I tried with and without a virtualenv for dependency isolation, using the following command:

fake2db --db postgresql --host 0.0.0.0 --port 5632 --password password --username user --rows 10 --name mydb

Which gave me:

File "/usr/local/bin/fake2db", line 7, in <module>
    from fake2db.fake2db import main
  File "/usr/local/lib/python3.5/site-packages/fake2db/fake2db.py", line 6, in <module>
    from custom import faker_options_container
ImportError: No module named 'custom'

But the custom.py file exists, it's just not finding it. Installed using pip 9.0.1 and python 2.7.10.

sudo seems to work to bypass the error. Not sure why it needs it, but this is probably unrelated to the specific package, so I'll close this.

I would like to create a table with a million rows or more to stress test a service I am evaluating.

Using a server hosted by aws (rds mysql 5.6), I'm currently using no cpu on my macbook so it is definitely network bound.

I'm able to load about 3000 rows per 5 minutes. This speed is really slow when you would like to generate a very large table (e.g. many millions of rows)

Some possible suggestions:

1. bulk insert
2. save to a csv, and then call a sql command to load from csv
3. create a pool of connections and insert through each connection

Let us specify (via a schema.json file or something similar) a custom schemata. Something compatible with https://github.com/topliceanu/mongoose-gen would be great.

SUMMARY: Followed README, attempted to run fake2db.py but encounter missing dependencies. The following replicates my steps to include confirmation of dependencies:

ERROR: _ Traceback (most recent call last): File "fake2db.py", line 6, in from .custom import faker_options_container ValueError: Attempted relative import in non-package _

STEPS:

# pip install fake2db Requirement already satisfied: fake2db in /Users/jason/Documents/Scripts/fake2db Requirement already satisfied: fake-factory>=0.5.3 in /usr/local/lib/python2.7/site-packages (from fake2db)

# pip install psycopg2 Requirement already satisfied: psycopg2 in /usr/local/lib/python2.7/site-packages

# cd /Users/jason/Documents/Scripts/fake2db *# ls LICENSE docs fake2db.egg-info setup.py README.md fake2db requirements.txt

# pip install -r requirements.txt Obtaining file:///Users/jason/Documents/Scripts/fake2db (from -r requirements.txt (line 3)) Requirement already satisfied: fake-factory>=0.5.3 in /usr/local/lib/python2.7/site-packages (from fake2db==0.5.2->-r requirements.txt (line 3)) Installing collected packages: fake2db Found existing installation: fake2db 0.5.2 Uninstalling fake2db-0.5.2: Successfully uninstalled fake2db-0.5.2 Running setup.py develop for fake2db Successfully installed fake2db

# cd fake2db # ls init.py custom.py mongodb_handler.py redis_handler.py base_handler.py fake2db.py mysql_handler.py sqlite_handler.py couchdb_handler.py helpers.py postgresql_handler.py

# python fake2db.py --db postgresql --host 127.0.0.1 --port 5432 --password fakepassword --username postgres --name TESTDB --rows 100 --custom address

Traceback (most recent call last): File "fake2db.py", line 6, in from .custom import faker_options_container ValueError: Attempted relative import in non-package

Name argument is not taken in consideration when creating mysql db. None should replace by args.name here https://github.com/emirozer/fake2db/blob/master/fake2db/fake2db.py#L161

$ fake2db --help
Traceback (most recent call last):
  File "/usr/local/bin/fake2db", line 9, in <module>
    load_entry_point('fake2db==0.2.2', 'console_scripts', 'fake2db')()
  File "/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/pkg_resources.py", line 357, in load_entry_point
    return get_distribution(dist).load_entry_point(group, name)
  File "/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/pkg_resources.py", line 2394, in load_entry_point
    return ep.load()
  File "/System/Library/Frameworks/Python.framework/Versions/2.7/Extras/lib/python/pkg_resources.py", line 2108, in load
    entry = __import__(self.module_name, globals(),globals(), ['__name__'])
  File "/Library/Python/2.7/site-packages/fake2db/fake2db.py", line 7, in <module>
    logger, extra_information = fake2db_logger()
  File "/Library/Python/2.7/site-packages/fake2db/helpers.py", line 13, in fake2db_logger
    local_ip = socket.gethostbyname(socket.gethostname())
socket.gaierror: [Errno 8] nodename nor servname provided, or not known

I don't think it's an internet issue, as I am connecting to the ip of a local vm. What are your thought?

This is a good technique to avoid typing the same packages names and versions in different places (setup.py and requirements files). Also it gives the ability to test different versions in development keeping the setup.py functional.

Cool idea; wanted to play with this.

OSX 10.9.5 MySQL 5.5.38 Python 2.7.5

Installed http://dev.mysql.com/downloads/connector/python/

Opened new terminal (iTerm2) session

Ran $ fake2db --rows 200 --db mysql

Got:

Traceback (most recent call last):
  File "/usr/local/bin/fake2db", line 8, in <module>
    load_entry_point('fake2db==0.1.5', 'console_scripts', 'fake2db')()
  File "/Library/Python/2.7/site-packages/fake2db/fake2db.py", line 106, in main
    fake_mysql_handler.fake2db_mysql_initiator(host, port, int(args.rows))
  File "/Library/Python/2.7/site-packages/fake2db/mysql_handler.py", line 42, in fake2db_mysql_initiator
    cursor.execute(tables[key])
AttributeError: 'NoneType' object has no attribute 'execute'

Any thoughts? Thanks!

When I use the following command:

fake2db --db sqlite --name test.db --rows 100

Instead of using test.db as the database file name, fake2db use a generated filename such as sqlite_JZMYNRKM.db.

I hope I had fixed the problem, but it is not full tested for other engine(postgresql, mysql, ...).

Thanks a lot.

Hi, I've installed fake2db successfully in ubuntu 14.04 and still, i'm unable to run cli commands as shown in readme.

Python 2.7.6 (default, Mar 22 2014, 22:59:56) [GCC 4.8.2] on linux2 Type "help", "copyright", "credits" or "license" for more information.

fake2db --rows 100 --db postgresql --name testdb File "", line 1 fake2db --rows 100 --db postgresql --name testdb ^ SyntaxError: invalid syntax

Tis ont erpefct, but now fake2db actually works for me now.

I tried to make this a minimal change, but there were some things I could not overlook, so there's some extra changes not related to the ability to pass in username/password.

Tested locally with:

$ fake2db --db postgresql --rows 1337 --host elk_postgres.local --password password --user docker
2015-09-02 17:19:25,854 crccheck      Rows argument : 1337
2015-09-02 17:19:36,184 crccheck      Database created and opened succesfully: postgresql_vsgjqwju
2015-09-02 17:19:38,134 crccheck      simple_registration Commits are successful after write job!
2015-09-02 17:19:39,864 crccheck      detailed_registration Commits are successful after write job!
2015-09-02 17:19:41,488 crccheck      companies Commits are successful after write job!
2015-09-02 17:19:42,958 crccheck      user_agent Commits are successful after write job!
2015-09-02 17:19:45,376 crccheck      customer Commits are successful after write job!

I found myself needing to do the following to replicate a table structure:

fake2db --rows 1500 --db mysql --name=test_speed --username root --password secret --custom date_time_this_year random_digit_not_null random_digit_not_null uuid4 word boolean boolean boolean random_number random_number word last_name word word word word last_name year

Then I found out that I couldn't use duplicate faker keys for my columns, so using random_digit_not_null twice is not possible.

I wrote some code in the mysql handler to append keys to the columns in order to allow duplicates (see this commit https://github.com/denitsa-cm/fake2db/commit/73df91998d6c3929e90d5c211f61b765c560510d)

I do think the mysql_handler is probably not the best place for this -> the unique columns should somehow be formatted further up and passed to all handlers. But it's the first time I'm touching python so I just hacked it a bit for my purposes.

Then the command above would result in a table structure like this:

Would be nice if something like this was supported out of the box for all database engines. Or perhaps even better, the option to name the columns in addition to providing the faker keys. This would alleviate the issue altogether.

I inserted 10000 to the db named fake. Then I want to insert another 10000. But error happened.

fake2db --rows 10000 --db postgresql --name fake --username postgres 2019-03-18 09:55:19,907 www Rows argument : 20000 2019-03-18 09:55:20,036 www database "fake" already exists

Traceback (most recent call last): File "/home/www/pyenv/aws/bin/fake2db", line 11, in sys.exit(main()) File "/home/www/pyenv/aws/local/lib/python2.7/site-packages/fake2db/fake2db.py", line 167, in main number_of_rows=args.rows, name=args.name, custom=custom) File "/home/www/pyenv/aws/local/lib/python2.7/site-packages/fake2db/postgresql_handler.py", line 18, in fake2db_initiator cursor, conn = self.database_caller_creator(number_of_rows, **connection_kwargs) File "/home/www/pyenv/aws/local/lib/python2.7/site-packages/fake2db/postgresql_handler.py", line 46, in database_caller_creator cur.execute('CREATE DATABASE %s;' % dbname) psycopg2.ProgrammingError: database "fake" already exists

Thanks for your great work.

Hi,

I just ran this command:

$ fake2db --rows 3 --db sqlite --custom name date country

And the result that I got:

I have 2 questions here:

1. How do I get auto-increment ID? For example, Francis get id 1 and Robert get id 3?
2. How do I rename the custom column names? For example, full_name instead of name and birth_date instead of date.

Thank you in advance for your help. This project is very cool and helpful. 🙂

Hey there – is there any way we could run this against an existing database (with a flag or something is fine) and not get errors? We'd like to add this to our migrations in our dev environments, but we end up having to create a new database, then import it into the primary db.

We'd be happy to make adjustments in a PR if you want to outline the easiest / best way to do so. Thanks :)