LaTeX Warning: You have requested, on input line 26, version `2018/02/21' of package expl3, but only version `2017/12/16 L3 programming layer (loader) ' is available.
! Package xparse Error: Support package l3kernel too old.
@JosephWright Ok, I will revert to the previous versions then... It's not a big deal for now, as I don't rely on 'features' of your current updates... (whatever they might be ;-))
Originally, the algorithm is too long to fit in the frame. If I use the \scalebox{0.8}{}, it can fit in the frame but doesn't look like to be centered in the frame. Here is the code and output. First picture is for the original code, second is for the code with scalebox. Are there any methods to ...
@alhelal that answer is talking about putting a minipage in scalebox, which is fine and your comment says you tried to use a block which is I assume a vertical construct you can not put any vertical construct such as displayed equations or center environments or lists etc in a horizontal box like scalebox
@alhelal :you can reference that question and show how your attempt to addapt the answer failed, but adding comments on old answers is not the way to do it (and I, and others, wouldn't follow the external links)
@alhelal I have no idea what your problem is other than you said you put a block in a scalebox (which is unlikely to work if you can not put a block in a \mbox)
@DavidCarlisle Yes, you are right. But, as it is a algorithm it is useful to show in a frame. If I split this in many slide audience understand nothing.
@DavidCarlisle Yes, in my mentioned link they use minipage.
@alhelal I don't know in detail what "this" is but if you want to scale some vertical material (say two paragraphs, or a list or anything) then you need to put that material in a box and a minipage or parbox are the usual way, so irrespective of being a beamer block or an algorithm, a minipage in a scalebox (as in the answer that you link to above) is likely to be the expected way, but if you can use a font change instead such as \small that is often better than scaling
@UlrikeFischer The fact that people choose various active chars is a pain: it would (I think) be better if a larger set of chars were active 'out of the box', and the kernel docs said 'Only use these ones'
@UlrikeFischer Of course, won't help with a package that was last updated in 2006
@JosephWright Well I would prefer as less active ascii as possible. But the main problem is that this takes you sometimes by suprise. Perhaps the kernel could check for active chars at begin document and issue an error "BAD PRACTICE" if it finds some? (And offer commands to en-/disable the check if someone really wants to use a package which does this sort of things.)
@UlrikeFischer if for example & was active and let to its catcode-whatever version then the tikz matrix thing could change its definition without restriction on being nested, and as far as I can see most things would work as before
(and no, that's not Google Translate on its glory. I asked a friend to help me with a proper translation. If he used Google Translate, that's outsourcing!)
You should definitely make a concerted effort to remove the cruft and deadwood from the document's preamble. The preamble you posted very much conveys a Cargo Cult feel, characterized chiefly by a "ritual inclusion of code or program structures that serve no real purpose". Do figure out what you really need, and then remove all directives that serve no purpose and/or are actually detrimental. — Mico28 mins ago
Anyone getting an issue with recent `xparse`: ? LaTeX Warning: You have requested, on input line 26, version `2018/02/21' of package expl3, but only version `2017/12/16 L3 programming layer (loader) ' is available.
@Hugh There is tds.zip on ctan: mirrors.ctan.org/install/macros/latex/contrib/l3kernel.tds.zip. Unzip it somewhere outside texlive and then attach the root with tlmgr conf auxtrees add /path/to/root. you can remove it tomorrow with tlmgr conf auxtrees remove /path...
@ChristianHupfer: some OS'es might use different paths when issued with sudo, so tlmgr --version and sudo tlmgr --version in Fedora will yield different things (the first being vanilla TL in my path, while the second is the system version from my root path).
@marmot -- no. the ams headquarters office is built on a concrete slab, so there is no basement. (the computer room, constructed when such things were necessary, does have a "raised floor", which really means the slab is about a foot lower, and the entire floor is level to walk on in this wing.) but the unix development machine i work on suddenly became unavailable, soon after the announcement was made that all windows workstations had to be rebooted on account of microsoft updates. fie.
@marmot @TorbjørnT. @CarLaTeX is there any way to force an arrow tip specification such as -{Straight Barb[]} within tikzpicture, such that all lines with \draw will use this tip automatically?
@ChristianHupfer try \documentclass{article} \usepackage{tikz} \usetikzlibrary{positioning} \begin{document} $a\to b$ \tikz[baseline=(a.base)]{\node(a){$a$};\node[right=10pt of a](b){$b$}; \draw[->](a)--(b);} \end{document} I think that the Ti*k*Z standard arrow is as ugly as the LaTeX arrow.
@ChristianHupfer But of course one may wonder why -latex does not produce the LaTeX \to. Probably it would have been better to switch the roles of these two.
@marmot Well, my first idea was finallymakethetikzstandarrowtipslooknice, but duckify is shorter and does reveal the purpose of the package immediately ;-)
@DavidCarlisle if some dumb package reserves two write registers it doesn't use is it ok to reset the allocation counter like this or there something better?
@egreg The commands are directly below the offending package (and it is not meant for a general solution, only for one document which fails due to "no room for a write"). I don't mind if I get errors if I was wrong and the package needs the writes, I only don't want some silent failures somewhere.
@egreg Oh, \mymessages is not my stream. It is a stream of the package which looks as if it is only meant for log messages. So I'm only trying not to kill (even if it looks unused) but to redirect it to the log-stream.
@egreg Somewhere in the code it (arabtex) makes an "openout" to a file for this stream. It wasn't used in my document but I wanted to force a real error just in case. The second is an example from xstring which contains "\newwrite\@xs@message" for as far as I can no real reason.
@DavidCarlisle Would you want to convert a 400 page document in latin1, with greek, arabic and hebrew text and four indices to luatex? ;-) I'm grateful I managed to remove morewrites without breaking it.
Interesting that not everything on this site is translated developer.mozilla.org/kab/docs/Web/SVG (just did a search for "svg" and it decided I needed to see it in ... whatever language that is)
@marmot: Given my knowledge about TikZ I suspect the release and uploading of duckify to CTAN in about 20 years, which might be yet before L3 is published ;-)
Hi! I have a localized issue, so I am guessing it's not worth asking a question. I did a tlmgr update --all and now my xparse.sty is asking for a expl3.sty that's newer than the one installed by tlmgr update l3kernel.
After the latest TeXLive updates I keep getting these errors:
LaTeX Warning: You have requested, on input line 26, version
`2018/02/21' of package expl3,
but only version
`2017/12/16 L3 programming layer (loader) '
is available.
! Package xparse Erro...
@MatthewLeingang See @HaraldHanche-Olsen's message. To find the revision number, issue the same command without the revision part, it will list all revisions found.
@HaraldHanche-Olsen It's not fixed until the correct CTAN version is fetched from TL -- for some reasons only the l3packages made it to TL, the kernel packages were not updated to TL