Re: lynx-dev [PATCH 2.8.3.dev6] various fixes

[Vlad Harchev]

On Tue, 17 Aug 1999, Klaus Weide wrote:

> On Mon, 16 Aug 1999, Vlad Harchev wrote:
> > On Tue, 17 Aug 1999, Klaus Weide wrote:
> > > Other cases to watch out for: [snipped ]
> > > with '|' denoting the point at which split_line gets called (in other 
> > > words,
> > > the point at which the line has reached maximum length).  Also with 
> > > several
> > > spaces instead of one.  Variations that have en-space, em-space, nbsp, 
> > > soft
> > > hyphen characters instead of, or in addition to, the normal spaces.
> > 
> >  I won't try e[nm]-space,nbsp, soft hyphen - these chars are absent in psrc
> > mode (since the entity's name is displayed, rather it's not rendered).
> 
> If the _raw bytes_ for theses characters occur in the input stream,
> are they converted to character entities or NCRs?

 They are not touched, no psrc-specific code will deal with them. So (as it
seems to me) they will be inserted literally.
 BTW, I didn't find HT_EM_SPACE definition. What 'em-space' are you talking?

 As for raw nbsp and en-space, they are converted to ' ' in 
HText_appendCharacter, so seems they won't survive with either plain src view
or psrc view.
 
> > As for
> > multiplie trailing whitespaces, I posted the patch yesterday (that prevented
> > trimming trailing whites in split_line).
> >  I'll try other variations, but seems they'll be ok - the problem my last
> > patch fixed was calculations of the line_pos and hightex2offset for anchors.
> >  Frankly speaking, I don't understand why you suggested such "testsuit" - 
> > IMO it should contain only breaks inside hightexts, the things you've 
> > proposed
> > don't contain them inside hrefs.
> 
> Those cases are "other cases" that could have problems.  I was assuming
> that you had already taken care of the cases of spaces inside.

 I tested them. It's alright.
 
>    Klaus
> 

 Best regards,
  -Vlad