sage-on-gentoo
sage-on-gentoo copied to clipboard
cschwan
missing map file "ckx.map"
First noticed this installing vanilla but it's also present on s-o-g. It may be of no consequence since it is not fatal.
From the build folder /var/tmp/portage/sci-mathematics/sage-doc-9999/work/build_doc/latex/ja/tutorial
# dvipdfmx -o tutorial-jp.pdf tutorial-jp.dvi
dvipdfmx:warning: Couldn't open font map file "ckx.map".
tutorial-jp.dvi -> tutorial-jp.pdf
[1
dvipdfmx:warning: Failed to load ToUnicode mapping: UTF8-UCS2
dvipdfmx:warning: Interpreting special command tounicode (pdf:) failed.
dvipdfmx:warning: >> at page="1" position="(72, 720)" (in PDF)
dvipdfmx:warning: >> xxx "pdf:tounicode UTF8-UCS2"
][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52][53][54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69][70][71][72][73][74][75][76][77][78][79][80][81][82][83][84][85][86][87][88][89][90][91][92][93][94][95][96][97][98][99][100][101][102][103][104][105][106][107][108][109][110][111][112][113][114][115][116][117][118][119][120][121][122][123][124][125][126][127][128][129]
702214 bytes written
ckx.map appears in /etc/texmf/dvipdfmx.d/dvipdfmx.cfg:226:f ckx.map but I cannot find the file anywhere. Some distros do include the file but I'm unable to determine its contents.
Blast from the past https://trac.sagemath.org/ticket/22252 we are commenting about this very file in there.
I had noticed that when "googling". Latest on Gentoo is texlive-2021. There is a 2022. Perhaps a correction is available with the newer texlive.
I am dubious, some of these map files have a long history of disappearing. Of course it may just need a tweak in the installation procedure for all I know. I am a bit concerned that I cannot find what looks like a source for it from simple googling, which tells me it may be generated.
Interestingly it is in the arch linux package https://archlinux.org/packages/extra/any/texlive-core/files/ so yes, may be going to 2022 will improve things. If not, digging for where it originates would be a job.
If I comment out f ckx.map in dvipdfmx.cfg there is
# dvipdfmx -o tutorial-jp.pdf tutorial-jp.dvi
tutorial-jp.dvi -> tutorial-jp.pdf
[1
dvipdfmx:warning: Failed to load ToUnicode mapping: UTF8-UCS2
dvipdfmx:warning: Interpreting special command tounicode (pdf:) failed.
dvipdfmx:warning: >> at page="1" position="(72, 720)" (in PDF)
dvipdfmx:warning: >> xxx "pdf:tounicode UTF8-UCS2"
No reference to the file. So maybe something deeper.
Since I don't do Japanese I have no idea if the rendered characters are correct.
While I know some japanese, I am not sure I could figure if everything is correct.
See https://oku.edu.mie-u.ac.jp/~okumura/texfaq/qa/52108.html for possible differences.
I am not sure what to do with it but it all looks legit to me - a mix of Kanji (chinese characters) and hiragana and curiously no katakana (the last two being purely japanese character sets similar to upper and lower case in design but not in usage).
I believe this is what is not being processed properly
sphinxoptionshyperref.sty: \AtBeginShipoutFirst{\special{pdf:tounicode UTF8-UCS2}}
We are now getting into sphinx - yuck. It looks like a lot of badly documented stuff.
From what I can find UCS2 is obsolete, although it is mentioned in the current dvipdfmx CTAN documentation. It may just be that sphinxoptionshyperref.sty should be modified to use something other than UTF8-UCS2. For example if I change UCS2 to UTF16 (a close cousin of UCS2) the warning goes away. This is probably nothing to worry about.
For reference, there are some fonts in tutorial-jp.pdf for which there is no explicit ToUnicode map in the PDF file
name type encoding emb sub uni object ID
------------------------------------ ----------------- ---------------- --- --- --- ---------
RQLDDN+TeXGyreHeros-Bold Type 1C Custom yes yes yes 4 0
MRIMFW+HaranoAjiGothic-Medium-Identity-H CID Type 0C Identity-H yes yes no 5 0
QGMOGC+TeXGyreHeros-BoldItalic Type 1C Custom yes yes yes 7 0
OXGXUO+TeXGyreTermes-Regular Type 1C Custom yes yes yes 15 0
NIUASB+TeXGyreTermes-Bold Type 1C Custom yes yes yes 16 0
ULCDQB+HaranoAjiMincho-Regular-Identity-H CID Type 0C Identity-H yes yes no 18 0
GGYCEQ+CMMI10 Type 1C Builtin yes yes yes 21 0
XWSANK+t1xtt Type 1C Builtin yes yes no 92 0
ENAOKT+t1xbtt Type 1C Builtin yes yes yes 101 0
JRNHNA+TeXGyreTermes-Regular Type 1C Custom yes yes yes 112 0
VUFNBZ+CMSY10 Type 1C Builtin yes yes yes 113 0
VEGDEZ+TeXGyreTermes-Italic Type 1C Custom yes yes yes 117 0
QNBHZI+tcxtt Type 1C Builtin yes yes no 134 0
NWZIPW+CMR10 Type 1C Builtin yes yes yes 160 0
ZRFZGW+CMR7 Type 1C Builtin yes yes yes 161 0
QKTGBS+CMEX10 Type 1C Builtin yes yes yes 165 0
QFARXB+CMMI7 Type 1C Builtin yes yes yes 166 0
TVOBSE+CMR5 Type 1C Builtin yes yes yes 167 0
LOZXWM+CMSY7 Type 1C Builtin yes yes no 168 0
NMPFRV+CMMI5 Type 1C Builtin yes yes no 173 0
NDRZSX+CMSY5 Type 1C Builtin yes yes yes 174 0
FORGAN+NimbusRomNo9L-Regu Type 1C Custom yes yes yes 217 0
NQBPKO+CMBX10 Type 1C Builtin yes yes yes 239 0
OSAFKY+CMMI12 Type 1C Builtin yes yes yes 298 0
From what I can find
UCS2is obsolete, although it is mentioned in the currentdvipdfmxCTAN documentation. It may just be thatsphinxoptionshyperref.styshould be modified to use something other thanUTF8-UCS2. For example if I changeUCS2toUTF16(a close cousin ofUCS2) the warning goes away. This is probably nothing to worry about.
Yes, that was the kind of things I found out as well. The style sheet belongs to sphinx, so that could be an issue to raise with them.