Should we really crack IBrowse 2.4???

[wawrzon]

Indeed, a port of Dillo would be much more useful.

indeed, indeed.
afair radoslaw, the netbsd coder, has compiled it to run on his amiga hardware under netbsd.
porting it to over to amiga or aros demands one essential dependency which is a simple gui library afair. i looked at it some time ago.

[wawrzon]

@novacoder
i dont know why chris cannot do 68k adjustments himself, as best familiar with the code instead seeking to encourage others since years, which as we see leads to nothing. lets assume its just additional work and maintenance involved he doesnt want to be doomed to. alright.

im not a coder, but ive looked at the source once and think that it is not trivial to make it run on 68k, since there is a lot of os4 functional dependency afair. however if youd succeed id push your contribution upstreem. perhaps it would hold some time without breaking and without the need of big adjustments.

the other possibility is to check into aros owb on aros68k. it has compiled and started already to certain extent, but jason has gave it up due to more essential tasks. owb is far more functional than netsurf as it seems.

[wawrzon]

Anyone know if the 68k SDL port's sources have been published?  I might be able to use them for reference.

http://aminet.net/comm/www/netsurf-m68k-sources.lha

a bit old, but updates were not essential i think and the new versions seem to contain some bug that crashes them on aros68k, maybe not worth to reproduce.

[wawrzon]

A huge difference actually!

not when it comes to aros68k, then all is native again, even more native than on mos or od4 so to say.

@chris, nova: hope something comes out of that, even though i have my doubts if this will provide better speed that the current sdl port. tell me if i may be of any assistance.

[wawrzon]

voyager has a nice clean, simple and appealing gui, that on the plus side. but its so buggy underneath, i dont even know if it pays to open its sources. i have recently played a little with aweb, both under aros68k and os3.1 as this is the simplest working browser i know working on  both, and must correct my previous opinion about it. it doesnt look very funky, but considering the circumstances it is quite fast, stable and accurate even under plain amiga chipset. if it was possible to plug in css into it i could become much better usable nowadays.

edit: the problem is, its open source but i dont know where to look for them. likely they are kept hostage by big gun who apparently was the last who worked with it, and rescued it from lost repository or something the like.

[wawrzon]

okay matt, i will see if anything s left stuck from way back. i have devcpp setup here, ll try to revive it and come back to you.

desiv: yesss.. you and the mos boys are stuck in semantics. it doesnt pay to discuss it, whether it is emulation or not. does it?

[wawrzon]

Paid Voyager releases are actually very stable. It is those free evaluation versions which were from yesteryears having annoying crash bugs. Not very good strategy to get customers, though.

ohh. i see. i know only the eval versions. i stand corrected.

The biggest downside in Voyager is that it is using lot of private MUI calls. Getting it running on AROS (Zune) require lot of in depth knowledge about MUI. On OS3 it of course is not problem.

talking about aros, this is actually an advantage if voyager would get open sourced. zune has to be fixed also up to the private mui classes. thats what currently prevents it from running ibrowse. so if a tool using private non documented mui features has been open sources, fixing zune might become much easier i suppose. you are actually meking me hot about voyager.

itix, do you have any link to the developers? olaf schoenweiss has proven to be very effective in gathering contributions for his aros68k distibution, perhaps he can get involved. i think it would be an advantage not only for aros but at least for 68k amiga users to have voyager open as pfs3 was.

[wawrzon]

very anti-OS3.5 and 3.9. I hate to think his thoughts on OS4!

encouraging to hear. he might actually have had some point..

[wawrzon]

@chris: or perhaps he just wanted to keep his support efforts down, not much else than telling people bugs have to be replicated with clean os4 install. but i was joking.

[wawrzon]

@matthey:
about aweb: it will not build on crosscompiler due to those 68k tools that do not execute there. but it builds under my cubicide (winuae) im just completing headers..

check here:

http://www.favrin.net/txt/articoli/vari/awebapl_compile_faq.txt

titlebar headers seems to be a problem yet, classact is supposed to be a part of 3.9 ndk, but i cant find those headers there. there is os depot archive that contains it, but i suppose it isnt appriopriate for 68, so seems broadblues has broken 68k compatibility introducing os4 reaction classes to aweb. how convinient...

and also miamissl is missing.

MiamiSSL SDK can't be found on the web. It has to be
requested from the author. Since Holger Kruse has been
unreachable for months the only two options seems to be to
drop the support for MiamiSSL (see section 12) or to build a
dummy include file based on the source code's references of
the library. Juergen Lachmann followed this way and he
managed to compile AWeb and browse some SSL sites with
MiamiSSL.

i could ask jürgen, have been in contact with him because of scalos.

edit: damn, id love to have a more versatile shell on cubic than tcsh, also it is damn slow to build there.

[wawrzon]

@chris:
why, then we could have a good chance. this is not about os4, as os4 doesnt need voyager anyway. this is all about genuine amiga, even if it means up to 3.1 which is fine with me.

[wawrzon]

on all respect, despite what chris suggests its not just a walk by to port his netsurf back to 68k. if it was so easy he would best do that himself in an instant, gifted as he is and already familiar with the code and implementation details.

i still suspect the easiest, fastest and most effective would be to have odyssey building for aros68k. here is all in place for it imho, including the dependencies. just some dediacted coder with a knowledge of complex build processes involving cmake is missing. aros devs dont have time for it, there is already enough workload, and 68k isnt the main priority of course, except the users will somehow express and push that.

i have aros owb in ocassional use here, and while it isnt fast, its almost usable, faster than expected. and expect odyssey to be faster as aros owb is only a basic port, not even sure if it doesnt involve statically linked sdl.

once owb runs on aros68k the users may switch to it given aros is fast and dependable enough by the time or an attempt on backport might be undertaken. jason already has prepared some basic means to run aros apps on aros afair, just none has followed on that and its been covered in dust.

[wawrzon]

@chris:
here we have it as it ever was. doesnt look like anybody actually able to do the job is going to step up.

aros odyssey version has an advantage that the existing code shouldnt need to be modified to run on aros and an amiga with gfx card, it needs only to be made to build.
of course it will not give us planar potters magically, but i fear the same is true for netsurf.

[wawrzon]

i think porting software you can forget about anything else than gcc. if you roll off your own project you can tailor it for something else.

[wawrzon]

i dont even know if amidevcpp sports up to date c++ compiler. afair the last available version was around 3.4.x. on the other hand there is a c backend 4.5.0 by bernd, which makes it fine for a contemporary project written entirely in c as netsurf is, and likely therefore artur was able to compile it for amiga/sdl without major changes.

on the other hand whether we like it or not, a lot of stuff we desire is nowadays written in c++. i cant argue if c++ produces slow code, i know some claim that. the main question is though, do we want to try to port these apps even if just to see if they are slow or not, or do we resign on them entirely. from my experience with aros owb on 68k real hardware i must say it is not as slow as you might expect it.

as said, odyssey/owb/webkit is c++, so any kind of owb needs c++ toolchain. aros has it including 68k (4.6.2 afair). 68k amiga lacks it. perhaps someone can work out and provide appropriate devpacks for amidevcpp. id like to have amiga/aros68k 4.6.x. that might simplify the situation. but dont see anyone up to the task.