The current code fails to compile with:

<source>:3:54: error: invalid operands to binary expression ('enum E' and 'enum E')
const enum E neg_one_class_29 = (enum E) ((enum E) 0 - (enum E) 1);
                                          ~~~~~~~~~~ ^ ~~~~~~~~~~

This change reverts to the code that was before #2186 but silences erroneous GCC warning enabled by -Wconversion (which is not a part of -Wall).

Fixes #2749

warning C4244: 'initializing': conversion from 'double' to 'float', possible loss of data

Pointer buf cannot be NULL

Clean up insertion points after use in prange code since they conflict with the deep-copying that with-statements use for their finally-clause.
Closes GH-2780.

Fix exception messages for float division by zero, as well as their tests (which were wrong in Py2.6).

Closes #2820.

See #2807.

Closes #1982.

It fixes error 
```
src\lxml\etree.c(241210) : error C2275: 'PyObject' : illegal use of this type as an expression
```
when installing lxml 4.3.0 to pypy2 v6.0.0 on Windows.

Accessing the first element of a `std::vector` that is empty results in
a crash.

The __PYX_XDEC_MEMVIEW statements generated for try/finally statements
varied per build if there were multiple types of variables to be cleaned
up.

"convolve_cy.so" -> "compute_cy.so" to match imports used in this tutorial.

This fixes the following compiler warning:
  warning: cast discards ‘const’ qualifier from pointer target type [-Wcast-qual]

Prevent dynamically multiplied tuple constants from being deduplicated. Only the constant part is considered for them.

Prevent repeated initialisation of constants that overwrite the previously created objects without cleanup.
Closes #2750.

Closes #2736.

Closes #2731.