arogozhnikov

# to be executed from different notebook!
def check_notebook_reproducibility(filename):
    md5s = []
    for attempt in [1, 2]:
        !jupyter nbconvert --to notebook --inplace --execute $filename --ExecutePreprocessor.timeout=86400
        result = !md5sum $filename
        md5s.append(result)
        print result
    a, b = md5s
    assert a == b, 'hashes are different'

arogozhnikov

----> call function __getitem__
------> call function __contains__
<------ exit function __contains__
------> call function _getitem_column
--------> call function _get_item_cache
----------> call function get
------------> call function _get_items
<------------ exit function _get_items
------------> call function isnull
--------------> call function _isnull_new

arogozhnikov

arogozhnikov

# Run command with the same user ID as current user
# -v $(pwd):/tmp/mount - mount current directory to /tmp/mount/
# --env HOME="/tmp/" - some commands may need to be able to write to your home, se it to temporary folder
docker run -ti --rm -v $(pwd):/tmp/mount —user=$(id -u) --env HOME="/tmp/" debian:jessie

# Mount current users and group and be able to use them
# mount /etc/group and /etc/passwd read only
# set user from $USER
docker run -ti --rm -v $(pwd):/tmp/mount -w /tmp/hx -v /etc/group:/etc/group:ro -v /etc/passwd:/etc/passwd:ro —user=$USER debian:jessie

arogozhnikov

arogozhnikov

Notes on edgedb

I’ve made a test evaluation of edgedb for the purposes of a small company.

My order of priorities (starting from most important):

1. correctness of results (data and code)
2. developer convenience
3. minimal maintenance
4. efficiency/scalability