Bug 382790 - Unicode combining characters printed with wrong glyph
Summary: Unicode combining characters printed with wrong glyph
Status: CLOSED WONTFIX
Alias: None
Product: Platform
Classification: Eclipse Project
Component: SWT (show other bugs)
Version: 4.2   Edit
Hardware: PC Windows 7
: P3 normal (vote)
Target Milestone: ---   Edit
Assignee: Platform-SWT-Inbox CLA
QA Contact:
URL:
Whiteboard: stalebug
Keywords: needinfo
Depends on:
Blocks:
 
Reported: 2012-06-16 12:10 EDT by John Keck CLA
Modified: 2019-11-27 07:05 EST (History)
4 users (show)

See Also:


Attachments
The attachment is a chart for how to type Greek diacriticals on a Windows machine. (46.56 KB, application/pdf)
2012-06-18 09:04 EDT, John Keck CLA
no flags Details
the requested text file (6.20 KB, text/plain)
2012-06-28 21:37 EDT, John Keck CLA
no flags Details
snapshot of (partial) console output (33.68 KB, image/png)
2012-06-28 22:03 EDT, John Keck CLA
no flags Details
java code for the previous two (5.07 KB, application/octet-stream)
2012-06-28 22:06 EDT, John Keck CLA
no flags Details
snapshot from editor (22.07 KB, image/png)
2012-06-29 09:49 EDT, John Keck CLA
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description John Keck CLA 2012-06-16 12:10:04 EDT
Build Identifier: 20120216-1857

I observe this bug in my work with Greek characters. This ordering (for one) manifests the problem:

letter, accent, breathing mark (the last two being combining characters)

The result should be a single combined character, but instead it produces

accented letter, breathing mark

This is a problem in the console as well as the editor.

More explanation (and sample code) here:
http://www.eclipse.org/forums/index.php/mv/msg/209100/886756/#msg_886756

Dani Megert asked me to report this and to be copied on it, but not sure how to do the latter.

Reproducible: Always

Steps to Reproduce:
1. print letter-accent-breathing mark to console (or type into editor)
2.
3.
Comment 1 Dani Megert CLA 2012-06-18 07:24:39 EDT
Please attach a test file that contains the characters which aren't printed right.
Comment 2 John Keck CLA 2012-06-18 09:04:29 EDT
Created attachment 217489 [details]
The attachment is a chart for how to type Greek diacriticals on a Windows machine.
Comment 3 Dani Megert CLA 2012-06-18 09:18:36 EDT
I copied and pasted the stuff into the Eclipse editor and at a first glance it looks like it's correctly rendered, but maybe I missed something.

Just attach a text file (that can be opened in Eclipse) and explain which characters in that file aren't rendered correctly.
Comment 4 John Keck CLA 2012-06-18 10:26:46 EDT
If you can't reproduce the problem with even the characters I've given you, then it seems we have a prior problem.

Were you able to see the graphical clip to compare? Does the program not reproduce that output for you? Is it only the editor that seems to be correct?

If you think it's useful, I can put something together, but it may take a while.
Comment 5 Dani Megert CLA 2012-06-18 10:44:21 EDT
(In reply to comment #4)
> If you can't reproduce the problem with even the characters I've given you,
> then it seems we have a prior problem.
Well, maybe I just don't see the difference you're referring to. A simple UTF-8 file with the character(s) plus pictures that show what's wrong would help a lot here.

> Were you able to see the graphical clip to compare? Does the program not
> reproduce that output for you? Is it only the editor that seems to be 
> correct?
The program involves many other components. Since you wrote in comment 0 that you also see it in the editor, we first need to look at that simple scenario.
Comment 6 John Keck CLA 2012-06-28 21:37:04 EDT
Created attachment 218060 [details]
the requested text file

This is the requested text file. Notice how the diacriticals appear over the kappa. Diacriticals never appear over consonants in Greek.

You can also download this file from http://web.mit.edu/jwk/www/diacriticalTest_output.txt.
Comment 7 John Keck CLA 2012-06-28 22:03:10 EDT
Created attachment 218061 [details]
snapshot of (partial) console output

This is a snapshot of the output in my console, or at least part of it. Again, notice how the diacriticals appear over the kappa. Diacriticals NEVER appear over consonants in Greek.

What the columns represent:

The character to the left of the colon is how the combination of glyphs *should* look on that line. It only appears if any of the combinations appear properly on that line.

The first two columns (to the right of any colon and to the left of the first pipe) are the vowel, then two combining diacritics, and then kappa. The combining diacritics are in the order shown in the rightmost two columns (to the right of the last pipe and separated by a comma). Between the first and second column the ordering of the diacritics is reversed.

The next two columns (between the two pipes) are the same as the previous, but they've been run through java.text.Normalizer, form NFC. (I should mention that I've submitted a bug to the java folks, but haven't received a reply.)

Regarding the text file I forgot to mention that it runs through combinations with all seven Greek vowels (α, ε, η, ι, ο, υ, ω), as well as (gratuitously) the three with iota subscripts (ᾳ, ῃ, ῳ).

You can also download this file from http://web.mit.edu/jwk/www/diacriticalTest_output.PNG.
Comment 8 John Keck CLA 2012-06-28 22:06:31 EDT
Created attachment 218062 [details]
java code for the previous two

Fwiw, attached is the java code used to generate the previous two files. You can also download it from http://web.mit.edu/jwk/www/DiacriticalTest2.java.

I apologize for the delay in my reply. I had an important deadline to meet earlier this week.

I appreciate your attention.

Regards,

John
Comment 9 John Keck CLA 2012-06-29 09:49:41 EDT
Created attachment 218085 [details]
snapshot from editor

Since you asked about the editor, I pasted some of the output from the console into the editor and attached a snapshot.
Comment 10 Lars Vogel CLA 2019-11-27 07:05:19 EST
This bug hasn't had any activity in quite some time. Maybe the problem got
resolved, was a duplicate of something else, or became less pressing for some
reason - or maybe it's still relevant but just hasn't been looked at yet.

If you have further information on the current state of the bug, please add it.
The information can be, for example, that the problem still occurs, that you
still want the feature, that more information is needed, or that the bug is
(for whatever reason) no longer relevant.

If the bug is still relevant, please remove the stalebug whiteboard tag.