« first day (1325 days earlier)      last day (2722 days later) » 

6:08 AM
@AlanMunn Another brain teaser ;-) latex-community.org/forum/viewtopic.php?f=44&t=24883
 
6:42 AM
Can somebody check on that? tex.stackexchange.com/questions/185179/…
 
7:18 AM
@Johannes_B Looks similar to me
 
@JosephWright Thanks for having a look
 
7:38 AM
Good moring. I got the feeling that I am not able to ping people here, if they are not online at the moment.
is that a general distriction?
Or some privilege I may earn later?
"general restriction" (maybe my English is just too bad for this chat ;-)
 
@LaRiFaRi You can ping people who are around or have been around not too long ago. I don't know the exact time till the »pingability« expires.
 
@Johannes_B Ah ok. Thanks. Would be cool to have longer pingability (like the word). But well, getting contact in comments and delete them afterwards is ok as well.
 
8:37 AM
@LaRiFaRi if their name doesn't pop up when you start typing @ and the first letter then the ping won't work unless you have super mod ping powers. They have to have been logged on to chat "recently"
 
@Johannes_B wow!
 
 
1 hour later…
9:41 AM
Spam
-1
A: drawing a matrix with its minors

olivia   muhumuzaregistration information system's matrix output process output Admission selection allocation of course student's details verification students ID and approval latter

 
10:04 AM
@Joseph: hey, when was your birthday? I sense we missed it!
 
@egreg undeserved alignment tick:-)
 
10:23 AM
@DavidCarlisle I'm unsure the OP really should do that, but…
 
@DavidCarlisle Ok, good to know. It seems as if I don't have super mod ping powers. What a pity. Thanks for your reply
 
10:44 AM
What's the penalty for apostasy?
0
Q: How to change (Mac)Vim's syntax highlighting rules for (La)TeX files?

pkazmierczakI'm a long-time AUCTeX/Emacs user that switched to Vim/LaTeX-Box recently. I got everything working the way I want and grew pretty fond of Vim, but there's still one unusual thing that bothers me: Vim's syntax highlighting rules for LaTeX seem to be different and more "noisy" than those of AUCTeX...

 
@egreg Trying to quit vim? :)
 
@PauloCereda I was thinking to full erasure of the apostate's hard disk, backups included.
 
@egreg You are mean. :)
 
11:02 AM
@PauloCereda 28th
 
@JosephWright May?
 
11:13 AM
@PauloCereda June
 
@JosephWright ooh! :)
 
@LaRiFaRi you need to have a coup and dispose @JosephWright as mod and get yourself voted in.
2
@DavidCarlisle: Is it big, red, flashy enough? — Heiko Oberdiek 48 mins ago
4
 
11:39 AM
@DavidCarlisle haha, no no. I've got the feeling he is doing quite a good job. I can live without a diamond behind my name.
 
11:53 AM
@LaRiFaRi :-)
 
@JosephWright If I need a super urgent ping, I just ping you for a re-ping. (And now I am going to google, if "pinging" is even an existing verb)
 
@LaRiFaRi Yes
 
ring, ping, jingle, pink, tinkle
cool
so, should be working. Ciao
 
12:15 PM
@PauloCereda Oh! And now I'm sure I've missed yours.
 
@egreg Don't worry, it was just 2 days ago. :) The cake is missing though. :)
 
@PauloCereda happy birthday, now you're grown up you can start using a real editor!
 
@DavidCarlisle Thank you! So I can start using pico? :)
 
@PauloCereda Belated happy birthday!
@PauloCereda And merry unbirthday.
 
@egreg Thank you! :) Now I feel wiser (or dumber, one of the two). :)
@egreg ooh Alice FTW!
 
12:21 PM
@PauloCereda The former (or the latter).
 
@egreg Or both. :)
 
@PauloCereda If you're lucky we might let you use notepad
 
12:49 PM
@PauloCereda Belated happy birthday, too. Hopefully you have recovered from the seizure? Blame @David, he wanted it flashy. ;-)
 
1:17 PM
@HeikoOberdiek Thank you, Heiko! :) I'm feeling better, @David is mean. :)
 
2:00 PM
is this on topic?
0
Q: Common fonts used in writing a thesis

SumanI am writing my thesis but I am not happy with the default fonts of the template that I am using. So kindly inform me what are the common fonts used in thesis along with the latex commands for those fonts.

 
@FifaEarthCup2014 Wow, that's a lot of ducks! :)
 
@PauloCereda :-)
 
2:27 PM
@PauloCereda Even more than the Japanese on my train this morning!
@PauloCereda Are you ready for the match? Mexico is not Croatia.
 
@egreg :)
@egreg Nope, I'm scared. :)
 
@HeikoOberdiek A duck army!
 
@egreg \medskip/\bigskip overlapping comments:-)
 
@DavidCarlisle \bigskip is too much.
 
@egreg funny I thought you'd say "I'm wrong: your comment is much better than mine" :-)
 
I find NIST math render on the web nice. Like here dlmf.nist.gov/19.30 I do not think they are use tex4ht. But one very annoying thing is this: They have this "I" little circle next to each equation on the right side of the page. Yet the page slider in the browser is also on the right. Each time I go to slide the page down more, the mouse moves over one of them, and a pop up menu comes up. very annoying. Bad design
 
3:03 PM
@DavidCarlisle That's absurd, of course. :P
 
They should put these on the LEFT side, away from the slider ! So they do not interfer with the hand motion. Basic 101 web page design. or make them clickable so they do not pop up each time the mouse just moves over them
 
@Nasser they use latexml, that convertor was written for that project
@Nasser don't use the mouse to scroll:-)
 
@DavidCarlisle nice looking math. I tried latexml once, could not get things to work on linux. Was one year ago or more.
@DavidCarlisle hard not to use the mouse if you want to scroll down the page :)
 
@Nasser page down, down arrow or space buttons are easier:-)
@Nasser I suspect latexml is currently the most actively maintained latex to html convertor, as well as dlmf for which it was written they are plugging away at the arxiv converting an ever growing proportion of that catalog link.springer.com/chapter/10.1007%2F978-3-540-85110-3_46#page-1
 
@DavidCarlisle do you if version 0.8 is the latest version? LaTeXML 0.8.0! the one I tried while ago was 0.7 something. I worry about trying it again and end up wasting days on it. I see little discussion on latexml and conversion to HTML anywhere.
Are there other latexml->html examples on the web one can see, I wonder.
if it is better than tex4ht, I can change to it. but it took me more than one year to change things from latex2html to using tex4ht.
The link you gave: "Transforming the arχiv to XML", why do they not simply use tex4ht for this? I wonder.
 
3:22 PM
@Nasser seems like 0.8 is still latest it dlmf.nist.gov/LaTeXML/Changes, not saying you should change as you have things set up and a new convertor won't be set up, "The link you gave: "Transforming the arχiv to XML", I think they found the math conversion to mathml was a lot better for their purposes.
@Nasser Fred has a firefox addon that lets you put tex in the page and uses the latexml cloud service to get mathml, I'll see if I can find it...
 
@DavidCarlisle I'll try latexml V 0.8 again on few HW's I have with lots of math, and compare with tex4ht and see what I find. May be 0.8 will work better for me than 0.7 did.
 
This is a probably dumb, and not even really a LaTeX one, but people here should lmpw the answer. Anyway, terminology question. I have a line in bold above a LaTeX table, and some text below it. Would I be justified in referring to the former as a title and the later as a caption?
 
3:47 PM
15
Q: How can we make First Posts review actually useful?

Shog9I'm about fed up with the First Posts review queue. Oh, it was a nice enough idea in theory: give folks a chance to welcome new users to the site, without the heavy constraints that folks complained about with Low Quality review. The problem is, the review system was designed around the idea ...

Might be worth contributing to
 
@FaheemMitha The things are some combination of "title" "caption" "legend" but which is which (and for figures if any of them should be a "key") seems to be a matter of deep confusion, I tried a google search but didn't find any page worth referencing, they all just said the terminology is confused:-) basically, no one understands English:-)
 
@DavidCarlisle Ok. Thanks for checking it out.l I guess I'll just declare the words below (in normal font, and more of them) as a caption vs the words above (in bold font and less of them) as a title.
This is made a little more complicated by the fact that the R function I'm using to generate the tables seems to think the captions are above the table. My impression is that below is more standard.
 
@FaheemMitha I think captions below figures but above tables is common, especially as tables may be multi-page and you might need to know what the numbers are about before getting to page 50.
 
@DavidCarlisle there is a nice latex to html editor on the web that shows the conversion on the fly ! latexml.mathweb.org/editor the math looks really nice.
it is better than what I get now with tex4ht + images for math. But I read latexml does not support all latex packages, so I am sure this will break my build if I use it, as I use so many latex packages.
"Caveats: It isn’t finished, there are gaps in the coverage, particularly in missing implementations of the many useful LaTeX packages"
but I think if they can support all of the latex packages one day, then it will be the choice for me over tex4ht.
 
4:09 PM
@Nasser if you use images for maths the result isn't readable so it doesn't matter how many packages the converter supports (but you know that's my view:-)
 
@DavidCarlisle yes, I know. I switched back to images due to current problems with mathml/tex4ht on some math, and due to \SI not working in math mode with tex4ht when using mathml+mathjax. tex.stackexchange.com/questions/161907/… when these get sorted out, will switch back to mathml/matjax route
 
@DavidCarlisle hmm
@DavidCarlisle So how about title, then caption below that, then the table?
 
user image
7
@DavidCarlisle So I went to London this week…
 
4:25 PM
@FaheemMitha yes
@HenriMenke perhaps I own that part of London.
 
@HenriMenke That one is pretty old
 
 
1 hour later…
5:39 PM
Any Windows TeXLive wizards around? See this comment tex.stackexchange.com/questions/24753/…
 
@AlanMunn Not sure, but I believe the error might be actually unrelated to that.
 
@PauloCereda I guess that's also possible.
 
6:03 PM
@DavidCarlisle @JosephWright fyi, tried latexml with sinuitx package, it does not support it either. Here is the result. The on-line editor is latexml.mathweb.org/editor
 
@PauloCereda I read you had birthday a few days ago? My best wishes! :)
 
@cgnieder 2 days ago. :) Thank you. :)
 
user image
5
@PauloCereda So this is what your cake looks like? ^^^
 
6:22 PM
@cgnieder Oh my!
 
@Nasser not that surprising:-)
 
I am installing texlive 2014 now on linux :)
 
6:37 PM
opps! Problems
Installing [0430/2841, time/total: 09:02/50:37]: chktex [223k]
Installing [0431/2841, time/total: 09:03/50:40]: chktex.i386-linux [48k]
Installing [0432/2841, time/total: 09:03/50:40]: chletter [150k]
./tlpkg/installer/xz/xzdec.i386-linux: (stdin): File is corrupt
tar: Unexpected EOF in archive
tar: Unexpected EOF in archive
tar: Error is not recoverable: exiting now
untar: untarring /usr/local/texlive/2014/temp/chletter.doc.tar failed (in /usr/local/texlive/2014/texmf-dist)
untarring /usr/local/texlive/2014/temp/chletter.doc.tar failed, stopping install.
This in on linux mint 17, fresh installation. something is wrong in that texlive 2014 tar file?
These are the steps in case someone wants to try it
gunzip install-tl-unx.tar.gz
tar -xvf install-tl-unx.tar
cd to the new folder created and type
sudo ./install-tl
 
Hi,
I have a paper for the conference, They asked me that the pdf file should be saved with embeded fonts.
I don't know how to do that.
could anyone tell me how I save a pdf file with embeded fonts?
thanks
 
7:11 PM
@barznjy The PDF files produced by pdftex contain all fonts (as embedded subsets); requiring full embedding may be a breach of the font license.
@Nasser Network problem, I guess.
 
@egreg I think you are right. I am running it again (from the start) and now I do not see the error. it is still installing as we speak. keeping fingers crossed.
 
@egreg: around? :)
 
@PauloCereda Watching the match.
 
@egreg :) Mind if I ask a silly math question?
 
@PauloCereda Fire off.
 
7:17 PM
@egreg Thanks! :) I have a sequence W. I need to remove the first element of this sequence, so I wrote this: q \leftarrow \pi_1 (W), W \leftarrow W \setminus \{ q \}. Bad?
 
@PauloCereda Dubious, to say the least. A sequence is not a set. I'd simply borrow cdr from Lisp and say W\gets\cdr(W).
Second offside for Fred. ;-)
 
@egreg :)
@egreg Thanks! Where's \cdr?
 
@PauloCereda There isn't: \DeclareMathOperator{\cdr}{cdr}
 
@egreg LOL thanks. :)
 
@PauloCereda I know that Fred is your favorite player.
 
7:23 PM
@egreg Of course. :P
 
leo
7:34 PM
good game so far :-)
 
Fourth law: listen @egreg. — Sigur 3 mins ago
@leo Hi, Leo! What do you think about next Friday?
According to the Italian announcer, in Mexico they speak Mexican
@leo Probably in Costa Rica you speak Costarican, don't you?
 
leo
@egreg I think we'll have draw, but I'm an optimist :-)
@egreg indeed
I don't like mexican dubbings
And think we'll win against England
 
I think it will be fun time now, trying to debug all the problems trying to build my tree with tex4ht and texlive 2014 :)
Here is first one. oh boy...
(/usr/local/texlive/2014/texmf-dist/tex/generic/pgf/modules/pgfmoduleshapes.cod
e.tex)
(/usr/local/texlive/2014/texmf-dist/tex/generic/pgf/modules/pgfmoduleplot.code.
tex)
(/usr/local/texlive/2014/texmf-dist/tex/latex/pgf/compatibility/pgfcomp-version
-0-65.sty
! Undefined control sequence.
\pgfsys@svg@newline ->\Hnewline

l.190   \pgfusepathqfill}
 
@leo Don't tell it to @DavidCarlisle
 
good thing, is that I kept my old Linux virtual box with texlive 2013 running. so I have valid setup to use there.
 
leo
7:44 PM
@egreg ow
he he I won't
 
@leo TOO LATE
 
leo
@DavidCarlisle oh... sorry ;-)
 
@DavidCarlisle Maybe you can lose also to Uruguay.
 
@egreg Fifth law: back to the fourth law. :)
 
leo
Uruguay will lose all their matches
 
7:55 PM
My 3 cents' worth:
 
@PauloCereda Did you read the first three laws?
 
@egreg No the English press have explained it all to us, no need to panic we can still get through to the next stage, apparently losing to Italy didn't count as Italy can play quite well or something, anyway that didn't count.
 
@Mathematician: Yes, the spacing looks fine to you in the separate environments and that's good. In a similar light, the discussion in What is the Necessity of $…$ Around Numbers? would probably seem superfluous to you... and that's fine as well. It's just that a majority of LaTeX users have conformed to not using eqnarray for reasons already mentioned. Just like I enjoy cream in my coffee doesn't mean that everyone has to follow what I do, so be happy with your choice and accept that (most) others have a different opinion. — Werner 1 min ago
 
@Werner Cream in the coffee? Oh, that's terrible!
 
Is there anyone here who uses auctex preview? If so, can you test a file that breaks it?
 
leo
7:56 PM
black without sugar
 
@leo That's what I mean, bro!
 
@egreg I did. :)
 
leo
@egreg :-)
 
@egreg I didn't grow up in Italy the way you did... where neat triple-shot Espressos were the order of the day at age 14, sorry. :)
 
@egreg no cream, no milk, no sugar – just coffee :)
 
leo
7:58 PM
was OK, they went to WC 2010, but they won't pass first round in 2014
 
@cgnieder So long as it's not German style coffee. ;-)
 
leo
sweet revenge
 
@egreg lol :) French style actually
 
leo
French press :-)
 
But I do enjoy coffee the Italian way, too. Especially after lunch
 
leo
8:00 PM
excuse me, which is the Italian way?
 
@leo How did you know? :)
@leo Espresso?
 
leo
@cgnieder oh, didn't know
@cgnieder they say is the best way to make coffee
 
@leo The only one. :P
 
leo
he he
 
@leo I know. I'm not sure I agree with best way but it's certainly a good way...
 
8:01 PM
@leo Actually, also Turkish style coffee can be very good.
 
leo
@egreg how is it?
 
A little question about LaTeX3. After reading this question tex.stackexchange.com/q/185346/21930 I remembered something I thought some time ago. What is the “correct way” of writing the signature in those cases? \test:w #1 \q_stop has clearly one argument :w but shouldn't \test:w #1 / #2 / #3 \q_stop be \test:www? I haven't seen any command in latex3 with more than one w in the signature, but I don't know, I doubt...
 
leo
we all should go to Europe
:-)
 
@Manuel :w is for “weird”.
 
8:04 PM
@Manuel My interpretation is that, whenever you see :w, you should really refer to the documentation to see what it's supposed to be doing.
 
I know, but there are three “weird” arguments. Why not indicate that in the signature?
 
Throughout the entirety of the manual, I don't think I've ever seen more than one w
I get what you're saying, though, and I would call it (with my limited experience) a gray area
 
@SeanAllred I haven't seen one, as I said… but I don't know, I think it's more “congruent” to have :www.
@SeanAllred That's why I brought the gray area here, to see what you think :P
 
@Manuel I'm reading texdoc expl3 to see what it says; perhaps it inadvertently sheds some light on this
@Manuel
> The ⟨arg-spec⟩ consists of a (possibly empty) list of letters, each denoting one argu- ment of the function. The letter, including its case, conveys information about the type of argument required.
So it would seem that :www is correct.
 
leo
@egreg I want Turkish coffee
 
8:10 PM
@leo Of course, in Greece they call it “Greek coffee”, but it's exactly the same. I had it in Wien, accompanying a big slice of Sachertorte at the Hotel Sacher. Yummy!
 
@SeanAllred Great :P The question came to my mind after seen for first time :wN and :nw and something like that. Which means that w is just one of the arguments. Then I thought, if there are more w arguments, then indicate that.
 
@Manuel It seems consistent with expl3's goals of creating a consistent naming scheme.
 
@Manuel You're right: it should be :www; example \__cs_show:www in expl3-code.tex
 
leo
@egreg sounds delicious
 
@Manuel Well w-type can be 'up to and including anything': something like wN is more defined
 
8:21 PM
@Manuel well partly because :w is needed at the lowest level but at higher levels you shouldn't be doing that:-)
@SeanAllred probably you shouldn't use delimited argumenst at all (and it would be more in the spirit of the question just to parse three numbers ending after the third
 
@DavidCarlisle but what if I (or someone else) would wish to provide only a two-digit year?
Wouldn't TeX throw a not-enough-arguments error?
 
@DavidCarlisle but if that kind of command is used, then the correct way is to indicate that, like that command \__cs_show:www.
 
@Manuel Up to you to some extent: it's really a warning that Hic sunt leones
 
@DavidCarlisle Sorry, I still don't know how to control what message I'm answering to :S
@JosephWright Okey, that's the point of my question. I thought it would be correct to say :www but since I haven't seen any command with more than one :w I thought that may be, as you say, the :w meant “up to and including anything”. However, I still think it's more clear (and does not harm) to use :www in that case.
 
@Manuel Unlike the other cases, w doesn't have to map to exactly one argument
 
8:34 PM
@JosephWright Yeah, I understand that. But since some of you “are LaTeX3” I wanted to see your opinion. :w can map to more than one argument… but is :www right? Is it even better in this case? Would you, from the LaTeX3 “authority” recommend using :w or :www in this case?
 
@Manuel Originally we used just w, but more recently have tried to be a bit more nuanced. Tends to be cases more of the form wnN (i.e. stuff after w)
 
@SeanAllred I added different code that doesn't rely on weird arguments
@PauloCereda Marcelo tried the same trick as Fred. ;-)
 
@egreg Ahh! Never thought about using sequences! And if history has taught me anything, clearer is almost always better than faster. Computers will get faster, but people may not get smarter. ;)
 
@egreg :)
@SeanAllred Beg to differ, the search for faster make you become eager to be clear on your thoughts. :)
 
8:51 PM
@Manuel the arrow on the right or "reply" from the drop down on the left:-)
 
End of game! :)
 
@SeanAllred pick up three integers and check it's less than 100
 
@DavidCarlisle Cool!
 
@PauloCereda One minute to finish downloading of MacTeX.
 
@egreg ooh! I need to download a copy!
 
8:59 PM
@PauloCereda The search for clearer leads to a solution that makes sense, and the clear solution is straightforward in its own right and thus faster. Now, the problem comes in with lazy solutions. Not thinking about your problem yields these, and these are the bottlenecks of software.
(I'm kind-of hand-waving nester for loops and such; those should be a red flag by default.)
@DavidCarlisle Oh—clever :)
 
@SeanAllred Theory is female dog sometimes. :) Since you are dealing with beautiful constructs, real scenarios don't count. :)
 
@SeanAllred I could post an answer along those lines and try to get that tick off egreg
 
9:23 PM
!!/help
@PauloCereda Where's Psmith when one needs him?
 
@egreg Just a minute. :)
 
@PauloCereda don't bother he doesn't want help he wants battle
 
@DavidCarlisle :)
 
!!/help
@PauloCereda Psmith, the TeX bot: Good night! I'm Psmith, the friendly TeX bot - the p in my name is silent, as in pshrimp. I'm here as a companion to our fellow users in the typographic land. As you probably noticed, I always reply under Paulo's account, but do not despair, I say, my replies are always preceded by my own name. Enjoy your stay at TeX.sx! If you need any help, just ask our chat residents. Cheerio!
 
!!/battle
 
9:27 PM
@egreg Psmith, the TeX bot: The current score is egreg 380 vs. 215 David. So far, egreg is winning.
 
!!/eightball Will England make it against Uruguay?
 
@egreg Psmith, the TeX bot: The great 8-ball says: only when emacs and vim learn to live in harmony.
 
@PauloCereda that's no then
 
OK, so @DavidCarlisle is settled up.
 
@DavidCarlisle You can try, but I'm very partial to expl3 answers :)
 
9:38 PM
@SeanAllred I could probably do an l3 version:-)
 
@DavidCarlisle One thing that occurred to me: If HTML5 validity is specified by describing the "parsing algorithm", what about decidability? Has this been established?
 
@DavidCarlisle Now we're talking XD We shall see :) No promises, though—egreg has a much cooler name.
@StephanLehmke Did HTML5 introduce a significant change in its language class?
(Or something that would give you pause?)
 
@SeanAllred I don't know anything about these matters. I learn and wonder.
But apparently there is no formal grammar for HTML5, just a prosa description of the parsing algorithm.
 
@StephanLehmke yes it's pretty clearly decidable I think as it's a finite state machine with finite lookahead with certain states declared invalid
@SeanAllred It replaced an SGML DTD by English prose, I'm not sure how that fits into a language classification:-)
@SeanAllred he hides behind a pseudonym because he's so embarrassed by his answers (but we know who he is really)
 
@DavidCarlisle Er, in that case there would trivially exist a grammar representation (say, EBNF). Why don't they write that down instead of a bunch of rules made of bullet lists?
@DavidCarlisle Is there a paper on this with a proof somewhere?
 
9:49 PM
@StephanLehmke mostly because they are more concerned with the invalid than valid startes. The parser algorithm gives a defined parse tree for every input string.
@StephanLehmke you are kidding, that isn't the way the whatwg works:-)
 
@DavidCarlisle With finite you mean "finite but longer than the longest admissible HTML file"? How would one check for proper tag nesting otherwise?
Also I get validation errors of the form "a column established by this <col> element has no cell starting in it". Do you really think this is the kind of thing checked by a finite state machine?
 
@StephanLehmke finite lookahead for tokenization If you start a tag <aa you only need to read one character at a tine to decide wher e the the tag name ends
 
@DavidCarlisle So how does the finite state machine check for tag nesting then?
 
@StephanLehmke well need to be careful in what you mean by valid, the html spec has a parser that is effectively a finite state machine with some states invalid, it also has several other prose descriptions of author conformance that authors should or should not do, yes well as I say tokenization (which is in fact most of it:-) has finite states you need some stacks for tag nesting of course and also for random element re-arrangements inflicted on tables and math:-)
 
@DavidCarlisle A finite state machine with two stacks is equivalent with a Turing machine, so decidability is not automatic considering this.
 
9:59 PM
@StephanLehmke yes mainly thinking of the character by character parsing to determine that tags (I think that is really finite state) although since it ow written in english and interleaved with other stuff hard to be sure:-)
 
@DavidCarlisle By valid I mean "I feed this HTML in and it says whether it's valid or not" (as for instance the W3C checker does). This is either decidable or not.
@DavidCarlisle Isn't this a sufficiently interesting property that someone would write a masters on it?
 
@StephanLehmke not so clear as it isn't formally specified what a validator needs to or can check, the authoring requirements are written in English and the v.nu parser approximates that in Java
 
10:11 PM
@DavidCarlisle Ok, this is really strange to me, but it's real obviously. Thanks for clarifying the muddiness.
 
@StephanLehmke with reference to the stacks you can't put arbitrary data on the stack just the stack of open elements (and a similar one for forms and stuff: the spec says:
Implementations must act as if they used the following state machine to tokenise HTML. The state machine must start in the data state. Most states consume a single character, which may have various side-effects, and either switches the state machine to a new state to reconsume the same character, or switches it to a new state to consume the next character, or stays in the same state to consume the next character. Some states have more complicated behavior and can consume several characters before switching to another state. In some cases, the tokenizer state is also changed by the tree cons
@StephanLehmke so basically a finite state machine builds a stream of open and close tags with attributes and text data and passes that to the tree construction stage that builds the tree and does need to maintain stacks of open elements, but the tokenization and tree building stages (and the rendering of that tree) are all interleaved so it's complicated (and document.write('<p>') means the whole lot is re-entrant and can start a nested parsing context at more or less any point...
@StephanLehmke so the quiz is, what tree results from
<math>
 <mfrac>
  <p>x</p>
  <mi>y</mi>
 </mfrac>
</math>
in an html parser (eg if you put that in an html file, select all and view selection source in frefox etc?
<math>
 <mfrac>
 </mfrac>
</math>
<p>x</p>
<mi>y</mi>
@StephanLehmke ^^^ the resemblance to XML syntax can be deceptive:-)
 
Hai there. Does anyone know how to typeset a double prime (equivalent to unicode character 2033) in LaTeX?
 
10:26 PM
@eiterorm $f''$
 
@eiterorm Or $\f^{\prime\prime}$
 
@SeanAllred Why doing it the hard way?
 
@egreg Ah. Thanks. I tried using the \prime command, but that didn't work.
I also tried a typewriter quote, assuming that would be the correct character for a double prime.
 
@eiterorm \prime should be used as a superscript; an apostrophe in math mode is equivalent to ^{\prime}, two to ^{\prime\prime} and so on.
 
@egreg I see. Thanks for clarifying.
 
10:30 PM
@egreg Ah, I didn't know they were equivalent :) They certainly look the same :) Certainly much easier than typing \prime over and again.
Out of curiosity, why do you suppose the control sequence exists? Is it just actively replaced by mathmode to typeset the right math character? (230 in this case, by the way)
 
@SeanAllred ' is “math active”, defined as ^\bgroup\prim@s; you might enjoy looking at the definition of \prim@s in latex.ltx
 
\prim@s:
macro:->\prime \futurelet \next \pr@m@s
 
Is there any reason why the character isn't available in text mode, or is it just not font supported?
 
@eiterorm It's present only in math fonts.
 
@egreg I find that a bit awkward, seeing as the character is fairly commonly used in regular text.
To denote seconds, inches, etc.
 
10:38 PM
@eiterorm Which are math.
 
As is currency, for instance, but there's support for that.
 
@SeanAllred Now look at \pr@m@s
 
\pr@m@s:
macro:->\ifx '\next \let \nxt \pr@@@s \else \ifx ^\next \let \nxt \pr@@@t \else \let \nxt \egroup \fi \fi \nxt


\pr@@@s:
macro:#1->\prim@s


\pr@@@t:
macro:#1#2->#2\egroup
 
@SeanAllred Now you know how it works. ;-)
A superscript is begun and a prime is added. Then the macro adds as many primes as apostrophes it finds; if a ^ follows, it is gobbled and the explicit superscript is added to the already open one; finally \egroup finishes off.
 
11:00 PM
@egreg It took me about eight read-throughs, but I think I finally get it… woah.
 
@SeanAllred It's taken directly from plain.tex
 
@SeanAllred if you replaced all those @ by _ it would be so much more readable:-)
 
@DavidCarlisle If you replaced the letters by words, maybe ;)
 
leo
11:54 PM
Intriguing:
! Package babel Error: Unknown option `spanish'. Either you misspelled it
(babel) or the language definition file spanish.ldf was not foun
d.
See the babel package documentation for explanation.
Type H <return> for immediate help.
yesterday it worked
@egreg given a positive rational number $r$, chances are that $\sqrt[k]{r}$ is irrational, right?
for integers it's known
 

« first day (1325 days earlier)      last day (2722 days later) »