Skip to content
Update Wiki (MoinMoin) authored by Gerald Combs's avatar Gerald Combs
......@@ -9,16 +9,21 @@ We have about [1200 active pages](https://wiki.wireshark.org/TitleIndex), but so
Flag day: Unscheduled. Can be done independently of Gerrit, Bugzilla, and Buildbot.
We have a [migration script](https://gitlab.com/wireshark/gitlab-migration/-/tree/master/moin) which is for the most part ready to go and should be sufficient for our purposes. A test migration can be found [here](https://gitlab.com/geraldcombs/ws-wiki-migration-2020-04-13/-/wikis/home).
## Details
MoinMoin has its own markup language that predates Markdown, which is what GitLab uses.
It looks like the most viable path would be to dump the wiki as HTML, then convert that to Markdown.
MoinMoin has its own markup language that predates Markdown, which is what GitLab uses. Our migration script fetches each wiki page as HTML and converts it to CommonMark using pandoc.
Other migration scripts can be found at:
* https://github.com/tetelle/moin
* http://moinmo.in/MoinDump
* http://webseitz.fluxent.com/wiki/ConvertingMoinmoinTextToMarkdown
We might be able to preserve our history by migrating Moin to a git repository, then converting the markup: https://github.com/mgaitan/moin2git. However, that would likely result in publishing spam that we've accumulated over the years.
### Migration Considerations
We might be able to preserve our history by migrating Moin to a git repository, then converting the markup: https://github.com/mgaitan/moin2git. However, that would likely result in publishing spam that we've accumulated over the years. Our migration script does not preserve history for this reason.
We run MoinMoin 1.9, but 2.0 supports [exporting to MarkDown](https://github.com/moinwiki/moin/commit/5edfbc2a3e3a8720fad8b2158d67133c2e7b6393#diff-42217d84820d3bab44de675d13077ae9).
Perhaps we could upgrade a copy of our current wiki to Moin 2.0, and then export?
We could upgrade a copy of our current wiki to Moin 2.0, and then export, but it was easier just to fetch each page as HTML.
\ No newline at end of file