Kasper Peeters
Kasper Peeters
The code in `pattern::match` runs `Ex_comparator::equal_subtree`. This is a generic routine which builds a replacement map even when (in cases such as this) a replacement map is not required. Add...
As the subject says. It works fine if the kernel is started before the last block of property declarations; probably something interferes which then confuses `evaluate`.
It may sometimes be useful to make the sum over index values explicit for a particular set only, e.g. to expand \partial_{\mu}{ A_{a b} } \partial^{\mu}{ A_{b a} } by...
At the moment a `μ` will not match `\mu`. As the display routines already turn the latter into the former when necessary, it is probably best to auto-convert all unicode...
Reproduce with ex:= \\bla; The preprocessor turns this into `\prod{\}{\bla}` and the parser then goes bananas on the `\}`.
When _declaring properties_, the notation for a trace is T::Trace( A_{m n} ); and this means $T = A_{m m}$. However, when you do {m,n,p}::Indices(vector); Tr{#}::Trace(indices=vector); and then write ex:=...
Evaluating a tensor equation leads to a `\components` node containing an equation for every component of the original tensor. Would be useful to have a helper which turns this into...
The `read_manual` function in `py_helpers.cc` is very basic, and does not in general produce readable text (it skips maths and leaves LaTeX markup in the string). This needs some generic...
It is not very difficult to add logic to be able to write ex:= substitute( sortproduct( A C B ), B->Q ) + D; and have this expanded immediately. Just...
The receiving end in `ComputeThread` expects it in the root; is that the logical place? Fix and document.