doublecount | A double entry accounting system/library for Rust

First, you can't mutate store's state directly. The only way to do that is to commit a mutation:

The compiler seems to be unable to infer what type the braced initializer list ist supposed to have. You can however use the direct list initializer of the inner std::vector

In the comparison functions, the pointers (x and y in your case) are pointers to elements of the array. If the array is an int, then it's a pointer to int (i.e. int *). If it's an array of structures, then they are pointers to the structure (like e.g. product *).

To access the members of the structure you of course needs to use the proper pointer-to-structure access, like e.g. the arrow operator ->.

It's also easier if you save the pointers in proper variables so you don't have to do all the casting all the time. In fact, if you want to use the passed data as pointers and not values there's no need to cast at all, because void * is implicitly convertible to almost any other pointer type (except pointers to functions).

So for your structure comparison function, you could do e.g.

I ended up using Pytest-Django as that testing framework actually allows for doing exactly what I wanted.

Using the --nomigrations flag, it takes my models which are only managed by django in tests and creates the appropriate table name for them (using their db_table attribute) in the test database. Then I can use factory_boy to create mock data and test it up!

I'm not sure what's causing your error but you are certainly not meant to be directly manipulating state outside of a mutation.

At no point should your code ever assign anything directly to a state property. For example, this is bad

Note   This post addresses the question of finding words spread out within a window, as asked. It does not consider the far more involved issues of general text parsing or language analysis.

The code below searches for the first word and then continues with another regex for the other two. There it scans the text word by word and keeps a counter so it can stop at 12 words. It uses pos to control where it should continue after checking the window.

The 12-long window is taken to start with word expect once it is found, as clarified in comments. The search continues from after the completed phrase, for the next one.

If the phrase is not found within the next 11 words the engine is returned to the position after expect to carry on with the search (as there may be another expect within the checked 11 words).

1. According to the docs of strtod:

   The functions sets the pointer pointed to by str_end to point to the character past the last character interpreted. If str_end is NULL, it is ignored.

   So your pointer buffer2 is still NULL and after you do buffer1= buffer2; - buffer1 now also is NULL (Btw, here is memory leak as the data is lost).

2. do I have th delete the 2nd buffer too?

   In this case - no, because deleting a NULL pointer is no-operation.

Solution:

Take a look at the example provided in docs for strtod function, here is similar according to your code: