Open yarikoptic opened 4 months ago
Hey @yarikoptic, thanks for the suggestion! Right now it would prevent the export from working on the enterprise instances. I suspect, the workaround would be to "observe" the edge api and making a custom call to the edge APIs.
It looks like we have the code here, but it is not being actively used in the codebase.
I even know who wrote it 😂 It's just a stray one from one of the experimental branches that I did not want to lose. Did not go far with it. It will need to be revised.
Known restrictions:
1. error getting conversation users: callback error: enterprise_is_restricted
2. error getting members for [my test channel]: callback error: enterprise_is_restricted
Experimented with this over the weekend - findings:
Slack does not allow to seek past Page 100, when page > 100 requested, Slack returns the first page (Page=1). Seems to be an internal limitation. The workaround would be to use the Query parameter, to be more specific about the channel names, but to get all channels, this would require iterating through all 65536 runes of unicode give or take the special characters.
I'll take a break from this, maybe some idea comes up.
Hey @yarikoptic , with the great help of a slackdump user in Telegram, we were able to get it going, you can check the "v3" branch.
git checkout v3
The only thing you need to do, is to use the -enteprise
flag.
In the examples I will use the "record" format which is a new way to save the Workspace information implemented in v3. It is already quite stable, so the plan is for future versions to support it unchanged. I'll show how to convert to export further, please keep reading.
To export full workspace (in the "chunk" format):
go run ./cmd/slackdump archive -enterprise
To export some conversations:
go run ./cmd/slackdump archive -enterprise CHANNEL_ID1 CHANNEL_ID2
This will create "slackdump_
To get the chunk format spec:
go run ./cmd/slackdump help chunk
"View" allows to view the slackdump record archives, Slack and slackdump export ZIP files and directories, and Slackdump dump archives. It will detect the format automatically.
To view the archive, run:
go run ./cmd/slackdump view slackdump_record_dir
Then head to http://localhost:8080
If you wish for some reason to convert to slack export format, run this command:
go run ./cmd/slackdump convert -o my_export_file.zip slackdump_record_dir
Any feedback is welcomed.
Help is still in progress, but most of the commands will provide the flags and their description.
To get the list of commands:
go run ./cmd/slackdump help
To get help for the command or read a help topic:
go run ./cmd/slackdump help <topic or command>
To get help on a subcommand:
go run ./cmd/slackdump help command subcommand
for example
go run ./cmd/slackdump help list users
Thanks for your hard work on this @rusq ! Really appreciated it! I can confirm that it works on my side as well. I checked out v3
branch, ran make arm_darwin
to generate a binary, then used ./arm_darwin workspace
to login and ./arm_darwin dump -enterprise C0xxxxxx
to dump all messages in a channel. I'd like to report that the record
command is not found on my v3 checkout(ffd4532d)
Hey @xiaket, thank you for the feedback! Sorry about that, v3 is in heavy development, I have renamed "record" to "archive" to make the purpose more clear, I'll update my prev msg.
@rusq I have also verified this works as advertised off the v3-wiz
branch against my work enterprise Slack account. However I did have to make one small change as the Slack URL my workspace uses is not compatible with the current test. Ours is of the format:
https:///company.enterprise.slack.com/...
To get past this I updated the regexp used in url_parse.go
to verify the URL to the following:
var slackURLRe = regexp.MustCompile(`^https:\/\/[a-zA-Z0-9]{1}[-\w]+(\.enterprise)?\.slack\.com\/archives\/[A-Z]{1}[A-Z0-9]+(\/p(\d+))?$`)
Thanks for all you work on this.
Hey @hutchiko thanks for letting me know, I'll update the code and tests, when able! Glad that you were able to make it work.
Is your feature request related to a problem? Please describe.
Tried to auth against our dandiarchive instance which is on mit's enterprise slack instance.
after telling slackdump to dump list of conversations and going through authentication in a browser just getting
Previously successfully fetched a list for another slack workspace on a "regular" slack instance