Python：使用 mysqldump + gridfs 存储 mysql 转储

Question

I'm trying to make a python script that (in order)

1. dumps a mysql db into a file in /tmp/x.sql
2. opens up that file
3. writes that file into mongo gridfs

Motivation : Many different versions of a database for testing

My problem : Step 1 does not complete writing to the file before Step 2 runs, hence an empty file is read

What I've got

# filename is a generated uuid4

def step_1(filename):
    d = dict(environ)
    d["MYSQL_PWD"] = "root_pass"
    file = open(filename, "wb")

    subprocess.Popen(
        ["mysqldump", "-h", "0.0.0.0", "-u", "root", "database"], stdout=file, env=d
    )

def step_2(filename):
    with open(filename, "rb") as f:
        file = f.read()       <- This is always empty
        mongo_gridfs.put(data, content_type = 'file/txt')
        mongo_gridfs.save()

Right now my "solution" is just calling sleep(1)

There doesn't seem to be a method in the file that tells you if it's being written or if writing has completed, so I'm just kinda looking for a way to know if the file has been written so that I don't have to call sleep(1)

If there is a way to pipe this directly into a ByteIO object and read that into gridfs that would be even better

Answer 1

Well found two excellent answers in 1 minutes

doing

    process = subprocess.Popen(
        ["mysqldump", "-h", "0.0.0.0", "-u", "root", "s2r-db"], stdout=PIPE, env=d
    )

    process.wait() <- add this

made it work, but also adding PIPE to the stdout and reading it into a BytesIO instead of writing a file