path: root/old/oldmail/200207.txt

[Dillo-dev]Re: Patch: META http-equiv=refresh is BROKEN a little :(

From: Nikita V. Borodikhin <eliterr@tk...> - 2002-07-31 07:30

I'm so sorry...

Patch is broken a little (DRC_TOKEN_USE_META_REFRESH placed in pref.c
instead of DRC_TOKEN_ALLOW_META_REFRESH)...


Here is the correct patch:

diff -pruN dillo.orig/dillorc dillo/dillorc
--- dillo.orig/dillorc Thu May 30 17:21:08 2002
+++ dillo/dillorc Wed Jul 31 14:07:07 2002
@@ -15,6 +15,9 @@ geometry=640x550
# If you have a lot of memory and a slow CPU, use YES, otherwise use NO
use_dicache=NO

+# We do not allow http-equiv=refresh by default because it's not
+# conform to official HTML 4.01 standard
+allow_meta_refresh=NO

#-------------------------------------------------------------------------
# RENDERING SECTION
diff -pruN dillo.orig/src/html.c dillo/src/html.c
--- dillo.orig/src/html.c Mon Jul 1 10:13:37 2002
+++ dillo/src/html.c Wed Jul 31 14:10:52 2002
@@ -2512,15 +2512,74 @@ static void Html_tag_close_form(DilloHtm

/*
* Handle <META>
- * We do not support http-equiv=refresh because it's non standar,
- * (the HTML 4.01 SPEC recommends explicitily to avoid it), and it
- * can be easily abused!
+ * We disable http-equiv=refresh by default because it's non standard,
+ * (the HTML 4.01 SPEC recommends explicitily to avoid it),
+ * and it can be easily abused!
*/
static void Html_tag_open_meta(DilloHtml *html, char *tag, gint tagsize)
{
+ const gchar *http_equiv;
+ const gchar *content;
+ const gchar *url_string;
+ DwPage *page;
+ DwStyle style_attrs, *link_style;
+ DilloUrl *url;
+ gint dummy;
+
+ page = DW_PAGE(html->dw);
+ style_attrs = *(html->stack[html->stack_top].style);
+
/* only valid inside HEAD */
- if (html->InFlags & IN_HEAD)
- return;
+ if (! (html->InFlags & IN_HEAD))
+ DEBUG_HTML_MSG ("META tag is not inside HEAD\n");
+
+ if ((http_equiv = Html_get_attr(html, tag, tagsize, "http-equiv")))
+ {
+ if ((strcasecmp (http_equiv, "refresh") == 0) && prefs.allow_meta_refresh
+ && (content = Html_get_attr(html, tag, tagsize, "content")))
+ {
+ url_string = strstr (content, "URL=");
+ if (strlen (url_string) > 4)
+ url_string += 4;
+ else
+ url_string = NULL;
+
+ if (url_string && (url = a_Url_new(url_string,
+ URL_STR_(html->linkblock->base_url),
+ 0, 0)))
+ {
+ a_Dw_page_add_text (page, g_strdup ("This page uses META-refresh to "),
+ html->stack[(html)->stack_top].style);
+
+ if (a_Cache_url_read(url, &dummy)) /* visited frame */
+ style_attrs.color = a_Dw_style_color_new
+ (html->linkblock->visited_color, html->bw->main_window->window);
+ else /* unvisited frame */
+ style_attrs.color = a_Dw_style_color_new
+ (html->linkblock->link_color, html->bw->main_window->window);
+
+ style_attrs.uline = TRUE;
+ style_attrs.link = Html_set_new_link(html, &url);
+ link_style = a_Dw_style_new (&style_attrs,
+ html->bw->main_window->window);
+
+ a_Dw_page_add_text(page, g_strdup(url_string), link_style);
+
+ a_Dw_style_unref(link_style);
+ }
+ else
+ {
+ a_Dw_page_add_text (page, g_strdup ("Strange META-refresh argument: "),
+ html->stack[(html)->stack_top].style);
+ a_Dw_page_add_text (page, g_strdup (content),
+ html->stack[(html)->stack_top].style);
+ }
+
+ a_Dw_page_add_parbreak(page, 9, html->stack[(html)->stack_top].style);
+ }
+ }
+
+ return;
}

/*
diff -pruN dillo.orig/src/prefs.c dillo/src/prefs.c
--- dillo.orig/src/prefs.c Thu Apr 11 12:37:35 2002
+++ dillo/src/prefs.c Wed Jul 31 13:43:44 2002
@@ -53,6 +53,7 @@ static const struct {
{ "limit_text_width", DRC_TOKEN_LIMIT_TEXT_WIDTH },
{ "font_factor", DRC_TOKEN_FONT_FACTOR },
{ "use_dicache", DRC_TOKEN_USE_DICACHE },
+ { "allow_meta_refresh", DRC_TOKEN_ALLOW_META_REFRESH },
{ "show_back", DRC_TOKEN_SHOW_BACK },
{ "show_forw", DRC_TOKEN_SHOW_FORW },
{ "show_home", DRC_TOKEN_SHOW_HOME },
@@ -178,6 +179,9 @@ static guint Prefs_parser(GScanner *scan
case DRC_TOKEN_USE_DICACHE:
prefs.use_dicache = (strcmp(scanner->value.v_string, "YES") == 0);
break;
+ case DRC_TOKEN_ALLOW_META_REFRESH:
+ prefs.allow_meta_refresh = (strcmp(scanner->value.v_string, "YES") == 0);
+ break;
case DRC_TOKEN_SHOW_BACK:
prefs.show_back = (strcmp(scanner->value.v_string, "YES") == 0);
break;
@@ -339,6 +343,7 @@ void a_Prefs_init(void)
prefs.limit_text_width = FALSE;
prefs.font_factor = 1.0;
prefs.use_dicache = FALSE;
+ prefs.allow_meta_refresh = FALSE;
prefs.show_back=TRUE;
prefs.show_forw=TRUE;
prefs.show_home=TRUE;
diff -pruN dillo.orig/src/prefs.h dillo/src/prefs.h
--- dillo.orig/src/prefs.h Thu Apr 11 12:37:35 2002
+++ dillo/src/prefs.h Wed Jul 31 13:44:01 2002
@@ -43,6 +43,7 @@ typedef enum {
DRC_TOKEN_SHOW_ALT,
DRC_TOKEN_LIMIT_TEXT_WIDTH,
DRC_TOKEN_USE_DICACHE,
+ DRC_TOKEN_ALLOW_META_REFRESH,
DRC_TOKEN_SHOW_BACK,
DRC_TOKEN_SHOW_FORW,
DRC_TOKEN_SHOW_HOME,
@@ -85,6 +86,7 @@ struct _DilloPrefs {
gboolean limit_text_width;
gdouble font_factor;
gboolean use_dicache;
+ gboolean allow_meta_refresh;
gboolean show_back;
gboolean show_forw;
gboolean show_home; 



[Dillo-dev]Patch: META http-equiv=refresh

From: Nikita V. Borodikhin <eliterr@tk...> - 2002-07-31 07:16

Hello all Dillo users !

I thought that META http-equiv=refresh parameter in fully not standard
but recently I found Web Design Group's page about META tag
(see http://www.htmlhelp.com/reference/html40/head/meta.html).

They wrote <META http-equiv=refresh content='10; URL=>
is usual tag but not all browsers can handle it, so I made patch that
adds support for it (disabled by default) that adds ability to see
URL refreshing to as link (like dillo's frame support does).

Sincerely yours,
Nikita V. Borodikhin



Here is the patch:

diff -pruN dillo.orig/dillorc dillo/dillorc
--- dillo.orig/dillorc Thu May 30 17:21:08 2002
+++ dillo/dillorc Wed Jul 31 14:07:07 2002
@@ -15,6 +15,9 @@ geometry=640x550
# If you have a lot of memory and a slow CPU, use YES, otherwise use NO
use_dicache=NO

+# We do not allow http-equiv=refresh by default because it's not
+# conform to official HTML 4.01 standard
+allow_meta_refresh=NO

#-------------------------------------------------------------------------
# RENDERING SECTION
diff -pruN dillo.orig/src/html.c dillo/src/html.c
--- dillo.orig/src/html.c Mon Jul 1 10:13:37 2002
+++ dillo/src/html.c Wed Jul 31 14:10:52 2002
@@ -2512,15 +2512,74 @@ static void Html_tag_close_form(DilloHtm

/*
* Handle <META>
- * We do not support http-equiv=refresh because it's non standar,
- * (the HTML 4.01 SPEC recommends explicitily to avoid it), and it
- * can be easily abused!
+ * We disable http-equiv=refresh by default because it's non standard,
+ * (the HTML 4.01 SPEC recommends explicitily to avoid it),
+ * and it can be easily abused!
*/
static void Html_tag_open_meta(DilloHtml *html, char *tag, gint tagsize)
{
+ const gchar *http_equiv;
+ const gchar *content;
+ const gchar *url_string;
+ DwPage *page;
+ DwStyle style_attrs, *link_style;
+ DilloUrl *url;
+ gint dummy;
+
+ page = DW_PAGE(html->dw);
+ style_attrs = *(html->stack[html->stack_top].style);
+
/* only valid inside HEAD */
- if (html->InFlags & IN_HEAD)
- return;
+ if (! (html->InFlags & IN_HEAD))
+ DEBUG_HTML_MSG ("META tag is not inside HEAD\n");
+
+ if ((http_equiv = Html_get_attr(html, tag, tagsize, "http-equiv")))
+ {
+ if ((strcasecmp (http_equiv, "refresh") == 0) && prefs.allow_meta_refresh
+ && (content = Html_get_attr(html, tag, tagsize, "content")))
+ {
+ url_string = strstr (content, "URL=");
+ if (strlen (url_string) > 4)
+ url_string += 4;
+ else
+ url_string = NULL;
+
+ if (url_string && (url = a_Url_new(url_string,
+ URL_STR_(html->linkblock->base_url),
+ 0, 0)))
+ {
+ a_Dw_page_add_text (page, g_strdup ("This page uses META-refresh to "),
+ html->stack[(html)->stack_top].style);
+
+ if (a_Cache_url_read(url, &dummy)) /* visited frame */
+ style_attrs.color = a_Dw_style_color_new
+ (html->linkblock->visited_color, html->bw->main_window->window);
+ else /* unvisited frame */
+ style_attrs.color = a_Dw_style_color_new
+ (html->linkblock->link_color, html->bw->main_window->window);
+
+ style_attrs.uline = TRUE;
+ style_attrs.link = Html_set_new_link(html, &url);
+ link_style = a_Dw_style_new (&style_attrs,
+ html->bw->main_window->window);
+
+ a_Dw_page_add_text(page, g_strdup(url_string), link_style);
+
+ a_Dw_style_unref(link_style);
+ }
+ else
+ {
+ a_Dw_page_add_text (page, g_strdup ("Strange META-refresh argument: "),
+ html->stack[(html)->stack_top].style);
+ a_Dw_page_add_text (page, g_strdup (content),
+ html->stack[(html)->stack_top].style);
+ }
+
+ a_Dw_page_add_parbreak(page, 9, html->stack[(html)->stack_top].style);
+ }
+ }
+
+ return;
}

/*
diff -pruN dillo.orig/src/prefs.c dillo/src/prefs.c
--- dillo.orig/src/prefs.c Thu Apr 11 12:37:35 2002
+++ dillo/src/prefs.c Wed Jul 31 13:43:44 2002
@@ -53,6 +53,7 @@ static const struct {
{ "limit_text_width", DRC_TOKEN_LIMIT_TEXT_WIDTH },
{ "font_factor", DRC_TOKEN_FONT_FACTOR },
{ "use_dicache", DRC_TOKEN_USE_DICACHE },
+ { "allow_meta_refresh", DRC_TOKEN_USE_META_REFRESH },
{ "show_back", DRC_TOKEN_SHOW_BACK },
{ "show_forw", DRC_TOKEN_SHOW_FORW },
{ "show_home", DRC_TOKEN_SHOW_HOME },
@@ -178,6 +179,9 @@ static guint Prefs_parser(GScanner *scan
case DRC_TOKEN_USE_DICACHE:
prefs.use_dicache = (strcmp(scanner->value.v_string, "YES") == 0);
break;
+ case DRC_TOKEN_ALLOW_META_REFRESH:
+ prefs.allow_meta_refresh = (strcmp(scanner->value.v_string, "YES") == 0);
+ break;
case DRC_TOKEN_SHOW_BACK:
prefs.show_back = (strcmp(scanner->value.v_string, "YES") == 0);
break;
@@ -339,6 +343,7 @@ void a_Prefs_init(void)
prefs.limit_text_width = FALSE;
prefs.font_factor = 1.0;
prefs.use_dicache = FALSE;
+ prefs.allow_meta_refresh = FALSE;
prefs.show_back=TRUE;
prefs.show_forw=TRUE;
prefs.show_home=TRUE;
diff -pruN dillo.orig/src/prefs.h dillo/src/prefs.h
--- dillo.orig/src/prefs.h Thu Apr 11 12:37:35 2002
+++ dillo/src/prefs.h Wed Jul 31 13:44:01 2002
@@ -43,6 +43,7 @@ typedef enum {
DRC_TOKEN_SHOW_ALT,
DRC_TOKEN_LIMIT_TEXT_WIDTH,
DRC_TOKEN_USE_DICACHE,
+ DRC_TOKEN_ALLOW_META_REFRESH,
DRC_TOKEN_SHOW_BACK,
DRC_TOKEN_SHOW_FORW,
DRC_TOKEN_SHOW_HOME,
@@ -85,6 +86,7 @@ struct _DilloPrefs {
gboolean limit_text_width;
gdouble font_factor;
gboolean use_dicache;
+ gboolean allow_meta_refresh;
gboolean show_back;
gboolean show_forw;
gboolean show_home; 



[Dillo-dev]Bug 254, Table width is too large.

From: Matias Aguirre <yo_soy@fa...> - 2002-07-31 07:16

Hi list, Im new to dillo and you maked a great work.

In 254 bug describes "table width is too large" but this is not the reason of this bug.
The reason is that if you read the source of http://linux.oreillynet.com/pub/a/linux/2001/11/15/learnunixos.html you found <IFRAME> tag, and this dont put in dillo sources.

This is a <IFRAME> referense: http://www.htmlhelp.com/reference/html40/special/iframe.html

Greetings

ps: sorry by my bad english 



[Dillo-dev]Nav push consistency

From: Eric GAUDET <eric@rt...> - 2002-07-28 22:45

Hi all,

I just occured to me that the way Dillo handles URL is somewhat inconsist=
ent.
Choosing an already visited URL from the back history pushes the old page=
, and
remembers the position in this page. However, clicking on a link for an a=
lready
visited URL pushes the same link, only a the top of the page.
What=B4s really inconsistent, is if you have several time the same page i=
n the
back history, the remembered position in the page is unique (the last
position), no matter what history entry you choose.

I understand that there is two different codepath and only one cache entr=
y, be
still.

To be more polished, we have to choose between two consistent behaviors:
- the position is stored in the back history, not in the cache entry. Tha=
t way,
every back history remembers its own position, and newly clicked link ope=
n the
page at the top
- the position remains stored in the cache entry, and is always used no m=
atter
what. All nav push to this URL positions the page at this unique position=
,
regardless of if it is clicked from the back history, a link in the page,=
or
entered in the location bar.

I think I prefer the latter myself. What do you guys think?

Best,
Eric 



Re: [Dillo-dev]Copy/paste in Dillo

From: Eric GAUDET <eric@rt...> - 2002-07-28 03:52

There is no problem asking for new features, but before doing that, it mi=
ght be
wise to read le web site, especially the bug track. I am sick and tired o=
f the
new visitors asking the same question again and again, even if there is a
published todo list.

John was right on: Dillo is open source and contributions have always bee=
n
welcome.

If you feel like a feature is missing, go head and code it.=20

If you don=B4t want/can=B4t do it, just wait. It might not be there eithe=
r because
it is difficult, or because Dillo=B4s developers have other priority, or =
are too
busy doing something else. We are all coding Dillo on our free time, and =
we can
not answer to individual demands. Dillo has been progressing a lot since =
the
very being, only driven by the needs of contributors.

If you don=B4t want to wait, use another browser.

As for copy/paste, it is actually the oldest entry in the bug-track engin=
e, and
I agree this is an important one. I filed it, then I volunteered to imple=
ment
it. However, many issues are in the way. First of all, Dillo=B4s internal
representation has changed a lot and keeps changing. Secondly, I didn=B4t=
have
time so far to give a try to the new iterators.

In short, this is far from being =A8a simple task=A8.

You want to know when: when somebody codes it! Simple as that. I am doing=
it,
slowly (don=B4t hold your breath). If somebody else does it before me, mo=
re power
to him!

Just sit tight and wait.

Best,
Eric

-- En reponse de "Re: [Dillo-dev]Copy/paste in Dillo" de Jamin W. Collins=
, le
27-Jul-2002 :
> On Sat, 27 Jul 2002 13:42:21 -0500 (CDT)
> John Utz <john@ut...> wrote:
>=20
>> We await your patch!
>>=20
>> (which is a nice way of saying, if you wont fix it yourself, stop=20
>> bitching)
>=20
> Curious, is there a problem with users of Dillo asking for or expressin=
g
> interested in having a feature added? I didn't take the OP's request a=
s
> "bitching".
>=20
> --=20
> Jamin W. Collins
>=20
>=20
> -------------------------------------------------------
> This s...net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Dillo-dev mailing list
> Dillo-dev@li...
> https://lists.so....net/lists/listinfo/dillo-dev

------------------------------------------------------------------------
Eric GAUDET <eric@rt...>
Le 27-Jul-2002 a 20:30:25
"Parler pour ne rien dire et ne rien dire pour parler sont les deux
principes majeurs et rigoureux de tous ceux qui feraient mieux de la
fermer avant de l'ouvrir."
------------------------------------------------------------------------ 



Re: [Dillo-dev]libdillo

From: Jonathan Gardner <gardnerj@pr...> - 2002-07-28 01:48

On Tue, 23 Jul 2002 09:06:26 +0200
Melvin Hadasht <melvin.hadasht@fr...> wrote:

> Mean while, I made a small patch for the same reason but with another
> approach, if you're interested:
> http://melvin.hadasht.free.fr/home/dillo/sylpheed/ 

Speaking of this patch. It still worked as of the 0.7.8claws release of
Sylpheed-claws but now as of the 0.8.0claws release it doesn't patch so
well and the make doesn't work.
Any plans on updating the patch?
Thanks much for it. It really makes Sylpheed even more nifty than it
already was.

Cheers,
Jonathan

-- 
#define __INTELLECTUAL_CHRISTIAN__

#include <JESUS><
#include ><DARWIN>
L L 



Re: [Dillo-dev]Copy/paste in Dillo

From: John Utz <john@ut...> - 2002-07-28 01:19

umm, i overspoke.

sorry.

On Sat, 27 Jul 2002, Jamin W. Collins wrote:

> On Sat, 27 Jul 2002 13:42:21 -0500 (CDT)
> John Utz <john@ut...> wrote:
> 
> > We await your patch!
> > 
> > (which is a nice way of saying, if you wont fix it yourself, stop 
> > bitching)
> 
> Curious, is there a problem with users of Dillo asking for or expressing
> interested in having a feature added? I didn't take the OP's request as
> "bitching".
> 
> 

-- 

John L. Utz III
john@ut...

Idiocy is the Impulse Function in the Convolution of Life 



Re: [Dillo-dev]Copy/paste in Dillo

From: Jamin W. Collins <jcollins@as...> - 2002-07-27 19:54

On Sat, 27 Jul 2002 13:42:21 -0500 (CDT)
John Utz <john@ut...> wrote:

> We await your patch!
> 
> (which is a nice way of saying, if you wont fix it yourself, stop 
> bitching)

Curious, is there a problem with users of Dillo asking for or expressing
interested in having a feature added? I didn't take the OP's request as
"bitching".

-- 
Jamin W. Collins 



Re: [Dillo-dev]Copy/paste in Dillo

From: John Utz <john@ut...> - 2002-07-27 18:42

We await your patch!

(which is a nice way of saying, if you wont fix it yourself, stop 
bitching)

On 27 Jul 2002, mightyunclean wrote:

> Anyone know when support for text copy/paste will be incorporated into
> Dillo ?
> 
> Or if there is a patch available to enable support for this ?
> 
> Have been following Dillo's development from about 0.4.0, but there
> still hasn't been any support for this.
> I wonder if anyone has even thought of this simple but important
> feature.
> It is so frustrating being forced to use some other browser all the time
> simply because Dillo lacks basic functionality.
> Apart from this, everything is fine & dandy.
> 
> Keep up the good work fellas.
> 
> 
> 
> http://www.sold.com.au - SOLD.com.au
> - Find yourself a bargain!
> 
> 
> -------------------------------------------------------
> This s...net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Dillo-dev mailing list
> Dillo-dev@li...
> https://lists.so....net/lists/listinfo/dillo-dev
> 

-- 

John L. Utz III
john@ut...

Idiocy is the Impulse Function in the Convolution of Life 



[Dillo-dev]Copy/paste in Dillo

From: mightyunclean <mightyunclean@ne...> - 2002-07-27 07:52

Anyone know when support for text copy/paste will be incorporated into
Dillo ?

Or if there is a patch available to enable support for this ?

Have been following Dillo's development from about 0.4.0, but there
still hasn't been any support for this.
I wonder if anyone has even thought of this simple but important
feature.
It is so frustrating being forced to use some other browser all the time
simply because Dillo lacks basic functionality.
Apart from this, everything is fine & dandy.

Keep up the good work fellas.



http://www.sold.com.au - SOLD.com.au
- Find yourself a bargain! 



Re: [Dillo-dev]libdillo

From: Melvin Hadasht <melvin.hadasht@fr...> - 2002-07-23 07:06

Hi erez@sa...,

on Tue, 23 Jul 2002 03:18:37 -1:00
erez@sa... wrote:

> I was wondering if it is possible to
> split dillo into a gtk-widget-lib
> 
> (libdillo) and an application, 
> so people could use libdillo and have
> html in their apps

that would be great. 

Mean while, I made a small patch for the same reason but with another approach,
if you're interested:
http://melvin.hadasht.free.fr/home/dillo/sylpheed/ 

Namlala unono (Comoran: Goodbye)


-- 
Melvin Hadasht 



[Dillo-dev]libdillo

From: <erez@sa...> - 2002-07-23 06:17

hello

I`m currently writing a mail client for
linux/ipaq, and I need a html widget.

I thought of using dillo for this
( porting it`s source into my app)
but as I already have dillo, this will
make the same code twice, and it`s a
waste of flash,and I will have to
port again when there will be a new
vesion of dillo.

I was wondering if it is possible to
split dillo into a gtk-widget-lib

(libdillo) and an application, 
so people could use libdillo and have
html in their apps


ps,
I`m not on this mailing list so please
also CC to me.

thanks,
erez. 



RE: [Dillo-dev]ecmascript, javascript and CSS 

From: floorzat floorzat <floorzat@ho...> - 2002-07-22 21:58

Um.....
The codebase structure severely restricts the possibility of reasonable 
implementation of script controlled DOM.
New Eventloop structures and highly complex semaphor systems would be 
required and a complete ground up recode would probably be simpler.



_________________________________________________________________
Join the world’s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com 



Re: [Dillo-dev]Pixmap GTK themes

From: Patrick Glennon <pglennon@me...> - 2002-07-21 17:31

Attachments: Message as HTML      

I have done a lot of pixmap themeing with dillo with no problems, from
v. 0.6.2 up to relatively current CVS builds. Also used thinice engine
and xenophilia with no problems...

do you have gtk themes properly set up? are your path names to the
engine libraries properly referenced in the gtkrc file? Also, do you
have the gtk/gdk libs installed? another thing to check is to make
sure that the theme engine libraries are in ( usually )
/usr/lib/gtk/themes/engines

libpixmap.so - pixmap theme engine
libxeno.so - xenophilia theme engine
libthinice.so - thinice theme engine
( etc.. )

hope this helps, and is not just a regurgitation of the obvious...

Cheers,
-Patrick

On Sat, 2002-07-20 at 21:59, Mark Schreiber wrote:

Does anyone else have any problems using pixmap GTK themes with dillo?
I'm not seeing any pixmaps in pixmap engine gtk themes being drawn in
a dillo (current CVS) window, though all the other GTK apps I use work
fine. I'd like to see whether this is just my copy of dillo first,
though.

If anyone has a pixmap gtk theme on their computer, could you try
switching to it and see whether the pixmaps are visible in your dillo
window?

Thanks!

-- 
Best of luck,
Mark Schreiber 



[Dillo-dev]Pixmap GTK themes

From: Mark Schreiber <mark7@an...> - 2002-07-21 02:59

Attachments: Message as HTML      

Does anyone else have any problems using pixmap GTK themes with dillo?
I'm not seeing any pixmaps in pixmap engine gtk themes being drawn in
a dillo (current CVS) window, though all the other GTK apps I use work
fine. I'd like to see whether this is just my copy of dillo first,
though.

If anyone has a pixmap gtk theme on their computer, could you try
switching to it and see whether the pixmaps are visible in your dillo
window?

Thanks!

--=20
Best of luck,
Mark Schreiber 



[Dillo-dev]Re: Reading gzipped files

From: Vikas G P <vikasgp386@ho...> - 2002-07-17 13:20

On Mon,15 Jul 2002 07:23 PM Ken Hayber wrote,

>My point was that I believe that when http servers like apache support
>gzip they do not change the filename, they just compress the data on the
>fly. I could be wrong too, but that's what I believe is the case. It is
>the content-coding attribute that tells the browser it is compressed, not
>the filename.
>
>Maybe they're just two complementary ways to accomplish the same thing:
>.gz for local files, content-coding for http-served files. Would it be
>hard to implement both

Ah,
I get it now. What Ken is saying is that we need to support on-the-fly decompession of content sent via http, by checking their
content-encoding attribute. The reason I did it only for local files is :
I don't think there are many sites out there which put up index.html.gz. And gzipped files on the server should probably be
downloaded, and I didn't know how to do it.
BTW, most other browsers offer to download a .html.gz sent over http.

On Mon, 15 Jul 2002 08:04 PM madis wrote,

>On Mon, 15 Jul 2002, Ken Hayber wrote:

>> Maybe they're just two complementary ways to accomplish the same thing:
>> ..gz for local files, content-coding for http-served files. Would it be
>> hard to implement both?

>I looked for it now, seems that it requires modifing IO/http.c and cache.c
>(at least). Implementing this way would be *the right thing* imho, but it
>requires fully understanding how these two things talk with each other
>(what i didn't bothered). Main problem is that a part of http traffic is
>gzip'ed (body), but another part is not (header).

Yes, IMHO it is the *right thing* to do.

Vikas G P 



[Dillo-dev]gzipped HTML files

From: Geoff Lane <zzassgl@tw...> - 2002-07-17 07:46

A few months ago I posted some patches for gzipped HTML support. These can
be found at
http://twirl.mcc.ac.uk/dillo/dillo_gzip.html
There is also a very brief discussion about the requirements and possible
improvements. The patches will probably need to be re-worked for the recent
dillo releases. Anyone who want's to grab a copy and work up a proper patch
for dillo, please do so.

I see that Apache2 contains mod_gzip as a standard module - there's probably
going to be a lot more compressed web pages in the future.

-- 
/\ Geoff. Lane. /\ Manchester Computing /\ Manchester /\ M13 9PL /\ England /\

Govt investigations contribute more to amusement than knowledge. 



Re: [Dillo-dev]RE: Dillo-dev digest, Vol 1 #539 - 1 msg

From: madis <madis@cy...> - 2002-07-16 19:53

On 16 Jul 2002, Patrick Glennon wrote:

> Go into ~/.dillo/dillorc and make changes to these entries, depending on
> how much you want to strip it down:
> 
> show_back=NO
> show_forw=NO
> show_home=NO 
> show_reload=NO
> show_save=NO
> show_stop=NO
> show_menubar=NO
> show_clear_url=NO
> show_url=NO
> show_progress_box=NO
> 
and...
panel_size=tiny

-- 
mzz 



Re: [Dillo-dev]RE: Dillo-dev digest, Vol 1 #539 - 1 msg

From: Melvin Hadasht <melvin.hadasht@fr...> - 2002-07-16 19:43

Hi Jedediah,

on Tue, 16 Jul 2002 16:31:41 -0300 "Jedediah Ferguson"
<jFerguson@tr...> wrote:

> I was wondering if there is a means to suppress the navigation bar.
> I am only running a 1/4 VGA display and it would be advantageous is
> the naviagtion bar could be removed or hidden upon initialization as
> it would make for a cleaner GUI.

You can double-click inside the window to go into full-window mode.
There is no command line option to run it from the start in that mode. But I
have a preliminary patch here:

http://melvin.hadasht.free.fr/home/dillo/fullwindow/index.html

Cheers

-- 
Melvin Hadasht 



Re: [Dillo-dev]RE: Dillo-dev digest, Vol 1 #539 - 1 msg

From: Jamin W. Collins <jcollins@as...> - 2002-07-16 19:41

On Tue, 16 Jul 2002 16:31:41 -0300 "Jedediah Ferguson"
<jFerguson@tr...> wrote:

> I was wondering if there is a means to suppress the navigation bar. I
> am only running a 1/4 VGA display and it would be advantageous is the
> naviagtion bar could be removed or hidden upon initialization as it
> would make for a cleaner GUI.

Double-click the background of the browser's page.

-- 
Jamin W. Collins 



Re: [Dillo-dev]RE: Dillo-dev digest, Vol 1 #539 - 1 msg

From: Patrick Glennon <pglennon@me...> - 2002-07-16 19:38

Attachments: Message as HTML      

Go into ~/.dillo/dillorc and make changes to these entries, depending on
how much you want to strip it down:

show_back=NO
show_forw=NO
show_home=NO 
show_reload=NO
show_save=NO
show_stop=NO
show_menubar=NO
show_clear_url=NO
show_url=NO
show_progress_box=NO

I also use it on an embedded system, and have also found it helpful to
blank the X cursor... let me know if you would like a patch for
that... it just amounts to replacing the gdk cursor call with an inline
blank gdk cursor creation, and it is just two files that need to be
modified....

Cheers!

-Patrick

On Tue, 2002-07-16 at 14:31, Jedediah Ferguson wrote:

Dillo-Dev,

Dillo is an impressive web browser. I am currently running it on a minimal Debian system on a single board computer in the hopes of using it as a GUI. Dillo runs much more quickly than other browsers I have tried. It can even render complex sites like Slashdot almost as fast as the video can refresh.

I was wondering if there is a means to suppress the navigation bar. I am only running a 1/4 VGA display and it would be advantageous is the naviagtion bar could be removed or hidden upon initialization as it would make for a cleaner GUI.

Thanks for a Great Program.


Best Regards,

Jedediah Ferguson
Computer Engineering Co-op

mailto:Jedediah.Ferguson@tr...
http://www.trecan.com

4049 St. Margaret's Bay Road
Hubley, Nova Scotia, B3Z 1C2, Canada

Phone: 902 876-0457
Fax: 902 876-8275





-------------------------------------------------------
This s...net email is sponsored by: Jabber - The world's fastest growing 
real-time communications platform! Don't just IM. Build it in! 
http://www.jabber.com/osdn/xim
_______________________________________________
Dillo-dev mailing list
Dillo-dev@li...

https://lists.so....net/lists/listinfo/dillo-dev 



[Dillo-dev]RE: Dillo-dev digest, Vol 1 #539 - 1 msg

From: Jedediah Ferguson <jFerguson@tr...> - 2002-07-16 19:31

Dillo-Dev,

Dillo is an impressive web browser. I am currently running it on a =
minimal Debian system on a single board computer in the hopes of using =
it as a GUI. Dillo runs much more quickly than other browsers I have =
tried. It can even render complex sites like Slashdot almost as fast as =
the video can refresh.

I was wondering if there is a means to suppress the navigation bar. I =
am only running a 1/4 VGA display and it would be advantageous is the =
naviagtion bar could be removed or hidden upon initialization as it =
would make for a cleaner GUI.

Thanks for a Great Program.


Best Regards,

Jedediah Ferguson
Computer Engineering Co-op

mailto:Jedediah.Ferguson@tr...
http://www.trecan.com

4049 St. Margaret's Bay Road
Hubley, Nova Scotia, B3Z 1C2, Canada

Phone: 902 876-0457
Fax: 902 876-8275 



Re: [Dillo-dev]Netscape bookmarks conversion?

From: Sebastian Geerken <sgeerken@st...> - 2002-07-16 15:14

On Thu, Jul 11, 2002 at 01:16:53PM -0500, Paul Chamberlain wrote:
> I think you set my expectations too high. I have a LOT of
> bookmarks in mozilla. I'm not sure which one caused it,
> but when dillo tried to load, it said:
> 
> Loading bookmarks...
> 
> GLib-ERROR **: could not allocate -75 bytes
> aborting...
> Aborted (core dumped)

Funnily, I had them in the booksmark.html file, i.e. something like

---
...
<li><A HREF="...">...</A></li>

** CRITICAL **: file web.c: line 56 (a_Web_dispatch_by_type): assertion `dw != NULL' failed.
<li><A HREF="...">...</A></li>

** WARNING **: 11 styles left

** WARNING **: 2 fonts (11 references) left

** WARNING **: 2 colors (11 references) left

** WARNING **: 1 shaded colors (8 references) left
<li><A HREF="...">...</A></li>
...
---

I cannot reproduce this, someone who can should insert an entry in to
the bug tracking engine. (I'm not even sure if these bugs still exist,
I rarely use the "View Bookmarks" function.)

> This s...net email is sponsored by:ThinkGeek
> PC Mods, Computing goodies, cases & more
> http://thinkgeek.com/sf

Seems that is time to switch the list hoster.

Sebastian 



Re: [Dillo-dev]Re:Reading gzipped files

From: Ken Hayber <khayber@so...> - 2002-07-15 14:48

I looked a bit more at the RFCs and see that there is also a 
transfer-coding attribute that seems to be more directed at what I am 
talking about. It seems a bit confusing; maybe I'll take a look at the 
apache code and see if I can tell what they are doing. It seems that 
content-coding may be intended to describe the content (not modify it) and 
transfer coding describes how it is being transferred (on the fly, e.g. 
chunked, gzip, etc).

I found a small discussion indicating that a content-coding of gzip might 
indicate to the browser (UA) to save the file rather than display it, 
whereas the transfer-encoding would indicate that the UA should decode it.

(http://ftp.ics.uci.edu/pub/ietf/http/hypermail/1999/0197.html)



On Monday, July 15, 2002, at 07:52 AM, madis wrote:

> On Mon, 15 Jul 2002, Ken Hayber wrote:
>
>> My point was that I believe that when http servers like apache support
>> gzip they do not change the filename, they just compress the data on the
>> fly. I could be wrong too, but that's what I believe is the case. It is
>> the content-coding attribute that tells the browser it is compressed, not
>> the filename.
>>
>> Maybe they're just two complementary ways to accomplish the same thing:
>> ..gz for local files, content-coding for http-served files. Would it be
>> hard to implement both?
>
> I looked for it now, seems that it requires modifing IO/http.c and cache.
> c
> (at least). Implementing this way would be *the right thing* imho, but it
> requires fully understanding how these two things talk with each other
> (what i didn't bothered). Main problem is that a part of http traffic is
> gzip'ed (body), but another part is not (header).
> 



Re: [Dillo-dev]Re:Reading gzipped files

From: madis <madis@cy...> - 2002-07-15 14:35

On Mon, 15 Jul 2002, Ken Hayber wrote:

> My point was that I believe that when http servers like apache support 
> gzip they do not change the filename, they just compress the data on the 
> fly. I could be wrong too, but that's what I believe is the case. It is 
> the content-coding attribute that tells the browser it is compressed, not 
> the filename.
> 
> Maybe they're just two complementary ways to accomplish the same thing: 
> ..gz for local files, content-coding for http-served files. Would it be 
> hard to implement both?

I looked for it now, seems that it requires modifing IO/http.c and cache.c 
(at least). Implementing this way would be *the right thing* imho, but it 
requires fully understanding how these two things talk with each other 
(what i didn't bothered). Main problem is that a part of http traffic is 
gzip'ed (body), but another part is not (header).

--
mzz 



Re: [Dillo-dev]Re:Reading gzipped files

From: Ken Hayber <khayber@so...> - 2002-07-15 13:53

Well, if you're trying to add support to open .html.gz files like some 
other apps (I've seen others like .ps.gz, ...?) can do then I think you're 
fine.

My point was that I believe that when http servers like apache support 
gzip they do not change the filename, they just compress the data on the 
fly. I could be wrong too, but that's what I believe is the case. It is 
the content-coding attribute that tells the browser it is compressed, not 
the filename.

Maybe they're just two complementary ways to accomplish the same thing: 
.gz for local files, content-coding for http-served files. Would it be 
hard to implement both?

On Sunday, July 14, 2002, at 11:22 PM, Vikas G P wrote:

> On Friday, July 12, 2002, Ken Hayber wrote,
>
>> Not to be negative, but isn't the proper way to handle gzip in a browser
>> to check the content-coding attribute and not the filename/ext?
>
> As I said, the patch only works for _local_ files, so we get it's type 
> from it's extension, and if the ext is .gz,
> it is stripped and the resulting name is examined again.
>
>> I thought
>> the filename stays the same.
>
> Where does the filename change ?
> Or am I getting this wrong ?
>
> Vikas G P
>
>
>
>
> -------------------------------------------------------
> This s...net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Dillo-dev mailing list
> Dillo-dev@li...
> https://lists.so....net/lists/listinfo/dillo-dev
>
>

"Nothing is impossible for anyone impervious to reason."
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Ken Hayber
mailto:khayber@so...
http://khayber.dyndns.org 



[Dillo-dev]Re:Reading gzipped files

From: Vikas G P <vikasgp386@ho...> - 2002-07-15 07:00

On Friday, July 12, 2002, Ken Hayber wrote,

> Not to be negative, but isn't the proper way to handle gzip in a browser 
> to check the content-coding attribute and not the filename/ext? 

As I said, the patch only works for _local_ files, so we get it's type from it's extension, and if the ext is .gz,
it is stripped and the resulting name is examined again.

> I thought 
> the filename stays the same.

Where does the filename change ?
Or am I getting this wrong ?

Vikas G P 



Re: [Dillo-dev]Netscape bookmarks conversion?

From: Tomas Guemes <tomas@pa...> - 2002-07-14 02:20

On 13 Jul 2002 09:56:31 -0500
"Lars Clausen" <lrclause@cs.uiuc.edu> wrote:

> 
> On Sat, 13 Jul 2002, Tomas Guemes wrote:
> 
> > I made a patch so dillo can support subdirectories in the bookmarks
> > menu, cause otherwise if you have a lot of entries, part of the menu
> > will be out of the screen and could not be reached with the actual
> > flat bookmarks scheme. Also I have modified the python script
> > xbel2html, to fit the format of the patch.
> 
> Ohhh! I wanted to see that! Thank you.
> 
> > You can find the patch and the script xbel2dbm in
> > http://golem/tomas/dillo/
> 
> Uhm... golem? That must be a new top-level domain that I've never
> heard of before:)
> 

Sorry, this is the internal name :)
http://pasky.dhs.org/tomas/dillo/

> -Lars
> 
> -- 
> Lars Clausen (http://shasta.cs.uiuc.edu/~lrclause)| Hårdgrim of
> Numenor"I do not agree with a word that you say, but I 
> |---------------------------- will defend to the death your right to
> |say it." | Where are we going, and --Evelyn Beatrice Hall
> |paraphrasing Voltaire | what's with the handbasket?
> 
> 
> -------------------------------------------------------
> This s...net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Dillo-dev mailing list
> Dillo-dev@li...
> https://lists.so....net/lists/listinfo/dillo-dev
> 



Re: [Dillo-dev]Netscape bookmarks conversion?

From: Lars Clausen <lrclause@cs...> - 2002-07-13 14:56

On Sat, 13 Jul 2002, Tomas Guemes wrote:

> I made a patch so dillo can support subdirectories in the bookmarks
> menu, cause otherwise if you have a lot of entries, part of the menu
> will be out of the screen and could not be reached with the actual flat
> bookmarks scheme. Also I have modified the python script xbel2html, to
> fit the format of the patch.

Ohhh! I wanted to see that! Thank you.

> You can find the patch and the script xbel2dbm in
> http://golem/tomas/dillo/

Uhm... golem? That must be a new top-level domain that I've never heard of
before:)

-Lars

--=20
Lars Clausen (http://shasta.cs.uiuc.edu/~lrclause)| H=E5rdgrim of Numenor
"I do not agree with a word that you say, but I |------------------------=
----
will defend to the death your right to say it." | Where are we going, and
--Evelyn Beatrice Hall paraphrasing Voltaire | what's with the handbas=
ket? 



Re: [Dillo-dev]Netscape bookmarks conversion?

From: Tomas Guemes <tomas@pa...> - 2002-07-13 13:14

On Thu, 11 Jul 2002 03:16:21 +0200 (MET DST)
"Lawrence Mayer" <Lawrence.Mayer@ds...> wrote:

> Hi,
> 
> I use Dillo, Links, Lynx, and Netscape. So far, all my bookmarks are
> in Netscape, but I would love to be able to share my bookmarks between
> these 4 browsers.

> 2. Any utilities which CONVERT bookmarks back and forth between the
> above browsers, or at least some of them?

I found a way to convert through xbel. 

Xbel is the XML format used by konqueror, there are some programs to
convert xbel to other formats and viceversa, the debian packages are
xbel and xbel-utils you can read more and find them in
http://pyxml.so....net/topics/xbel/

I made a patch so dillo can support subdirectories in the bookmarks
menu, cause otherwise if you have a lot of entries, part of the menu
will be out of the screen and could not be reached with the actual flat
bookmarks scheme. Also I have modified the python script xbel2html, to
fit the format of the patch.

You can find the patch and the script xbel2dbm in
http://golem/tomas/dillo/

greetings

Tomas 



Re: [Dillo-dev][PATCH] reading gzipped files

From: Ken Hayber <khayber@so...> - 2002-07-12 13:35

Not to be negative, but isn't the proper way to handle gzip in a browser 
to check the content-coding attribute and not the filename/ext? The rfc 
says on-the-fly encoding/decoding between server and client. I thought 
the filename stays the same.

bug #156:
Impact: feature fault
At: run time
Type: misbehaviour
Comments: Dillo doesn't render gzipped html's. Sourceforge (once) put
up a gzipped inde.html and all I got was binary :-(
This is kind of new, but nice!
HTReproduce: Just gzip a local .html. Try to look at it in Netscape and
then in Dillo.


On Friday, July 12, 2002, at 12:38 AM, Vikas G P wrote:

> Hi everyone,
>
> Here's a small patch to make dillo read gzipped files[BUG #156]. It only 
> works only on local files, since a .gz file on a server is
> probably there for download. It sends the content-length as 0, which is 
> ugly, but seems to work fine.
>
> Happy coding,
> Vikas 



[Dillo-dev][PATCH] reading gzipped files

From: Vikas G P <vikasgp386@ho...> - 2002-07-12 07:33

Hi everyone,

Here's a small patch to make dillo read gzipped files[BUG #156]. It only works only on local files, since a .gz file on a server is
probably there for download. It sends the content-length as 0, which is ugly, but seems to work fine.

Happy coding,
Vikas

--------------------
--- dillo-0.6.6/src/IO/file.c Wed Apr 10 06:29:42 2002
+++ dillo-0.6.6.patch/src/IO/file.c Thu Jul 11 18:08:23 2002
@@ -29,6 +29,7 @@
#include <stdio.h>
#include <signal.h>
#include <math.h> /* for rint */
+#include <zlib.h>

#include <errno.h> /* for errno */
#include "Url.h"
@@ -80,7 +81,8 @@ static gint File_get_file(const gchar *F
static gint File_get_dir(const gchar *DirName);
static char *File_dir2html(DilloDir *Ddir);
static void File_not_found_msg(DilloWeb *web, const char *filename, int fd);
-
+static gint File_ext(const char *filename,const char *ext);
+static char *File_content_type_gz(const char *filename);

/*
* Allocate a DilloFile structure, and set working values in it.
@@ -162,6 +164,7 @@ static void *File_transfer_file(void *da
char buf[8192];
DilloFile *Dfile = data;
ssize_t nbytes;
+ gzFile gzipfile;

/* Set this thread to detached state */
pthread_detach(Dfile->th1);
@@ -172,7 +175,8 @@ static void *File_transfer_file(void *da

/* Send File Size info */
if (Dfile->FileSize != -1) {
- sprintf(buf, "Content-length: %ld\n", Dfile->FileSize);
+ sprintf(buf, "Content-length: %ld\n",
+ File_ext(Dfile->Filename,"gz") ? 0l : Dfile->FileSize );
write(Dfile->FD_Write, buf, strlen(buf));
}
/* Send end-of-header */
@@ -181,8 +185,23 @@ static void *File_transfer_file(void *da


/* Append raw file contents */
- while ( (nbytes = read(Dfile->FD, buf, 8192)) != 0 ) {
- write(Dfile->FD_Write, buf, nbytes);
+
+ /* gzipped */
+ if( File_ext(Dfile->Filename,"gz")){
+ if( (gzipfile = gzdopen(Dfile->FD,"rb")) ){;
+ while( (nbytes = gzread(gzipfile, buf, 8192)) != 0){
+ write(Dfile->FD_Write, buf, nbytes);
+ }
+ }
+ else{
+ g_print("Error: Could not decompress gzipped file\n");
+ }
+ }
+ /* not gzipped */
+ else{
+ while ( (nbytes = read(Dfile->FD, buf, 8192)) != 0 ) {
+ write(Dfile->FD_Write, buf, nbytes);
+ }
}

close(Dfile->FD);
@@ -258,8 +277,21 @@ static char *File_content_type(const cha
} else if (File_ext(filename, "html") || File_ext(filename, "htm") ||
File_ext(filename, "shtml")) {
return "text/html";
+ } else if(File_ext(filename, "gz")){
+ return File_content_type_gz(filename);
}
return "text/plain";
+}
+
+/*
+ * Return content_type of the uncompressed file
+ */
+static char *File_content_type_gz(const char *filename)
+{
+ gchar *original_name = g_strndup(filename,rindex(filename,(int)'.')-filename);
+ gchar *content_type = File_content_type(original_name);
+ g_free(original_name);
+ return content_type;
}

/* 



Re: [Dillo-dev]Netscape bookmarks conversion?

From: Paul Chamberlain <tif@ti...> - 2002-07-11 18:17

Ross J. Reedstrom wrote:
> Well, for dillo and netscape, you can just put a symlink in one or the other's
> dot directory to the others bookmarks.html file. I just tested this by
> symlinking ~/.diullorc/bookmarks.html to ~/.mozilla/default/bookmarks.html

I think you set my expectations too high. I have a LOT of
bookmarks in mozilla. I'm not sure which one caused it,
but when dillo tried to load, it said:

Loading bookmarks...

GLib-ERROR **: could not allocate -75 bytes
aborting...
Aborted (core dumped)
--
Paul Chamberlain, tif@ti... 



Re: [Dillo-dev]Netscape bookmarks conversion?

From: Pat Shanahan <pat@em...> - 2002-07-11 11:34

* Ross J. Reedstrom <reedstrm@ri...> [07-11-02 05:05]:
> On Thu, Jul 11, 2002 at 03:16:21AM +0200, Lawrence Mayer wrote:
> > Hi,
> > 
> > I use Dillo, Links, Lynx, and Netscape. So far, all my bookmarks are in
> > Netscape, but I would love to be able to share my bookmarks between these
> > 4 browsers.
> 
> Well, for dillo and netscape, you can just put a symlink in one or the other's
> dot directory to the others bookmarks.html file. I just tested this by
> symlinking ~/.diullorc/bookmarks.html to ~/.mozilla/default/bookmarks.html
> 
> It worked fine: the 'View bookmarks" function in dillo handled the page just
> fine. with sections and everything.

You may also make bookmarks in netscape/dillo for each, ie:
file:///home/yurName/.dillo/bookmarks.html in Netacape &
file:///home/yurName/.mozilla/default/bookmarks.html in Dillo

-- 
Patrick Shanahan
Registered Linux User #207535 
@ http://counter.li.org 



Re: [Dillo-dev]Netscape bookmarks conversion?

From: Ross J. Reedstrom <reedstrm@ri...> - 2002-07-11 05:49

On Thu, Jul 11, 2002 at 03:16:21AM +0200, Lawrence Mayer wrote:
> Hi,
> 
> I use Dillo, Links, Lynx, and Netscape. So far, all my bookmarks are in
> Netscape, but I would love to be able to share my bookmarks between these
> 4 browsers.

Well, for dillo and netscape, you can just put a symlink in one or the other's
dot directory to the others bookmarks.html file. I just tested this by
symlinking ~/.diullorc/bookmarks.html to ~/.mozilla/default/bookmarks.html

It worked fine: the 'View bookmarks" function in dillo handled the page just
fine. with sections and everything.

Ross 



[Dillo-dev]Netscape bookmarks conversion?

From: Lawrence Mayer <Lawrence.Mayer@ds...> - 2002-07-11 01:16

Hi,

I use Dillo, Links, Lynx, and Netscape. So far, all my bookmarks are in
Netscape, but I would love to be able to share my bookmarks between these
4 browsers.

Can anyone offer any advice?

1. Will Netscape's bookmark import do me any good here?

2. Any utilities which CONVERT bookmarks back and forth between the above
browsers, or at least some of them?

3. Any utilities which SYNCHRONIZE bookmarks between the above browsers,
or at least some of them?


I would be thankful for your advice or tips.

Friendly Greetings,
Lawrence Mayer <lawmay@ki.se>
Ume=E5, Sweden 



Re: [Dillo-dev]Cookies don't work

From: Jorgen Viksell <jorgen.viksell@te...> - 2002-07-10 13:49

> Allo,

Hi,
=20
> I compiled dillo-0.6.6, passing the parameter --enable-cookies, and got n=
o
> errors; made sure my ~/.dillo/cookiesrc says:
> DEFAULT ACCEPT
>=20
> yet cookies don't appear to work. ~/.dillo/cookies remains empty (0 byte=
)
> and web pages I visit don't retain my login information. Tested against
> http://www.livejournal.com, http://www.w30wnzj00.com, http://www.slashdot.org. =20

Slashdot works for me.=20
http://www.w30wnzj00.com seems to send a malformed expiry date and gets
ignored.=20
Unfortunately, I can't test livejournal ATM.

> Please advise; is there anything I did wrong, or any logfiles I should be=
looking at?
> Thanks.

You could change the line (in cookies.c) which says:
#define DEBUG_LEVEL 8
to:
#define DEBUG_LEVEL 1

Then it would tell you pretty much everything it does. A good starting
point would be to check if the date strings are parsed correctly.

>=20
> matthew@de...
>=20

Cheers,
J=F6rgen 



[Dillo-dev]Cookies don't work

From: matthew <matthew@de...> - 2002-07-10 12:59

Allo,

I compiled dillo-0.6.6, passing the parameter --enable-cookies, and got no
errors; made sure my ~/.dillo/cookiesrc says:
DEFAULT ACCEPT

yet cookies don't appear to work. ~/.dillo/cookies remains empty (0 byte)
and web pages I visit don't retain my login information. Tested against
http://www.livejournal.com, http://www.w30wnzj00.com, http://www.slashdot.org. Please advise;
is there anything I did wrong, or any logfiles I should be looking at?
Thanks.


matthew@de... 



[Dillo-dev]dillo user notes

From: Abc Xyz <abc@an...> - 2002-07-10 10:35

fbsd 4.5 dillo .66

i didn't use this more than a few minutes, but this is what
i was missing severely enough to stop ...

no keyboard navigation (esp backward foreward - Left/Right arrows).
(i hate having to reach for my mouse needlessly).
animated gifs apparently broken.
reload HTML meta tags don't work.
missing backwards/forewards in right-click mouse menu.

minor stuff ...

would be nice if "view source" opened to the same size as parent window.
i dunno if it does YahooMail and Hotmail, but that would be nice.
hopefully it uses ~/.mailcap ...
hopefully it allows https? for online ordering?
HTML warning: hexadecimal color lacks leading '#' - oh pooh.
WARNING: Cache_stop_client, inexistent client - lots of these.

wish list:

UP-ARW - up one link (or half/most of a page if no link).
DN-ARW - same, but downwards (like lynx - invert link colors).

ALT-UP-ARW - smooth scrolling up.
ALT-DN-ARW - smooth scrolling down.
ALT-LT-ARW - smooth scrolling left.
ALT-RT-ARW - smooth scrolling right.

i guess i would like to see that the user interface is
as snappy and hassle free as the browser :)

but all in all, this browser is small and snappy,
and renders *extremely* nicely. i can't wait till
i can use it and toss my 14MB Navigator!
with 40MB of Linux emulation :) 



Re: [Dillo-dev]ecmascript, javascript and CSS

From: Freya <Freya@he...> - 2002-07-09 11:31

> > Is there an intention to implement these scripts?
> 
> Both JavaScript and Java would greatly increase the size of Dillo and

Incidently, I've never felt the need to have java, I've never come
across any site that really required it that much on the client side
beyone a spectrum emulator and some really clever audio/video
technology. I think most Java is done through servlets and the like, I
do hope so anyway! :)

> require numerous hooks into the existing code to obtain the
> functionallity seen in other browsers. For example, in DHTML, you
> perform special effects by tagging various events with a fragment of
> JavaScript so, for example, when the mouse moves over a specific area
> of the screen a button lights up or additional text appears. This has
> some deep implications about the way that Dillo paints the window
> images.

I've been wondering about this, Mozilla of course has preety good
javascript support and the thing is, is that it uses some kind of plugin
architechture for the scripting, so I was wondering if we could just
provide support for the plugin architexture and just use their scripting
engine, or whatever scripting engines are available, or even no
scripting engine.

The trouble I have with this idea, is that all this talk of Mozilla just
having a javascript plugin, is I suspect a real oversimplification of
things. It surely must talk directly to the rendering engine at a really
low and nasty level. Whats worse, is I don't even think it's as simple
as having implications in the way that dillo paints the window images,
as I understand it you would have to implement some kind of object model
in the rendering engine too? This seems to me like a massive task unless
the object model is implemented in the js engine which I strongly
suspect it isn't, it's too specific to rendering and I think their
javascript engine is also used in the netscape webserver, or at least
used to be.

Personally I can also live without things like DHTML support. I'd really
just like support something like that of Netscape 3, enough to let me
log in to yahoo mail and the like, however the more I think about it the
more nasty it seems. 


> > how about flash?

I've often thought it would be nice to just have a flash browser
seperately from my webbrowser that could just open flash site if I felt
the need or show animations, I don't see much point in it being part of
the browser and I generally hate flash as it's mostly just used for
really annoying adverts.

> If anyone has some spare time, one project that would usefully
> demonstrate the kind of changes that would be needed in the Dillo
> image code is implementing GIF and PNG animations. You would need
> some kind of regular"update animation" event and some state
> information associated with animation files. A thread[1] per animated
> image using some kind of "server push" scheme may work; or more
> crudely, you could get the cache to manage it all and set up some kind
> of list of "repaint screen area x,y,w,h" events.

IF anyone does implement this please also implement an option to turn it
off! I think this is kind of low priority myself as it doesn't affect
the usability of a website much. 
CSS was mentioned in the subject line, but no-one has really talked
about it here. Of all the technologies unimplimented, this is in a way
the nicest, as it's not actually fudamentally bad technology in the way
that javascript or flash is. From the point of view of dillo
implementing good technology really well, my vote would be for css but
of course, thats just looking at creating a good browser based on nice
technology as opposed to the real world of websites. :(

love

Freya 



Re: [Dillo-dev]ecmascript, javascript and CSS

From: Geoff Lane <zzassgl@tw...> - 2002-07-09 07:59

> Is there an intention to implement these scripts?

Both JavaScript and Java would greatly increase the size of Dillo and
require numerous hooks into the existing code to obtain the functionallity
seen in other browsers. For example, in DHTML, you perform special effects
by tagging various events with a fragment of JavaScript so, for example,
when the mouse moves over a specific area of the screen a button lights up
or additional text appears. This has some deep implications about the way
that Dillo paints the window images.

> how about flash?

The flash "interpreter" is quite small and for simple images you could run
it as an external helper (I have actually thought about doing this) but you
wouldn't get any of the interactive or animation features. Again, proper
implementation of flash would require some extensive changes in the way
Dillo paints the window images.

If anyone has some spare time, one project that would usefully demonstrate
the kind of changes that would be needed in the Dillo image code is
implementing GIF and PNG animations. You would need some kind of regular
"update animation" event and some state information associated with
animation files. A thread[1] per animated image using some kind of "server
push" scheme may work; or more crudely, you could get the cache to manage it
all and set up some kind of list of "repaint screen area x,y,w,h" events.

BTW, for those of you waiting for the external helper code - sorry, Real
World events have delayed any progress recently. Work is continuing, just
very, very slowly.

[1] True threads acting as "data generators" is a very powerful design idea
for browsers and can hugely simplify the code in some cases.


-- 
/\ Geoff. Lane. /\ Manchester Computing /\ Manchester /\ M13 9PL /\ England /\

Syntax is another name for conscience money. 



Re: [Dillo-dev]ecmascript, javascript and CSS

From: Scott Barnes <reeve@du...> - 2002-07-08 19:45

Attachments: Message as HTML      Message as HTML      





[Dillo-dev]ecmascript, javascript and CSS

From: jonathan chetwynd <j.chetwynd@bt...> - 2002-07-08 16:26

Is there an intention to implement these scripts?
how about flash?

thanks

jonathan 



[Dillo-dev]A quick Thankyou

From: Freya <Freya@he...> - 2002-07-08 14:26

Madis sent me a binary of the latest CVS with some extra stuff patched.
Thankyou Madis!

I'm rather taken aback at how much faster and stable it is than my old
copy. So far it's never crashed at all and I'm able to use a lot more
websites etc now that I have cookies! I'm not sure where the speed
increase is coming from. Madis stripped it before sending it to me but
I've been led to believe that this shouldn't make a lot of difference.
Perhaps it does when you only have 8mb!

Anyway thankyou to you all for your help which makes my nasty old laptop
almost as usable as the Duron downstairs.

love

Freya 



Re: [Dillo-dev]valign="middle"

From: Freya <Freya@he...> - 2002-07-06 17:24

A bit of further investigation shows this is definitely a dillo thing.
Hard breaks also don't seem to render in dillo.

I booted a computer using konoppix, (a version of linux that runs
directly off a cd-rom with a read only live file system) and tried a
number of webbrowsers that it comes with. I also tried some pages in IE
under windows.

The results were kind of interesting. All of them seem to render valign
properly. My page seemed to render best in Mozzilla. Konquerer a close
second, but what shocked me was how ghastly everything looked in IE, and
I'm not just talking about my page but a lot of sites I looked at.
Interestingly, dillo seems to my eye to make webpages look the nicest.
The valign bug even makes one page I use look much better! ;)

I guess theres still a few problems in the dillo rendering engine. It
tends to render pages so well usually that I always just kind of assume
that part of dillo is perfect! :)

love

Freya


> I'm writing a webpage and I'm trying to centere the text vertically in
> a cell of a table. I'm using valign="middle" in a td element but it
> seems to have absolutely no effect. Is this supported in dillo or is
> there something very wierd about my webpage?
> 
> I don't presently have the latest dillo, so it might be related to
> that? I'm currently running 0.62, if anyone has a more recent compiled
> binary they could e-mail me that would be great as I don't presently
> have a working compiler. :(
> 
> love
> 
> Freya
> 
> 
> -------------------------------------------------------
> This s...net email is sponsored by:ThinkGeek
> Got root? We do.
> http://thinkgeek.com/sf
> _______________________________________________
> Dillo-dev mailing list
> Dillo-dev@li...
> https://lists.so....net/lists/listinfo/dillo-dev
> 



[Dillo-dev]valign="middle"

From: Freya <Freya@he...> - 2002-07-06 10:05

I'm writing a webpage and I'm trying to centere the text vertically in a
cell of a table. I'm using valign="middle" in a td element but it seems
to have absolutely no effect. Is this supported in dillo or is there
something very wierd about my webpage?

I don't presently have the latest dillo, so it might be related to that?
I'm currently running 0.62, if anyone has a more recent compiled binary
they could e-mail me that would be great as I don't presently have a
working compiler. :(

love

Freya 



Re: [Dillo-dev]Dillo scaling hack

From: Pigeon <pigeon@pi...> - 2002-07-03 13:34

> No ARM binaries to test, I'm afraid. Do you have a cross-compile setup
> working for dillo? I'd love to know how you set it up.

I have both cross compiling and native compiling environment setup. For cross compiling I basically use the one found on handhelds.org, the skiff one.


Pigeon. 



[Dillo-dev]ematic is down again :(

From: Jorge Arellano Cid <jcid@in...> - 2002-07-02 18:34

Hi there,

Once again, when at crossroads, the ematic server goes down (I'll
begin to think about a correlation there ;).

Well, just send me email to [jcid at inf.utfsm.cl]


Cheers
Jorge.-


--