Is there a way to remove a chapter title entirely? I'm adding the title page to the table of contents via \chapter because the title page is supposed to be page i in the toc, and I removed the page number via \pagestyle{empty} but I can't figure out how to hid the text created by \chapter{Abstract Title} on the page itself
@StefanKottwitz @JosephWright This question contains the answer, so cannot really be answered. Dragging out to a CW answer? tex.stackexchange.com/questions/417378/…
@JosephWright looking at sourceforge.net/p/xetex/code/ci/… is interesting (a) it's a change as recently as last month and (b) it's back porting a change already applied in texlive.... Perhaps the thing to do is make a git pull/merge request directly to sourceforge once you are ready?
@DavidCarlisle However, I'm still battling with the primitives that need C functions: the RNG is OK (although that got messed up by trying to reorder other parts of the WEB), but everything else is still not working. I'm trying yet another different tack today ...
@JosephWright yes I can't do anything before then anyway. If you run make in Work/texk/web2c then it just compiles the files you have changed and skips configure and skips tests so a few seconds rather than a few tens of minutes on my machine
@egreg with latin as main language I get either an error (' in text) or a loop (' in math) depending on when unicode-math is loaded. Imho the shorthand definition is wrong but I don't know if there is better version. If you want to look I added a few more examples at github.com/wspr/unicode-math/issues/462.
@DavidCarlisle I don't think it's Unicode as the changes I'm struggling with are simply moving lines and/or stuff to do with the timer code (which is simple)
@DavidCarlisle I see that enabling the C code seems to work, so the issue is likely at the WEB end
@skull Well: (1) No vaccination exists. (2) Preventive antimalarials cause you similar symptoms as the disease itself. (3) I won't stay in the highly affected regions. (4) I stay in Brazil only for 3 weeks. Conclusion: I won't do anything beforehand, it I catch it, I'll try to get over it until I'm back home.
@StefanKottwitz -- is there any likelihood that texdoc.net will be reinstated (updated)? (i've been recommending it to people in my function as ams technical support.)
4
@UlrikeFischer -- i have to backtrack on what i said about using the lightface roman math font (\symup) only for single-letter symbols. unicode tech report #25 says clearly (p.8) that lightface roman is for "operator names (sin, cos, etc.), some constants, units". so i stand corrected. (there won't be any ligatures though, at least not with direct unicodes, nor precomposed accented forms (p.9).)
@DavidCarlisle Building locally handy (though I have to work out where the files are stored so I can edit them outside of nano, and set up Git commits from the subsystem)
@barbarabeeton but I think tr25 is wrong:-) Or at least it's wrong given the current unicode math fonts (TR25 doesn't say the math fonts should not have ligatures, I think?) (I think it would be a lot better if they did, and the bold fonts)
@JosephWright yes but you will trash the file permissions and it's really not recommended:-) (I used the ubuntu git to check out a second copy of texlve-source and ubuntu emacs to edit in there (you can never have too many emacs installed:-)
@JosephWright it is file saving that's the compatibility issue...
@JosephWright I once read the description of what cygwin has to do to make file permissions in its posix shell and file permissions for windows line up. It's not the sort of thing you want to read twice (and it warns at the top that you never want to read it at all:-)
@JosephWright it may be but I wouldn't rely on it:-) The cygwin page was scary stuff and the microsoft guidelines were pretty explicit not to mix (and in the first release they prevented you from starting any native windows app from the subshell)
@DavidCarlisle Locally if you like, but Travis-CI will fail if you check them in, until TL gets an update in a day or so (we could check-in a fixed script, but I'd rather not)
@DavidCarlisle Ah: better than nothing I guess (I meant it's a shame that the subsystem can't access 'shared' data on the parent, most obviously the ssh agent)
@DavidCarlisle -- tr25 may not say explicitly that the math fonts "should not" have ligatures, but they will never be given unicodes. (it does say explicitly that there will not be precomposed accented forms, as already mentioned. so no slashed o nor a-ring.)
@DavidCarlisle -- i know that. with otf fonts, there are other ways of identifying such glyphs, but with type 1 fonts, if a unicode id is used, that forces them into the pua, which is a big mess.
@JosephWright I don't know if it's possible to assign unix process id's to running windows applications, if the process identified by SSH_AGENT_PID says the ssh key is good then it's good, it may be possible to hook that to pageant but I somehow doubt it. (I haven't used putty for years:-)
@barbarabeeton but type 1 fonts don't really apply to tr25 (or unicode-math typesetting) do they?
@DavidCarlisle -- probably not, but some of us still have to live with them. (the stix 2 fonts use unicode ids, and they're being re-cast for use with pdftex.)
@barbarabeeton it's defined by unicode-math (oh unless the stix2 package defines a version?) (which seems not to be the case)
@barbarabeeton there isn't any way to make the distinction between \symbf and \mathbf in pdftex. \mathbf switches fonts, \symbf uses the same font but changes character codes up to the math alphabets. In pdftex changing font is the only possibility
@barbarabeeton well from a technical point \symup is perhaps not the best options as it doesn't handle accented chars. And imho e.g. spanish document like to use an i with accent in \lim. But I still think that math should be typeset with the math font. And if ligatures are wanted they should be added to the math font. (I'm not sure that I would want it but e.g. with latin modern math I would adjust the kerning a bit).
@DavidCarlisle -- i think it was probably the intention to equate the names, so that input could be shifted easily between xetex and pdftex. that would make sense (to me at least), and could account for my confusion.
@barbarabeeton I'm not sure that would be good, it isn't so easy to document how they differ in unicode-math and if they are identical in pdftex I am not sure if that helps or hinders, I suppose it could be OK
@UlrikeFischer -- you're certainly correct that spanish or swedish (or any number of other european languages) could want operator names with accented letters. that simply wasn't considered. i'll put it on my list to ask murray sargent and asmus freytag. (while looking at tr25 this morning, i found a number of "infelicities" that need to be corrected.)
@DavidCarlisle -- while ams is intending to switch to xetex for production, it can't happen until a reliable and foolproof method is found for automated graphics checking that can be applied to full-document pdf files. until then, dvi is needed.
@DavidCarlisle -- well, i didn't know that existed. i will query the people who are actually working on it. (and what horse did you think was the "right" one. luatex is not yet stable enough to consider, when you take into consideration that some books are in production for several years.)
@barbarabeeton yes but for a several year project you want software that's still there at the end, and I have my doubts.
@barbarabeeton luatex is an extended pdftex, but xetex is an extended tex and makes dvi (called xdv) output that gets made into pdf by using the (x)dvipdfmx dvi driver
@barbarabeeton as I said to some in an answer the other day, given that the underlying engine (in luatex and xetex) is different, it isn't a matter of them breaking it, the default position would be that it doesn't work at all and the amount to which it mostly works is testament to the effort they have put in to compatibility.
@barbarabeeton thinking about that, the statement is really odd:-) a dvi file doesn't contain graphics, so any checking is presumably on the separately supplied eps files? you could still use pdflatex but require external images in eps format.
@DavidCarlisle -- we do use pdflatex (in dvi mode) and the images are in eps format, but the `\includegraphcs' calls often involve scaling, so the checking (done by a separate program) takes that into effect, looking at font size, line thickness, and other elements in context to determine whether the resulting image will hold up when printed. i'm not positive, but i think this check is guided by the specials that are written into the dvi file.
@DavidCarlisle -- you're going to get latex to determine the thickness of a line in an eps file produced by mathematica and scaled by \includegraphics so that the graphic is .85\textwidth ?
@barbarabeeton but as the dvi only contains only a reference to the graphic the best you can get from it is the name of the graphic and the scaling factor. Why shouldn't latex be able to write this to a file?
@UlrikeFischer -- the graphics-checking routine looks at both the scaling and the content of the eps file, and lists the various elements checked. a skull-and-crossbones icon is placed next to any element that needs to be inspected more carefully. any graphics report that is free of such icons needs no more attention. this can save hours on a book with several hundred graphics. (cont'd)
(cont'd) if only the graphic name and scaling factor are given, the human needs to do calculations and visual inspection. human time is expensive; computer time is relatively cheap.
@barbarabeeton no I mean the only information that the dvi \special will give you other than the filename is the x and y scale factors used, and it gets those from latex, so even if you were using luatex or pdftex and had no dvi file you could get latex to write out the infomation to a text file as well as putting it directly in the pdf
@UlrikeFischer oh I just go home answered @barbarabeeton's ping then saw you had said same while I was driving:-)
@barbarabeeton so all it needs is the scaling factors 9which you get from the dvi but latex could write to a file) and the external graphics files which you have in anycase.
@DavidCarlisle -- i didn't design the graphics checking routine. the person who did, designed it as he thought best. (i'm clearly not a graphics guru, and have been discouraged from sticking my nose in.)
@barbarabeeton yes understood, but anyway I could imagine that such a routine means you have restrictions on graphics formats (eg eps not pdf) but it's not really usable as a reason for restricting which tex engine is in use.
@DavidCarlisle -- i'm afraid that's decided at a higher pay grade than mine. (but really, there are still too many uncertainties about luatex to adopt it for production without serious testing. that'll be done, i'm sure, but at the moment, there's more emphasis on implementing electronic presentation as something other than fixed pdf, than on getting things that work more "up to date".)
@AlanMunn Oh, I didn't know that. I thought Mt. Washington is near Seattle. But now I googled and yes, you're right. Where can I read about the cafe? But there is even hut named about us...
@barbarabeeton yes of course "we don't have the resources to test other tex engines in production" is a perfectly valid reason for not supporting anything you don't support. What i mean is that that is better than a more technical sounding reason such as "we need dvi to check image inclusion" which fixing is probably more a minor irritant rather than a technical problem if there was actually an intention to use another engine.
@marmot It's at the bottom of the page I linked to (the image was from that page). The American Mt Washington is on the other side of the country, in New Hampshire.
@AlanMunn Oh, I see. (So far I only hiked mountains in Europe, Nepal, California and Colorado...) BTW, I am reading "House prepared pizza by the slice, hot dogs, soft drinks, snacks & affordable combos." so that must be fake marmots.
@marmot British Columbia is beautiful. I haven't done much hiking there though, most of my hiking has been on the East coast on the Appalachian Trail or the Long Trail.
@AlanMunn In some cases at least it's a question seeing how far one can push ones limits, which is an interesting challenge for many. But to each his own.
@AlanMunn I also agree with @TorbjørnT. and do not think these people are (necessarily) crazy. Of course, you could say "what's the point of running a marathon" but one can really enjoy it and I cannot see anything wrong with it, nor with ultra hikes. (Not that I am doing these things these days.)
@UlrikeFischer what the... @JosephWright I guess you have a windows texlive as well... is there any pattern why ü fails and other non ascii letters wok?
@DavidCarlisle -- i've just discovered a most delightful reason why align* shouldn't be used for one-line displays. adding \qedhere causes the box to overprint whatever is at the right-hand end of the line. changing this to equation* (of course) fixes it. i've added an answer to align vs equation
@UlrikeFischer could you send me the the log file (so I can check what bytes are in the error message (I don't trust the display as filtered through mail systems)
@DavidCarlisle Yes but I thought that pdflatex would read the file name before. Btw: the pattern is not that ü musst be the first, begräxxüung fails too.
@UlrikeFischer the switch happens in \everyjob so as soon as it sees a character it thinks isn't a filename letter, which for some reason is here. But the texput.log you sent shows no error? just **\stop didn't the file not found error get logged?
@UlrikeFischer what happens with pdflatex \input\detokenize{begräxxüung } or whatever quoting the windows shell needs for that form?
@UlrikeFischer yes the error goes to the terminal only as the log file not open yet:( I may reply on list with detokenize workaround just so the user has something.
@UlrikeFischer really., no chinese filenames? all non ascii bytes should be catcode 12 until the filename is read so it really shouldn't matter, I have a horrible feeling that the 8+3 doss names BEGRN~1 or whatever are getting invoked and so ~ is being expanded. But it makes no sense that I can work out.
@DavidCarlisle If I should guess I would say the difference between ä and ü is that ä is binary 1110 0100 while ü (and ú that breaks too) start with more 1: 1111 1100
@DavidCarlisle now you know why I'm a hardcode ascii-file-name user. One can spent hours trying to figure out how the damn chars are encoded in the file system and during the various compilations steps ;-)
@UlrikeFischer and I lied above. I was thinking that the non ascii characters were catcode 12 and I flipped them to decode utf8 in \everyob but I don't: they are decoding utf8 but for example \long\def\UTFviii@two@octets#1#2{\string#1\string#2} so in \everyjob I only need to flip those four macros to the real "non \string" versions. But that means if the filename is not valid utf8 then bang....
@UlrikeFischer I guess pdflatex \UseRawInputEncoding \input begrüßung.tex works as well?
\usepackage{fancyhdr} % Required for header and footer configuration
\pagestyle{fancy} \renewcommand{\chaptermark}[1]{\markboth{\sffamily\normalsize\bfseries\chaptername\ \thechapter.\ #1}{}} % Chapter text font settings \renewcommand{\sectionmark}[1]{\markright{\sffamily\normalsize\thesection\hspace{5pt}#1}{}} % Section text font settings
How do I decrease in the code that I have attached the font of the part indicated with the red rectangle and add a copyright on each page of the book?
@marmot Hi, do you know Legrand orange book?
@DavidCarlisle Hi, Sir David. But Christian Hupfer where is? When I write @.... I not see your name
NTFS stores filenames in UTF16, however fopen is using ANSI (not utf8).
In order to use an UTF16-encoded file name you will need to use the Unicode versions of the file open calls. Do this by defineing UNICODE and _UNICODE in your project. Then use the CreateFile call or the wfopen call.
if miktex is working I guess it uses the unicode file api....
@marmot lucky you are happy, I'm glad. :-) I can not decrease the font size in the header (red rectangle in the image) and add a number of your choice with the \ copyright command
@Sebastiano you can only ping people who have been here in last few days, if @ does not prompt the name you can not ping them) but you should know that code questions are better with a real example and asked as a question
@DavidCarlisle Then I will ask a question on the site. Surely there will be someone who will vote negative my question :-) I have made many attempts but I have not succeeded.
@DavidCarlisle yes. Btw I found an old bug report (from me ...) where the fact that the ü is more problematic than ä appeared too: sourceforge.net/p/miktex/bugs/2068
Hi, I want to make my own custom note template which is somewhere in between article and book.
(1) I want the book-style organizations with TOC and such but individual page style will follow mainly that of article. (2) I want it to support Unicode. (3) I am going to define custom environments for each kind of text piece in my note (theorem, proof, definition, example, etc.) which I would be able to reuse. (4) I want to separate each chapter or each section of the chapter into different latex files if possible while keeping the the chapter relations between those files. But, I want to put t…
@Sebastiano Sorry, I have not played with this at all. An orange book on the foundations of special relativity? ;-) (You could of course be very courageous and do \renewcommnd{\textcopyright}{\empty} or something like that in the preamble, but I would not do that for many reasons.
@marmot I'm writing on book together with a full professor of Sperimental Physic on "lezioni di fondamenti di fisica moderna". Meanwhile, I'm writing it alone and there are many reasons for discouragement. After two months my wish to write it came back. When I have finished my prof. he will correct it. For the figures I use Adobe illustrator.
@Sebastiano I see. I was using pstricks and TikZ for all the Minkowski diagrams and so on. I find this more fun, and also the labels have the same font as in the ambient document.
@Penguinlay This is pretty standard LaTeX stuff. For (1) book or report + titlesec would be fine, or one of the general classes (KOMA (scrbook etc) or memoir). I'm partial to memoir because the English documentation is much better than KOMA. For (2) use either XeLaTeX or LuaLaTeX. For (3) amstheorem is pretty standard. (4) \include and \includeonly does this. For (5) make your own package, or if the documents are all the same type, your own class.
How can I decrease the font of the part indicated with the red rectangle and to add a copyright with a number on each page of the book for the structure.tex of the Legrand Orange Book?
I believe that to decrease the character and to be able to add the \copyright symbol you can do in the code tha...
@Sebastiano Surely you should know by now that most questions require a MWE (a compilable, complete document for people to play with.) You've posted just a fragment, and much of that is probably not actually relevant to the question.
I was posting some question about this strange thing LaTeX is doing when I try to compile my thesis. Someone asked me to provide a minimal example that reproduces the problem. My thesis is now a few hundred pages long and spans along ten different source files! How am I supposed to know what or w...
\renewcommand{\chaptermark}[1]{\markboth{\sffamily\normalsize\bfseries\hspace{20pt}\chaptername\
\thechapter.\ #1\hfill \textcopyright Sebastiano\hfill}{}} % Chapter text font settings
\renewcommand{\sectionmark}[1]{\markright{\sffamily\small\thesection\hspace{5pt}#1\hfill
\textcopyright Sebastiano\hfill}{}} % Section text font settings
@Sebastiano And I did not downvote you .... (I only downvoted 4 times, the two of the downvotes I forgot the reason and for the other two I remember all too well... ;-)
@egreg "collective responsibility" mean anything in Italian?
@egreg I have three cygwin texlives and an ubuntu one on this windows box and none of them suffer this problem, why do people insist on using windows programs in windows?
@UlrikeFischer boo well I guess tomorrow i'll get a version of that latex.ltx into the github sources and get some more testing then push out a pl5 at the weekend. (@JosephWright)
@UlrikeFischer thanks for the help (no thanks to @egreg of course:-)
@marmot not responsible for causing it. but jointly responsible for cleaning up
@Sebastiano not in any way suggested by the link I gave about MWE though. there is no usable code there. I don't downvote, but if you get some downvotes don't complain here.
@Sebastiano you ask what is wrong with your questions, Alan and I both suggested not having a MWE was a major issue, and now the question has a code section that is 489 lines long (not minimal) has no \documentclass or \end{document} so not working and does not produce the output shown in your image (so not an example) so it is not a MWE it fails to meet all three parts of that acronym.
NTFS stores filenames in UTF16, however fopen is using ANSI (not utf8).
In order to use an UTF16-encoded file name you will need to use the Unicode versions of the file open calls. Do this by defineing UNICODE and _UNICODE in your project. Then use the CreateFile call or the wfopen call.
