Upcoming release
168 views
Skip to first unread message
Neil Hodgson
Apr 18, 2024, 10:34:24 PMApr 18
to Scintilla mailing list, scite-i...@googlegroups.com
New versions of Lexilla (5.3.2), Scintilla (5.5.0), and SciTE (5.5.0) will be released in a few days - likely 23rd.
This release contains fixes and features for Lexilla, Scintilla, and SciTE.
SciTE removes support for older Windows versions.
Lexilla sees changes to COBOL, HTML, Lua, Ruby, and Rust.
Changes:
• SciTE
• Remove support for Windows XP and Windows Vista.
• Requires a C++20 compiler. GCC 11.2 and MSVC 2022 are supported.
• Fix read failure for UTF-16LE files where surrogate pair straddles block boundary and lead surrogate is a multiple of 256.
• On GTK, treat message box close as cancel. This is safer when closing a file with unsaved changes. Bug #2434.
• Lexilla 5.3.2
• COBOL: Stop string literal continuing over line end. Issue #229.
• COBOL: Stop doc comment assigning different styles to \r and \n at line end. Issue #229.
• COBOL: Recognize keywords that start with 'V'. Issue #230.
• COBOL: Recognize comments after tag or that start with '/'. Issue #231.
• HTML: Implement substyles for tags, attributes, and identifiers SCE_H_TAG, SCE_H_ATTRIBUTE, SCE_HJ_WORD, SCE_HJA_WORD, SCE_HB_WORD, SCE_HP_WORD, SCE_HPHP_WORD.
• HTML: Implement context-sensitive attributes. "tag.attribute" matches "attribute" only inside "tag".
• HTML: Match standard handling of comments. Issue #232.
• Lua: Implement substyles for identifiers SCE_LUA_IDENTIFIER.
• Ruby: Allow non-ASCII here-doc delimiters. Issue #234.
• Ruby: Allow modifier if, unless, while and until after heredoc delimiter. Issue #236.
• Rust: Recognize raw identifiers. Issue #239, Pull request #240.
• Scintilla 5.5.0
• Add elements for inactive additional selections SC_ELEMENT_SELECTION_INACTIVE_ADDITIONAL_TEXT and SC_ELEMENT_SELECTION_INACTIVE_ADDITIONAL_BACK. When not set these default to SC_ELEMENT_SELECTION_INACTIVE_TEXT and SC_ELEMENT_SELECTION_INACTIVE_BACK. Bug #2417.
• On Cocoa, avoid use of NSUserDefaults which will soon require justification when used in applications on the App Store.
• Fix Win32 IME crash in windowed mode. Bug #2433.
• Scale reverse arrow cursor for margins to match other cursors when user changes pointer size. Bug #2321.
The committed changes can be examined either in the repositories
git clone https://github.com/ScintillaOrg/lexilla
hg clone http://hg.code.sf.net/p/scintilla/code scintilla
hg clone http://hg.code.sf.net/p/scintilla/scite
or from
https://www.scintilla.org/scite.zip Source
https://www.scintilla.org/wscite.zip Windows executable (64-bit)
Neil
This release contains fixes and features for Lexilla, Scintilla, and SciTE.
SciTE removes support for older Windows versions.
Lexilla sees changes to COBOL, HTML, Lua, Ruby, and Rust.
Changes:
• SciTE
• Remove support for Windows XP and Windows Vista.
• Requires a C++20 compiler. GCC 11.2 and MSVC 2022 are supported.
• Fix read failure for UTF-16LE files where surrogate pair straddles block boundary and lead surrogate is a multiple of 256.
• On GTK, treat message box close as cancel. This is safer when closing a file with unsaved changes. Bug #2434.
• Lexilla 5.3.2
• COBOL: Stop string literal continuing over line end. Issue #229.
• COBOL: Stop doc comment assigning different styles to \r and \n at line end. Issue #229.
• COBOL: Recognize keywords that start with 'V'. Issue #230.
• COBOL: Recognize comments after tag or that start with '/'. Issue #231.
• HTML: Implement substyles for tags, attributes, and identifiers SCE_H_TAG, SCE_H_ATTRIBUTE, SCE_HJ_WORD, SCE_HJA_WORD, SCE_HB_WORD, SCE_HP_WORD, SCE_HPHP_WORD.
• HTML: Implement context-sensitive attributes. "tag.attribute" matches "attribute" only inside "tag".
• HTML: Match standard handling of comments. Issue #232.
• Lua: Implement substyles for identifiers SCE_LUA_IDENTIFIER.
• Ruby: Allow non-ASCII here-doc delimiters. Issue #234.
• Ruby: Allow modifier if, unless, while and until after heredoc delimiter. Issue #236.
• Rust: Recognize raw identifiers. Issue #239, Pull request #240.
• Scintilla 5.5.0
• Add elements for inactive additional selections SC_ELEMENT_SELECTION_INACTIVE_ADDITIONAL_TEXT and SC_ELEMENT_SELECTION_INACTIVE_ADDITIONAL_BACK. When not set these default to SC_ELEMENT_SELECTION_INACTIVE_TEXT and SC_ELEMENT_SELECTION_INACTIVE_BACK. Bug #2417.
• On Cocoa, avoid use of NSUserDefaults which will soon require justification when used in applications on the App Store.
• Fix Win32 IME crash in windowed mode. Bug #2433.
• Scale reverse arrow cursor for margins to match other cursors when user changes pointer size. Bug #2321.
The committed changes can be examined either in the repositories
git clone https://github.com/ScintillaOrg/lexilla
hg clone http://hg.code.sf.net/p/scintilla/code scintilla
hg clone http://hg.code.sf.net/p/scintilla/scite
or from
https://www.scintilla.org/scite.zip Source
https://www.scintilla.org/wscite.zip Windows executable (64-bit)
Neil
Csaba Ráduly
May 10, 2024, 2:23:29 AMMay 10
to scite-i...@googlegroups.com
On 19/04/2024 04:34, 'Neil Hodgson' via scite-interest wrote:
[...]
Csaba
--
Life is complex, with real and imaginary parts.
[...]
> • Fix read failure for UTF-16LE files where surrogate pair straddles block boundary and lead surrogate is a multiple of 256.
That must have been fun to track down (for small values of fun).
Csaba
--
Life is complex, with real and imaginary parts.
Neil Hodgson
May 10, 2024, 8:42:39 PMMay 10
to scite-interest
Csaba Ráduly:
>> • Fix read failure for UTF-16LE files where surrogate pair straddles block boundary and lead surrogate is a multiple of 256.
>
> That must have been fun to track down (for small values of fun).
It was a bit of an accident as I wrote a unit test before modifying the loading code and chose a character that just happened to be a multiple of 256.
Neil
>> • Fix read failure for UTF-16LE files where surrogate pair straddles block boundary and lead surrogate is a multiple of 256.
>
> That must have been fun to track down (for small values of fun).
Neil
jenny...@163.com
May 18, 2024, 6:09:39 AMMay 18
to scite-interest
Dear Neil:
i want colorrise label in my script.
it needs to search backward when the lexer working, such as a label,
syntax like efcc2088, this is a hexadecimal number,
and efcc2088: is a label.
I modified the underlying function of lexer to implement the effect before.
I think this approach is not good and I read Some lexer codes, not found any code that implements similar functions
so I would like to ask you how to implement StyleContext look back.
--You received this message because you are subscribed to the Google Groups "scite-interest" group.To unsubscribe from this group and stop receiving emails from it, send an email to scite-interes...@googlegroups.com.To view this discussion on the web visit https://groups.google.com/d/msgid/scite-interest/AC191197-E9A7-4979-A7F3-82AB93AC411E%40me.com.
Neil Hodgson
May 20, 2024, 7:18:09 PMMay 20
to scite-interest
jenny.venus:
> i want colorrise label in my script.
> it needs to search backward when the lexer working, such as a label,
> syntax like efcc2088, this is a hexadecimal number,
> and efcc2088: is a label.
It's common to set the current state to 'identifier' then when the ':' is seen, call ChangeState to say its really a label.
>
> I modified the underlying function of lexer to implement the effect before.
> I think this approach is not good and I read Some lexer codes, not found any code that implements similar functions
>
> so I would like to ask you how to implement StyleContext look back.
You can examine the previous characters with GetRelative or use the underlying LexAccessor methods like SafeGetCharAt. You can also look at the previously set style with StyleAt and BufferStylesAt.
Neil
> i want colorrise label in my script.
> it needs to search backward when the lexer working, such as a label,
> syntax like efcc2088, this is a hexadecimal number,
> and efcc2088: is a label.
>
> I modified the underlying function of lexer to implement the effect before.
> I think this approach is not good and I read Some lexer codes, not found any code that implements similar functions
>
> so I would like to ask you how to implement StyleContext look back.
Neil
jenny...@163.com
Jun 1, 2024, 5:07:44 AMJun 1
to scite-interest
Dear Neil:how to clear all change history without saveing files?
i have add change history in my tools, i want clear change history manuly when file is modified and not saved, but when i call setsavepoint + disable change history + enable change history , change history is not work.
pic 1 - file open
pic 2 - modify it
pic 3 - Format and align script, every line is modified
pic 4 - clear change state without save file
then the change history not work.
thanks very much!
jenny...@163.com
Jun 2, 2024, 4:07:46 AMJun 2
to scite-interest
Dear Neil:
I don't need the solution of clearing the change history.
By optimizing the formatting script and configuring the indicator color, this problem has been perfectly solved.
thanks very much.
--
You received this message because you are subscribed to the Google Groups "scite-interest" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scite-interes...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/scite-interest/2024060117072346251522%40163.com.
Neil Hodgson
Jun 2, 2024, 6:57:33 AMJun 2
to scite-interest
jenny...@163.com:
> how to clear all change history without saveing files? i have add change history in my tools, i want clear change history manuly when file is modified and not saved, but when i call setsavepoint + disable change history + enable change history , change history is not work.
Change history is closely aligned with undo history and can only be started when there is no undo history, that is, at start or just after undo history has been discarded. It's possible that enabling change history with some undo history could be implemented but this is complex code.
If it was me implementing the feature, it would require significant time and effort. Perhaps someone else is interested in working in this area.
Neil
> how to clear all change history without saveing files? i have add change history in my tools, i want clear change history manuly when file is modified and not saved, but when i call setsavepoint + disable change history + enable change history , change history is not work.
If it was me implementing the feature, it would require significant time and effort. Perhaps someone else is interested in working in this area.
Neil
jenny...@163.com
Jun 2, 2024, 8:23:55 AMJun 2
to scite-interest
--You received this message because you are subscribed to the Google Groups "scite-interest" group.To unsubscribe from this group and stop receiving emails from it, send an email to scite-interes...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/scite-interest/A4FDBE99-A81C-4D6F-A470-0E97F3D3701A%40me.com.
jenny...@163.com
Jun 25, 2024, 1:43:32 AMJun 25
to scite-interest
Dear Neil:
I found that sometimes the block editing has some problems, I took a screenshot of these, I hope you can try it when you have time.
--You received this message because you are subscribed to the Google Groups "scite-interest" group.To unsubscribe from this group and stop receiving emails from it, send an email to scite-interes...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/scite-interest/A4FDBE99-A81C-4D6F-A470-0E97F3D3701A%40me.com.
Neil Hodgson
Jun 30, 2024, 6:32:54 PMJun 30
to scite-interest
jenny.venus:
> I found that sometimes the block editing has some problems, I took a screenshot of these, I hope you can try it when you have time.<001.gif><002.gif>jenny...@163.com
I can't tell what the problem is. Its difficult to decipher videos without a clear explanation of what the expected behaviour is.
Neil
> I found that sometimes the block editing has some problems, I took a screenshot of these, I hope you can try it when you have time.<001.gif><002.gif>jenny...@163.com
I can't tell what the problem is. Its difficult to decipher videos without a clear explanation of what the expected behaviour is.
Neil
jenny...@163.com
Jun 30, 2024, 9:02:16 PMJun 30
to scite-interest
Dear Neil:
use scite 550, paste the text to it.
1 =
22 =
333 =
4444 =
55555 =
666666 =
7777777 =
88888888 =
999999999 =
and enter to block edit mode( alt + shift + arrow up )
type "123456789" , sometimes you can see the text is confusion.
--You received this message because you are subscribed to the Google Groups "scite-interest" group.To unsubscribe from this group and stop receiving emails from it, send an email to scite-interes...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/scite-interest/4A03BDFE-AE51-49EF-B4A8-9EC8C3CA9C8F%40me.com.
John Yeung
Jul 2, 2024, 3:18:06 PMJul 2
to scite-i...@googlegroups.com
Probably the issue is that you're using a proportional font, so
columns are not going to line up exactly.
When moving the cursor up or down a line, I believe SciTE tries to
maintain the horizontal pixel position of the cursor as much as
possible. This means the cursor may be at a different *character*
position from one line to the next.
John Y.
columns are not going to line up exactly.
When moving the cursor up or down a line, I believe SciTE tries to
maintain the horizontal pixel position of the cursor as much as
possible. This means the cursor may be at a different *character*
position from one line to the next.
John Y.
Reply all
Reply to author
Forward
0 new messages