Updated converters/dvi2tty to 6.0.0nb1
Updated fonts/afm2pl to 0.7.1nb3
Updated fonts/gsftopkk to 1.19.2nb8
Updated fonts/lcdftypetools to 2.106
Updated fonts/ps2pkm to 1.8nb1
Updated fonts/ttf2pk to 2.0nb3
Updated fonts/ttfdump to 0.5.5nb4
Updated print/cjkutils to 4.8.4nb2
Updated print/dvidvi to 1.1nb1
Updated print/dviljk to 2.6.5nb8
Updated print/dvipdfmx to 20170318
Updated print/dvipsk to 5.997
Updated print/kpathsea to 6.2.3
Updated print/lacheck to 1.28nb2
Updated print/luatex to 1.0.4
Updated print/ptexenc to 1.3.5
Updated print/seetexk to 1.1nb3
Updated print/tex4ht to 0.0.20170523
Updated print/vlna to 1.5nb3
Updated print/web2c to 2017
Updated print/xdvik to 22.87.03nb1
Updated print/xetex to 0.99997
Updated print/xpdfopen to 0.86nb2
Updated textproc/makeindexk to 2.15nb9
Updated textproc/mendexk to 2.6fnb7
pdfTEX: Support JPEG Exif as well as JFIF; do not even emit a warning if
\pdfinclusionerrorlevel is negative; sync with xpdf 3.04.
LuaTEX: New library newtokenlib for scanning tokens; bug fixes in the normal
random number generator and other places.
XeTEX: Image handling fixes; xdvipdfmx binary looked for first as a sibling
to xetex; internal XDV opcodes changed.
MetaPost: New numbersystem binary; new Japanese-enabled upmpost and
updvitomp programs, analogous to up*tex.
Infrastructure: The fmtutil script has been reimplemented to read
fmtutil.cnf on a per-tree basis, analogous to updmap. Web2C mktex* scripts
(including mktexlsr, mktextfm, mktexpk) now prefer programs in their own
directory, instead of always using the existing PATH.
* fread.c (copy_multibyte_char): Avoid ambiguous operation.
* fread.c: Add copy_multibyte_char() to
copy multibyte character correctly.
* fread.c: Wide char is not always even bytes in getestr().
* convert.c: Add (unsigned char*) cast arg1 of UTF8StoUCS().
* fwrite.c: Declare fprint_euc_char() as static.
* main.c: Add TeX Live version.
* {main,convert,fread,fwrite,styfile}.c, kana.h,
mendex.doc, mendex.1: Add a new option -I to switch
internal buffers for key words to EUC-JP or UTF-8.
* mendex.1: Fix usage.
* convert.c: Allow longer words in dictionary.
* fwrite.c: Make a buffer size larger.
* fwrite.c: Replace sprintf with snprintf.
* fread.c: Allow lines with only a line ending in mfgets(),
because dictionary files with DOS line endings can't be read.
This is an upper-compatible (but slow) substitute for the makeindex
program which forms part of standard TeX distribution. Mendex features
better Japanese (EUC) support and more.