Bug 425976

Summary: GC.drawText() & GC.drawString() - incorrect rendering of ASCII controls (=UNICODE C0 controls)
Product: [Eclipse Project] Platform Reporter: Andre Saibel <andre.saibel>
Component: SWTAssignee: Platform-SWT-Inbox <platform-swt-inbox>
Status: NEW --- QA Contact:
Severity: normal    
Priority: P3 CC: andre.saibel, ericwill
Version: 4.3Keywords: triaged
Target Milestone: ---   
Hardware: PC   
OS: Linux-GTK   
Whiteboard:

Description Andre Saibel CLA 2014-01-17 07:58:46 EST 
Some of the ASCII control characters (also defined in UNICODE http://www.unicode.org/charts/PDF/U0000.pdf) are rendered as a block containing the hexadecimal codepoint info (even if there is a 'real' glyph within the used font), for instance the four NFORMATION SEPARATOR controls 'unit separator' \u001F, 'record separator' \u001E, 'group separator' \u001D and 'file separator' \u001C... and maybe some more.

It seems to be special to Linux' SWT implementation, since the same content is shown correctly on Windows platforms and also on Linux with other apllications e.g. browsers.

public void paintControl(PaintEvent pe)
{
     pe.gc.drawText("abc\u001Exyz", 0, 0);
}
Comment 1 Alexander Kurtakov CLA 2016-12-21 19:19:51 EST 
As it's working for the real characters, I don't consider this an issue. Can you point to font with such glyphs to be tried?
Comment 2 Andre Saibel CLA 2017-01-09 04:42:00 EST 
The actual misbehaviour is that there's a block with a hex number at all – as well the font doesn't contain special glyphs for the control characters (which is common) as it does.
Comment 3 Eric Williams CLA 2019-09-17 16:10:03 EDT 
Reproducible on GTK3.24, Fedora 30, and SWT from master as of today.