Red output text from SQLite?

[Edifice]

When doing this:

$iRval = _SQLite_GetTable2d (-1, "SELECT * FROM datas", $aResult, $iColumns, $iRows)
    If $iRval = $SQLITE_OK Then
        _SQLite_Display2DResult($aResult)
    Else
        MsgBox(0,"ERROR","Query error! 2")
    EndIf

To my database, and hitting the F5, I get the output from the database on-screen.

That's all fine. The weird thing is that some of the text-output in written in red, and the red stops in the middel of a colum.

Anyone know why?

Edited February 27, 2009 by Edifice

[PsaltyDS]

When doing this:

$iRval = _SQLite_GetTable2d (-1, "SELECT * FROM datas", $aResult, $iColumns, $iRows)
    If $iRval = $SQLITE_OK Then
        _SQLite_Display2DResult($aResult)
    Else
        MsgBox(0,"ERROR","Query error! 2")
    EndIf

To my database, and hitting the F5, I get the output from the database on-screen.

That's all fine. The weird thing is that some of the text-output in written in red, and the red stops in the middel of a colum.

Anyone know why?

I have seen that, and not only with SQLite data. The SciTE console interprets certain escape seqences for text color changes, font, bold, etc. Sometimes the strings you pass to ConsoleWrite() just coincidentally contain those sequences.

I never looked that deeply into the SciTE docs, but if you track down the exact control data the console uses you could even use that on purpose to color tag your text.

[James]

Same. I was once outputting a STDOUT and the first few lines were blue

[martin]

Same. I was once outputting a STDOUT and the first few lines were blue

The easiest way is just by the first character of a line

ConsoleWrite("! somethjing red" & @CR)
ConsoleWrite("> something bluue" & @CRLF)
ConsoleWrite("- something orange" & @CRLF)
ConsoleWrite("+ something green" & @CRLF)

Edited February 28, 2009 by martin

[James]

The easiest way is just by the first character of a line

ConsoleWrite("! somethjing red" & @CR)
ConsoleWrite("> something bluue" & @CRLF)
ConsoleWrite("- something orange" & @CRLF)
ConsoleWrite("+ something green" & @CRLF)

But I was only outputting things like:

1536272
1092939
2030303
2728304
5695894

(Example).

[PsaltyDS]

But I was only outputting things like:

1536272
1092939
2030303
2728304
5695894

(Example).

Yep.

ConsoleWrite(1536272)
ConsoleWrite(1092939)
ConsoleWrite(2030303)
ConsoleWrite(2728304)
ConsoleWrite(5695894)

Output:

+>17:38:38 AU3Check ended.rc:0

>Running:(3.3.0.0):C:\Program Files\AutoIt3\autoit3.exe "C:\Program Files\AutoIt3\Scripts\Test_1.au3"

15362721092939203030327283045695894+>17:38:39 AutoIT3.exe ended.rc:0

+>17:38:40 AutoIt3Wrapper Finished

>Exit code: 0 Time: 2.458

Interesting.

@martin: Did you just poke around and find that by trial and error, or did you find a reference in the SciTE docs?

Edited February 28, 2009 by PsaltyDS

[martin]

Yep.

ConsoleWrite(1536272)
  ConsoleWrite(1092939)
  ConsoleWrite(2030303)
  ConsoleWrite(2728304)
  ConsoleWrite(5695894)

Output:

Interesting.

@martin: Did you just poke around and find that by trial and error, or did you find a reference in the SciTE docs?

I just noticed that when lines are red there is usually '!' as the first character so I tried different characters.

In Scintilla there are different character sets or styles that can be defined. Each set can have, among other things, a colour. So if you can find out which set has the colour red, then I imagine that you could send a message to SciTE to switch to that set then send the text. I got as far as finding

SCI_STYLEGETFORE(int styleNumber)

but I could find the definitions for the messages.

I searched for posts about coloured text in SciTE and found an old post which was similar to mine but the characters gave different colours, so I assume Jos has changed the colours of the defined sets at some time, or the post was just wrong.

BTW, I get a pale purple colour not red for the ConsolWrite code you posted, but it seems to be that colour for any text posted without @CR.

Eg

ConsoleWrite("THERE IS NO END TO IT")

Edited March 1, 2009 by martin

[James]

Weird huh! I understand a little bit better why though, still possibly a bug?

[PsaltyDS]

Weird huh! I understand a little bit better why though, still possibly a bug?

If so, it would be on Scintilla SciTE, not SciTE4AutoIt3.

Maybe Jos would offer an opinion...

[Zedna]

Here are my older notes:

ConsoleWrite('! = Red' & @LF)
ConsoleWrite('> = Blue' & @LF)
ConsoleWrite('- = Orange' & @LF)
ConsoleWrite('+ = Green' & @LF)
ConsoleWrite('(5) : = Red (jump to line 5 when double-clicked)' & @LF)
ConsoleWrite('Start with String or Integer then ' & @TAB & '6' & ' = Pink (jump to line 6 when double-clicked)' & @LF)


ConsoleWrite('(' & @ScriptLineNumber & ') : = Red (jump to line ' & @ScriptLineNumber & ' when double-clicked)' & @CRLF)

[James]

Zedna, that's fine however the output I had started with an integer but no @tab.

[Zedna]

Zedna, that's fine however the output I had started with an integer but no @tab.

I just gave you my notes without any conclusion so it may help you.

Sorry I don't know more details about that.

Edited March 2, 2009 by Zedna

[Jos]

As stated by Martin: there are a couple of "first characters" that SciTE uses to give a Outputpane line a color.

On top of that SciTE is trying to recognise errors generated by compiler/linker programs and giving them a color.

The reason this line :

15362721092939203030327283045695894+>17:38:39 AutoIT3.exe ended.rc:0

has the start of it in red is simply because it adheres to one of these formats.

Specificallly:

// Look for one of the following formats:

// GCC: <filename>:<line>:<message>

// Microsoft: <filename>(<line>) :<message>

// Common: <filename>(<line>): warning|error|note|remark|catastrophic|fatal

// Common: <filename>(<line>) warning|error|note|remark|catastrophic|fatal

// Microsoft: <filename>(<line>,<column>)<message>

// CTags: \t<message>

// Lua 5 traceback: \t<filename>:<line>:<message>

// Lua 5.1: <exe>: <filename>:<line>:<message>

and the color changes when you change this line in Others.properties:

# gcc Error

style.errorlist.2=fore:#800080

We use this feature in au3check and tidy.

For all details you could have a look at RecogniseErrorListLine() in LexOthers.cxx.

Edited March 2, 2009 by Jos

[herewasplato]

... something bluue" ...

U brits with your colours :-)

[James]

U brits with your colours :-)

You Americans with your missing letters. You ruined our language

Also, has anyone sent a bug report to the developers of ScITE.

[PsaltyDS]

You Americans with your missing letters. You ruined our language

Also, has anyone sent a bug report to the developers of ScITE.

Look again at Jos's post.

I don't think it would be accepted as a bug, because it is a configurable pattern match. If you don't like the behavior, you could change the patterns that get highlighted.

[martin]

U brits with your colours :-)

Putting two u's in blue was a bit over the top I admit. Glad youu slotted that.

[James]

Putting two u's in blue was a bit over the top I admit. Glad youu slotted that.

Martin, you're letting the team down.

[herewasplato]

Since martin is from the "uk" - he be a brit

Since JamesBrooks is from the "United Kingdom" - he be a Brit

...but we all know where I be from...

[Jos]

Look again at Jos's post.

I don't think it would be accepted as a bug, because it is a configurable pattern match. If you don't like the behavior, you could change the patterns that get highlighted.

Yea this particular bug is commonly referred to as a feature