[texhax] LaTeX amsmath: two equations aligned left-right

[Olivier Cailloux]

Le 25/03/2011 17:04, Lars Madsen a écrit :
> Olivier Cailloux wrote:
>> Hello,
>>
>> I can't find how to display properly some equations the way I would 
>> to. It is a bit difficult to explain, please compile this and observe 
>> the result. Or read below.
>>
>> \documentclass[a4paper]{article}
>> \usepackage[utf8]{inputenc}
>> \usepackage{amssymb}
>> \usepackage{amsmath}
>> \usepackage{amsthm}
>>
>> \begin{document}
>> \section{Two equations}
>> \begin{multline}
>>     \label{eq:eq1}
>> \forall j \in J, a \in A, 1 \leq h \leq k-1 :\\
>> \frac{1}{M_j+ \delta_j} ((g_j(a) - g_j(b_h))+\delta_j)
>> \leq C_j(a, b_h).
>> \end{multline}
>> \begin{multline}
>>     \label{eq:eq2}
>> \forall j \in J, a \in A, 1 \leq h \leq k-1 :\\
>> C_j(a,
>> b_h) \leq \frac{1}{M_j} (g_j(a) - g_j(b_h)) + 1.
>> \end{multline}
>>
>> \section{Now better}
>> \begin{multline}
>> \forall j \in J, a \in A, 1 \leq h \leq k-1 :\\
>> \left\{\begin{gathered}
>> %     \label{eq:eq1}
>> \frac{1}{M_j+ \delta_j} ((g_j(a) - g_j(b_h))+\delta_j)
>> \leq C_j(a, b_h);\\
>> C_j(a,
>> b_h) \leq \frac{1}{M_j} (g_j(a) - g_j(b_h)) + 1.
>> \label{eq:eq2}
>> \end{gathered}\right.
>> \end{multline}
>> \end{document}
>>
>> In the first section I present the two equations I need. I need two 
>> separate equation numbers to be able to refer to each of them 
>> independently.
>>
>> I would like to display them all together as in the second section, 
>> which would allow me to avoid repeating the first, common, part. In 
>> order to do that, I need in total three lines. The first line holds 
>> the common beginning; the second and third lines are enclosed in a 
>> left bracket to make it clear that they both are the continuation of 
>> the first part. The second and third line must be aligned on the 
>> right margin while the first line is left aligned. All of this I get 
>> with the "multline" and embedded "gathered", as in the sample code. 
>> However what I don't get is two different equation numbers. I would 
>> like a number on the second line, to refer to that equation 
>> specifically, and a different (consecutive) number on the third line.
>>
>> Any suggestion?
>> Olivier
>>
>> _______________________________________________
>> TeX FAQ: http://www.tex.ac.uk/faq
>> Mailing list archives: http://tug.org/pipermail/texhax/
>> More links: http://tug.org/begin.html
>>
>> Automated subscription management: 
>> http://tug.org/mailman/listinfo/texhax
>> Human mailing list managers: postmaster at tug.org
>
> I would not write it like that at all (and I would never use multline, 
> it often ends up looking really bad, especiellay if one afterwards 
> change the margins)
>
> \documentclass[a4paper]{article}
> \usepackage[utf8]{inputenc}
> \usepackage{amssymb}
> \usepackage{amsmath}
> \usepackage{amsthm}
> \usepackage{empheq}
> \begin{document}
>
> \section{Now better}
> $\forall j \in J, a \in A, 1 \leq h \leq k-1 :$
> \begin{empheq}[left=\empheqlbrace]{align}
> &\frac{1}{M_j+ \delta_j} ((g_j(a) - g_j(b_h))+\delta_j)
> \leq C_j(a, b_h);\\
> &C_j(a,
> b_h) \leq \frac{1}{M_j} (g_j(a) - g_j(b_h)) + 1.
> \end{empheq}
> \end{document}
>
> The $\forall...$ is the premise (is that the correct word?) and I 
> would not write that as a part of the equation it self, it is 
> understood from the context that the equation is under those conditions
Thanks for your suggestion. It is simple and it answers the question. Is 
it possible to get rid of empheq (that's yet an other package I would 
have to learn)? I guess no...
Olivier
>
>
>
>
>
>
>
>