`./manage.py import` does not take a tarball; it takes a directory.
Making a separate tarball is a waste of CPU time and disk, as it is
never used.
This was included in the commit of the initial Slack conversion code
in 5b37c5562b and propagated from there into every conversion tool.
Remove the unnecessary tarball creation.
Black 23 enforces some slightly more specific rules about empty line
counts and redundant parenthesis removal, but the result is still
compatible with Black 22.
(This does not actually upgrade our Python environment to Black 23
yet.)
Signed-off-by: Anders Kaseorg <anders@zulip.com>
The naive solution #23465 creates situations where the same user can have
multiple reactions as the base emojis are not unique, e.g. +1::skin2
and +1::skin4 would both reduce to +1 but the userlists are separate.
This solution handles the reduction, merges the same-base reactions,
and deduplicates the userlist.
Co-authored-by: Alex Vandiver <alexmv@zulip.com>
Co-authored-by: rht <rhtbot@protonmail.com>
Previously, emoji.json was read from
"$ZULIP_PATH/node_modules/emoji-datasource-google/emoji.json".
This path doesn't exist in production when installing from scratch from
a release tarball. And so, we ensure emoji.json exists by copying it to
`static/generated/emoji`.
With tweaks to comments by tabbott.
Fixes: #23469
This commit adds the OPTIONAL .realm attribute to Message
(and ArchivedMessage), with the server changes for making new Messages
have this set. Old Messages still have to be migrated to backfill this,
before it can be non-nullable.
Appropriate test changes to correctly set .realm for Messages the tests
manually create are included here as well.
build_message has a lot of arguments, so it's hard to verify correctness
of callers that just try to get the order right. It's much clearer to be
explicit via kwargs. mattermost.py and rocketchat.py already do this, so
let's bring slack.py and gitter.py up to par.
Because Slack emoji naming is different from Zulip's.
According to https://emojipedia.org/slack/, Slack's emoji shortcodes are
derived from https://github.com/iamcal/emoji-data.
There are probably some deviations from that dataset, but this PR should
at least catch the ones that are identical to iamcal's.
If there are more than 1 room with the same set of users, the import
will fail due to a unique constraint on the huddle_hash. Figuring out
why and which room is causing this database error is kinda difficult.
We deduplicate those cases here and simply merge the rooms together.
Note however, that the deduplication does not work as expected so we
simply ignore them all together for now and only raise an exception
along some logging output. At least this way, it is pretty clear what is
wrong and you do not have to wait to get a database error during the
actual import.
We also ignore empty huddle rooms since those are the duplicates that
caused problems for me and if they are empty, ignoring them is easier
than trying to get the merge to work.
Not sure where those channels come from since we discovered this with
production data.
Signed-off-by: Florian Pritz <bluewind@xinu.at>
Not sure where those come from since we discovered this with production
data. Somehow there were reactions with usernames that were old and no
longer existed.
Signed-off-by: Florian Pritz <bluewind@xinu.at>
Not sure where those come from since we discovered this with production
data.
There only was a single instance of this in my entire batch of data in
an old message from the time when we started using Rocket.Chat. This
might be an old issue or it might require some special settings that
were later changed.
Signed-off-by: Florian Pritz <bluewind@xinu.at>
Users will only be able to login via GitHub, because imported users
get GitHub's generated noreply email addresses - so this should be the
only auth method enabled at first, to avoid confusion.
Only ["id"] is accessed on the dicts (representing the external tool
users). Given that for some tools the id may be under a different name
etc. due to different user dicts format, it's best to just pass those
ids to the function so that it can stay generalized and not reliant
on a specific user dict format.
get_timestamp_from_message was extracted in the previous commit. We can
deduplicate and the code a bit cleaner by using it where appropriate
instead of message["ts"].
message["ts"] is slack-specific. For this to be a general util function
it needs to take a callable that will grab a timestamp from the message
dict (which has varying formats depending on what we're importing from).
It is apparently possible to have a mention of a user who is not (or
no longer?) in the `users.bson` table.
Skip such mention for the purposes of Zulip import; there's nothing
better for us to do.
This is likely an error somewhere in rocketchat's MongoDB "eventual
consistency," but there is no problem with skipping the chunks at this
step.
In the one case where this was observed so far, the upload-id was not
referenced in any message -- if it is referenced and has chunks, but
has no metadata, we will fail later, at that reference.
We construct model instances in the import tool solely for the purpose
of serializing them with the `model_to_dict` helper that returns a
dictionary. Passing `float` to these models' DateTimeField is not
accepted by the type checker. Modifying the dictionary instead avoids
this typing issue.
Signed-off-by: Zixuan James Li <p359101898@gmail.com>
Now that we can assume Python 3.6+, we can use the
email.headerregistry module to replace hacky manual email address
parsing.
Signed-off-by: Anders Kaseorg <anders@zulip.com>
Alex Vandiver
Anders Kaseorg
M@
rht
Mateusz Mandera
Florian Pritz
Zixuan James Li