In[197]:= ClearAll["Global`*"]
a = {{4, 16, 4}, {7, 25, 8}, {17, 63, 18}};
b = {{2, -2, -2}, {7, -71, 2}, {14, -114, 0}};
Last[HermiteDecomposition[a]] == Last[HermiteDecomposition[b]]
MatrixForm /@ {LatticeReduce[a], LatticeReduce[b]}
During evaluation of In[197]:= Set::wrsym: Symbol a is Protected.
During evaluation of In[197]:= Set::wrsym: Symbol b is Protected.
During evaluation of In[197]:= HermiteDecomposition::latm: Matrix contains an entry that is not rational.
During evaluation of In[197]:= HermiteDecomposition::latm: Matrix contains an entry that is not rational.
Out[200]= a == b
Out[201]= {\!\(\*
TagBox[
RowBox[{"LatticeReduce", "[", "a", "]"}],
Function[BoxForm`e$,
MatrixForm[BoxForm`e$]]]\), \!\(\*
TagBox[
RowBox[{"LatticeReduce", "[", "b", "]"}],
Function[BoxForm`e$,
MatrixForm[BoxForm`e$]]]\)}
As you can see, I’ve used the command ClearAll[“Global`*”] at the beginninf of my code, but still the Symbol a is Protected problem is triggered. Any hints for fixing this problem?
ClearAll does not affect symbols with the attributes Locked or Protected. »
You could Unprotect your variables or localize them. Or do not Protect symbols in the global context. Protecting them should indicate that you do not want to use them except in the protected way. That seems to be false in your situation.
If you’ve also locked the variables, then use different variables.
In my own workflows, I do not ClearAll["Global`*"] nor do I Protect symbols in the Global` context. I also do not set single-letter variables to a value, except localized in Module, Block, With, or Pattern — or in scratch work in a Documentation notebook, since cell groups get a unique context in them. That’s all a matter of personal style, though, and probably idiosyncratic. Mostly they avoid mistakes I commonly make. One danger I live with is that when a new documentation page loads, it wipes out your scratch work, even if you go back (I suspect many people wouldn’t like that danger).
