Well, that didn't work either.  Sorry for the big message with all the
junk.  I hope some of you will be able to open the attachment.

Bill


On Tue, Oct 29, 2013 at 9:49 AM, William Hanson <whan...@umn.edu> wrote:

> Dear All,
>
> Many thanks to Ernesto Posse for the very detailed and helpful response.
> However, what I'm trying to do is eliminate *all* the lines between nodes
> *except* those that indicate a genuine branching of the tree.  In other
> words, I want the overall structure of the tree to look like the following
> (which I tried to attach to my previous messages, but which apparently
> could not be opened).  As usual, any and all help is appreciated.
>
> Bill
>
>
>
>
>
>
> On Mon, Oct 28, 2013 at 3:45 PM, Ernesto Posse <epo...@cs.queensu.ca>wrote:
>
>> I'm unable to see your attachment, but if I understand what you want, the
>> simplest approach is to create "phantom" nodes in the tree. This can be
>> achieved with "child [missing]" as in the following examples:
>>
>> First, a simple tree with two nodes: A and B; A is the root, and B is
>> directly below it:
>>
>> \begin{tikzpicture}
>>
>> \node {A}
>>
>>        child {node {B}};
>>
>> \end{tikzpicture}
>>
>> Then with node B towards the left:
>>
>> \begin{tikzpicture}
>>
>> \node {A}
>>
>>        child {node {B}}
>>
>>        child [missing];
>>
>> \end{tikzpicture}
>>
>> and now, with B towards the right:
>>
>> \begin{tikzpicture}
>>
>> \node {A}
>>        child [missing]
>>
>>        child {node {B}};
>>
>> \end{tikzpicture}
>>
>> Now, with your example (I changed the numbers in the nodes, for
>> reference):
>>
>> \begin{tikzpicture}
>>
>> \node {$1.~\neg ((p \lor (p \land q)) \limp p)$}
>>
>>        child {node {$2.~ p \lor (p \land q)$}
>>
>>               child [missing]
>>
>>               child {node {$3.~ \neg p $}
>>
>>                      child {node {$4.~ \ p $}}
>>
>>                      child {node {$5.~ p \land q$}
>>
>>                             child [missing]
>>
>>                             child {node {$6.~ p $}
>>
>>                                    child {node {$7.~ q $}}
>>
>>                                    child [missing]}}}}
>>
>>        child [missing];
>>
>> \end{tikzpicture}
>>
>>
>> As indicated in the examples above, the position of the "child [missing]"
>> relative to its siblings determines where you get the child nodes. Of
>> course you can also add any number of missing children, which increases the
>> angle:
>>
>>
>> \begin{tikzpicture}
>>
>> \node {A}
>>        child [missing]
>>        child [missing]
>>
>>        child {node {B}};
>>
>> \end{tikzpicture}
>>
>> Furthermore, you can control the distance between sibling nodes:
>>
>> \begin{tikzpicture}[sibling distance=4cm]
>>
>> \node {A}
>>
>>        child {node {B}}
>>
>>        child {node {C}};
>>
>> \end{tikzpicture}
>>
>>
>> and even the distance between levels:
>>
>>
>> \begin{tikzpicture}[sibling distance=4cm,level distance=5cm]
>>
>> \node {A}
>>
>>        child {node {B}}
>>
>>        child {node {C}};
>>
>> \end{tikzpicture}
>>
>>
>> On Mon, Oct 28, 2013 at 4:06 PM, William Hanson <whan...@umn.edu> wrote:
>>
>>> Dear LyX Colleagues,
>>>
>>> I'm still trying to create tableau proofs, which are branching columns
>>> of text, as illustrated in the attachment.
>>>
>>> Ernesto Posse's sample tableau (below) is helpful, but it contains two
>>> features I don't want:
>>>
>>> 1.  Two sentences at a node, separated by commas.  I want just one
>>> sentence at each node, as in the attached sample.  I've been figured out
>>> how to solve this problem by modifying Ernesto's code as follows:
>>>
>>> \def\land{\wedge}
>>> \def\lor{\vee}
>>> \def\limp{\to}
>>> \begin{tikzpicture}
>>> \node {$1\neg ((p \lor (p \land q)) \limp p)$}
>>> child {node {$ 1 p \lor (p \land q)$}
>>> child {node {$1 \neg p $}
>>> child {node {$1 \ p $}}
>>> child {node {$1 p \land q$}
>>> child {node {$1 p $}
>>> child {node {$1 q $}}}}}};
>>> \end{tikzpicture}
>>>
>>> 2.  But the foregoing code retains another feature I don't want:
>>> vertical lines from node to node when there is no branching.  I want only
>>> the (approximately) 45 degree (and 315 degree) lines that indicate
>>> branching, as on the attached sample.
>>>
>>> I've used LyX for several years, but I don't know LaTeX.
>>>
>>> Any and all help appreciated.
>>>
>>> Bill
>>>
>>>
>>>
>>>
>>>
>>> [image: Sample 
>>> Tableau.pdf]<https://mail.google.com/mail/u/0/?ui=2&ik=fcb7343f58&view=att&th=141e0e8e8016ae7f&attid=0.1&disp=safe&realattid=f_hn3bnlm80&zw>
>>> *Sample Tableau.pdf*
>>> 146K   
>>> View<https://docs.google.com/viewer?a=v&pid=gmail&attid=0.1&thid=141e0e8e8016ae7f&mt=application/pdf&authuser=0&url=https://mail.google.com/mail/u/0/?ui%3D2%26ik%3Dfcb7343f58%26view%3Datt%26th%3D141e0e8e8016ae7f%26attid%3D0.1%26disp%3Dsafe%26realattid%3Df_hn3bnlm80%26zw&sig=AHIEtbQpDR5qvKd2TSh_O5cOhrpoG-Owmg>
>>> Download<https://mail.google.com/mail/u/0/?ui=2&ik=fcb7343f58&view=att&th=141e0e8e8016ae7f&attid=0.1&disp=safe&realattid=f_hn3bnlm80&zw>
>>>
>>> Ernesto Posse <epo...@cs.queensu.ca>
>>> Oct 22 (6 days ago)
>>>
>>>  to me, lyx-users
>>> Hello. The easiest (and nicest) way to do this is using the tikz
>>> package: in the preamble put
>>>
>>> \usepackage{tikz}
>>>
>>> and then, wherever you want the tableau, put in a TeX box the following:
>>>
>>> \def\land{\wedge}
>>>
>>> \def\lor{\vee}
>>>
>>> \def\limp{\to}
>>>
>>> \begin{tikzpicture}
>>>
>>> \node {$\{\neg ((p \lor (p \land q)) \limp p)\}$}
>>>
>>>     child {node {$\{p \lor (p \land q), \neg p\}$}
>>>
>>>         child {node {$\{p\}$}}
>>>
>>>         child {node {$\{p \land q\}$}
>>>
>>>             child {node {$\{p,q\}$}}}};
>>>
>>> \end{tikzpicture}
>>>
>>>
>>> Note that the structure of the tree depends on the grouping braces { ...
>>> }.
>>>
>>>
>>>  Richard Heck <rgh...@lyx.org>
>>> Oct 22 (6 days ago)
>>>
>>> to Ernesto, me, lyx-users
>>>  There are lots of useful resources about this here:
>>>     http://www.logicmatters.net/latex-for-logicians/trees/
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>>
>> --
>> Ernesto Posse
>>
>> Modelling and Analysis in Software Engineering
>> School of Computing
>> Queen's University - Kingston, Ontario, Canada
>>
>
>

Reply via email to