " ") If nArgs < 1 Then Return TRUE lError = 0 For i = 1 To nArgs lAwait = Peek(Long, StrPtr(sAwait) + StrPtrLen(StrPtr(sAwait)) - SizeOf(Long)) Select Case sArg(i) Case "EXIT" If nArgs > i Then Select Case sArg(i + 1) Case "FOR", "DO", "WHILE", "SUB", "FUNCTION" sArg(i + 1) = "" End Select EndIf Case "#IF" sAwait += MKL$(%await_DPlusENDIF) Case "#ELSE", "#ELSEIF" If lAwait <> %await_DPlusENDIF Then sFound = sArg(i) lError = 1 Exit For EndIf Case "#ENDIF" If lAwait = %await_DPlusENDIF Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) Else sFound = sArg(i) lError = 1 Exit For EndIf Case "END" If nArgs > i Then Select Case sArg(i + 1) Case "SUB", "FUNCTION" If lAwait = %await_EndSub Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) sArg(i+1) = "" ElseIf lAwait = %await_EndFunction Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) sArg(i+1) = "" Else lError = 1 sFound = "END " + sArg(i + 1) Exit For EndIf If i + 1 < nArgs Then sErrors += Hex$(Start, 8) + "expected: End of Line" + $CRLF Case "SELECT" If lAwait = %await_EndSelect Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) sArg(i+1) = "" Else lError = 1 sFound = "END SELECT" Exit For EndIf If i + 1 < nArgs Then sErrors += Hex$(Start, 8) + "expected: End of Line" + $CRLF Case "WITH" If lAwait = %await_EndWith Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) sArg(i+1) = "" Else lError = 1 sFound = "END WITH" Exit For EndIf If i + 1 < nArgs Then sErrors += Hex$(Start, 8) + "expected: End of Line" + $CRLF Case "TYPE" If lAwait = %await_EndType Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) sArg(i+1) = "" Else lError = 1 sFound = "END TYPE" Exit For EndIf If i + 1 < nArgs Then sErrors += Hex$(Start, 8) + "expected: End of Line" + $CRLF Case "UNION" If lAwait = %await_EndUnion Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) sArg(i+1) = "" Else lError = 1 sFound = "END UNION" Exit For EndIf If i + 1 < nArgs Then sErrors += Hex$(Start, 8) + "expected: End of Line" + $CRLF End Select Else sErrors += Hex$(Start, 8) + "END missing: identifier" + $CRLF EndIf Case "IF" If i < nArgs Then For j = i + 1 To nArgs If sArg(j) = "THEN" Then If j = nArgs Then sAwait += MKL$(%await_EndIf) Exit For EndIf Next j EndIf Case "ELSE", "ELSEIF" If lAwait <> %await_EndIf Then lError = 1 sFound = sArg(i) Exit For EndIf Case "ENDIF" If lAwait = %await_EndIf Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) Else lError = 1 sFound = sArg(i) Exit For EndIf Case "SUB" sAwait += MKL$(%await_EndSub) Case "FUNCTION" sAwait += MKL$(%await_EndFunction) Case "CALLBACK" If nArgs > 1 Then If sArg(2) = "FUNCTION" Then sAwait += MKL$(%await_EndFunction) sArg(2) = "" EndIf EndIf Case "DO" sAwait += MKL$(%await_Loop) If nArgs > i Then Select Case sArg(i+1) Case "WHILE", "UNTIL" sArg(i+1) = "" End Select EndIf Case "SELECT" sAwait += MKL$(%await_Case) If nArgs > i Then If sArg(i+1) = "CASE" Then sArg(i+1) = "" EndIf Case "CASE" If lAwait = %await_Case Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) + MKL$(%await_EndSelect) ElseIf lAwait = %await_EndSelect Then If nArgs > i Then If sArg(i+1) = "ELSE" Then sArg(i+1) = "" EndIf Else lError = 1 sFound = sArg(i) Exit For EndIf Case "LOOP" If lAwait = %await_Loop Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) Else lError = 1 sFound = sArg(i) Exit For EndIf If nArgs > i Then Select Case sArg(i+1) Case "WHILE", "UNTIL" sArg(i+1) = "" End Select EndIf Case "WHILE" sAwait += MKL$(%await_Wend) Case "WEND" If lAwait = %await_Wend Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) Else lError = 1 sFound = sArg(i) Exit For EndIf Case "REPEAT" sAwait += MKL$(%await_Until) Case "UNTIL" If lAwait = %await_Until Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) Else lError = 1 sFound = sArg(i) Exit For EndIf Case "FOR" lError = 300 If i >= 3 Then For j = 1 To i - 1 If sArg(j) = "ARRAY" Then lError -= 210 If sArg(j) = "SCAN" Then lError -= 90 If lError = 0 Then Exit For Next j EndIf Select Case lError Case 0 ' ignore array scan Case 300 ' for-next sAwait += MKL$(%await_Next) lError = 0 Case Else sErrors += Hex$(Start, 8) + "ARRAY SCAN ?" + $CRLF End Select Case "NEXT" If lAwait = %await_Next Then sAwait = Peek$( StrPtr(sAwait), StrPtrLen(StrPtr(sAwait)) - SizeOf(Long) ) Else lError = 1 sFound = sArg(i) Exit For EndIf Case "WITH" sAwait += MKL$(%await_EndWith) Case "TYPE" sAwait += MKL$(%await_EndType) Case "UNION" sAwait += MKL$(%await_EndUnion) Case Else ' collect unknown expressions If StrPtrLen(StrPtr(sArg(i))) Then bFound = FALSE If Peek(StrPtr(sArg(i))) = 38 Then If Peek(StrPtr(sArg(i))+1) = 66 Then '&B bFound = is_Binary(sArg(i)) ElseIf Peek(StrPtr(sArg(i))+1) = 72 Then '&H bFound = is_Hex(sArg(i)) EndIf EndIf If Not bFound Then If Array Scan sKeyword, Collate Ucase, = sArg(i) Then bFound = TRUE Else If nExpressions Then lFound = Array Scan sExpression, Collate Ucase, = sArg(i) If lFound Then lExpression(lFound) = 0 bFound = TRUE EndIf EndIf EndIf EndIf If Not bFound Then nExpressions += 1 If nExpressions > UBound(lExpression) Then ReDim Preserve lExpression(nExpressions * 2) ReDim Preserve sExpression(nExpressions * 2) EndIf sExpression(nExpressions) = sArg(i) lExpression(nExpressions) = Start EndIf EndIf End Select Next If lError = 0 Then Return TRUE Select Case lAwait Case %await_NONE sErrors += Hex$(Start, 8) + "expected none, but found " + sFound + $CRLF Case %await_EndFunction sErrors += Hex$(Start, 8) + "expected END FUNCTION, but found " + sFound + $CRLF Case %await_EndSub sErrors += Hex$(Start, 8) + "expected END SUB, but found " + sFound + $CRLF Case %await_EndWith sErrors += Hex$(Start, 8) + "expected END WITH, but found " + sFound + $CRLF Case %await_EndType sErrors += Hex$(Start, 8) + "expected END TYPE, but found " + sFound + $CRLF Case %await_EndUnion sErrors += Hex$(Start, 8) + "expected END UNION, but found " + sFound + $CRLF Case %await_DPlusENDIF sErrors += Hex$(Start, 8) + "expected #ENDIF, but found " + sFound + $CRLF Case %await_EndIf sErrors += Hex$(Start, 8) + "expected ENDIF, but found " + sFound + $CRLF Case %await_Case sErrors += Hex$(Start, 8) + "expected CASE, but found " + sFound + $CRLF Case %await_EndSelect sErrors += Hex$(Start, 8) + "expected END SELECT, but found " + sFound + $CRLF Case %await_Next sErrors += Hex$(Start, 8) + "expected NEXT, but found " + sFound + $CRLF Case %await_Wend sErrors += Hex$(Start, 8) + "expected WEND, but found " + sFound + $CRLF Case %await_Loop sErrors += Hex$(Start, 8) + "expected LOOP, but found " + sFound + $CRLF Case %await_Until sErrors += Hex$(Start, 8) + "expected UNTIL, but found " + sFound + $CRLF End Select End Function ' - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ' following two functions will help sort out binary and hex-expressions ' called from Check_Loops-function Function is_Binary(ByVal sString As String) As Boolean If StrPtrLen(StrPtr(sString)) < 3 Then Return FALSE Local i As Long For i = 3 To StrPtrLen(StrPtr(sString)) Select Case Peek(StrPtr(sString) + i - 1) Case 48, 49 Case Else Return FALSE End Select Next Function = TRUE End Function ' - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Function is_Hex(ByVal sString As String) As Boolean If StrPtrLen(StrPtr(sString)) < 3 Then Return FALSE Local i As Long For i = 3 To StrPtrLen(StrPtr(sString)) Select Case Peek(StrPtr(sString) + i - 1) Case 48 To 57 Case 65 To 70 Case Else Return FALSE End Select Next Function = TRUE End Function ' ------------------------------------------------------------------------------

Quote:

Originally Posted by ReneMiner

Edit: @Eros
(who else could answer this question?)

Is there a way to add in some user-defined function or utility as this to thinAir? So I could instantly run the spellcheck from within thinAir on the current active scripts text without cumbersome actions as save script, change to spellchecker, start it, (search folder of script to check) dragging that there - checking here and then back to thinAir for corrections

- I imagine like some menuitem on Tools\Utilities\"User defined" which can be choosen through thinAir-options - set User-defined-utility-Path + Command$ (OS_GetCommands) there then thinAir "knows" if to pass script-filename & path or plain text-string of current script or whatever (list of all current opened scriptnames, crlf-delimited?) as command to the user-utility...:talk38:... ok, I would rewrite it, so you pass some tB-script as String to the function and get some $crlf-delimited error-report back?

Maybe, maybe very soon there will be some surprises on running scripts into thinAir as native thinAir commands.

Rene, in the meantime have you considered using Tokenizer module?
This module has been thought to assist programmers having to "parse" big text giving each text token a meaning defined by the programmer depending from his/her needs.
There are some examples into \thinBasic\SampleScripts\Tokenizer\

There is also a crazy (I mean positive) thinBasic example by Michael Hartlef using Tokenizer module to translate a thinBasic source code into Pascal Source code: http://www.thinbasic.com/community/s...hlight=thinpas

@Bilbo
thinBasic 1.9.7.0 is on the way, maybe next week-end

No, I did not test tokenizer yet - especially because "tokens" confuse me- I don't know what "token" means for sure. On the other side I fear to use up users "private" memory mostly - that's why I didn't use "sort-out-powers" of dictionary ;) But- anyway- tomorrow I'll take a look at tokenizer.

Tokens are just piece of text like you are parsing in your script example you posted here:

FOR is a token
, (comma) is a token
"this is a quoted string" is a token
' (single apostrophe) is a token
...

The parser in tokenizer module (Tokenizer_GetNextToken) can scan a string buffer and returns at every token found giving back the token found and tokentype.
TokenType can change depending on what keywords you have loaded into your session (tokenizer_KeyAdd) and special chars setup (Tokenizer_Default_Set and other functions)

See examples:

\thinBasic\SampleScripts\Tokenizer\Tokenizer.tBasic
\thinBasic\SampleScripts\Tokenizer\Tokenizer_UserKeys.tBasic

Hope this can help.

Ciao
Eros

I had to do a little "cosmetics" to both - the script and the attachement - there were some typos inside quotes :D - also forgot to check for space in front of REM and - because I ran it on itself - found some useless, leftover variable.

And playing around with tokenizer: found some ...bugs? in help,

thinBasic Modules > Tokenizer > Tokenizer_KeyGetMainType, there's some description about Tokenizer_KeyGetName but does not tell anything about Tokenizer_KeyGetMainType
same to:

thinBasic Modules > Tokenizer > Tokenizer_KeyFind, where it tells about use of Tokenizer_KeyGetUserString but nothing about Tokenizer_KeyFind.

So I have to check other sources for informations about tokenizer now...

Thanks Rene for help bugs hunting.
There was an error only in the syntax where I copied the name of the keyword from another one. The title, the topic and the page are correct.

Will be fixed in next release.