In message <b76f275e4d.pnyoung(a)pnyoung.argonet.co.uk>
Dr Peter Young <pnyoung(a)argonet.co.uk> wrote:
>On 19 Apr 2005 "David J. Ruck" <druck(a)druck.org.uk> wrote:
>>What you need to do is to get a unicode True Type font, and convert it
>>for use under RISC OS, then NetSurf can find it, and use it.
>>Either download a free font such as CyberBit from
>>then get !TTF2f from http://moose.mine.nu:6888/ttf2f2.zip and follow
>>the instructions for converting the TTF font carefully.
I got off to a bad start when !FTPc did not find communicator in pub.
Changing the URL prefix to http: and Oregano2 obliged.
>Done all that, and after a brief read of the help file for TTF2f, I've
>put off doing anything more till another day.
>One thing I couldn't quite follow is where to put the converted font.
The converted font is a RISC OS font and it goes inside any !Fonts
application, mine went into !Boot.Resources.!Fonts
The awkward bit on the Iyonix is updating the Messages1 file. A new font
not referred to in that file will not be available for use. I used
!ScanFonts from Select which works with Aemulor. Otherwise the new font
could be extracted, put into a dummy !Fonts and then installed via the
Castle Iyonix, RISC OS 5.08.
In message <560f3a5e4d.Sendu(a)sendu.me.uk>
Sendu Bala <sendu(a)sendu.me.uk> wrote:
>On 19 Apr, David Pitt wrote:
>> In message <6bf4355e4d.tim(a)south-frm.demon.co.uk>
>> Tim Powys-Lybbe <tim(a)powys.org> wrote:
>> >Finally I looked for the new font through NetSurf's Choices but
>> >nothing was to be seen. Rebooting NetSurf did nothing. Got nowhere
>> >with the various Font... commands. Probably the machine needs
>> >rebooting to sort that out. That'll be done later.
>> The Messages1 file needs updating. See my other post.
>I don't even /have/ a Messages1 file...
>> Cyberbit is present in NetSurf's choices here and working.
>... but Cyberbit also works well here.
It will if there is no Messages1 file, what does not work is a Messages1
file that does not contain the new font. The Messages1 file tells the
FontManager what is available, if it is not present then it has to work
it out for itself. Messages1 is quicker which may not matter much unless
there are a lot of fonts.
Castle Iyonix, RISC OS 5.08.
Here's a summary of important changes over the last few weeks. Please try
the new features and let us know of problems!
* new entries on the iconbar menu to open URLs and show the history and
* changes to HTML processing which improve handling of forms and titles,
* many improvements to table cellpadding and borders
* <object> fallback implemented
* text selection (drag on text to select it; ctrl-drag selection to save
* drag scrolling (drag an empty area of a page, or shift-drag)
* drag-saving images (ctrl-drag an image to save it)
* text input with alphabets other than Latin1
* and many bug fixes and minor improvements
Thanks for all your feedback - we read all posts to the list, even if we
don't always respond.
In message <Marcel-1.53-0417074442-0b0zokP(a)TopOffice.nom>
Gordon F McLaren <info(a)wardlaw.demon.co.uk> wrote:
>Further to my post in Vol 1 #609 (2):-
>Done some tests with the offending html file that is causing the crash
>and stripped down to a bare minimum:-
><!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
><!-- 16.04.2005 WS REDUCED TO BARE MIN FOR NETSURF TESTING -->
><!-- Next line causes the crash (I think) -->
><FRAMESET COLS="168, *">
><FRAMESET ROWS="76, *, 32">
I can replicate the "running out of memory" error with the above
fragment on my 128M Iyonix. NetSurf claims all the available memory as
its Dynamic area, eg. 79456K.
Castle Iyonix, RISC OS 5.08.
18 April 2045 build, Iyonix 5.08
I have just been looking up some obscure words in the OED on-line and,
while pages were being awaited, there have been narrow horizontal streaks
flickering for a fraction of a second about one quarter way up the
left-hand side of the Netsurf window - from that edge to the left edge of
the display. The flickering streak appears and disappears very rapidly -
but not regularly - some 1 to 3/4 seconds between flickers.
This flickering does not appear in conjunction with other phases of
document retrieval/rendering, nor in conjunction with any other
application. There are a humungous number of content_destroy and
contetn-remove_user calls in the stderr file - but nothing else apart from
cache interaction reports.
[PGP key available if desired]
RISC OS Adjust 4.39
Virtual RPC Adjust
NetSurf test Build (18 April 2005 12:00)
On two of my sites NetSurf appears to truncate some content. It may well
do it on other sites as well but as I don't know what the should say, I
can't be sure.
At the foot of www.clubmans.org.uk/ there is a link starting "This site
is designed..." only the word "site" is truncated to "sit". The link
continues to work and <F8> shows that the source has not been corrupted.
On www.btinternet.com/~brian.jordan9/quiz/index.htm/ some numbers >1000
in the "Match (F)" column of the second table have lost their last digit.
Again the downloaded sorce contins all of the data.
Can anyone else confirm this? I know the HTML isn't very elegant but it
works on Ovation 2 and Firefox here.
>From somewhere in North Hampshire. England
On Tue, 19 Apr 2005, Stefan Bellon wrote:
> This can even lead to inconsistencies between different compilation
> units if one includes one order and the other includes the other order.
> I think the proper fix is to include errno.h from within iconv.h rather
It did this before. I'm not entirely sure why it was changed, tbh
(although it appears to have been part of Nick's work making errno thread
> than defining EILSEQ but I wasn't courageous enough to do it this way
> because I cannot overlook the side effects this change may have.
I'm not entirely convinced that's even needed in UnixLib (as our errno.h
already defines EILSEQ). I only included in the original stubs as the SCL
headers supplied with Norcroft don't define it.
It Netsurf's own (bundled) hotlist, (access via F6) the link
for Netsurf's test builds is still
http://netsurf.strcprstskrzkrk.co.uk. Isn't that URL
O, varium et mutabile semper alpha.