Drew from Infra fetched and sanitized the dump, and I was able use it directly instead of fetching the data from UI.
I ran the first import, and here's how it looks: https://github.com/vlsi/tmp-jmeter-issues/issues It looks more-or-less fine to me. Note: last time I imported, there was a connection reset, so there will be only 3710 or so bugs in GitHub this time. The full import would take several hours (we would need to make both Bugzilla and GitHub read-only during that time) Images are displayed inline. You could try searching issues with images via https://github.com/vlsi/tmp-jmeter-issues/issues?q=is%3Aissue+attachment+png Small text files are displayed inline as well. You could find them via https://github.com/vlsi/tmp-jmeter-issues/issues?q=is%3Aissue+attachment+preview Sample issue: https://github.com/vlsi/tmp-jmeter-issues/issues/1457 I plan to improve some things, however, it would be great if you could check the conversion results and let me know what could be improved. Please consider reporting the issues via https://github.com/vlsi/bugzilla2github (or https://forms.gle/3DzVsre2ytismoxx6) Known issues: * "reporter and commenter" is the same everywhere. It will be ASF robot account or something like that. Unfortunately, this can't be fixed, as INFRA says we can't ask for GitHub support, so real reporters and commenters are included in the first lines. * reporters and commenters are not linked to their GitHub accounts. I will add the username to email mapping soon, so if you want to have your GitHub profile linked, please submit your GitHub username via https://forms.gle/zjnrGUo4VWPoWMVh9 (see https://lists.apache.org/thread/x8hwob37pvwrwk2xslpk3ton76bkr5o9 ) * "cross-references" (e.g. issue#x duplicates issue#y) are not represented for now. The problem is the exact issue id is not known before the issue is imported, so it is hard to print all those "duplicates/duplicated by" references. We might probably workaround that by assuming the ids will be sequential, then we could try waiting for each issue import to fully complete. * formatting is weird sometimes. Of course, we can't fix all the cases, however, it might be worth adding workarounds if possible. * milestones are not imported yet. I think we could try creating milestones (~fixed in version) in GitHub, and use them for issues that have "target milestone" set. The attachments live in a separate repository ( https://github.com/vlsi/tmp-jmeter-attachments), so issues could reference them. Note: the attachments repository would be read-only, and it is only for attachments migrated from Bugzilla. Vladimir
