buggy parser - // comments

[Martin crysman Zahradník]

Hi,

recently I've updated BF from svn.

look at this:

(PHP language)

I've had tried F5 to refresh, not helped.

As I see, it is not only the highlighter, but the parser as a whole,
because it is buggy even on using CTRL+SHIFT+C.

Where is the problem?

Thanks
McZ


PS: my BF info: bluefish 2.2.4 rev8133M

[Martin crysman Zahradník]

Please, how to properly make bluefish completely "default"? I mean all
the preferences, syntax highlighters, user settings... everything.
"total pure clean install". What to remove from $HOME? Or another way?
Thank you.

I had been using both binnary package and svn compiled versions in the
past. Now I got into troubles with my fresh Ubuntu 13.10 install (see
the mail below - but not only that, there are more issues with the
syntax highlightning)... So I would like to try "purge" BF and make a
clean install.

Thanks
McZ

/Martin crysman Zahradník/
/www:/ crysmanovo.blogspot.com <http://crysmanovo.blogspot.com>
/e-mail:/ crysman-***@public.gmane.org
/jabber:/ crysman-/eSpBmjxGS4dnm+***@public.gmane.org
/Google talk/hangout:/ crysmanovo-***@public.gmane.org
/facebook:/ no way!
<http://crysmanovo.blogspot.cz/2013/04/proc-koncim-na-facebooku.html>
/Skype:/ no way!
<http://crysmanovo.blogspot.cz/2013/05/proc-nepouzivam-skype-resp-proc.html>
using Ekiga <http://ekiga.net> - SIP:crysman-***@public.gmane.org
running GNU/Linux Xubuntu <http://xubuntu.org/>

Post by Martin crysman Zahradník
Hi,
recently I've updated BF from svn.
(PHP language)
I've had tried F5 to refresh, not helped.
As I see, it is not only the highlighter, but the parser as a whole,
because it is buggy even on using CTRL+SHIFT+C.
Where is the problem?
Thanks
McZ
PS: my BF info: bluefish 2.2.4 rev8133M

[Andrius]

I think You just need remove $HOME/.bluefish folder. It is hidden folder, so probably You first would need to make hidden folder visible.
After removal, bluefish will start with default settings.
Andrius

________________________________
Sent: Thursday, October 31, 2013 1:11 PM
Subject: Re: buggy parser - // comments
Please, how to properly make bluefish completely "default"? I mean all the preferences, syntax highlighters, user settings... everything. "total pure clean install". What to remove from $HOME? Or another way? Thank you.
I had been using both binnary package and svn compiled versions in

the past. Now I got into troubles with my fresh Ubuntu 13.10
install (see the mail below - but not only that, there are more
issues with the syntax highlightning)... So I would like to try
"purge" BF and make a clean install.

Thanks
McZ
 Martin crysman Zahradník
www: crysmanovo.blogspot.com
facebook: no way!
running GNU/Linux Xubuntu
Hi,
recently I've updated BF from svn.
(PHP language)
I've had tried F5 to refresh, not helped.
As I see, it is not only the highlighter, but the parser as a

whole, because it is buggy even on using CTRL+SHIFT+C.

Where is the problem?
Thanks
McZ
PS: my BF info: bluefish 2.2.4 rev8133M

[Olivier Sessink]

Post by Martin crysman Zahradník
Hi,
recently I've updated BF from svn.
(PHP language)
I've had tried F5 to refresh, not helped.
As I see, it is not only the highlighter, but the parser as a whole,
because it is buggy even on using CTRL+SHIFT+C.
Where is the problem?

can you send me that file? There is a related bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108 but I cannot reproduce
it somehow.

Olivier

[Martin crysman Zahradník]

It's evidently not about that file, it does the same today with a bash
script - I am sending it to You...


(why the hack BF thinks the 9 line is also a comment??)

McZ

PS: thanks for the reset to defaults, I will try

/Martin crysman Zahradník/
/www:/ crysmanovo.blogspot.com <http://crysmanovo.blogspot.com>
/e-mail:/ crysman-***@public.gmane.org
/jabber:/ crysman-/eSpBmjxGS4dnm+***@public.gmane.org
/Google talk/hangout:/ crysmanovo-***@public.gmane.org
/facebook:/ no way!
<http://crysmanovo.blogspot.cz/2013/04/proc-koncim-na-facebooku.html>
/Skype:/ no way!
<http://crysmanovo.blogspot.cz/2013/05/proc-nepouzivam-skype-resp-proc.html>
using Ekiga <http://ekiga.net> - SIP:crysman-***@public.gmane.org
running GNU/Linux Xubuntu <http://xubuntu.org/>

Post by Olivier Sessink

Post by Martin crysman Zahradník
Hi,
recently I've updated BF from svn.
(PHP language)
I've had tried F5 to refresh, not helped.
As I see, it is not only the highlighter, but the parser as a whole,
because it is buggy even on using CTRL+SHIFT+C.
Where is the problem?

can you send me that file? There is a related bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108 but I cannot
reproduce it somehow.
Olivier
--
Bluefish websitehttp://bluefish.openoffice.nl/
Bloghttp://oli4444.wordpress.com/

[Olivier Sessink]

What I requested on the bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108

Two questions:

1) can somebody with this problem on latest SVN send the contents of their
~/.bluefish directory to see if it is related to some setting?

2) can somebody on latest SVN edit bftextview2_private.h and set

#define DBG_SCANNING g_print

recompile, and post the full terminal output after loading the simple
example
from the original bugreport:

<?php
// Print something
if ($test == false) {
echo "something";
}

echo "done";
?>

I hope this gives some clues why this goes wrong (and why I cannot
reproduce it).

thanks,
Olivier

Post by Martin crysman Zahradník
It's evidently not about that file, it does the same today with a bash
script - I am sending it to You...
(why the hack BF thinks the 9 line is also a comment??)
McZ
PS: thanks for the reset to defaults, I will try
/Martin crysman Zahradník/
/www:/ crysmanovo.blogspot.com <http://crysmanovo.blogspot.com>
/facebook:/ no way!
<http://crysmanovo.blogspot.cz/2013/04/proc-koncim-na-facebooku.html>
/Skype:/ no way!
<http://crysmanovo.blogspot.cz/2013/05/proc-nepouzivam-skype-resp-proc.html>
running GNU/Linux Xubuntu <http://xubuntu.org/>

Post by Olivier Sessink

Post by Martin crysman Zahradník
Hi,
recently I've updated BF from svn.
(PHP language)
I've had tried F5 to refresh, not helped.
As I see, it is not only the highlighter, but the parser as a whole,
because it is buggy even on using CTRL+SHIFT+C.
Where is the problem?

can you send me that file? There is a related bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108 but I cannot
reproduce it somehow.
Olivier
--
Bluefish website http://bluefish.openoffice.nl/
Blog http://oli4444.wordpress.com/

[Stephen Doonan]

Post by Olivier Sessink
2) can somebody on latest SVN edit bftextview2_private.h and set
#define DBG_SCANNING g_print

Should one uncomment (remove the # sign) from that line of code, after
editing it to specify g_print, before compiling bluefish?

[Olivier Sessink]

Post by Stephen Doonan

Post by Olivier Sessink
2) can somebody on latest SVN edit bftextview2_private.h and set
#define DBG_SCANNING g_print

Should one uncomment (remove the # sign) from that line of code, after
editing it to specify g_print, before compiling bluefish?

no, just change the line to "#define DBG_SCANNING g_print" including the
hash sign.

Olivier

[Jim Hayward]

Post by Olivier Sessink
What I requested on the bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108
1) can somebody with this problem on latest SVN send the contents of their
~/.bluefish directory to see if it is related to some setting?

I already considered this and tried deleting my .bluefish directory. The
issue still existed with default settings.

Post by Olivier Sessink
2) can somebody on latest SVN edit bftextview2_private.h and set
#define DBG_SCANNING g_print
recompile, and post the full terminal output after loading the simple
example

I posted the output to the original bug report.


Regards,
Jim H

[Stephen Doonan]

Post by Olivier Sessink
2) can somebody on latest SVN edit bftextview2_private.h and set
#define DBG_SCANNING g_print
recompile, and post the full terminal output after loading the simple
example
<?php
// Print something
if ($test == false) {
echo "something";
}
echo "done";
?>

OK-- I downloaded the latest Bluefish from svn (bluefish 2.2.4 rev8134M)
and edited bftextview2_private.h as Olivier asked, then started Bluefish
from the command line with "bluefish &"

I then loaded a small text file composed of the PHP snippet above.
Attached is a screenshot of that snipped after being loaded, and the
output in the terminal window--

My environment-- Linux Mint 15, language English (utf-8). Could the bug
in question have to do with changing the colors or highlighting theme
within bluefish? If I remember correctly, the original post about this
issue had a screenshot in which this PHP snippet was colored
differently. I have no idea though--

Steve


--- --- --- terminal output of Bluefish --- --- ---

bfwin_main_ui_init, call lang_mode_menu_create(bfwin=0x2338510)
fb2_init, started for bfwin=0x2338510, fb2=0x2806570,
fb2->filebrowser_viewmode=0
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext
0(0Kb) = 0Kb
bflang2scan_finished, call lang_mode_menu_create() for bfwin 0x2338510
fb2_file_is_opened, called for
/home/xscd/Desktop/bluefish-comment-bug-test.php with mimetype (null)
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext
0(0Kb) = 0Kb
fb2_focus_document,doc /home/xscd/Desktop/bluefish-comment-bug-test.php
change_focus_to_uri, called for
/home/xscd/Desktop/bluefish-comment-bug-test.php
Possible error in language file, id - / pattern } has ends_context=2,
but has only 2 parent contexts
Possible error in language file, id end-style-tag / pattern </style> has
ends_context=3, but has only 3 parent contexts
Language statistics for PHP from
/usr/local/share/bluefish/bflang//php.bflang2
reference size 861.46 Kbytes
largest table 32354 ( 8088.50 Kbytes)
total tables 62109 ( 15527.25 Kbytes)
contexts 333 ( 15.61 Kbytes)
matches 9699 ( 454.64 Kbytes)
blocks 14 ( 0.44 Kbytes)
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext
0(0Kb) = 0Kb
bftextview2_find_region2scan returned region 0:85
start scanning at start iter
scanning from 0 to 85
scanning offset 0 pos 0 < (context=1).. got newpos 2
scanning offset 1 pos 2 ? (context=1).. got newpos 3
scanning offset 2 pos 3 p (context=1).. got newpos 4
scanning offset 3 pos 4 h (context=1).. got newpos 5
scanning offset 4 pos 5 p (context=1).. got newpos 6
scanning offset 5 pos 6
(context=1).. got newpos 0
we have a match from pos 0 to 5
found_match for pattern 1 <(\?|\?php) at charoffset 0,
starts_block=1,ends_block=0, nextcontext=2 (current=1)
found_match, apply tag 0x231b190 from 0 to 5
no nextfound, so enlarge scanning region to end iter
found_context_change, new fcontext 0x3325b60 with context 2 onto the
stack, parent=(nil)
after match context=2
scanning offset 5 pos 0
(context=2).. got newpos 0
scanning offset 6 pos 0 / (context=2).. got newpos 32333
scanning offset 7 pos 32333 / (context=2).. got newpos 32350
scanning offset 8 pos 32350 (context=2).. got newpos 0
we have a match from pos 6 to 8
found_match for pattern 4631 // at charoffset 6,
starts_block=0,ends_block=0, nextcontext=8 (current=2)
found_match, apply tag 0x230ed40 from 6 to 8
no nextfound, so enlarge scanning region to end iter
found_context_change, new fcontext 0x320b6a0 with context 8 onto the
stack, parent=0x3325b60
after match context=8
scanning offset 8 pos 0 (context=8).. got newpos 1
scanning offset 9 pos 1 P (context=8).. got newpos 1
scanning offset 10 pos 1 r (context=8).. got newpos 1
scanning offset 11 pos 1 i (context=8).. got newpos 1
scanning offset 12 pos 1 n (context=8).. got newpos 1
scanning offset 13 pos 1 t (context=8).. got newpos 1
scanning offset 14 pos 1 (context=8).. got newpos 1
scanning offset 15 pos 1 s (context=8).. got newpos 1
scanning offset 16 pos 1 o (context=8).. got newpos 1
scanning offset 17 pos 1 m (context=8).. got newpos 1
scanning offset 18 pos 1 e (context=8).. got newpos 1
scanning offset 19 pos 1 t (context=8).. got newpos 1
scanning offset 20 pos 1 h (context=8).. got newpos 1
scanning offset 21 pos 1 i (context=8).. got newpos 1
scanning offset 22 pos 1 n (context=8).. got newpos 1
scanning offset 23 pos 1 g (context=8).. got newpos 1
scanning offset 24 pos 1
(context=8).. got newpos 0
scanning offset 24 pos 0
(context=8).. got newpos 2
scanning offset 25 pos 2 i (context=8).. got newpos 0
we have a match from pos 24 to 25
found_match for pattern 4632 (
|
) at charoffset 24, starts_block=0,ends_block=0, nextcontext=-1 (current=8)
no nextfound, so enlarge scanning region to end iter
found_context_change, should pop 1 contexts, curfcontext=0x320b6a0
pop_and_apply_contexts, end context 8 at 8:24, has tag 0x230ed40 and
parent 0x3325b60
after match context=2
scanning offset 25 pos 0 i (context=2).. got newpos 728
scanning offset 26 pos 728 f (context=2).. got newpos 10180
scanning offset 27 pos 10180 (context=2).. got newpos 0
we have a match from pos 25 to 27
found_match for pattern 4529 if at charoffset 25,
starts_block=0,ends_block=0, nextcontext=0 (current=2)
found_match, apply tag 0x231b0a0 from 25 to 27
found_match, pattern does not start block or context, return
after match context=2
scanning offset 27 pos 0 (context=2).. got newpos 0
scanning offset 28 pos 0 ( (context=2).. got newpos 32331
scanning offset 29 pos 32331 $ (context=2).. got newpos 0
we have a match from pos 28 to 29
found_match for pattern 4589 ( at charoffset 28,
starts_block=1,ends_block=0, nextcontext=0 (current=2)
found_match, apply tag 0x231b4a0 from 28 to 29
no nextfound, so enlarge scanning region to end iter
after match context=2
scanning offset 29 pos 0 $ (context=2).. got newpos 32346
scanning offset 30 pos 32346 t (context=2).. got newpos 32347
scanning offset 31 pos 32347 e (context=2).. got newpos 32348
scanning offset 32 pos 32348 s (context=2).. got newpos 32348
scanning offset 33 pos 32348 t (context=2).. got newpos 32348
scanning offset 34 pos 32348 (context=2).. got newpos 0
we have a match from pos 29 to 34
found_match for pattern 4598 $[a-zA-Z_][a-zA-Z0-9_]* at charoffset 29,
starts_block=0,ends_block=0, nextcontext=0 (current=2)
found_match, apply tag 0x231b1e0 from 29 to 34
found_match, pattern does not start block or context, return
after match context=2
scanning offset 34 pos 0 (context=2).. got newpos 0
scanning offset 35 pos 0 = (context=2).. got newpos 0
scanning offset 36 pos 0 = (context=2).. got newpos 0
scanning offset 37 pos 0 (context=2).. got newpos 0
scanning offset 38 pos 0 f (context=2).. got newpos 1070
scanning offset 39 pos 1070 a (context=2).. got newpos 4714
scanning offset 40 pos 4714 l (context=2).. got newpos 32316
scanning offset 41 pos 32316 s (context=2).. got newpos 32317
scanning offset 42 pos 32317 e (context=2).. got newpos 32318
scanning offset 43 pos 32318 ) (context=2).. got newpos 0
we have a match from pos 38 to 43
found_match for pattern 4549 false at charoffset 38,
starts_block=0,ends_block=0, nextcontext=0 (current=2)
found_match, apply tag 0x231b140 from 38 to 43
found_match, pattern does not start block or context, return
after match context=2
scanning offset 43 pos 0 ) (context=2).. got newpos 32332
scanning offset 44 pos 32332 (context=2).. got newpos 0
we have a match from pos 43 to 44
found_match for pattern 4590 ) at charoffset 43,
starts_block=0,ends_block=1, nextcontext=0 (current=2)
found_match, apply tag 0x231b4a0 from 43 to 44
no nextfound, so enlarge scanning region to end iter
after match context=2
scanning offset 44 pos 0 (context=2).. got newpos 0
scanning offset 45 pos 0 { (context=2).. got newpos 32327
scanning offset 46 pos 32327
(context=2).. got newpos 0
we have a match from pos 45 to 46
found_match for pattern 4585 { at charoffset 45,
starts_block=1,ends_block=0, nextcontext=0 (current=2)
found_match, apply tag 0x231b4a0 from 45 to 46
no nextfound, so enlarge scanning region to end iter
after match context=2
scanning offset 46 pos 0
(context=2).. got newpos 0
scanning offset 47 pos 0 (context=2).. got newpos 0
scanning offset 48 pos 0 (context=2).. got newpos 0
scanning offset 49 pos 0 e (context=2).. got newpos 126
scanning offset 50 pos 126 c (context=2).. got newpos 32235
scanning offset 51 pos 32235 h (context=2).. got newpos 32236
scanning offset 52 pos 32236 o (context=2).. got newpos 32237
scanning offset 53 pos 32237 (context=2).. got newpos 0
we have a match from pos 49 to 53
found_match for pattern 4519 echo at charoffset 49,
starts_block=0,ends_block=0, nextcontext=0 (current=2)
found_match, apply tag 0x231b0a0 from 49 to 53
found_match, pattern does not start block or context, return
after match context=2
scanning offset 53 pos 0 (context=2).. got newpos 0
scanning offset 54 pos 0 " (context=2).. got newpos 32349
scanning offset 55 pos 32349 s (context=2).. got newpos 0
we have a match from pos 54 to 55
found_match for pattern 4599 " at charoffset 54,
starts_block=1,ends_block=0, nextcontext=7 (current=2)
found_match, apply tag 0x230ed90 from 54 to 55
no nextfound, so enlarge scanning region to end iter
found_context_change, new fcontext 0x31e7e40 with context 7 onto the
stack, parent=0x3325b60
after match context=7
scanning offset 55 pos 0 s (context=7).. got newpos 91
scanning offset 56 pos 91 o (context=7).. got newpos 1
scanning offset 57 pos 1 m (context=7).. got newpos 1
scanning offset 58 pos 1 e (context=7).. got newpos 1
scanning offset 59 pos 1 t (context=7).. got newpos 1
scanning offset 60 pos 1 h (context=7).. got newpos 1
scanning offset 61 pos 1 i (context=7).. got newpos 1
scanning offset 62 pos 1 n (context=7).. got newpos 1
scanning offset 63 pos 1 g (context=7).. got newpos 1
scanning offset 64 pos 1 " (context=7).. got newpos 0
scanning offset 64 pos 0 " (context=7).. got newpos 115
scanning offset 65 pos 115 ; (context=7).. got newpos 0
we have a match from pos 64 to 65
found_match for pattern 4630 " at charoffset 64,
starts_block=0,ends_block=1, nextcontext=-1 (current=7)
found_match, apply tag 0x230ed90 from 64 to 65
no nextfound, so enlarge scanning region to end iter
found_context_change, should pop 1 contexts, curfcontext=0x31e7e40
pop_and_apply_contexts, end context 7 at 55:64, has tag 0x230ed90 and
parent 0x3325b60
after match context=2
scanning offset 65 pos 0 ; (context=2).. got newpos 0
scanning offset 66 pos 0
(context=2).. got newpos 0
scanning offset 67 pos 0 } (context=2).. got newpos 32328
scanning offset 68 pos 32328
(context=2).. got newpos 0
we have a match from pos 67 to 68
found_match for pattern 4586 } at charoffset 67,
starts_block=0,ends_block=1, nextcontext=0 (current=2)
found_match, apply tag 0x231b4a0 from 67 to 68
no nextfound, so enlarge scanning region to end iter
after match context=2
scanning offset 68 pos 0
(context=2).. got newpos 0
scanning offset 69 pos 0
(context=2).. got newpos 0
scanning offset 70 pos 0 e (context=2).. got newpos 126
scanning offset 71 pos 126 c (context=2).. got newpos 32235
scanning offset 72 pos 32235 h (context=2).. got newpos 32236
scanning offset 73 pos 32236 o (context=2).. got newpos 32237
scanning offset 74 pos 32237 (context=2).. got newpos 0
we have a match from pos 70 to 74
found_match for pattern 4519 echo at charoffset 70,
starts_block=0,ends_block=0, nextcontext=0 (current=2)
found_match, apply tag 0x231b0a0 from 70 to 74
found_match, pattern does not start block or context, return
after match context=2
scanning offset 74 pos 0 (context=2).. got newpos 0
scanning offset 75 pos 0 " (context=2).. got newpos 32349
scanning offset 76 pos 32349 d (context=2).. got newpos 0
we have a match from pos 75 to 76
found_match for pattern 4599 " at charoffset 75,
starts_block=1,ends_block=0, nextcontext=7 (current=2)
found_match, apply tag 0x230ed90 from 75 to 76
no nextfound, so enlarge scanning region to end iter
found_context_change, new fcontext 0x3245a60 with context 7 onto the
stack, parent=0x3325b60
after match context=7
scanning offset 76 pos 0 d (context=7).. got newpos 19
scanning offset 77 pos 19 o (context=7).. got newpos 1
scanning offset 78 pos 1 n (context=7).. got newpos 1
scanning offset 79 pos 1 e (context=7).. got newpos 1
scanning offset 80 pos 1 " (context=7).. got newpos 0
scanning offset 80 pos 0 " (context=7).. got newpos 115
scanning offset 81 pos 115 ; (context=7).. got newpos 0
we have a match from pos 80 to 81
found_match for pattern 4630 " at charoffset 80,
starts_block=0,ends_block=1, nextcontext=-1 (current=7)
found_match, apply tag 0x230ed90 from 80 to 81
no nextfound, so enlarge scanning region to end iter
found_context_change, should pop 1 contexts, curfcontext=0x3245a60
pop_and_apply_contexts, end context 7 at 76:80, has tag 0x230ed90 and
parent 0x3325b60
after match context=2
scanning offset 81 pos 0 ; (context=2).. got newpos 0
scanning offset 82 pos 0
(context=2).. got newpos 0
scanning offset 83 pos 0 ? (context=2).. got newpos 32352
scanning offset 84 pos 32352 > (context=2).. got newpos 32353
scanning offset 85 pos 32353 (context=2).. got newpos 0
we have a match from pos 83 to 85
found_match for pattern 4635 ?> at charoffset 83,
starts_block=0,ends_block=1, nextcontext=-1 (current=2)
found_match, apply tag 0x231b190 from 83 to 85
no nextfound, so enlarge scanning region to end iter
found_context_change, should pop 1 contexts, curfcontext=0x3325b60
pop_and_apply_contexts, end context 2 at 5:83, has tag (nil) and parent
(nil)
after match context=1
scanned from 0 to position 85, (end=85) which took 0.003475
microseconds, loops_per_timer=1000
scanning run 1 (3 ms): 0, 0, 0, 3; from 0-85, loops=105,chars=85,blocks
5/5 (0) contexts 4/4 (0) scancache 12
memory scancache 12(0Kb+0Kb) found 5(0Kb) fcontext 4(0Kb) = 1Kb
average 28333 chars/s 85 chars/run
scancache integrity check done in 0.010000 ms.
change_focus_to_uri, called for
/home/xscd/Desktop/bluefish-comment-bug-test.php

[Olivier Sessink]

Post by Stephen Doonan
OK-- I downloaded the latest Bluefish from svn (bluefish 2.2.4 rev8134M)
and edited bftextview2_private.h as Olivier asked, then started Bluefish
from the command line with "bluefish &"
I then loaded a small text file composed of the PHP snippet above.
Attached is a screenshot of that snipped after being loaded, and the
output in the terminal window--

exactly the same output as I have - and perfect highlighting.

Post by Stephen Doonan
My environment-- Linux Mint 15, language English (utf-8). Could the bug
in question have to do with changing the colors or highlighting theme
within bluefish? If I remember correctly, the original post about this
issue had a screenshot in which this PHP snippet was colored
differently. I have no idea though--

not likely, the output that Jim posted in the bugreport has a different
scanning result when scanning offset 25 (the newline), so it's reallt
got to do with the language file, or the scanning code (or both).

Olivier

[Martin crysman Zahradník]

OK, here it is.

1) - in the attachment - before any changes
2) terminal output:

$ bluefish bftest.php

bfwin_main_ui_init, call lang_mode_menu_create(bfwin=0x274a380)

cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb

fb2_file_is_opened, called for /tmp/bftest.php with mimetype (null)

cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb

fb2_focus_document,doc /tmp/bftest.php

Possible error in language file, id - / pattern } has ends_context=2, but has only 2 parent contexts

Possible error in language file, id end-style-tag / pattern </style> has ends_context=3, but has only 3 parent contexts

Language statistics for PHP from /usr/local/share/bluefish/bflang//php.bflang2

reference size 861.46 Kbytes

largest table 32354 ( 8088.50 Kbytes)

total tables 62109 ( 15527.25 Kbytes)

contexts 333 ( 15.61 Kbytes)

matches 9699 ( 454.64 Kbytes)

blocks 14 ( 0.44 Kbytes)

cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb

scanning run 1 (0 ms): 0, 0, 0, 0; from 0-87, loops=100,chars=87,blocks 2/2 (0) contexts 3/3 (0) scancache 6

memory scancache 6(0Kb+0Kb) found 2(0Kb) fcontext 3(0Kb) = 0Kb

average 0 chars/s 87 chars/run

scancache integrity check done in 0.008000 ms.

delete_async, failed to delete: Error removing file: No such file or directory

(bluefish:15041): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed


screenshot:



3) I've just tried deleting the .bluefish dir.
terminal output:

***@crysman-U36SD:tmp$ mv ~/.bluefish/ ~/.bluefish.bak
***@crysman-U36SD:tmp$ bluefish bftest.php
error reading list 1 Error opening file: No such file or directory

** (bluefish:15100): WARNING **: no configfile rcfile-2.0, try to convert config files from older versions

config file migration error 1:Error opening file: No such file or directoryerror reading list 1 Error opening file: No such file or directory
error reading list 1 Error opening file: No such file or directory
bfwin_main_ui_init, call lang_mode_menu_create(bfwin=0x88eda0)
fb2_init, started for bfwin=0x88eda0, fb2=0xe51e60, fb2->filebrowser_viewmode=0
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb
bflang2scan_finished, call lang_mode_menu_create() for bfwin 0x88eda0
fb2_file_is_opened, called for /tmp/bftest.php with mimetype (null)
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb
fb2_focus_document,doc /tmp/bftest.php
change_focus_to_uri, called for /tmp/bftest.php
Possible error in language file, id - / pattern } has ends_context=2, but has only 2 parent contexts
Possible error in language file, id end-style-tag / pattern </style> has ends_context=3, but has only 3 parent contexts
change_focus_to_uri, called for /tmp/bftest.php
Language statistics for PHP from /usr/local/share/bluefish/bflang//php.bflang2
reference size 861,46 Kbytes
largest table 32354 ( 8088,50 Kbytes)
total tables 62109 ( 15527,25 Kbytes)
contexts 333 ( 15,61 Kbytes)
matches 9699 ( 454,64 Kbytes)
blocks 14 ( 0,44 Kbytes)
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb
scanning run 1 (0 ms): 0, 0, 0, 0; from 0-87, loops=100,chars=87,blocks 2/2 (0) contexts 3/3 (0) scancache 6
memory scancache 6(0Kb+0Kb) found 2(0Kb) fcontext 3(0Kb) = 0Kb
average 0 chars/s 87 chars/run
scancache integrity check done in 0,008000 ms.


screenshot:



I hope it helps you to fix it...

McZ

/Martin crysman Zahradník/
/www:/ crysmanovo.blogspot.com <http://crysmanovo.blogspot.com>
/e-mail:/ crysman-***@public.gmane.org
/jabber:/ crysman-/eSpBmjxGS4dnm+***@public.gmane.org
/Google talk/hangout:/ crysmanovo-***@public.gmane.org
/facebook:/ no way!
<http://crysmanovo.blogspot.cz/2013/04/proc-koncim-na-facebooku.html>
/Skype:/ no way!
<http://crysmanovo.blogspot.cz/2013/05/proc-nepouzivam-skype-resp-proc.html>
using Ekiga <http://ekiga.net> - SIP:crysman-***@public.gmane.org
running GNU/Linux Xubuntu <http://xubuntu.org/>

Post by Olivier Sessink
What I requested on the bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108
1) can somebody with this problem on latest SVN send the contents of their
~/.bluefish directory to see if it is related to some setting?
2) can somebody on latest SVN edit bftextview2_private.h and set
#define DBG_SCANNING g_print
recompile, and post the full terminal output after loading the simple
example
<?php
// Print something
if ($test == false) {
echo "something";
}
echo "done";
?>
I hope this gives some clues why this goes wrong (and why I cannot
reproduce it).
thanks,
Olivier

Post by Martin crysman Zahradník
It's evidently not about that file, it does the same today with a bash
script - I am sending it to You...
(why the hack BF thinks the 9 line is also a comment??)
McZ
PS: thanks for the reset to defaults, I will try
/Martin crysman Zahradník/
/www:/ crysmanovo.blogspot.com <http://crysmanovo.blogspot.com>
/facebook:/ no way!
<http://crysmanovo.blogspot.cz/2013/04/proc-koncim-na-facebooku.html>
/Skype:/ no way!
<http://crysmanovo.blogspot.cz/2013/05/proc-nepouzivam-skype-resp-proc.html>
running GNU/Linux Xubuntu <http://xubuntu.org/>

Post by Olivier Sessink

Post by Martin crysman Zahradník
Hi,
recently I've updated BF from svn.
(PHP language)
I've had tried F5 to refresh, not helped.
As I see, it is not only the highlighter, but the parser as a whole,
because it is buggy even on using CTRL+SHIFT+C.
Where is the problem?

can you send me that file? There is a related bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108 but I cannot
reproduce it somehow.
Olivier
--
Bluefish website http://bluefish.openoffice.nl/
Blog http://oli4444.wordpress.com/

[Martin crysman Zahradník]

PS: I do not understand why there is "with libgucharmap ... no", because
I certainly have it installed in my Xubuntu. Look:

***@crysman-U36SD:tmp$ sudo apt-get install libgucharmap-2-90-dev
Reading package lists... Done
Building dependency tree
Reading state information... Done
libgucharmap-2-90-dev is already the newest version.
0 upgraded, 0 newly installed, 0 to remove and 7 not upgraded.

McZ
/Martin crysman Zahradník/
/www:/ crysmanovo.blogspot.com <http://crysmanovo.blogspot.com>
/e-mail:/ crysman-***@public.gmane.org
/jabber:/ crysman-/eSpBmjxGS4dnm+***@public.gmane.org
/Google talk/hangout:/ crysmanovo-***@public.gmane.org
/facebook:/ no way!
<http://crysmanovo.blogspot.cz/2013/04/proc-koncim-na-facebooku.html>
/Skype:/ no way!
<http://crysmanovo.blogspot.cz/2013/05/proc-nepouzivam-skype-resp-proc.html>
using Ekiga <http://ekiga.net> - SIP:crysman-***@public.gmane.org
running GNU/Linux Xubuntu <http://xubuntu.org/>

Post by Martin crysman Zahradník
OK, here it is.
1) - in the attachment - before any changes
$ bluefish bftest.php
bfwin_main_ui_init, call lang_mode_menu_create(bfwin=0x274a380)
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb
fb2_file_is_opened, called for /tmp/bftest.php with mimetype (null)
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb
fb2_focus_document,doc /tmp/bftest.php
Possible error in language file, id - / pattern } has ends_context=2, but has only 2 parent contexts
Possible error in language file, id end-style-tag / pattern </style> has ends_context=3, but has only 3 parent contexts
Language statistics for PHP from /usr/local/share/bluefish/bflang//php.bflang2
reference size 861.46 Kbytes
largest table 32354 ( 8088.50 Kbytes)
total tables 62109 ( 15527.25 Kbytes)
contexts 333 ( 15.61 Kbytes)
matches 9699 ( 454.64 Kbytes)
blocks 14 ( 0.44 Kbytes)
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb
scanning run 1 (0 ms): 0, 0, 0, 0; from 0-87, loops=100,chars=87,blocks 2/2 (0) contexts 3/3 (0) scancache 6
memory scancache 6(0Kb+0Kb) found 2(0Kb) fcontext 3(0Kb) = 0Kb
average 0 chars/s 87 chars/run
scancache integrity check done in 0.008000 ms.
delete_async, failed to delete: Error removing file: No such file or directory
(bluefish:15041): GLib-GObject-CRITICAL **: g_object_unref: assertion 'G_IS_OBJECT (object)' failed
3) I've just tried deleting the .bluefish dir.
error reading list 1 Error opening file: No such file or directory
** (bluefish:15100): WARNING **: no configfile rcfile-2.0, try to convert config files from older versions
config file migration error 1:Error opening file: No such file or directoryerror reading list 1 Error opening file: No such file or directory
error reading list 1 Error opening file: No such file or directory
bfwin_main_ui_init, call lang_mode_menu_create(bfwin=0x88eda0)
fb2_init, started for bfwin=0x88eda0, fb2=0xe51e60, fb2->filebrowser_viewmode=0
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb
bflang2scan_finished, call lang_mode_menu_create() for bfwin 0x88eda0
fb2_file_is_opened, called for /tmp/bftest.php with mimetype (null)
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb
fb2_focus_document,doc /tmp/bftest.php
change_focus_to_uri, called for /tmp/bftest.php
Possible error in language file, id - / pattern } has ends_context=2, but has only 2 parent contexts
Possible error in language file, id end-style-tag / pattern </style> has ends_context=3, but has only 3 parent contexts
change_focus_to_uri, called for /tmp/bftest.php
Language statistics for PHP from /usr/local/share/bluefish/bflang//php.bflang2
reference size 861,46 Kbytes
largest table 32354 ( 8088,50 Kbytes)
total tables 62109 ( 15527,25 Kbytes)
contexts 333 ( 15,61 Kbytes)
matches 9699 ( 454,64 Kbytes)
blocks 14 ( 0,44 Kbytes)
cleanup_scanner, memory scancache 0(0Kb+0Kb) found 0(0Kb) fcontext 0(0Kb) = 0Kb
scanning run 1 (0 ms): 0, 0, 0, 0; from 0-87, loops=100,chars=87,blocks 2/2 (0) contexts 3/3 (0) scancache 6
memory scancache 6(0Kb+0Kb) found 2(0Kb) fcontext 3(0Kb) = 0Kb
average 0 chars/s 87 chars/run
scancache integrity check done in 0,008000 ms.
I hope it helps you to fix it...
McZ
/Martin crysman Zahradník/
/www:/ crysmanovo.blogspot.com <http://crysmanovo.blogspot.com>
/facebook:/ no way!
<http://crysmanovo.blogspot.cz/2013/04/proc-koncim-na-facebooku.html>
/Skype:/ no way!
<http://crysmanovo.blogspot.cz/2013/05/proc-nepouzivam-skype-resp-proc.html>
running GNU/Linux Xubuntu <http://xubuntu.org/>

Post by Olivier Sessink
What I requested on the bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108
1) can somebody with this problem on latest SVN send the contents of their
~/.bluefish directory to see if it is related to some setting?
2) can somebody on latest SVN edit bftextview2_private.h and set
#define DBG_SCANNING g_print
recompile, and post the full terminal output after loading the simple
example
<?php
// Print something
if ($test == false) {
echo "something";
}
echo "done";
?>
I hope this gives some clues why this goes wrong (and why I cannot
reproduce it).
thanks,
Olivier

Post by Martin crysman Zahradník
It's evidently not about that file, it does the same today with a bash
script - I am sending it to You...
(why the hack BF thinks the 9 line is also a comment??)
McZ
PS: thanks for the reset to defaults, I will try
/Martin crysman Zahradník/
/www:/ crysmanovo.blogspot.com<http://crysmanovo.blogspot.com>
/facebook:/ no way!
<http://crysmanovo.blogspot.cz/2013/04/proc-koncim-na-facebooku.html>
/Skype:/ no way!
<http://crysmanovo.blogspot.cz/2013/05/proc-nepouzivam-skype-resp-proc.html>
running GNU/Linux Xubuntu<http://xubuntu.org/>

Post by Olivier Sessink

Post by Martin crysman Zahradník
Hi,
recently I've updated BF from svn.
(PHP language)
I've had tried F5 to refresh, not helped.
As I see, it is not only the highlighter, but the parser as a whole,
because it is buggy even on using CTRL+SHIFT+C.
Where is the problem?

can you send me that file? There is a related bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108 but I cannot
reproduce it somehow.
Olivier
--
Bluefish websitehttp://bluefish.openoffice.nl/
Bloghttp://oli4444.wordpress.com/

[clansco]

On Mon, 04 Nov 2013 12:29:14 +0100

Post by Martin crysman Zahradník
PS: I do not understand why there is "with libgucharmap ... no", because

you have libgucharmap_2 ;) it's all right!

with libgucharmap_2... yes

Regards,
ff

[Olivier Sessink]

Martin,

can it be that you forgot to recompile, or perhaps you forgot to edit
bftextview2_private.h?

I don't see the output of the scanner.

Olivier

[Andrius]

Might be this bug related to the end of line symbols? I attached three files with different EOL. On Mac all files has correct highlightning, but maybe other users will see the difference?
Andrius

________________________________
Sent: Saturday, November 2, 2013 11:34 AM
Subject: Re: buggy parser - // comments
What I requested on the bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108
1) can somebody with this problem on latest SVN send the contents of their
~/.bluefish directory to see if it is related to some setting?
2) can somebody on latest SVN edit bftextview2_private.h and set
#define DBG_SCANNING g_print
recompile, and post the full terminal output after loading the simple
example
<?php
// Print something
if ($test == false) {
  echo "something";
}
echo "done";
?>
I hope this gives some clues why this goes wrong (and why I cannot
reproduce it).
thanks,
Olivier

Post by Martin crysman Zahradník
It's evidently not about that file, it does the same today with a bash
script - I am sending it to You...
(why the hack BF thinks the 9 line is also a comment??)
McZ
PS: thanks for the reset to defaults, I will try
 
/Martin crysman Zahradník/
/www:/ crysmanovo.blogspot.com <http://crysmanovo.blogspot.com>
/facebook:/ no way!
<http://crysmanovo.blogspot.cz/2013/04/proc-koncim-na-facebooku.html>
/Skype:/ no way!
<http://crysmanovo.blogspot.cz/2013/05/proc-nepouzivam-skype-resp-proc.html>
running GNU/Linux Xubuntu <http://xubuntu.org/>

Post by Olivier Sessink

Post by Martin crysman Zahradník
Hi,
recently I've updated BF from svn.
(PHP language)
I've had tried F5 to refresh, not helped.
As I see, it is not only the highlighter, but the parser as a whole,
because it is buggy even on using CTRL+SHIFT+C.
Where is the problem?

can you send me that file? There is a related bugreport
https://bugzilla.gnome.org/show_bug.cgi?id=704108 but I cannot
reproduce it somehow.
Olivier
--
Bluefish website http://bluefish.openoffice.nl/
Blog http://oli4444.wordpress.com/

--
Bluefish website http://bluefish.openoffice.nl/
Blog http://oli4444.wordpress.com/
--
To unsubscribe from this list: send the line "unsubscribe bluefish-users" in
List Web Page: http://www.ems.ru/cgi-bin/listargate.cgi
Main Web Page: http://bluefish.openoffice.nl/

[Olivier Sessink]

Post by Andrius
Might be this bug related to the end of line symbols? I attached three
files with different EOL. On Mac all files has correct highlightning,
but maybe other users will see the difference?
Andrius

I managed to reproduce it on Fedora 19.

but it is really weird:

the loop at bftextview2_patcompile.c line 520

for (j = 0; j <=NUMSCANCHARS ; j++)

ran far beyond 127 (the value of NUMSCANCHARS) which caused an incorrect state
table.

I fixed it by reversing the loop

for (j = NUMSCANCHARS; j >=0 ; j++)


but how could this have happened ???????????????????

Olivier

[Lindsay Haisley]

Post by Olivier Sessink
the loop at bftextview2_patcompile.c line 520
for (j = 0; j <=NUMSCANCHARS ; j++)
ran far beyond 127 (the value of NUMSCANCHARS) which caused an incorrect state
table.
I fixed it by reversing the loop
for (j = NUMSCANCHARS; j >=0 ; j++)
but how could this have happened ???????????????????
Olivier

Olivier, I don't know the code at this point, but normally rewriting the
loop as you have done would generate an infinite loop if NUMSCANCHARS is
a positive number. REVERSING the loop would require j-- instead of
j++. This is bizarre!
--
Lindsay Haisley | "Everything works if you let it"
FMP Computer Services |
512-259-1190 | --- The Roadie
http://www.fmp.com |

[Olivier Sessink]

Post by Olivier Sessink

Post by Olivier Sessink
the loop at bftextview2_patcompile.c line 520
for (j = 0; j <=NUMSCANCHARS ; j++)
ran far beyond 127 (the value of NUMSCANCHARS) which caused an

incorrect state

Post by Olivier Sessink
table.
I fixed it by reversing the loop
for (j = NUMSCANCHARS; j >=0 ; j++)
but how could this have happened ???????????????????
Olivier

Olivier, I don't know the code at this point, but normally rewriting the
loop as you have done would generate an infinite loop if NUMSCANCHARS is
a positive number. REVERSING the loop would require j-- instead of
j++. This is bizarre!

Sorry, I did not copy paste, yes I changed j++ to j--, you are totally right.

Olivier

[Andrius]

Would it be better to reverse to r8134 and use "<" instaead of "<="? That would save us some processor time...

________________________________
I fixed it by reversing the loop
for (j = NUMSCANCHARS; j >=0 ; j++)

[Olivier Sessink]

I don't think it makes much difference, if we use the reversed loop or
the normal loop, but yes, I think you can revert that change.

Olivier

Post by Andrius
Would it be better to reverse to r8134 and use "<" instaead of "<="?
That would save us some processor time...
------------------------------------------------------------------------
**
I fixed it by reversing the loop
for (j = NUMSCANCHARS; j >=0 ; j++)