teTex on the LN server

Started by Toruk Makto, September 30, 2013, 05:51:59 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Toruk Makto

#20
Additional: I also copied the .tex file to http://files.learnnavi.org/NaviDictionary_ru.tex so you can see the actual input file to see the offending line numbers.

EDIT:  Additionally additional... The .pdf was actually created by pdflatex. I have copied it to http://files.learnnavi.org/NaviDictionary_ru.pdf. You may want to take a look to see if it is viable, since I can't read Cyrillic.

Lì'fyari leNa'vi 'Rrtamì, vay set 'almong a fra'u zera'u ta ngrrpongu
Na'vi Dictionary: http://files.learnnavi.org/dicts/NaviDictionary.pdf

`Eylan Ayfalulukanä

I took a good look at the text in line 2287, which is the line that seems to be making EE hang via pdftex. There is something that the compiler doesn't like about one of the two quotes on this line. I don't know much about cyrillic, either, but on the surface, the quotes look OK. There is an unmatched quote in line 2328, that also contains a line of untranslated text. I also do not see what is causing the cluster of error on line 2236.

I've noticed in troubleshooting problems with Dothraki that the actual problem can be several lines either side of a flagged line. But as far as I can tell (and it is easy to overlook things in code like this), everything otherwise looks fine several lines in either direction of these problem spots.

Yawey ngahu!
pamrel si ro [email protected]

baritone

Quote from: `Eylan Ayfalulukanä on October 08, 2013, 02:48:51 AM
I took a good look at the text in line 2287, which is the line that seems to be making EE hang via pdftex. There is something that the compiler doesn't like about one of the two quotes on this line. I don't know much about cyrillic, either, but on the surface, the quotes look OK. There is an unmatched quote in line 2328, that also contains a line of untranslated text. I also do not see what is causing the cluster of error on line 2236.

I've noticed in troubleshooting problems with Dothraki that the actual problem can be several lines either side of a flagged line. But as far as I can tell (and it is easy to overlook things in code like this), everything otherwise looks fine several lines in either direction of these problem spots.
It is an LaTeX "feature".
Issue is in line 2287, if delete this line or replace {\it blah-blah-blah} with \textit{blah-blah-blah}, compilation goin on correctly.

It is some issues with fonts now, all bold and slanted cyrillic fonts in compiled dictionary replaced with his regular variants. Section list with russian names is empty now.

Toruk Makto

#23
I will manually install all LH T2A fonts today.  Perhaps that will provide the tfm needed for italics and all.



EDIT: I have compiled and installed the entire lh font set. t2a, t2b, t2c and t2d.

Did this fix the italic/bold problem?

Lì'fyari leNa'vi 'Rrtamì, vay set 'almong a fra'u zera'u ta ngrrpongu
Na'vi Dictionary: http://files.learnnavi.org/dicts/NaviDictionary.pdf

`Eylan Ayfalulukanä

Will installing these new fonts have any effect on non-cryllic languages?

Ma Baritone, what is the difference between {\it blah-blah-blah} and \textit{blah-blah-blah} ? I see both of these used seemingly interchangeably in the Dothraki dictionary, as well.

Yawey ngahu!
pamrel si ro [email protected]

Toruk Makto

Quote from: `Eylan Ayfalulukanä on October 08, 2013, 04:46:54 PM
Will installing these new fonts have any effect on non-cryllic languages?

Ma Baritone, what is the difference between {\it blah-blah-blah} and \textit{blah-blah-blah} ? I see both of these used seemingly interchangeably in the Dothraki dictionary, as well.

It shouldn't affect anything else.

Lì'fyari leNa'vi 'Rrtamì, vay set 'almong a fra'u zera'u ta ngrrpongu
Na'vi Dictionary: http://files.learnnavi.org/dicts/NaviDictionary.pdf

baritone

Quote from: Toruk Makto on October 08, 2013, 07:27:08 AM
I will manually install all LH T2A fonts today.  Perhaps that will provide the tfm needed for italics and all.

EDIT: I have compiled and installed the entire lh font set. t2a, t2b, t2c and t2d.

Did this fix the italic/bold problem?
Sorry, this is not fix the problem. And in my home machine with texlive there is the some thing with bold and italic. It is necessary to try out further at my home LaTeX, test the LaTeX packages/parameters.

Quote from: `Eylan Ayfalulukanä on October 08, 2013, 04:46:54 PM
Ma Baritone, what is the difference between {\it blah-blah-blah} and \textit{blah-blah-blah} ? I see both of these used seemingly interchangeably in the Dothraki dictionary, as well.
Yes, it does. But sometimes LaTeX macros and contexts can do unexpected things. It is out of my ability explain this on my english...  :-\

Toruk Makto

OK,  let me know what you find out.  I can add whatever is needed on the LN server.

Lì'fyari leNa'vi 'Rrtamì, vay set 'almong a fra'u zera'u ta ngrrpongu
Na'vi Dictionary: http://files.learnnavi.org/dicts/NaviDictionary.pdf

baritone

Quote from: Toruk Makto on October 09, 2013, 12:33:40 AM
OK,  let me know what you find out.  I can add whatever is needed on the LN server.
All emphasizing are OK now.
\usepackage[utf8]{inputenc}
\usepackage[T2A]{fontenc}
\hypersetup{unicode=true}  % For hyperlinks on Russian
\renewcommand{\rmdefault}{cmr}  % To cure font issues
\SetSymbolFont{operators}{normal} {T2A}{cmr} {m}{n}    % To cure font issues
\usepackage[russian]{babel}

But I can not understand why the hyperref package has not created a table of contents as so as it have done in my home machine?
Maybe it will be done, if you erase all *_ru.out and *_ru.aux files,
rm *_ru.out  *_ru.aux  and run pdflatex two or three times?



Toruk Makto

Table of contents?  Is it supposed to?  The English version doesn't.

Lì'fyari leNa'vi 'Rrtamì, vay set 'almong a fra'u zera'u ta ngrrpongu
Na'vi Dictionary: http://files.learnnavi.org/dicts/NaviDictionary.pdf

baritone

Quote from: Toruk Makto on October 21, 2013, 05:29:14 PM
Table of contents?  Is it supposed to?  The English version doesn't.
Sorry, may be I have use wrong word.
I have mean the list of links to appendicies, that appears in left when someone view the pdf file.
This list is generated by the hyperref package.

Toruk Makto

Oh,  right.  I'm afraid I don't have foggiest how that works...

Lì'fyari leNa'vi 'Rrtamì, vay set 'almong a fra'u zera'u ta ngrrpongu
Na'vi Dictionary: http://files.learnnavi.org/dicts/NaviDictionary.pdf

baritone

Quote from: Toruk Makto on October 22, 2013, 03:08:15 PM
Oh,  right.  I'm afraid I don't have foggiest how that works...
I know that  *_ru.out and *_ru.aux files are generated at *_ru.tex files compilation, and it used for making the list of appendicies in the dictionary. Sometimes that list are generated with errors and in this circumstances the TeX thugs recomend one to delete this files and run latex two or three times after the files deletion.
Now the hyperref package leaves the list of appendicies on Russian dictionary empty despite of the \hypersetup{unicode=true} instruction.

There is another problem: the ä and ì letters in Russian dictionary are combined from two ¨a and symbols respectivelly, which makes impossible to find some words in the dictionary with this letters. This can be solved with xelatex, which is part of new TexLive package, because the xelatex can use unicode fonts (and generates pdf by default). I can compile the dictionary with xelatex at home computer but I don`t know what to do with pdflatex on LN server.

`Eylan Ayfalulukanä

I'd really like to see xetex get installed as well. On the Dothraki/Valyrian side of the house, I have problems with vowels (and one consonant) that have macrons. I have a workaround for them but it makes it such that you cannot search on the database for words with letters containing macrons.

I think, though, this is a Tuiq thing.

Yawey ngahu!
pamrel si ro [email protected]

Tuiq

Quote from: `Eylan Ayfalulukanä on October 29, 2013, 02:40:25 PM
I think, though, this is a Tuiq thing.


☑ Web Interface
☑ Database
☑ Translator membership administration
☑ LaTeX generation
☑ calling pdflatex on the generated LaTeX
☑ SQL generation
☑ other output generation
...
☐ god beware anything that goes beyond `pdflatex Dictionary.tex` and has to do with ps/pdf generation
Eana Eltu: PDF/TSV/jMemorize

Toruk Makto

 My biggest concern is that XeTex installs to the same destinations as TeTex, so I will have to remove TeTex first. If XeTex either doesn't work as expected, or the look of the dictionary is altered, I will have to revert back. This doesn't sound like a big deal, but with the cockamamie way TeTex has scripts and fonts all over the damned place, I don't want to make a hash of our Tex capabilities by yanking and replacing stuff and leaving crumbs on the floor. Also, the current vernacular in the .tex file may not work. I simply don't know enough about it to predict any outcomes.

However, if anyone already has XeTex and can run a sample pdf generation of the current NaviDictionary.tex, we can see if it works right before I replace the Tex system on the LN server.  Anybody up for it?

-M.

Lì'fyari leNa'vi 'Rrtamì, vay set 'almong a fra'u zera'u ta ngrrpongu
Na'vi Dictionary: http://files.learnnavi.org/dicts/NaviDictionary.pdf

Tuiq

Eana Eltu: PDF/TSV/jMemorize

Toruk Makto


Lì'fyari leNa'vi 'Rrtamì, vay set 'almong a fra'u zera'u ta ngrrpongu
Na'vi Dictionary: http://files.learnnavi.org/dicts/NaviDictionary.pdf

`Eylan Ayfalulukanä

I'm willing to try. First, I have to figure out how it works! ;)

Yawey ngahu!
pamrel si ro [email protected]

Tuiq

If we are scared about changing stuff on a production server (and rightfully so), then best solution to find out whether it works or not would be to try it in a safe environment - this could be a VM (which are, with limited capabilities, even free - and for a server, they should suffice I hope) or just another computer/server to try it out.

Worrying about switching to xetex or doing the upgrade and then having to painfully revert back to something that did work are both not exactly solutions and don't bring us forward... if anything, the latter would be a major setback.
Eana Eltu: PDF/TSV/jMemorize