Naive truncations of complexes are functorial. Given a map of chain complexes, this method returns the canonical map from the naive truncation of the source to the naive truncation of the target. The degree of f is used to determine the appropriate naive truncation of the target complex.
First, we define some non-trivial maps of chain complexes.
i1 : R = ZZ/101[a..d]; |
i2 : C = (freeResolution coker matrix{{a,b,c}})[1] 1 3 3 1 o2 = R <-- R <-- R <-- R -1 0 1 2 o2 : Complex |
i3 : D = freeResolution coker matrix{{a*b,a*c,b*c}} 1 3 2 o3 = R <-- R <-- R 0 1 2 o3 : Complex |
i4 : E = freeResolution coker matrix{{a^2,b^2,c*d}} 1 3 3 1 o4 = R <-- R <-- R <-- R 0 1 2 3 o4 : Complex |
i5 : f = randomComplexMap(D, C) 1 o5 = -1 : 0 <----- R : -1 0 1 3 0 : R <------------------------------------------------------- R : 0 | 24a-36b-30c-29d 19a+19b-10c-29d -8a-22b-29c-24d | 3 3 1 : R <---------------------- R : 1 {2} | -38 21 -47 | {2} | -16 34 -39 | {2} | 39 19 -18 | 2 1 2 : R <--------------- R : 2 {3} | -13 | {3} | -43 | o5 : ComplexMap |
i6 : g = randomComplexMap(E, D) 1 1 o6 = 0 : R <----------- R : 0 | -15 | 3 3 1 : R <---------------------- R : 1 {2} | -28 2 45 | {2} | -47 16 -34 | {2} | 38 22 -48 | 3 2 2 : R <----- R : 2 0 o6 : ComplexMap |
i7 : h = g * f 1 3 o7 = 0 : R <------------------------------------------------------- R : 0 | 44a+35b+46c+31d 18a+18b+49c+31d 19a+27b+31c-44d | 3 3 1 : R <---------------------- R : 1 {2} | -41 32 24 | {2} | 2 22 -25 | {2} | -32 28 38 | o7 : ComplexMap |
We use these maps to illustrate naive truncation.
i8 : tf = naiveTruncation(f, 0, 1) 1 3 o8 = 0 : R <------------------------------------------------------- R : 0 | 24a-36b-30c-29d 19a+19b-10c-29d -8a-22b-29c-24d | 3 3 1 : R <---------------------- R : 1 {2} | -38 21 -47 | {2} | -16 34 -39 | {2} | 39 19 -18 | o8 : ComplexMap |
i9 : tg = naiveTruncation(g, (0, 1)) 1 1 o9 = 0 : R <----------- R : 0 | -15 | 3 3 1 : R <---------------------- R : 1 {2} | -28 2 45 | {2} | -47 16 -34 | {2} | 38 22 -48 | o9 : ComplexMap |
i10 : th = naiveTruncation(h, (0, 1)) 1 3 o10 = 0 : R <------------------------------------------------------- R : 0 | 44a+35b+46c+31d 18a+18b+49c+31d 19a+27b+31c-44d | 3 3 1 : R <---------------------- R : 1 {2} | -41 32 24 | {2} | 2 22 -25 | {2} | -32 28 38 | o10 : ComplexMap |
i11 : assert all({tf, tg, th}, isWellDefined) |
i12 : assert(th == tg * tf) |
i13 : t2f = naiveTruncation(f, -infinity, 1) 1 3 o13 = 0 : R <------------------------------------------------------- R : 0 | 24a-36b-30c-29d 19a+19b-10c-29d -8a-22b-29c-24d | 3 3 1 : R <---------------------- R : 1 {2} | -38 21 -47 | {2} | -16 34 -39 | {2} | 39 19 -18 | o13 : ComplexMap |
i14 : assert(t2f == naiveTruncation(f,, 1)) |
i15 : assert(tf != t2f) |
It is also possible to truncate the source and target independently.
i16 : t2f = naiveTruncation(f, (0,1), (1,2)) 3 3 o16 = 1 : R <---------------------- R : 1 {2} | -38 21 -47 | {2} | -16 34 -39 | {2} | 39 19 -18 | o16 : ComplexMap |
i17 : assert(source t2f == naiveTruncation(C, (1,2))) |
i18 : assert(target t2f == naiveTruncation(D, (0,1))) |
There is another type of truncation, canonical truncation, which induces an isomorphism on (a range) of homology.