py3: some small care for matrix

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​f8d7805

py3: some care for matrix

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​bb5184e	Merge branch 'u/chapoton/26716' in 8.5.b4
​6d94e11	trac 26716 fix the pyflakes warnings

[jdemeyer]

1. This change looks suspicious, can you justify it:

   @@ -1927,6 +1928,7 @@ def block_matrix(*args, **kwds):
        sub_matrices = args[0]
    
        if is_Matrix(sub_matrices):
   +        M = sub_matrices
            # a single matrix (check nrows/ncols/ring)
            if (nrows is not None and nrows != 1) or \
               (ncols is not None and ncols != 1):
   

If it's a bug fix, then a doctest should be added.

2. Replace if ring == QQ or ring == ZZ: by if ring is QQ or ring is ZZ:

3. PEP 8 recommends to write

   entries = lambda i, j: 1 / (i + j + 1)
   

as

def entries(i, j):
    return 1 / (i + j + 1)

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​574f478

trac 26716 adding a missing doctest, plus details

[chapoton]

About 1. Pyflakes said that M was not defined anywhere. And indeed, this part of the code was not doctested.

2. and 3. Done

[chapoton]

ready for review and the bot is green

[vklein]

Looks good to me.

Side Note : This ticket is actually a bug fix with the M = sub_matrices.