@OleksandrR. Disagree with site getting "downvote happy" - ref'd question is one of several by same that seem to show little if any effort in utilizing and understanding documentation. Myself, I will downvote many "be my automated documentation source" questions.

On another note, is anyone aware of a MMA implementation of the Goulden-Jackson clustering method for pattern-avoiding words? I'd rather grab an existing package vs reinvent the wheel...

@ciao, none of my searches seemed to turn up anything useful, so you may have to roll your own implementation. In which case: I presume you've seen Doron Zeilberger's work?

By default MaTeX's output will have a height of at least a full line width, regardless of how small the text is. Does anyone have a need to be able to turn this off and make it behave as in the bottom example below?

@Guesswhoitis. Of course - and the interesting generalizations by Kupin and Pudwell. I don't have to tell you (if you've viewed their implementations), they're mathematicians, not coders :-) , so was just checking if anyone had seen an mma implementation before I start on one.

@OleksandrR. According to data.stackexchange.com/mathematica/query/1718/…: 2676 dv. the last seven days v. 745 dv. the previous seven days. The ratio of up/down has decreased from 248-367 to 38-48. The total number of votes has decreased from 213693 to 123775. It could be that someone has put us in a bad mood and/or driven away the happy people.

There'll be of the order of a hundred downvotes from ChrisJJ's posts, which is nowhere near the two thousand disparity.

@ciao well, in my opinion, it can be discouraging for inexperienced users if they gain downvotes when they are merely struggling rather than having asked an unreasonable question. Of course it is up to everyone to vote as they see fit, but my view would be that questions like this should certainly be closed, but it doesn't benefit anyone to downvote as well. I tend to downvote only when I am trying to discourage the user from asking ill-formed/"do my work for me"/rudely demanding questions

@PatrickStevens I think meta votes don't count

@MichaelE2 thanks. It is quite a big difference by the numbers

@ciao, a bit harsh. :P He does have Shalosh Ekhad at his side, so there we are… Maple has its charms, tho. :)

@OleksandrR. I was surprised, too. I thought I was misinterpreting the results, but AFAICT, that's what it says.

Curious - why is a bug in 10.2 regarding Global marked as a duplicate, when the answer to the linked question is for 10.0.0, and in that case been fixed as of 10.0.1? The problems, although similar apply to different versions?i

People voted to close it and it got closed.

Well, I understand that bit!

Yes, I wouldn't have voted, but I think it's no big deal.

Does anyone have better ideas for returning multiple results from LibraryLink? -->

@OleksandrR. Do you use LibraryLink?

What really bothers me about LibraryLink is that it's a lot of work to write the code. I was thinking if the pain could be eased by generating some of the code automatically.

We would have a "template", perhaps slightly similar to MathLink templates, that would define an interface not to functions, but a C++ class.

Managed library expressions would be used to create and manage more than one instance of that class.

Returning multiple results would be solved by calling separate member functions. The LibraryLink code to call these would be auto-generated, and it would spare me a lot of trouble.

The member function "template" could be something similar to what is being used in LibraryFunctionLoad now, just a Mma expression. On the C++ side we could have easy-to-use data structures (classes) which map directly to the types Mma can pass.

Any comments/ideas? Do you see anything that will go wrong?

My main motivation is to make it faster and more convenient to write LibraryLink code, perhaps at the cost of some flexibility.

@Szabolcs not seriously, no. I also have not tried anything in version 10 yet. It is definitely a problem to return inhomogeneous results, but I tend to look on LibraryLink as sort of similar to compiled code. It's unfortuate that it is very limited compared to what can be done inside the kernel.