12:09 AM
It's a pitty tabularx doesn't support line breaks :-(
http://tex.stackexchange.com/questions/182863/long-tabularx-doesnt-break-rows

@wget: You mean page breaks? ;-)

@ChristianHupfer Yep this is what I meant ;-)
@ChristianHupfer ltxtable could come in handy this time be would require me to use another file, according to the manual.

@wget: No, not really. You could keep it in the same file as your main document, using \begin{filecontents}{averysophisticatedname} \begin{longtable}....\end{longtable}\end{filecontents} ... it requires of course the filecontents package

@ChristianHupfer Ok from #latex on freenode, I've maybe found a simper solution --> use the more recent tabu package. Gonna check that. Thanks for your idea btw.

@wget: I expect a report ;-)

12:32 AM
@ChristianHupfer no, it does not require the package.

12:47 AM
@ChristianHupfer For the report ;-) this is working nicely with longtabu, which allowed me to keep my tabularx macro implementations. Had just to remove the first tabularx parameter and replace the env name obviously.

1:45 AM
@wget the examples there like 99% of similar requests for that feature would be better set as a list then page breaking is automatic
@wget tabu can't break a page mid row either unless it has changed recently
@ChristianHupfer you only need filecontents package if you want to put the filecontents mid-document, if you put it at the start it is in the latex format

@DavidCarlisle or if you want it to overwrite an existing file, which is what would be needed in this case, assuming he'll be working on the table "live".

@DavidCarlisle: Thanks, I've seen solutions with filecontents only so far... never used ltxtable so far ;-)

2:30 AM
@DavidCarlisle Unfortunately, with a list, I cannot separate the values from my keys with a colon. wrt. tex.stackexchange.com/questions/227259/…

@wget of course you can, why not?
@PaulGessler ah yes there is that. Of course ltxtable doesn't really have to have a separate file I could have used a macro except I took it for granted no one had enough memory to hold a long table in macro space. ltablex and tabu use the same code but assume a bigger machine:-)

3:04 AM
@DavidCarlisle I always wondered the reason for that, makes sense now that I think about it (that data has to be stored somewhere). First I've heard of ltablex, actually.

3 hours later…
6:30 AM
Are there any liquid substances that become solidified when they are heated? I just found eggs. Do you have any ideas?

3 hours later…
9:11 AM
@stalkingisn'ttolerated you just found eggs?

9:24 AM
@DavidCarlisle First thing Peter Capaldi said was kidneys. :)

@PauloCereda must have given his mother a shock

@DavidCarlisle But he could fix it later on, after all he's the Doctor. :) Unless of course that event being time locked...

9:37 AM
Do you love falling dominoes? Video posted on ELU: teachertube.com/video/honda-rube-goldberg-269615 (no onebox preview, unfortunately)

@DavidCarlisle Case changing mainly done: a few tweaks/doc fixes only outstanding

10:00 AM
@DavidCarlisle Yes. Is there any other?

10:12 AM
@stalkingisn'ttolerated Eggs are not a liquid: complex solution/colloid going on
@stalkingisn'ttolerated You are basically looking at a supercooled state by the sound of things
@stalkingisn'ttolerated For example, one can cool liquid water well below zero if it is not moving, which can then lead to a sudden freeze when a nucleation site is introduced
@DavidCarlisle Will do another CTAN update of case code once I've addressed the loose ends: I want to get people to drop \tl_expandable_(lower|upper)case:n (there are a few uses in the wild)

But I want to heat rather than cool.

The enthalpy of fusion also known as (latent) heat of fusion is the change in enthalpy resulting from heating a given quantity of a substance to change its state from a solid to a liquid. The temperature at which this occurs is the melting point. The 'enthalpy' of fusion is a latent heat, because during melting the introduction of heat cannot be observed as a temperature change, as the temperature remains constant during the process. The latent heat of fusion is the enthalpy change of any amount of substance when it melts. When the heat of fusion is referenced to a unit of mass, it is usually called...
@stalkingisn'ttolerated See the part about exceptions: the usual suspect (helium) is the only one

@JosephWright OK. Thanks, I am reading... :-)

@Rico No
@Rico TeX requires an explicit open-brace/close-brace pair

11:00 AM
Morning, got a quick question, probably too specific for a proper question: I was using titlesec to reformat the paragraph command to use it like a subsubsubsection title, but I was just wanting to switch to biblatex and now I see it's not compatible with titlesec. I don't think I can do the same thing with sectsty - is there another alternative?

@FionaSmith they are incompatible only partially, are you sure you need those specific incompatible options?

@Rico No, it isn't: \bgroup can't be used for delimiting a mandatory argument.

@yo' I was just wondering that - if it is only incompatible with refsection I can probably live without it (might actually have been useful in this application but an extra level of sectioning is more critical (yes I know, the document is probably poorly structured but I can't see a way round it...)

@FionaSmith you can create the extra sectioning level manually, or use the other package. I'm, however, not sure whether it's not also incompatible.

@yo' yes I know there are other methods, but I think when I looked into it before titlesec was the easiest way (I'm not very experienced and would prefer not to add too much indecipherable mess...)

11:09 AM
@FionaSmith titlesec is good. biblatex is good but really messy in some sense

@yo' Ha! Maybe you're putting me off my current project to convert. Just totally fed up with .bst files - can't find a bibliography style that I like; quite a few don't even seem to work with my .bib file (possibly because of a critical paper with 42 co-authors).

@FionaSmith you mean 42 authors in the bibliography? Have you tried AUTHOR = "Doe, J. et al."
you don't wanna list all 42 author in the References section, do you?

@FionaSmith 42 authors?! Holy duck!
The \maketitle thingy is bigger than the paper content itself. :)

@yo' 42 co-authors in one reference. The problem is that they are pretty much all authors on this also, and might get a bit narked to be left off. I had to do it in alphabetical order because it wasn't really possible to separate main contributors...

@FionaSmith that's a plain wrong approach IMHO. First you cannot modify the order of the authors, second the copy editor will very likely reject this as well, third nobody expects (I hope) that any paper with more than 5 authors will ever have all names mentioned in the references.

11:22 AM

@PauloCereda there are papers that have ~1000 authors I think. (CERN, I look at you)

@yo' - I meant when I wrote the original paper the authors were in alphabetical order, so some of those who are my current co-authors are way down the bottom because their names begin with S/T/whatever

@yo' No way! You are making fun of me! :)

@FionaSmith then make it First Author et al. There's no other way to go.

@yo' and take all the credit for myself (being first author).... Hmm.

11:24 AM
@FionaSmith No: what makes you think that?

@yo' Anyway, it's all fine with certain .bst, just means I can't be too flexible about my choice of formatting.

@FionaSmith If you want to highlight someone else, then 'X and coworkers' by hand (typical in chemistry: lead author last)

@JosephWright what makes me think what?

@FionaSmith but it is plain wrong. I truly hope that if you do it, it'll get marked in the copy editorial for shortening.

@FionaSmith That putting 'X et al.' is emphasising the first author: using et al. is just an way of abbreviating things

11:26 AM

@yo' well sorry to say, this is not for a journal, this is a "science plan" not for public consumption

@FionaSmith well, then the standard way is: Smith, F. et al (the ABC Team), which simply says: I was the first author and it was made at this department.

@yo' or like I said 'ABC and coworkers'

@JosephWright I'm only talking about my reference list here, not the citation in the text itself. I already have this reference in my thesis, in my opinion it's fine, and I'd rather all the co-authors were listed than arbitrarily cut off after 5 or something.

@FionaSmith Give all the authors then (sorry, I don't agree on the 'you can truncate' idea)

11:29 AM
@JosephWright now I'm losing you. You don't agree on 'you can truncate' or 'you cannot truncate'?

@yo' I don't think you should cut out authors in the references section: et al. is for the body of the document

@JosephWright yes I agree with you.
Anyway, this is a bit beside the point - wish I'd never mentioned it!

@JosephWright well, the reference is intended to be a unique ID of the cited paper. I agree that in a project it's different from a paper. But in general, you don't even need to include the first author. The following is a valid full reference: ApJ 5:7102

@JosephWright I do think the same. Sadly some journals do "cut" authors when the number goes higher than 4.

(of course given there's an article starting on page 7102 of volume 5 of the Astrophysical Journal)

11:32 AM
@yo' Certainly one sees that approach in older papers, but that's really 'lab talk' to me

@JosephWright I think we won't agree on this :)

I don't like DOI's, ISBN's, ISSN's and other witchcraft! :)
Soon we will have a Facebook Like nearby each reference. Bah.

@yo' Borderline library science, so not my area of expertise :-)
@yo' Not so good when there is a typo :-)

@FionaSmith The reference section is meant for identifying cited papers and books, not for giving credit.

@PauloCereda I like to have all the proper information myself, but I do find the DOI useful nowadays - you can just type that into google and get straight to the article.

11:35 AM
@FionaSmith Indeed. Useful when your colleagues don't know how to cite papers. :)

@egreg To some extent, but as @yo' says by that argument even historically (pre-DOIs) you wouldn't include authors for papers (or titles in areas that do that)

@FionaSmith note that DOI is not currently considered a full citation. It'll likely be in a close future, though, I think.

@everyone... I think it's pretty clear we have differences of opinion on whether one should list all authors in a reference list, but given that I have complete control of this document and it does little harm, they're staying in. Anyway, the point is that I think that is why I am having issues with certain .bst

@JosephWright Cereda, PRM. The quantum duck says quark! Journal of Theoretical Ducks, 7:302. 2015.

@FionaSmith Which one?

11:38 AM
@JosephWright Under the dominant “cite-as-much-as-you-can” policy, credit is given to authors by the automated scanning.

@egreg :)

@egreg Like I said, I was thinking more historically

@JosephWright there could be other problems with my .bib file; it may not be that. But I've tried about 25 different options, and I'd say 8 won't run (complain that the format is not compatible with Author,Date)

@JosephWright currently, the problem is as big in some branches: A paper has 20 pages. The first 15 are authors and the last 4 are references.

@JosephWright I am pretty sure it is that though, because I have constructed a MWE to help me find a .bst that gives output I like (very personal, obviously), and is capable of generating 2009a, 2009b. nature.bst is an example of one that doesn't like it. I was thinking of giving up on natbib and moving to biblatex.

12:39 PM
New question idea for main: Why is it so complicated for me to have tikz-pgf lay out my flowchart with blocks snapped to the golden spiral?

1:03 PM
@PaulGessler that one would be off-topic, since it makes too much sense.

@yo' ;-) ok, good to know these things ahead of time. I will stop preparing my MWE now. :-)

1:46 PM
open bounty:
1

When I paste a text inside quotation marks into Emacs+AUCTeX, AUCTeX formats it with a soft red color (see the screenshot below at the first line). I like this color. However, I'm using csquotes package to handle (multilingual) quotations marks. The \enquote commands (provided by csquote package...

ending today

@yo' @JosephWright Pretty happy now. Has taken me 2 hours to switch to biblatex (though I sorted out how to run biber in TeXworks a couple of days back) and fiddle with all the options so my citations and reference list now appear exactly how I want them to. I declare biblatex a success (famous last words).

2:02 PM
@FionaSmith wait until a journal will ask for "a separate .bbl file". ;)

@yo' sure, I know it's not for everything, but if writing for a journal I am not going to be annoyed that the formatting is not what I like since you just have to do it as they want. It's pretty easy to switch anyway.

@FionaSmith and I suppose you don't post your stuff on arCiv right? Their biblatex support is zero (not an epsilon more), as far as I know.
s/arCiv/arXiv/ of course

3:00 PM
@ChristianHupfer tex.stackexchange.com/questions/200721/… Haarspalter

3:17 PM
Hm the feeling of being completely stuck, having spent a long time writing a minimal example (without solving the problem), having asked the question and hoping someone will give some nice pointers while at the same time feeling that although one is completely stuck one should try to find another solution... Is there a word for that feeling? :)

@Johannes_B: Gern geschehen... (Upvote war übrigens von einem Schwarzwälder, der nicht näher genannt werden möchte :D)

@ChristianHupfer Had the feeling that it could have been you. How are you?

@Johannes_B: Busy.... preparing for school job, Abitur starts in three weeks ... and how about you?

@ChristianHupfer Busy, janitor in real world, janitor in the real LaTeX world as well. And of course all the other stuff going on.

3:23 PM
@ChristianHupfer Still going on.
@ChristianHupfer That is the other stuff

@Johannes_B... but somehow sleeping ? ;-)
@Johannes_B: You should not act as the template police ;-) Will bring much spare time back ;-)

@ChristianHupfer Ja, you are right. Will cut back on this a bit.

@Johannes_B: I will have to cut down my TeX.SX time in near future too... much work to do

@ChristianHupfer Abitur, ohje.

@Johannes_B: Unter anderem... Fortbildungsmanagement... Bewerbung um Seminarstelle... Mentor für einen Physikreferendar... mein Tag hat 28 Stunden ;-)

3:30 PM
@ChristianHupfer I know the feeling. And still, we are here, chatting about being too busy. Contradiction ;-)

@Johannes_B: Well, school holidays is my excuse, what's your's? :-P

@ChristianHupfer Short break ;-)

@Johannes_B: Entered the grades into a table ;-)
@Johannes_B: Seen this: tex.stackexchange.com/questions/228804/… ???

@ChristianHupfer I have found that there seem to be a few kind of users. Some improve their skills with every new question, problems get more complicated. Others, do just the same, gimme stuff approach, most likely not posting a MWE, even if the person has been asked for an MWE with every single question so far. The problems keep being solvable by a two minute google search. And then there are the persons, so seemingly not even read/process/understand the solutions and explanations by helpers.
3

@Johannes_B: I can imagine to which group you categorize the OP of the question I linked ;-)

3:44 PM
@ChristianHupfer That was unrelated.

@Johannes_B: Anyway ;-)

@ChristianHupfer have you opened his network profile?

@yo': Now I have ... I am not impressed

@JosephWright @egreg @DavidCarlisle Concerning language support, will there be any consensus? tug.org/pracjourn/2007-1/gregorio/gregorio.pdf mentions pros and cons of giving the language as a global option. If you pass the option directly to babel, translator won't see it, and hence siunitx. Where does language come in in the layer system of L3?
2

@ChristianHupfer I don't say you should be impressed, I just think it's worth noting.

3:59 PM
@ChristianHupfer btw, i would say the letter â does not exist in the german alphabet.

@Johannes_B: Yes, that's what I wrote (well, I wrote German language)

@ChristianHupfer Not the same my friend

@Johannes_B: Yes ;-)

@ChristianHupfer you should have seen me, finger in the air saying Du, Du, Du!

@Johannes_B: scared to death

4:05 PM
@ChristianHupfer You should be, i can through u+1e9e (LATIN CAPITAL LETTER SHARP S), more dangerous than a shuriken. ;-)
@ChristianHupfer Other questions i was confused of tex.stackexchange.com/questions/163242/… and tex.stackexchange.com/questions/210441/…

:)
@Johannes_B Das Unikodeletter!
I should've gotten a DANTE membership instead of UK-TUG. :)

@PauloCereda Looking at my post, i think throw $\neq$ through :-(
@PauloCereda Careful, the letter isn't part of the official german alphabet.

@Johannes_B ooh!

@PauloCereda It seems the germans are the only latin based alphabet, that don't have a matching number of uppercase/lowercase letters.
On Topic?
1

I see many slides of my study materials which are done by WYSIWYG and colored with many colours to improve the contrast. To improve contrast with colors, I think is not necessarily the best way to go. I think this kind of text is difficult to read. Is there any package which considers the contr...

@PauloCereda got an arara/texworks question...
sorry, should say I have got...

4:21 PM
@FionaSmith Be careful, @PauloCereda is jut the programmer, he doesn't use arara. ;-)

@Johannes_B well he probably can't help then ;-)

@FionaSmith I'd like to help you, but my arara knowledge is more than limited.

@FionaSmith On site or here? :)
@Johannes_B Truer words were never spoken. :)

@PauloCereda here I think: It'd only be closed if I asked elswhere... Anyway, my question (to whoever can answer).... My document is pretty buggy at the moment (partly because I just switched to biblatex). I find if running arara that when the pdflatex part gets stuck and I click on the "stop typesetting" button, that arara stops but leaves pdflatex.exe hanging
Anyway to circumvent this? I keep having to have TaskManager open...

arara 5.0 will have an integrated TeX editor (with L3 features!). :)
@FionaSmith Windows. :(
I think I answered that somewhere, hold the phone.

4:24 PM
Does anyone know how I can replicate the style and format of the AMS Notices, e.g. ams.org/notices/200409/what-is-illusie.pdf?

@PauloCereda Yeah I know. I am stuck between a rock and a hard place: at work I have linux with a latex distro circa 1996... supposedly will be "more recent" when we get RedHat7. At home I am too scared to have linux (and I don't buy into Apple), so I'm stuck with windows...

And use the location from TL instead. Mine is /usr/local/texlive/2014/texmf-dist/scripts/arara/. :)

@Johannes_B I have some ideas but they need development: certainly this should be a kernel-managed concept

@JosephWright Good to hear that someone is thinking about it. Thanks.

@Johannes_B I need to talk to other people on the team (most obviously Javier) about it but there are all sorts of other decisions needed (for example the entire Unicode question)

4:33 PM
@JosephWright The whole thing is quite a task, but with Javier in the boat as well, i am confident that there will be a good solution. Better to take some time than to hurry into a decision.

@Johannes_B Certainly
@Johannes_B There's the entire business of language versus locale to sort out: babel rather conflates the two

@PauloCereda sadly doesn't work... Texworks says it can't find the file (and yes, I have triple checked I've typed the location correctly - if I try to select it from the "browse" button it says it's not executable and won't even allow me to select it :-(

@FionaSmith Try the other way: Select java.exe and add -jar and other stuff in the parameters.

Also, does anyone know of any free fonts very similar to Lucida Bright?

@JamalS -- sorry, that's not a publicly distributed class. the one used in-house has idiosyncracies and often requires manual "tuning" of the content. not in any condition to distribute since support requirement would be impossibly heavy.

4:39 PM
@PauloCereda Hi, Paulo! Did you try the code with the loop?

@barbarabeeton Any tips on how I can replicate the format of the title?

@PauloCereda I don't know Java - not 100% sure of the syntax you're suggesting. "-jar /blah/texlive/blah/arara.jar"?

@GonzaloMedina Hi Gonzalo! :) Here!

@PauloCereda Is that so? Please tell me you'll be able to quit the editor in a reasonable way and not like in that editor in which login out is a puzzle ;)

@GonzaloMedina (hold on, please :) )

4:42 PM
@PauloCereda sorry I see that's in the answer to the previous Q

@FionaSmith this should be the first parameter to the java.exe call, yes

@PauloCereda ahh! <3

@GonzaloMedina: ^^ This is using the raster images. I have Mary totally vectorized, but the background was still too difficult for me to grasp in just one go. I'll deal with it later on and we will have a 100% TikZ powered solution. :)

@PauloCereda with the test text there, I smell a programmer is involved :)

4:44 PM
@JamalS -- i presume you mean the title of feature articles. a minipage to the desired width, \flushright

@FionaSmith Here we go

@PauloCereda Looks nice!

Executable: C:\path\to\java.exe
Parameters:
1. -jar
2. c:\path\to\arara.jar
3. <arara flag 1>
4. <arara flag 2>
...
n. file name
@PaulGessler Can't wait until @percusse uses @DavidCarlisle's aircraft to run over Mary in the sky. :)

@PauloCereda I'm lost. Are those lines of code you posted for me?

@barbarabeeton I'll give it a go; thanks.

4:47 PM
@GonzaloMedina No, those were for @FionaSmith, sorry. :) I meant it was your code that helped produce this awesome keynote! <3
@GonzaloMedina: I will surely send you the final TikZ version of this beamer template. :)

@PauloCereda I have the perfect place to use a variation of your idea...

@PauloCereda don't use the word template when @Johannes_B is around. :-p :-)
Nobody expects the template inquisition
3

@JamalS -- (accidentally hit return, and edit didn't work.) \flushright or \flushleft depending on odd or even page. make sure each element is ended with \par to freeze the correct baselines. set up font size with \fontsize{...}{...}\selectfont.

@PaulGessler LOL
@Johannes template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template template
:)
Power shortage, uh-oh!

@PaulGessler I don't have a problem with templates, they have a problem with me. ;-)

4:52 PM
@Johannes_B :-)

@PauloCereda I was invited to ExpoLaTeX 2015 in Peru to give a mini-couse. I'm thinking of using it there...

@PaulGessler Shall call my first son Gundar, hunter destroyer of templates
3

@PauloCereda By the way, we could meet there. It's midway between Colombia and Brazil.

@Johannes_B ooo! You just thought of a great character for @Paulo's new HBO series: A Game of Macros:
20

Here's my humble attempt to answering this question. Since David asked me in the chatroom whether my answer would include ducks, here's one for obvious reasons: :) Prologue I personally believe that complexity is in the eye of the beholder. I commented this somewhere in this site, but IMHO it'...

@PaulGessler LOL

4:58 PM
I'll be the food hunter for a while, don't do any crazy stuff while i am away.

@PauloCereda seems to be working in as much as I can now typeset again :-) Thanks very much. Of course pdflatex is now running to completion so I'm not sure whether this method is killing off the process properly or not...
@PauloCereda just had cause to kill it mid-compilation, seems to have not left anything hanging around :-)

5:15 PM
@Johannes_B: First son? Do you know something we don't know just now? ;-)

@ChristianHupfer No, nothing in sight right now.

5:45 PM

Howdy, perhaps an odd request, perhaps not: what's my best bet for getting an unbreakable box in a subfigure environment? The purpose is to allow a graphic to be larger in height than would usually be allowed without breaking the page (since a figure would appear on the next page if the height of its contents were too large to be fit on the first page).
I've tried embedding both the contents of the subfigure as well as the subfigure altogether in a minipage. Didn't change anything. :-(

6:05 PM
@PauloCereda July 20-24.

@GonzaloMedina -- but if @paulo attends expolatex in peru, he can't attend tug 2015 in darmstadt. they're scheduled for the same time, so to attend both would violate the laws of physics. paulo may have extraordinary powers, but i don't think that's one of them.

@barbarabeeton Ah, it's a shame.

@Gonzalo, @barbara: Oh no! What should I do?

@barbarabeeton why did you remove all that cursing ^^ :-)

@GonzaloMedina -- not cursing, actually. was just saying that the dates for both conferences are shown on the tug.org home page. but i directed it to the wrong person, and i don't like to confuse people unintentionally. (intentionally is a different matter entirely.)

6:12 PM
@Johannes_B Done

@Gonzalo: it sounds like a very good alternative if I could not make to Darmstadt, but I wonder if I could actually fully enjoy it since I don't speak Spanish. I believe I can understand 30% of what people say. :(

@PauloCereda Since 70% of the speeches will be (La)TeX technicalities, you will understand everything!

@GonzaloMedina -- and since you'll be attending, i think i'll put a note in my tugboat editor's nag folder to hold you up for a conference report.

@GonzaloMedina I'm more worried. :)

@barbarabeeton I was confused :-p But now, everything is clear. :-)
@egreg +1 :-)

6:16 PM
@Johannes_B -- gee, i'll have to try harder next time! (glad it worked out.)

@barbarabeeton My pleasure! What exactly is expected from a conference report?

@barbarabeeton I am confused most of the day anyway, no need to worry :-)

@barbarabeeton Yay news from South America! Can we have a column in the TUGboat named News from a duck where we report everything TeX-related in South America? :)
This week, we will have a pie eating contest! It's strictly forbidden to use \expandafter's. :)

@PauloCereda That sounds interesting. I have a feeling that LaTeX is experiencing a boom here in SouthAmerica, I have some proposals to make to the hosts of ExpoLaTeX and I'm sure they will also have many more.

@GonzaloMedina Cool! I want to have TUG in here in 2017. :)
I can even ask IPTV from USP to broadcast the event live. :)

6:21 PM
@GonzaloMedina -- conference report: name and topic(s), highlights of interesting presentations, who was there, "reportable" extracurricular activities (or, why would this be a nice place to visit, even if there is no tex conference going on). if you feel so inclined, you can read some of the reports from previous conferences in tugboat. all are online in the proceedings issues, and all but the most recent are open to the world.
@PauloCereda -- no \expandafters could be a disadvantage. too much pie, and one probably needs to loosen one's belt. (i suppose applying "rubber lengths" might be a possibility.)

@barbarabeeton oooooh! :)

@barbarabeeton I can't help myself, but omitting \expandafter doesn't mean omitting \output
oh what I did now :D

why hasn't anyone written a nice language to program with rather than latex, but still use the same core tex algorithms? I mean, tex algorithms can be implemented in any language, right? so can't someone come up with new easy to use language, but still use the same engine algorithms?

@Nasser You could take the chance.

@PauloCereda I am not an expert in this area.

6:30 PM
@barbarabeeton I'll read them tonight at home.

@Nasser -- other front ends have been created -- eplain and lollipop for two. but users always want more. so the pressure never ends.

@Nasser it was created, it's named LuaTeX

@michal.h21 I have not seen any lua language related questions here. I wonder if it is used in programming latex? I understand one can use lua to make a package with, and embed code in the latex document using lua, but one has to use lualatex to compile it with. I looked at lua and it looks like a nice language. May be tikz should have been written in lua then? I find tikz code hard to read, because it is written in latex syntax.

6:49 PM
@Nasser some lua related questions are here, see tex.stackexchange.com/questions/tagged/luatex you can use lua to output LaTeX code, but more interesting is ability to work with TeX nodes. you can completely bypass TeX with that. see speedata.github.io/publisher for direct conversion from xml to pdf, using TeX algorithms, but without TeX code. btw. some parts of tikz are coded in Lua, see the graph drawing library

@michal.h21 here is log of tex4ht building on my pc now. It takes about 2 seconds per one svg image to generate.

System call: dvisvgm -n -p 2830 -c 1.2,1.2 -s report.idv > re2402x.svg
processing page 2830
page size: 5495.3pt x 47.4805pt (1931.38mm x 16.6875mm)
page written to <stdout>
1 of 6975 pages converted in 1.3268 seconds
System return: 0
System call: dvisvgm -n -p 2831 -c 1.2,1.2 -s report.idv > re2403x.svg
processing page 2831
page size: 1169.87pt x 37.5007pt (411.162mm x 13.18mm)
or between 1 and 2 seconds per image. This is for one formula.
I wonder if dvisvgm is just slow, or it is just the way it is supposed to be.

@Nasser maybe the size of the dvi file play some role in this as well. maybe gnu parallel could help if you have multicore processor? or to use different options for dvisgm and output all the files from one dvisvgm call?

@michal.h21 I do not know, the idv file is is about 4 to 10 MB depending.
@michal.h21 do not know anything about gnu parallel., I run this in Vbox actually. The VBox is on windows 7. I give the VBox 2 GB RAM. But it seems too slow for 1.5 second on average to generate one svg image. Given I have over 100,000 formulas or more
@michal.h21 are you saying, if I have a PC with multicore, and run Linux directly on it (no VBox) then this will run faster? i.e. each image generation can be done on separate cores?

7:06 PM
@Nasser I will try something
@Nasser in theory maybe: gnu.org/software/parallel

@michal.h21 fyi, I wrote the notes on using SVG for tex4ht, and configuring the .env file all here 12000.org/my_notes/faq/LATEX/htch3.htm#x5-7100025
@michal.h21 I see about the gnu paraller. As long as I do not have to change makefiles and such, and just use a new shell, I could handle this. WIll look at it., but I am sure I can't use this in VBox, and have to have separate PC for this. Since VBox can't reach cores ofcourse.

@Nasser you can use make4ht -u index.tex "nma,htm,pic-align,notoc*"  insetad of make4ht index.tex "nma,htm,pic-align,charset=utf-8,notoc*" " -cunihtf -utf8"

@michal.h21 oh, so I do not need " -cunihtf -utf8" at end? Ok, is this just short cut notation? I mean the "- u"? Or will it do something else? I can change my makefiles.
@michal.h21 I saw your docs here now. github.com/michal-h21/make4ht , it is -u,--utf8 For output documents in utf8 encoding thanks.

@Nasser yes it is shorthand for charset=utf8" " -cunihtf -utf8". and I also use usually -m draft to use just one LaTeX compilation when I don't use references and TOC

7:51 PM
Hey all. How would I accomplish the following; set the width of an image to the paperwidth + 3 mm on each side? width=\paperwidth+6mm doesn't work and just prints out '+6mm' in text.

@Nasser so basically you are doing dvisvgm -n -p %%2-c1.2,1.2 -s %%1 > %%3 to convert #1.dvi to #3.svg every time. It must be possible to replace that by a script that says if #3 exists do nothing else dvisvgm -n -p %%2-c1.2,1.2 -s %%1 > %%3 fi
@WalterB \dimexpr\paperwidth+6mm\relax (wider than the paper?)

@DavidCarlisle Thanks, will try that now! (I want to make my image a bit wider in order to have enough 'bleed' for when the printer cuts the edges of the paper)

@PauloCereda typical local bird as met this afternoon:

Would anybody mind testing the following and looking at the pdf. Is the parindent the same and my eyes are weird?
\documentclass{scrbook}
\newlength{\saveparindent}
\setlength{\saveparindent}{\parindent}
\showthe\parindent
\usepackage{classicthesis}
\showthe\parindent
\usepackage{blindtext}
\deffootnote{0em}{\saveparindent}{\thefootnotemark}%
\showthe\saveparindent
\begin{document}
\chapter{Der Ententeich}
\blindtext\marginpar{Duckweed is yummy}\par
\blindtext\footnote{\blindtext\par\blindtext}
\end{document}

@Johannes_B which paringdent is same as which? the one in the body same as one on the footnote?

8:04 PM
@DavidCarlisle exactly

@Johannes_B ^^

@DavidCarlisle Thank you, so not the same. I am a bit confused. Let me take a look at the source.

@Johannes_B footnote is indented 20pt body 10.95 pt
@Johannes_B see \showthe\parindent
!\blindtext\footnote...
> 10.95pt.
l.14 \showthe\parindent

@DavidCarlisle I see, but ... am i running out of juice? Do i need a break? A KitKat?
@DavidCarlisle Now i got it.

8:22 PM
@Johannes_B dunno I've been out all day and just skitting around. @wipet has just pointed out one of my answers is totally wrong (despite having been accepted:-) so I'd better fix that first:-)

@DavidCarlisle You are not right. \string\zzz X produces \zzzX and, of course \expandafter\string\csname\endcsname X produces \csname\endcsnameX. The bug correction in the last version of TeX is concerned to the \write or \message output without usage of \string. For example \message{\csname\endcsname X} gives \csname\endcsnameX in TeX version 3.1415926 (and previous versions) and it gives \csname\endcsname X in TeX version 3.14159265. — wipet 9 mins ago

@DavidCarlisle Found it :-)

@DavidCarlisle awwww <3
@DavidCarlisle zoo?
@JosephWright How do you like my Mary Poppins beamer template? :)

@PauloCereda well "wildlife park" but it comes to the same thing.

8:33 PM
@DavidCarlisle Yay! Did M enjoy?

@PauloCereda seems like it, he's been before although I've never actually been to the one we went to today cotswoldwildlifepark.co.uk

@Nasser try to run dvisvgm -n -c 1.2,1.2 -p 1- filename.idv

@DavidCarlisle Cool! :)

I asked for clarification, but once again, the edited post is just as unclear.
0

I want to make own paragraphs for algorithms by numbering them appropriately. I do not need any fancy things - white background, no borders and black font are ok. I am trying to write in Springer style which is different from styles of Cormen and Dasgupta in their textbooks. Code \documentcla...

8:49 PM
@Johannes_B indeed. Now that it is open to any publisher, the answer might as well be: use the code in your post.

@PaulGessler I'd upvote this as an answer ;-)

@Johannes_B oops, never mind; the code doesn't even compile... haha!

[13729][boettchj.Pearl: testing_tex]$ls -l masiAlgorithmVerbose.tex -rw-rw-r-- 1 boettchj boettchj 1067 Dez 1 08:18 masiAlgorithmVerbose.tex [13730][boettchj.Pearl: testing_tex]$ cat masiAlgorithmVerbose.tex
\documentclass{article}
\usepackage{mathtools}
\usepackage{listings}
\newcommand{\caphl}[1]{{\bfseries#1.}}
\begin{document}
\section{Under Pressure}
\subsection{Serial Computation}
\begin{algorithm}
\caption{\textbf{Baseline method}. This algorithm lorem. Pseudocode.}

@Johannes_B I'm lost...

@PaulGessler As you can see, not the first time i have spent time on this.

8:58 PM
@Johannes_B ah, I see, this is stuff on your machine
didn't realize it at first
@Johannes_B My constant thinking that you might be Johannes Braams can now be stopped, as well: now I know for sure that this is not the case... :-)

@PaulGessler No, i am really not.
:-D

@PaulGessler I am not a power user ;-)

@Johannes_B /s/power/template/
;-)
@Johannes_B voted

@PaulGessler :-)
5 hours ago, by Johannes_B
On Topic?

9:22 PM
@michal.h21 I will try dvisvgm -n -c 1.2,1.2 -p 1- filename.idv and see if it helps. thanks!
@DavidCarlisle you said: It must be possible to replace that by a script that says if #3 exists do nothing But how do I know that the image happens to represent the same math expression in the document? If I modify the latex document and move things around, now the image file names on disk no longer will match the order of equations in the document. These image files are all names like
parse_lg process file: re4298x.svg
parse_lg process file: re4299x.svg
parse_lg process file: re4300x.svg
parse_lg process file: re4301x.svg

@Nasser you must be able to tag them in some way, either the dvi fragment or tex fragment or whatever. If filename isn't enough use cmp to compare them so the logic is if re4298x.dvi has changed run dvisvg else dont

@DavidCarlisle I do not know how to do this, as this is handled by tex4ht internal coding. It runs the math conversion of each math expression, so all of this is something if possible has to be done by tex4ht? may be with an option? but I do not think it is something a user can implement. At least, I do not know how.

@Nasser you are specifying dvisvg in your config file that you linked to above so you can replace that by your own script. as a simple test replace it by a script that does nothing and see how long it takes to do the conversion to html while not making new svg, and just pick up your existing ones. If that isn't much faster it's not worth considering further, but if it is faster you should be able to make it that fast normally

9:38 PM
@Nasser I think this deserves it's own question, I have a idea for .mk4 file

@michal.h21 I think your command is missing the dvi file? This is what I have now in tex4ht.env

Gdvisvgm -n -p %%2 -c 1.2,1.2 -s %%> %%3

@Nasser see ^^ @michal.h21 has the advantage that he knows what he's talking about. I'm just asserting things must be possible based on general principles of computing:-)

I can't replace the above with your suggestion:
dvisvgm -n -c 1.2,1.2 -p 1- filename.idv
I get an error
System call: dvisvgm -n -c 1.2,1.2 -p 1- re32x.svg
DVI error: invalid DVI file
System return: 0

@Nasser don't add that to the env file, just run it on the command line

@michal.h21 oh, I see
ok, that works on its own. seems much faster !

9:43 PM
@DavidCarlisle all images are stored in one big dvi file and dvi processing tools are called on that file, one call for each page. this is not normally problem, but with dvi file containing 8000 images it seems to be little bit slower :)
@Nasser yes, because it doesn't have to load fonts and other things over and over :)

@michal.h21 But how do I incorporate this in my tex4ht build? do I need to remove the SVG entry I added from my tex4ht.env? What should I replace it with? Where do I add your command? to my makefile? Before calling tex4ht? I am confused

@michal.h21 You could tell the user not to be silly and have 8000 single page files not 1 8000 page one. I tried that before without success:-)

@Nasser now the resulting files must be renamed to the correct form

@michal.h21 you are right. ALl the new files now have different names:
-rwxrwxrwx 1 me me  9765 Feb 18 15:41 report-34.svg
-rwxrwxrwx 1 me me  8617 Feb 18 15:41 report-35.svg
-rwxrwxrwx 1 me me 21101 Feb 18 15:41 report-36.svg
-rwxrwxrwx 1 me me   228 Feb 18 15:41 report-37.svg
-rwxrwxrwx 1 me me 10807 Feb 18 15:41 report-38.svg
-rwxrwxrwx 1 me me 24271 Feb 18 15:41 report-39.svg
-rwxrwxrwx 1 me me 10809 Feb 18 15:41 report-40.svg
-rwxrwxrwx 1 me me  8144 Feb 18 15:41 report-41.svg
-rwxrwxrwx 1 me me   228 Feb 18 15:41 report-42.svg
@michal.h21 SHould I post question on this, and you can show there what exactly one needs to change to incorporate this improvment to tex4ht? I see it is faster, but do not see how to use it in the actual build.

@Nasser I would bypass the env file completely and do it all from the .mk4 build file. this is why you should post a question, so the solution could be find in the future (as oposed to the chat :)

9:48 PM
@michal.h21 Sure. Should I post the question asking "How to speed up svg image processing with tex4ht?"
rigth now, I have no idea how to use your method at all.

@Nasser sure :)

@michal.h21 ok, will post the question now, please give me about 20 minutes

10:17 PM
@michal.h21 fyi, I posted it . thanks for any help you can provide.
0

Currently, I tell tex4ht to use svg instead of png for math images. This results in better image quality. However, it is slow in image generation, as it takes about 1-2 seconds for each image to be generated since it has to load the idv file each time to make each image for each math expression. ...

Is it possible, that the following is very well answerable, even with the little snippet of code? Did three guys, including myself, miss that?
0

Why it appears on the bottom? I use a .cls file where \tableofcontents is defined as: \def\tableofcontents{% \@starttoc{toc}\contentsname } \def\listoffigures{\@starttoc{lof}\listfigurename} \def\listoftables{\@starttoc{lot}\listtablename} \AtBeginDocument{% \@for\@tempa:=-1,0,1,2,3...

Unclear, or off-topic as it could be resolved using part? tex.stackexchange.com/questions/185141/…

10:42 PM
@Johannes_B In the end it is only a guess, so I would still say unclear.

@TorbjørnT. vv
\def\tableofcontents{%
\@starttoc{toc}\contentsname
}
That explains the strange behaviour, there is never a sectioning command and the contentsname is placed after the toc-contents in output.

@TorbjørnT. Definitely unclear. Surely the class is based on amsbook, as @barbarabeeton says. Probably something like parskip or tocloft is loaded.

@Johannes_B That is the exact definition from amsbook.cls, so barbara's comment could also be the answer.

@PauloCereda Don't hold your breath :)

@egreg you missed one:
0

The space only appears where you added it: \documentclass{article} \setlength{\parindent}{0pt} \begin{filecontents*}{test.csv} One,Two,Three,Four a,b,c,d a,b,c, a,b,, a,,, \end{filecontents*} \usepackage{datatool} \DTLloaddb{test}{test.csv} \begin{document} \section{Undesired output} \fre...

10:53 PM
@TorbjørnT. @egreg Didn't know that. So, unclear it is. :-)

@DavidCarlisle Damn!

11:17 PM
@DavidCarlisle this is an odd error. The output shows longtable at the end. And removing either of the \end{longtable}'s causes an error to show up.
I mean, the output shows longtable at the end with your command added.
But it still errors out if either of the end{longtables}'s are omitted. Strange.
And why would it be so sensitive to such a small change in the input?

@FaheemMitha yes sorry it's not the kind of thing you can just spot the error, needs tracing, but I don't know datatool and it's kind of late to start looking tonight:-)
@FaheemMitha longtable, like its author, should be treated delicately

@DavidCarlisle But do you agree this is an actual bug, not user error?

@FaheemMitha no idea:-)

@DavidCarlisle Oh

Back from the Ash Wednesday mass! :)
@egreg memento homine quia pulvis es et in pulvis reverteris! :)

11:27 PM
@PauloCereda By the way, do you know that there's an earthquake swarm in the surroudings of Monghidoro? bolognatoday.it/cronaca/…

@egreg Oh my! Mum was nearby when I opened the news, she helped me with the text. I hope people are ok.

@PauloCereda I think so. Not very strong shakes, but very frequent.

@egreg I'd be very scared. :(

@PauloCereda Ask @GonzaloMedina for help. He's used to this, I guess.

@egreg Ah he once mentioned a bad time in his region. :(

11:51 PM
@michal.h21 thanks for the answer. Will try it. But I am confused about your build make file. with .mk4 and how to actually invoke it. I thought I check with you in chat might be easier. I saw your page that talks about this build file, but I do not know how to actually use it. I use gnu make now. I do not know what command I need to use to call .mk4 and where to put the .mk4 file. do I need separate .mk4 file for each latex file? how to call it, etc.... Very basic user questions. I've .....
read your page here github.com/michal-h21/make4ht many times, and still can't figure it out :)
I think you call .mk4 as the Build file.

I don't understand why there is so much activity here, and almost none on U&L. Do you people take special vitamins?

@FaheemMitha I take one multi vitamin bill each day and lots of coffee also.

@Nasser there are two ways: you can name the build file same as the tex file with .mk4 extension, or you can use -e` option to specify one with different name

@Nasser coffee is not so healthy. I assume you know about the new coffee site? No doubt a tea site will follow shortly.