Ignore:
Timestamp:
Oct 10, 2012, 3:59:17 PM (7 years ago)
Author:
nmedfort
Message:

some progress

File:
1 edited

Legend:

Unmodified
Added
Removed
  • docs/Working/icXML/parfilter.tex

    r2289 r2429  
    44Several stages of standard XML processing include compressions
    55that filter out data from certain input byte positions
    6 in various ways.   These include the compression of multibyte
     6in various ways. These include the compression of multibyte
    77sequences in UTF-8 to UTF-16 transcoding, the compression
    88of CR-LF sequences to single LF characters in line break
     
    5050code units by 4, although each code unit is  now a 16-bit
    5151quantity.  This compression is carried out by the UTF-8 to UTF-16
    52 transcoder in Xerces.   Line 3 then shows the reduction of the
    53 two CR-LF sequences to simple LF values.  This is the line break
    54 normalization that takes place within the Reader???
     52transcoder in Xerces. Line 3 then shows the reduction of the
     53two CR-LF sequences to simple LF values. 
     54% This is the line break normalization that takes place within the Reader???
    5555Line 4 then shows the further compression by replacement of
    5656the character and general entity references by their corresponding
     
    6363deletion using parallel block-at-a-time operations, following
    6464for example, the parallel compress algorithm\cite{HackersDelight}.
    65 Figure \ref{Fig:compression2} shows three deletion masks corresponding
     65Figure \ref{fig:compression2} shows three deletion masks corresponding
    6666to the three compressions of Figure \ref{fig:compression1}.
    6767In the fourth line of this figure, the bitwise-or of these
Note: See TracChangeset for help on using the changeset viewer.