Now I can accept the answer to my yesterday's question :)

Hello! I am having trouble with a .dvi copy of a thesis I want to study. Can anyone help me by converting it to .pdf and mailing me the result?

@Arrow wht sort of trouble? if dvips doesn't work then most likely it refers to fonts or images that are not available, in which case anyone will have the same problem

@DavidCarlisle I get "METAFONT modes cannot be initialized beaus 'modes.mf' is missing." I have no idea how to solve this. I don't know latex, I'm a lyx simpleton :)

@Arrow oh your tex installation is incomplete, how did you get there? not that you want to be using metafont anyway, what does dvipdfmx do on the same file?

@DavidCarlisle I just installed the LyX bundle. I don't know what 'dvipdfmx' is, sorry.

@Arrow it's like dvips but makes pdf rather than postscript

@DavidCarlisle how do I use/install it?

@Arrow if you have a tex installation I'd expect you have it and run it the way you just ran dvips

@DavidCarlisle I just double clicked the file, that's all.

@Arrow ah :-)

if you send it to me I'll do it this once:-) (google for my name and you'll see my gmail address)

@DavidCarlisle thank you very much!

@DavidCarlisle ooh

@PauloCereda you could send your thesis, but I think my spam filter scrubs empty attachments

@DavidCarlisle you are mean

@DavidCarlisle oh

@DavidCarlisle done.

@Arrow speaking of spam filters that's where yours went:-) Not to worry I can get it out:-)

@DavidCarlisle :)

@DavidCarlisle ooh

@Arrow I think it replied, although it grumbled a bit about replying from a spam filtered mail, if you don't get it, I'll resend under a different subject line

@DavidCarlisle thank you very much! Have a nice day.

Funny timing. I got it.

I beat Mario and Luigi: Dream Team for Nintendo 3DS, yay!

If I'd be able to hack my own calculator, I'd surely be able to go well on an exam. :)

repology.org/metapackage/xindy/versions lists xindy version of 2.5.1, we have 2.4 on gentoo I wanted to update our gentoo package, but upstream has only 2.3 sourceforge.net/projects/xindy/files which magician can help me? ;-)

@JonasStein Install the vanilla TeX Live

@egreg this question is rather from the view as gentoo developer than as user. Gentoo distributes a texlive package, but also xindy as single install package. Where does the development of xindy take place?

@JonasStein That's a good question: nowhere, it seems.

@egreg Do you have an idea, where TeX Live vanilla takes it from?

the development of tex packages is a nightmare for distributions. 95% of the tex stuff is very badly maintained. I saw that our tex packages on Gentoo are in a really bad condition and try to help out there a bit.

@JonasStein I think that Karl Berry knows.

@egreg is he here?

Is it CarLaTeX?

@JonasStein Karl has an account, but it's easier to reach him by email.

@JonasStein A lot of the stuff nowadays is hosted in TL, in the sense MiKTeX picks up from the TL SVN for several tools

@JonasStein I'm a beginner LOL!

@CarLaTeX But are you Karl? :)

@PauloCereda No, I'm Carla LOL

;-)

@CarLaTeX Levenshtein of 2, you are still a suspect. :)

@PauloCereda LOL

@JonasStein As @PauloCereda says

I will try it there, thank you.

@JonasStein There is a mailing list, but when I needed help to sort out xindy for miktex I wrote Joachim Schrod and Akira Kakuto. If you want the newest version the best is to go to the texlive sources.

@UlrikeFischer Sounds right

@UlrikeFischer In other news, I have all of the font stuff sorted for siunitx v3 I hope. It's all very different from v2 ...

@UlrikeFischer ok. It sound a bit like we have to rethink all out packages which are outside texlive on gentoo... Do I understand it correct, that within the last years most projects develop actively on the SVN of texlive? I am scolling through the svn repo... Is there a gitlab/github mirror?

@JonasStein There is an unofficial mirror (Norbert Preining runs it): TL is too big for GitHub

We still have about hundred packages which think that the upstream website hosts the latest version. (Like I expected in the case of xindy) I fear now I opened a big box of work and we have to fix many many packages...

@JosephWright do you have a link?

@JonasStein tug.org/pipermail/tldistro/2017q1/000282.html

git.preining.info/texlive.git :-( none found

git.preining.info/texlive ah it moved

@JosephWright I got a reply about Xindy. Latest development is on TeXLive svn and backported from time to time to github.com/jschrod/xindy.ctan

@JonasStein Sounds about right

yes it was the reply from jschrod, I just wanted to give a feedback on the outcome.

I have a PDF created by LuaLaTeX that none of the text is copiable. If I try to export it to a Word or plain text document with Acrobat, I get gibberish. If I select some text in Skim and paste it into a plain text document I get garbage. How did this happen?

Top image is the PDF; bottom is export from Acrobat to Word.

@AlanMunn i suppose this again proves that word is evil?

that was a joke

@AGoldMan I think in this case it proves that LuaTeX is evil.

@AlanMunn maybe it's in cahoots with word

I wouldn't know, i use xetex

i feel like it's cheating if there's a more traditional programming language involved

@AGoldMan I use both, but this particular document has some randomization code in lua.

@AlanMunn It's not deliberate is it? Some 'actual text' anti-copy stuff?

@JosephWright Not on my part, no.

@AlanMunn care to share the code?

@AlanMunn have you tried what pdftotext outputs? I suppose it's the same, but just in case...

@yo' Yeah, it's pretty much the same.

@AlanMunn ok

hey @CarLaTeX, what do you think of tex.meta.stackexchange.com/questions/7472/…

@AlanMunn and can you see which fonts are used? Is it a full set or an embedded subset? Because for embedded subsets I can imagine luatex remaps the used glyphs while embedding.

@yo' Hmm. If I reduce the document to just the title and first paragraph it's ok. So it seems to be something to do with multicol as well. Output of pdffonts is:

name                                 type              emb sub uni object ID
------------------------------------ ----------------- --- --- --- ---------
YVKBCI+MinionPro-It                  CID Type 0C       yes yes no       9  0
QDEIWR+CMMI8                         Type 1            yes yes no      10  0
VHCBFA+MinionPro-Bold                CID Type 0C       yes yes no       7  0
FLCYMN+MinionPro-Regular             CID Type 0C       yes yes no       8  0

@AlanMunn so embedded, that could be it. But I wonder how multicol could influence it; maybe it introduces logical subpages and this is it?

But I dunno, these are basically half-educated guesses

@yo' I'm gradually making the document larger and larger to see when the problem arises. The first pair of examples doesn't show the problem... weird.

@AlanMunn i suppose it's luatex being evil again. It's on to you

@AGoldMan @yo' I've just figured it out. It's not LuaTeX that's evil, it's Mac Preview.

@AlanMunn It's a simple character shift.

@AlanMunn that's why I asked about pdftotext

The actual file was created by LuaTeX, but then I deleted the second page in Preview and saved it.

@StefanKottwitz But how did it happen?

@AlanMunn just goes to prove that apple is evil

@AGoldMan I'm not going to rise to that bait. :)

@AlanMunn I just see it's neither nonsense nor gibberish but "crypto" encoding ;-) So simple, that it could be a bug, or a weak virus. I wish every ransomware would be so easy to fix.

@AlanMunn it's never really works in here. not like mentioning vim when @DavidCarlisle is around

@AGoldMan :)

@AGoldMan vim usage never goes unpunished

@DavidCarlisle what's the emacs command for punishing vim users? c-x c-v c-i c-m?

M-x doctor

@StefanKottwitz I'm guessing bug then.

@DavidCarlisle Emacs will fix my weird file? Now that would be a neat trick.

hey @DavidCarlisle is there a way to check if a command is being expanded by \the?

meaning that the expansion changes if it's being expanded or assigned to

@AGoldMan I think you just need to know where it's appropriate to use \the.

@HaraldHanche-Olsen that much I know. I'm toying with a front end for working with \toks, and I want it to work as transparently as possible

so it'll process a += as an instruction to append, but if prefixed by \the it'll return the value

for example

@AGoldMan not sure what you mean by "expansion changes" things that take \the are not expandable (that's why you need \the to extract and expand the value

@DavidCarlisle apologies. I meant detect if it's trying to be extracted

@DavidCarlisle I mean make a macro that expands to a token register for value retrieval when prefixed with \the, and otherwise starts a parsing routine to allow for different syntax in assignments.

@AGoldMan not really, or at least only by inspecting the tokens before attempting the assignment, there is no way to insert a test as in the case you need a toks register your test would have to expand to nothing

@AGoldMan why would you need that? why cant your assignment syntax just do the right thing without the use using \the

@DavidCarlisle I want it to be transparent? maybe that's asking too much

@AGoldMan that's what I mean, you don't need to wait until \mytoks tokem is encountered to ask what is the context as if you are defining some syntax then you know the context in advance

@DavidCarlisle i just like the idea of being able to type \everypar+={something} to do the assignment. It makes it feel more homey

@AGoldMan well you could redefine \everypar to look for += and if so do \realeverypar\expandafter{\the\realeverypar something} but that seems very fragile \addtotoks\everypar{something} is so much more tex-like

@DavidCarlisle LuaTeX \toksapp ;)

@DavidCarlisle for the most part I use that, but there's something I like about the other type of syntax. you could even implement things like \everypar/*search*/*replace*/ and it'd be like using vim

@JosephWright or even 2e \addto@hook :-)

no luck with italics

@AGoldMan what have the Romans ever done for us.

@DavidCarlisle Hans has added lots of stuff

@DavidCarlisle Romanes eunt domus

@DavidCarlisle i'm fairly certain they destroyed the temple, so gotta hold that against them

@JosephWright he added Lua for a start:-)

@DavidCarlisle :)

@JosephWright back in the UK?

@AGoldMan This reminds of some really misguided notion I encountered ages ago: To make Pascal programmers feel at home in C, just do #define begin {, #define end }, and so on. You're really better off not trying to implement your favourite language in a different language.

@DavidCarlisle the roads!

@DavidCarlisle Nope, in a facility hostel in Switzerland

@HaraldHanche-Olsen i think TeX is intentionally freeform enough for me to do this but I guess you're right

@DavidCarlisle google.ch/maps/place/Paul+Scherrer+Institute/…

@HaraldHanche-Olsen Don't tell Bruno

@HaraldHanche-Olsen although there is a question somewhere on the site about writing gcc in tex, which is hilarious

here it is

@AGoldMan Upvoted! :):):)

@AGoldMan You could add an answer here: tex.meta.stackexchange.com/questions/430/text-building-blocks

@CarLaTeX it is the greatest article eva

just waiting for @DavidCarlisle to translate that to portugese

@AGoldMan Tudo o que você precisa saber sobre patos: eles são bons para comer.

@AGoldMan LOL

@PauloCereda ^^^

@PauloCereda could verify the translation

@DavidCarlisle a great truth indeed

i made the horrible mistake of booting into windows today

it changed my clock back an hour

@AlanMunn -- nah, "hot sauce" is tabasco. chili sauce is only mildly hot, and chunkier than ketchup. (mixed with mayo, makes decent thousand lisland dressing.) tabasco is what you put on shrimp and grits if you want to heat it up. (ketjap doesn't work so well with grits either, although it could be interesting with shrimp.)

@AlanMunn -- i'd be interested in your opinion, as a linguist, about the supposition that one can tell what part of the south someone comes from by how many syllables they put in "grits". (the most i've ever heard is three.)

@barbarabeeton I wasn't equating the two but using 'chili sauce' as a translation of hot sauce, i.e., hot sauce is a sauce made from chilis (and vineagar), since 'hot sauce' isn't really a very transparent name to non-Americans.

@barbarabeeton Well you've got to make up for the missing syllables in words like 'oil'. :)

@barbarabeeton I'm not much of a phonologist, and I don't know much about Southern American vowels. I do remember there being plenty of triphthongs in Baltimore and the Eastern Shore of Maryland. But I know very little about the deep south.

@barbarabeeton And for many, there are local brands of hot sauce and people don't like tabasco as a substitute.

@AlanMunn -- hmmm. as a bal'mer native, i can't really remember any tripthongs; maybe since i've moved north? but on the shore, the accents have always been rather more distinct.

@barbarabeeton How do you say 'Ocean City'. :)

@AlanMunn -- true! choice of hot sauce is definitely intensely personal. (my husband always carries a small bottle of tabasco with him in case he's somewhere that prefers something else. but he grew up in boston, so that probably says something.)

@barbarabeeton :D Yes indeed. Tabasco is my sauce of last resort.

@barbarabeeton This has gotten me hungry. I think I'll make shrimp and grits for dinner.

@AlanMunn -- pretty much "standard". my mother was a second grade teacher, so i got schooled in a non-local accent. when i landed in providence in college, a know-it-all claimed he could tell where a person was from by their accent, and placed me as someone from chicago. hmmmph. i do genuinely come from below the greasy/greazy line. "grease" is a noun, and "greaze" is a verb.

@barbarabeeton Ah, so that's how it works. Like 'house', then. (I only have the /s/ form, for 'grease').

@AlanMunn -- sounds good to me! (we enjoyed a most pleasant thanksgiving in quebec city. instead of turkey, supped on bison and wapiti. with good traditional pea soup as a starter.)

@barbarabeeton Nice. Québec City is a great tourist town I think. (I had to google 'wapiti' though.)

@AlanMunn -- yes. hadn't thought of the "house" comparison, but that's right on. when i was a kid, the local newspaper published a delightful "local" rendition of "the night before christmas", with "chimbley" and "pixture". wish i could find a copy of it now.

@barbarabeeton Do you also say 'Warshington DC'?

@AlanMunn -- you can see them in the flesh in yellowstone park.

@barbarabeeton What's weird to me though, is why the adjective form with –y should have the /z/ and not /s/ form. Since I think –y attaches to nouns, not verbs. e.g. girly, mousey, etc.

@AlanMunn -- a little closer to "worshington", but yes, i can do that. (however, i don't usually. my mother broke me of that habit pretty quickly.)

@barbarabeeton Playing her societal role well. :)

@AlanMunn -- i'll have to think about that. maybe i can come up with a counterexample.

@barbarabeeton Of course it's hard to tell nouns from verbs much of the time. What's really needed is an irregular verb which has a different nominal, like 'think'/'thought', but neither 'thinky' nor 'thoughty' are very good.

@AlanMunn -- that's a good start on a dinner-table conversation this evening. i'll report later on what we come up with.

@barbarabeeton Ok!

@AlanMunn -- maybe you can answer this ... do continental french people speak as fast as the canadian french? at the airport this morning, someone seated near us was speaking german on the phone, and my husband observed that he was speaking german as fast as the french canadians speak french. (i can't apeak english anywhere near that fast!)

@barbarabeeton if I can interfere into you chat, I can say that I know very fast speakers and also reasonable speakers both amongst French and Quebecois.

@barbarabeeton The Romance languages and the Germanic languages differ in their basic timing properties, mainly as a function of the ratio of consonants to vowels. Romance languages are 'syllable' timed and Germanic languages are 'stress timed'. This is because we allow more consonants per syllable than they do. So overall speech rate will be faster in French than in either German or English. I have no idea about a difference between Québec and Parisian French, though.

@barbarabeeton Here's nice paper comparing the three: ddl.ish-lyon.cnrs.fr/fulltext/pellegrino/Dellwo_2006.pdf

@AlanMunn -- neat! something for my husband to read. (well, me too.) his french is more marseille then paris; mine is the pitiful remains of three years in high school. don't get the chance to practice much.) thanks.

@barbarabeeton Here's a plot of %Vowels vs. StdDev of Consonants for a bunch of different languages. English/Dutch/Polish (stress timed), Spanish, Italian, French and Catalan(syllable timed) and Japanese (mora timed).

@barbarabeeton Even cooler is that 3day old babies can distinguish the three groups (but not within groups) with no exposure!