Re: Problem in Line breaking

From: Asmus Freytag (asmusf@ix.netcom.com)
Date: Tue Feb 26 2008 - 13:23:55 CST

  • Next message: Satoshi Nakagawa: "Re: Problem in Line breaking"

    I don't know what you two are fighting about here.

    The problem reported was:

    > In Japanese writing, [$B$3$?$($O!"(Banswer] should be breakable into
    > > lines like:
    > >
    > > $B$3$?$($O!"(B
    > > answer
    > >
    > > Because [$B!"(B](U+3001) and [$B!#(B](U+3002) in Japanese are used just
    > > like comma and period in English. We can break a line after
    > > comma or period in English.
    What looks like a space in front of the "a" of answer, is just empty
    space in the glyph of the "$B!"(B" character (ideographic comma), which can
    be seen when showing it between bars like this: |$B!"(B|

    Therefore, the rule in 5.0 CL X (AL | NU) has the effect of preventing a
    break which is a significant problem for Asian text that was overlooked
    when the rule change for 5.0.0 was accepted.

    I am told that this will be fixed in 5.1.

    A./

    PS: whether or not "before, after" allows a break after the comma and
    the following space (2 characters) is beside the point for Japanese
    text, which customarily does NOT have spaces.

    On 2/26/2008 1:06 AM, Satoshi Nakagawa wrote:
    > On 2008/02/25, at 6:22, Philippe Verdy wrote:
    >>>> LB30
    >>>> Do not break between letters, numbers, or ordinary symbols and opening
    >>>> or closing punctuation.
    >>>>
    >>>> CL $B!_(B (AL | NU)
    >>
    >> In fact the rules translate as CL % AL, i.e. it's an indirect break:
    >> CL $B!_(B AL
    >> but
    >> CL $B!_(B SP* $B!`(B AL
    >
    > I would not agree.
    >
    > See 7.3,
    >
    >> % denotes an indirect break opportunity: B % A is equivalent to B $B!_(B A
    >> and B SP+ $B!`(B A;
    >> in other words, do not break before A, unless one or more spaces
    >> follow B.
    >
    > We can only say CL % AL is equivalent to CL $B!_(B AL and CL $B!_(B SP* $B!`(B AL
    > here. So CL $B!_(B AL is not equivalent to CL % AL. Because the left hand
    > side doesn't have "and CL $B!_(B SP* $B!`(B AL" part.
    >
    > I would say in other words, if A $B!_(B B allowed a break, the A $B!_(B B rule
    > would be no information.
    >
    > I think CL $B!_(B AL doesn't allow a break between CL and AL, doesn't it?
    >
    > --
    > Satoshi Nakagawa
    >
    >
    >
    >



    This archive was generated by hypermail 2.1.5 : Tue Feb 26 2008 - 13:25:56 CST