@AlanMunn I recall now why I was using epstool instead of pdfcrop. I want to use a bounding box from a different EPS occasionally. Now I have to extract the bounding box using something like sed and pass that to pdfcrop since it takes a --bbox.
A quick search on the site suggested that it's not possible but I'd still like to ask: can the breqn package be switched off for the whole document except for where it's actually used?
I have this one equation I do not want to line wrap by hand.
But I don't want to buy this convenience in one single equation with the inconvenience of coping with all the strange error messages that breqn produces elsewhere.
Can I have both? And a pony? Or will I just have to face the hard truth that I'll have to hand wrap this equation?
Is there maybe an alternative?
some script that I put my equation into and it magically produces a beautified, line-wrapped and indented equation as LaTeX snippet?
@Qrrbrbirlbel Ok thanks, I'm going to ... but this was the very first equation in that chapter and it already failed ... that discouraged me somewhat from using the package :/
! Argument of \select@group has an extra }.
<inserted text>
\par
l.155 ...frac{y\maxsub}{\sqrt{S s_\mathit{max}^2}}
Just compiled without \includeonly and I got an earlier error
@Qrrbrbirlbel Your fix for the other error works though :)
But the apparent frequency of errors still conjures up this image of me sitting through the rest of the night fixing strange errors throughout my 300 page thesis that weren't there without breqn.
@Qrrbrbirlbel Well, even when compiling just that one chapter which works now, I get Package breqn Error: eq@repack penalty neq 0,1,2,3 for the equation in question. So it doesn't seem to be worth it :/
Seems that beast is too tough for Skynet and good old human brain is needed :/
As an experiment, I took some equations that I'd manually split using aligned like so:
\begin{equation}\label{eqn:fourierSeries:30}
\begin{aligned}
\int_a^b \phi(x) e^{-i \omega m x} dx
&= \sum c_k \int_a^b e^{i \omega (k -m) x} dx \\
&= c_m (b - a) + \sum_{k \ne m} \frac{e^{i \omega(k-m) b} - e...
The breqn package changes the definition of \sum (and of almost all other commands for math mode). The definition it uses is (using the output of \show)
> \sum=\protected\long macro:
->\@sym \sum \math_sym_COs:Nn \mg@cop {50}.
The macro \@sym is the same as \@gobble, at least between \[ and \]...
With no warranty of any kind!
\documentclass{article}
\usepackage{color}
\makeatletter
\def\colorizemath #1#2{%
\expandafter\mathchardef\csname orig:math:#1\endcsname\mathcode`#1
\mathcode`#1="8000
\toks@\expandafter{\csname orig:math:#1\endcsname}%
\begingroup
\lccode`~...
Using a variation of the technique Dissecting paragraphs with \lastbox
described in Section 5.9.6 of TeX by Topic, you can produce a blur effect; the idea is to use two copies for each line and typeset the line and its copies superimposed:
\documentclass{article}
\usepackage{lipsum}
\newbox\fli...
pdfcrop allows for a --bbox <left> <bottom> <right> <top> key. I'd like to extract the %%BoundingBox from one EPS file and use/port that as the --bbox key to pdfcrop.
Here is a simple example of some issues with latex->HTML conversions. I write latex document using \documentclass[twocolumn] When converted to HTML, it comes out as one column. So, to make it come out as if it was 2 columns in HTML, I have to write everything in a table with 2 columns. But I do not want to use table, I want page with 2 columns in the sense of \documentclass[twocolumn] So I have to make 2 version of the same latex document, one for HTML and for pdf
and when I change something in one, I have to remember to make the same changes in the second.
@Nasser do you realize that if your document is longer, your readers will have to scroll several screens up after reading first column? I don't think many of them will appreciate that :)
@Nasser but if you really want it, try something like:
@michal.h21 I have suffered with bad png quality for math for long time. I just found by trial and error a fix which improved the math png 100% I am still stunned by it
I can;t believe it
I changed this line
<dvipng>
G.png
Gdvipng -T tight -x 1400 -D 72 -bg Transparent -pp %%2:%%2 %%1 -o %%3
I do not know the details. I just know on my computer, the HTML pages now look so much better. I mean the math. I can make you a screen shot if you like to see.
I am now rebuilding my whole web site with this new setting
I just saw you have new texlive, can you test please if make4ht is running correctly? there is new version of luatex and it has some non backward compatible changes
@michal.h21 Yes, but texlive 2013 is still not on the mirror site. I just tried it. It still says 2012. May be tommorow it will be.
I will test everything when it is there. I have new Vbox machine just for this.
@michal.h21 You might like to consider updating your answer here tex.stackexchange.com/questions/43772/… and mention the possibility of trying 96 vs. 72 in your answer there. Since someone looking for help on this issue, will see this. That is what I saw also...
@Nasser you might also consider use of svg as is described at the end of the answer, maybe it can work for you. as svg is vector format it doesn't have problems with bad rendering
@michal.h21 OK. But i am now VERY HAPPY with texht and png images for math. with 96 dpi, it is very good. no need for mathjax or anything else. This was the main problem for me all along. Now it is solved :)
@michal.h21 YEs, I know, that is how I found about 96. I was reading that man page :) I was actually trying to see if I can change the bit-dept. It is now 4. I was trying to see if I can change it to 24 hoping that might improve things, when I noticed the dpi number, then said let me try that first
But your post about using dvipng instead of the convert was very useful.
$ sutlmgr update --all
[sudo] password for knsam:
tlmgr: The TeX Live versions supported by the repository
(2013--2013)
do not include the version of the local installation
(2012).
@Nasser you can increase the inout stack in texmf.cnf (search fro that error message on this site) but if you hit 5000 ypu may be in an infinite loop so making it 50000 just means it takes longer to crash
I have no idea why it happens only on this one file really. This file uses \usepackage{breqn} I wonder if this has anyting to do with it. I only include 4 images and then it dies. \includegraphics{prob1_parta.png}
@DavidCarlisle I always like to be on the front line !
if I can't figure this out, I'll remove breqn. The problem is I change all my equations from align* to dmath* and there are many of them. I did not keep a copy, so I have to redo them all. Not a problem. I still do not know if this is what causing this file to fail with tex exceeded capacity, but that is the only file that has this package. The PDF is fine, just the HTML is not coming up complete.
@Nasser what you ought to do id lose brekn and define dmath to just be [ in htl mode and then let mathjax or the mathml renderer to the line breaking later as it makes no sense to linebreak as an image in html to an unkown page size. But I may have mentioned that before. (And the implementation of line breaking in the browsers isn't really great yet)
@DavidCarlisle FOUND IT !! It was breqn package that was causing tex exceeded capacity with tex4ht !! I just removed the package, changed everything back from dmath to align and now the problem went away. No tex exceeded capacity on same file. That is all what I changed.
JUst a quuck question please In this:
%tex.stackexchange.com/questions/8434/…
%\DeclareMathSizes {t-size} {mt-size} {s-size} {ss-size}
%Basically, we instruct the use of a <mt-size> as the math
%text size, <s-size> as the script size and <ss-size> the
%scriptscript size to be used in math, when <t-size> is the
%current text size. You can adjust the parameters to suit.
I found that script size means superscript size, ie x^this
But what is ss-size used for ?
and if the the correct values to use for 12pt document please?
I am using this now \DeclareMathSizes{12}{11}{7}{7}
Ok, I found it. ss-size is for subscript x_this and s-size used for superscipt x^this
@Nasser that is an odd choice you are making the normal math and subscripts smaller than the default (and the surrounding text) but making sub-sub scripts large
@DavidCarlisle but {8}{6} there are hardcoded?? What if one uses 12pt vs. 10pt? Should these not be propertional? This is for HTML only. Not PDF. With PDF I do not touch this stuff. I just use 12pt for class, that is all. But for HTML I have to change it since math comes out too large or too small if I do not try to fix it.
@Nasser It's an image and baselines are all wrong, size doesn't match, it's the expected behaviour if you use images for math, honestly I really wouldn't bother fine tuning the images.
@DavidCarlisle I agree with you. But I do not want to use mathjax for now. (1) it takes long time to load, and also if I am connected to the net, it will switch to mathml I think or some other math. This way at least, it is the same all the time, and pages load up fast on my server.
i.e when I open my web page, it loads up fast. With mathjax, it will takes longer.
but will keep eye on mathjax, and keep trying. WHen it gets faster, I can always switch. I use makefiles, and easy to change.
@Nasser That actually isn't clear at all, if you have a lot of math (as we do) the size of the png files and the number of connections required to download them all can be comparable or slower than mathjax (if suitably configured)
@DavidCarlisle That is what I first thought. But in practice, the pages actualy load up faster. With Mathjax, the page loads, but I can see at the bottom that Mathjax is still working and sometimes it took a whole half a minute for the math to show. With .png, it is almost instantanous . May be the internet is getting good.
@Nasser actually if you turn off that mathjax message box the perceived time is a lot less. Browsers always incrementally render so you can look at the top of the page before it has finished laying out the rest, but mathjax tells you it is doing it. And no I don't want 300000 png files in my documentation distribution.
@Nasser or you can use jqmath (quicker than mathjax not as pretty) or you can just use mathml and tell people to use a browser that conforms to standards (firefox and safari and just about internet explorer+mathplayer then you get direct mathml layout with no javascript delay at all)
@DavidCarlisle if they can somehow bundle mathjax into the browser itself, so it is transparant, that will be much better. i.e. ability to render math, using mathjax like way, if this can be build-in, no configuration or connection. Have the Browser javascript engine itself render latex, just like mathjax does it. Then it will be much faster.
Mathjax folks should talk the HTML5 group and the browsers companies to work on this.
@Nasser Use MathML and then do a browser test to see if it can handle MathML natively. If not, send it MathJaX. Note that the MathML conversion that MathJaX does is much faster than when it has to convert from raw TeX.
wow! There was a latex to HTML translator in 1993 !
In those days Tim was sitting just a few offices down the corridor from where Sebastian and I were working; and already at the beginning of 1993 we had translated, with the active help of Tim, some rather complex LaTeX documents into HTML, using our own ad hoc set of LaTeX macros which translated LaTeX high-level commands into HTML equivalents (with formulae left as TeX inside the HTML).
@Nasser you are kidding right. We have been talking to the browser makers since 1998. With the honourable exception of mozilla they all hate mathematics (it is hard and they get more feedback asking to display cats on youtube)
@Nasser There was a time when latex2html output constituted a large fraction of the internet.
@DavidCarlisle I know about l2h, I wrote a document on its installation which is used by many. I used it for few years. 12000.org/my_notes/l2hwin/index.htm
this may sound like a weird question; but is there a way to put divisions in a preamble? You see I have so much stuff in there I'd love an easy way to just access the points via some sort of label via the struture tree.
@Nasser maybe not: i'm using texmaker and in the structure panel I'd like to see like an icon I can click to access that chuck of code in the preamble.
@Grey the preamble shouldn't be that big:-) you can put any re-usable code in my-package-for-page-setup.sty my-package-for-fonts-and-layout.sty and whatever units make sense then the preamble just needs \usepackage{my-package-for-page-setup,my-package-for-fonts-and-layout}
@egreg A lot. :) A friend of mine who lives in Recife told me that although there are lots of sharks, they are usually harmless, as long as their boundaries are respected. :)
@David sorry but breqn is an experimental package the author of which (who I counted as a friend) died. So if you use it you have to be prepared to dig into its internals if it doesn't do exactly what you want. If page breaking is a requirement you probably need to say that in the question. There are not many who really know breqns workings though, so you should not expect a speedy answer.
does anyone use sublime editor for their latex work? I'm trying it out, but I've got so used to texmakers structure panel; does sublime have an alternative to this?
Psmith, the TeX bot, in fixed font mode: Let's take a look at the last cricket results:
- England 57/2 * v South Africa 175/10
- Netherlands 123/5 * v Nottinghamshire
- Leicestershire v Glamorgan 148/3 *
- Kent v Sussex
Our cricket expert David might explain these results later on.
Oh my, Subversion 1.8 features first-class renames! /faints
@Jake
\addplot+[raw gnuplot, draw=red, mark=none, smooth] gnuplot {
f(x) = a*x/(b+x);
a=0;
b=0;
set fit errorvariables;
fit f(x) 'source6.dat' using 1:4 via a,b;
% Next, plot the function and specify plot range
% The range should be approx. the same as the test.dat x range
plot [x=0.1:1000] f(x);
set print "parameters.dat"; % Open a file to save the parameters into
print a, a_err; % Write the parameters to file
print b, b_err;
@AlanMunn I was once in a lab connected to a gigabit network. I was downloading a Solaris ISO, it took 10 secs. I was like, "c'mon, this image must be corrupted." Nope.
@Jake thanks :) there are still 2 open questions, 1. why does the legendentry of you example not work :D and 2. where did you get the equation for f(x)
@Rico Hm, I was thinking of using tex.stackexchange.com/questions/21740/… for finding the intersections with the axes, but that example doesn't work for me anymore (says it doesn't know a path named a...)
@Jake ok not a problem, I'll explain it within the caption, what bothers me there has to be a way to use my existing data for producing those 1/x and 1/y
@Christian There are a couple of issues: The syntax for declare function is declare function={f(\x,\y)=erf(\x)+erf(\y);} (you need to make the variables explicit); and you need to put % after all the lines in \pgfmathdeclarefunction, otherwise you introduce heaps of spurious white space.
No, actually it works for 3D plots as well. I was always a bit suspicious of just using x and y in the function definition and pgf would somehow figure out that these were the two variables to use
but as long as it worked I didn't question it too much
I'll turn this into a question if you think I might not be the only one with this problem.
@Christian Hehe, yeah! PGFPlots is pretty awesome. I just noticed that again after having to produce two plots in Excel recently. It's impossible to get them to line up nicely, so frustrating!
nah, here the image looks just crappy ... especially when you zoom in
gnumeric? Yeah, it's an open source spreadsheet, too. But it is optimized for speed, precision and has its own graph wizard. And it's relatively easy to manually fix things in the XML that gnumeric uses as a native file format .. if all else fails I mean.
@Christian Loved the title of one of the referenced works in that one: It’s Easy to Produce Chartjunk Using Microsoft Excel 2007 but Hard to Make Good Graphs.
This question is a direct follow-up to the question Can a \newcommand Definition Contain Braces as Substitution Text? I asked yesterday.
The following example is once again a counter-intuitive one, to demonstrate what I wish to achieve. I'm trying to split the parbox syntax across the begin cod...
The breqn package changes the definition of \sum (and of almost all other commands for math mode). The definition it uses is (using the output of \show)
> \sum=\protected\long macro:
->\@sym \sum \math_sym_COs:Nn \mg@cop {50}.
The macro \@sym is the same as \@gobble, at least between \[ and \]...
With no warranty of any kind!
\documentclass{article}
\usepackage{color}
\makeatletter
\def\colorizemath #1#2{%
\expandafter\mathchardef\csname orig:math:#1\endcsname\mathcode`#1
\mathcode`#1="8000
\toks@\expandafter{\csname orig:math:#1\endcsname}%
\begingroup
\lccode`~...
This question is a direct follow-up to the question Can a \newcommand Definition Contain Braces as Substitution Text? I asked yesterday.
The following example is once again a counter-intuitive one, to demonstrate what I wish to achieve. I'm trying to split the parbox syntax across the begin cod...
