[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: lynx-dev Forms based options (patch)
Re: lynx-dev Forms based options (patch)
Thu, 30 Jul 1998 17:09:49 +0400 (MSD)
> lynx-dev Forms based options (patch)
> * To: address@hidden (Lynx Development List)
> * Subject: lynx-dev Forms based options (patch)
> * From: address@hidden (Mike Castle)
> * Date: Wed, 29 Jul 1998 23:37:27 -0500 (CDT)
> * Reply-To: address@hidden
> * Sender: address@hidden
> Attached is my current patch against 2.8.1rel18 for having a forms based
> options implementation.
> It is still a work in progress, however it does appear to be mostly usable.
Oh, great for first try!
I compile in this patch, few problems were found:
1) when I change user mode from 'advanced' to 'novice' next time
I saw a form element '[ ]' ABOVE the statusline (yes, screen size changed).
2) while changing 'assume_charset' or 'display_charset' its value messed up:
it substituted with iso-latin-1 value ANYWAY for some unknown reason
(may be some variable not intitialized: MIMEnumber(iso-latin-1) equal to 0).
I will try to fix it tonight but maybe the author fix it before :-)
> There are a lot of FIXME's all over the code, so please read them.
> Especially those concerning the use of charsets and raw mode. I really
> have no clue what the hell is going on with those settings. I don't have a
> need for them, so I don't have experience with them and understanding how
> they are supposed to work.
> Most of the options screen is implemented. The following items are either
> not implemented or only partially implemented:
> * Any book mark stuff (I'm not sure how I'm going to handle this yet. It's
> probably last on the list).
> * DISPLAY: I forgot to do a putenv, and need special VMS handling stuff
> from regular options code too.
> * display character set: When going down the page, I just skipped this
> * User agent: Should be trivial.
> * Exec stuff: Another item I never use, so don't know the semantics yet.
> * Line editor keybindings: Is there support for that ANYWHERE?
> In the original system, when returning from the options submenu, a whole
> bunch of items are checked to see if they changed, and if so, force a
> reload of the page. I'm not sure how to handle that now (currently I think
> it seems random to actually load page before the options form).
> A lot of things that should be caught don't work, such as preventing book
> marking and gotos (though I can goto out of a print menu as well, so I
> assume I'm not setting up some option anyway, but I still doubt it will be
> stopped). I need to do some more analysis on this.
> I have stubs in place for handling some fairly trivial security (making
> sure the form that gets posted is the one that was generated internally).
> I'm not positive how to head off special url's not generated internally,
> so I'm put in some stuff that was inspired by old DOS telnet program I used
> to use (was it NCSA or CUTE?) using oneshot passwords.
> I write C code for a living, not HTML. So I'd like suggestions on how to
> better implement the actual form.
> It's definitely not ready for prime-time, but I think it's a good start.
> Also, this is definitely something you do not want to use without popups
> for select. On an 80x24 terminal, it's 2 pages with popups, 7 pages
> To actually use the silly thing: apply the patch to 2.8.1dev18, run
> configure, then in the makefile, add -DNEW_OPTIONS to SITE_DEFS.
> The only ifdefs are really in LYMainLoop.c. Once the patch gets applied,
> all the code is compiled, whether it's used or not. This is development
> after all. We're not looking for effeciency yet. :->
> Now for some bad news.
> I'm picking up keys for my new apartment tomorrow, and will start moving
> stuff over shortly, with an official move in on 10 Aug 1998. Soon after
> that, I'm moving my girlfriend from Cleveland here to Milwaukee. And not
> long after that, I'll be facing a deadline at work.
> So, I'm not sure when I'll have time to put a lot of thought into this.
> I should be able to finish up everything except the multibookmark stuff
> before I get moved. That item is what I forsee to be my biggest obstacle.
> Btw, this is also my first project I've put under CVS, so I'm still
> learning (I'm in the process of switching to it at work as well). So
> hopefully this patch will work as I intended for it to.
> PS: An indent configuration would be nice too. The indentation scheme we
> use at work is very different from lynx. Being able to have a default
> configuration for indent would be darned nice. ;->