janl opened a new pull request, #5314: URL: https://github.com/apache/couchdb/pull/5314
This PR combines work from #4305 (@nickva) #5146 (@big-r81) & and #5306 (@indutny) plus some more original work from @indutny to swap out our usage of our [homegrown `seal()`](https://github.com/apache/couchdb/blob/main/src/couch/priv/couch_js/86/main.cpp#L214-L228) function for [`Object.freeze()`](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Object/freeze#deep_freezing). Compiles and `make check` passes on macOS with SpiderMonkey 91, 102, 115 and 128. It’d be great if we can test this on other systems as well and double check that `Object.freeze()` has the security properties be want (but I think it does, it might even be stricter than before, not allowing reassignment of values to existing properties). The underlying issue is that starting with SpiderMonkey 102 one of our original `seal()` calls would seal the [`State.funs` array](https://github.com/apache/couchdb/blob/main/share/server/state.js#L16), which then makes a subsequent [`State.funs.push()`](https://github.com/apache/couchdb/blob/main/share/server/state.js#L32) fail with the type error reported in #4305 and #5146. (h/t to @indutny for teasing this out). This [should unblock support for Ubuntu 24.04](https://github.com/apache/couchdb-pkg/issues/136), which ships with SpiderMonkey 128. It feels like the build args lists in `src/couch/rebar.config.script` are due for a refactor, but I didn’t want to mess too much with the shape of the project just yet. Similarly, the differences in `couch_js/main.cpp` and `couch_js/util.cpp` from 91 to 128 are not a lot and one could argue we should start switching to `ifdef` the relevant bits, rather than duplicating a lot of code. On the other hand, this will make it easier to drop versions later, if we ever want to. Thanks all for doing the hard work here, I’m just pulling it all together to validate this is (hopefully) a way forward ✌️ -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
