I am new Geany user and in this mailing list so Hi all from Spain.
With your permission, i will start with a question.
I am running Geany 0.21 on Ubuntu.
I have setting "Line breaking column" in Prefs at 72 and "Line Breaking" option enabled. When i type a word past this limit Geany insert a carriage return at the end of the line and put the cursor on the line below. Now on the same paragraph if i do a "Reflow Lines/Block" the lines are cut at 70 position. The same for diferents "Line breaking column" values (cuts at 2 less "Line breaking column" value).
Is this a right behavior ? Thanks. (Sorry for my sorry English)
On 18 March 2012 09:42, Toni Wiki toniwiki@gmail.com wrote:
I am new Geany user and in this mailing list so Hi all from Spain.
With your permission, i will start with a question.
I am running Geany 0.21 on Ubuntu.
I have setting "Line breaking column" in Prefs at 72 and "Line Breaking" option enabled. When i type a word past this limit Geany insert a carriage return at the end of the line and put the cursor on the line below. Now on the same paragraph if i do a "Reflow Lines/Block" the lines are cut at 70 position. The same for diferents "Line breaking column" values (cuts at 2 less "Line breaking column" value).
Is this a right behavior ? Thanks. (Sorry for my sorry English)
Hi Toni,
This is known behaviour. The two ways of doing line breaking are performed differently.
The difference between the two is that line breaking on input is performed by Geany based on character columns, whereas reflow is performed by Scintilla (the editing component Geany uses) using line widths in pixels calculated based on the width of a space in the default style.
We are not sure why these are different (at least when everything uses fixed fonts) but it may be due to differences in counting columns (ie is column 72 on the left of the character or the right?), different handling of trailing spaces and off-by-one pixel counts cause the latter to break more conservatively than the former.
IIRC fiddling with offsets didn't seem to cure it, so until someone authoritatively finds why the difference there is no known cure.
Cheers Lex
Geany mailing list Geany@uvena.de https://lists.uvena.de/cgi-bin/mailman/listinfo/geany
On Sun, 18 Mar 2012 13:00:44 +1100 Lex Trotman elextr@gmail.com wrote:
On 18 March 2012 09:42, Toni Wiki toniwiki@gmail.com wrote:
I am new Geany user and in this mailing list so Hi all from Spain.
With your permission, i will start with a question.
I am running Geany 0.21 on Ubuntu.
I have setting "Line breaking column" in Prefs at 72 and "Line Breaking" option enabled. When i type a word past this limit Geany insert a carriage return at the end of the line and put the cursor on the line below. Now on the same paragraph if i do a "Reflow Lines/Block" the lines are cut at 70 position. The same for diferents "Line breaking column" values (cuts at 2 less "Line breaking column" value).
Is this a right behavior ? Thanks. (Sorry for my sorry English)
Hi Toni,
This is known behaviour. The two ways of doing line breaking are performed differently.
The difference between the two is that line breaking on input is performed by Geany based on character columns, whereas reflow is performed by Scintilla (the editing component Geany uses) using line widths in pixels calculated based on the width of a space in the default style.
We are not sure why these are different (at least when everything uses fixed fonts) but it may be due to differences in counting columns (ie is column 72 on the left of the character or the right?), different handling of trailing spaces and off-by-one pixel counts cause the latter to break more conservatively than the former.
IIRC fiddling with offsets didn't seem to cure it, so until someone authoritatively finds why the difference there is no known cure.
Hi guys.
First of all, I must tell that it was me who proposed a patch to implement split/join lines commands in Geany which eventually became "Reflow paragraph".
Second point, even I don't remember why I didn't reuse the code implementing line breaking option (which was already written then). Maybe, as I already used SCI_JOINLINES, I considered SCI_SPLITLINES first and it did the job rather well (but not perfectly, as I understood after quite some time).
Third point, there are already some bug reports on the same topic (just search the bug tracker for "reflow"):
1.https://sourceforge.net/tracker/index.php?func=detail&aid=3293009&gr... My summary: SCI_SPLITLINES breaks lines on style boundary. I think, this behaviour is unneeded and sometimes just wrong for "reflow"
2.https://sourceforge.net/tracker/index.php?func=detail&aid=2982136&gr... An older bug report on the same topic as (1)
3.http://sourceforge.net/tracker/index.php?func=detail&aid=2871809&gro... Line is broken too early (just what Toni talks about). I posted a long comment to this bug report (my nick on SourceForge is statc), particularly explaining why Scintilla does not work "properly".
4.http://sourceforge.net/tracker/index.php?func=detail&aid=2923338&gro... Seems to be a duplicate of (3)
Fourth point, for some time I was going to raise a topic about reimplementing "Reflow paragraph" myself. Two things stopped me from doing it: (1) an absence of free time; (2) the fact that I propose other little patches to Geany from time to time and I don't want to bother Geany developers / maintainers too much.
Some time ago I posted a bug report to Scintilla [1]. They don't want to refine SCI_SPLITLINES, so my proposal is to rewrite "Reflow paragraph" reusing the code which implements line breaking. I can do it myself, but not now.
Sorry for a long message. I'm now waiting for a reply on my proposal.
[1]:https://sourceforge.net/tracker/index.php?func=detail&aid=3475314&group_id=2439&atid=102439
-- Best regards, Eugene.
[...]
Hi guys.
Hi Eugene,
First of all, I must tell that it was me who proposed a patch to implement split/join lines commands in Geany which eventually became "Reflow paragraph".
Second point, even I don't remember why I didn't reuse the code implementing line breaking option (which was already written then). Maybe, as I already used SCI_JOINLINES, I considered SCI_SPLITLINES first and it did the job rather well (but not perfectly, as I understood after quite some time).
Third point, there are already some bug reports on the same topic (just search the bug tracker for "reflow"):
Damn, "reflow" :)
1.https://sourceforge.net/tracker/index.php?func=detail&aid=3293009&gr... My summary: SCI_SPLITLINES breaks lines on style boundary. I think, this behaviour is unneeded and sometimes just wrong for "reflow"
2.https://sourceforge.net/tracker/index.php?func=detail&aid=2982136&gr... An older bug report on the same topic as (1)
3.http://sourceforge.net/tracker/index.php?func=detail&aid=2871809&gro... Line is broken too early (just what Toni talks about). I posted a long comment to this bug report (my nick on SourceForge is statc), particularly explaining why Scintilla does not work "properly".
Explanation makes sense, line end marker would account for some of the "pessimism" about fitting text on the line, and as I said I think there may be an off by one difference too.
4.http://sourceforge.net/tracker/index.php?func=detail&aid=2923338&gro... Seems to be a duplicate of (3)
Fourth point, for some time I was going to raise a topic about reimplementing "Reflow paragraph" myself. Two things stopped me from doing it: (1) an absence of free time; (2) the fact that I propose other little patches to Geany from time to time and I don't want to bother Geany developers / maintainers too much.
Don't worry about the maintainers :) if you get time feel free to do it.
Some time ago I posted a bug report to Scintilla [1]. They don't want to refine SCI_SPLITLINES, so my proposal is to rewrite "Reflow paragraph" reusing the code which implements line breaking. I can do it myself, but not now.
Sorry for a long message. I'm now waiting for a reply on my proposal.
The reply was "No", but with a reasonable reason, sort of given Neil's performance focus.
Cheers Lex
-- Best regards, Eugene.
Geany mailing list Geany@uvena.de https://lists.uvena.de/cgi-bin/mailman/listinfo/geany