Hi James, My experience might be a bit old. I seem to recall, way back when, we did try to build some plugins outside of Drill itself and that there were issues. Maybe it was just the inconvenience of debugging? Perhaps the test libraries were not available? Development is fastest when you can write a unit test that fires up Drill, and exercises your plugin. You can then step through the code, see an error, fix it, and try again in a matter of seconds. Without that, you have to rebuild your jar, copy it to Drill, restart Drill, submit a query, and hope to figure out what is wrong when things blow up.
So, I wonder if we also publish test jars? If not, that would be a big help. UDFs also have issues since Drill doesn't actually run your code: Drill copies it. And, unless you know about the magic thingie, Drill won't even load your UDF. (Have to tell Drill not to load from cache, if I recall.) To test all this out, just build a demo plugin and demo UDF using the libraries. If it is smooth sailing, we're good to go. If not, figure out what's missing and fix it. Oh, and another issue: class loader isolation. As Drill includes ever more plugins, dependencies will conflict. That's why Presto/Trino loads plugins in a separate class loader: Trino may use version 5 of library X, but I might use 7. With class loader isolation, stuff just works. Without it, one lives in Maven dependency hell for a while. Thanks, - Paul On Tue, Jan 18, 2022 at 12:29 AM James Turton <dz...@apache.org> wrote: > For my part, I'd forgotten that GitHub does give users the opportunity > to attach binary distributables to releases. So my first thought of > "GitHub would mean using Git repositories to host Jar files" was off the > mark. > > Paul, setting aside the hosting and distribution for a moment, may I ask > about the statement "ensure plugins can be built outside of the Drill > repo"? Released versions of Drill's own libs are already published to > Maven. E.g. > > > https://mvnrepository.com/artifact/org.apache.drill.exec/drill-java-exec/1.19.0 > > Can a plugin writer not create a new project which lists the required > Drill libs in its pom.xml deps and proceed to build a plugin away from > the main tree? Interactive debugging without the Drill main tree should > even be possible by attaching a debugger to a running embedded Drill > with the storage plugin deployed to it, or am I wrong here? > > On 2022/01/18 00:32, Paul Rogers wrote: > > Hi Ted, > > > > Thanks for the explanation, makes sense. > > > > Ideally, the client side would be somewhat agnostic about the repo it > pulls > > from. In a corporate setting, it should pull from the "JFrog Repository" > > that everyone seems to use (but which I know basically nothing.) Oh, > lord, > > a plugin architecture for the repo for the plugin architecture? > > > > - Paul > > > > On Mon, Jan 17, 2022 at 1:46 PM Ted Dunning <ted.dunn...@gmail.com> > wrote: > > > >> Paul, > >> > >> I understood your suggestion. My point is that publishing to Maven > >> central is a bit of a pain while publishing by posting to Github is > nearly > >> painless. In particular, because Github inherently produces a > relatively > >> difficult to fake hash for each commit, referring to a dependency using > >> that hash is relatively safe which saves a lot of agony regarding keys > and > >> trust. > >> > >> Further, Github or any comparable service provides the same "already > >> exists" benefit as does Maven. > >> > >> > >> > >> On Mon, Jan 17, 2022 at 1:30 PM Paul Rogers <par0...@gmail.com> wrote: > >> > >>> Hi Ted, > >>> > >>> Well said. Just to be clear, I wasn't suggesting that we use > >>> Maven-the-build-tool to distribute plugins. Rather, I was simply > observing > >>> that building a global repo is a bit of a project and asked, "what > could we > >>> use that already exists?" The Python repo? No. The > Ubuntu/RedHat/whatever > >>> Linux repos? Maybe. Maven's repo? Why not? > >>> > >>> The idea would be that Drill might have a tool that says, "install the > >>> FooBlaster" plugin. It downloads from a repo (Maven central, say) and > puts > >>> the plugin in the proper plugins directory. In a cluster, either it > does > >>> that on every node, or the work is done as part of preparing a Docker > >>> container which is then pushed to every node. > >>> > >>> The key thought is just to make the problem simpler by avoiding the > need > >>> to create and maintain a Drill-specific repo when we can barely have > enough > >>> resources to keep Drill itself afloat. > >>> > >>> None of this can happen, however, unless we clean up the plugin APIs > and > >>> ensure plugins can be built outside of the Drill repo. (That means, > say, > >>> that Drill needs an API library that resides in Maven.) > >>> > >>> There are probably many ways this has been done. Anyone know of any > good > >>> examples we can learn from? > >>> > >>> Thanks, > >>> > >>> - Paul > >>> > >>> > >>> On Mon, Jan 17, 2022 at 9:40 AM Ted Dunning <ted.dunn...@gmail.com> > >>> wrote: > >>> > >>>> I don't think that Maven is a forced move just because Drill is in > Java. > >>>> It may be a good move, but it isn't a forgone conclusion. For one > thing, > >>>> the conventions that Maven uses are pretty hard-wired and it may be > >>>> difficult to have a reliable deny-list of known problematic plugins. > >>>> Publishing to Maven is more of a pain than simply pushing to github. > >>>> > >>>> The usability here is paramount both for the ultimate Drill user, but > >>>> also for the writer of plugins. > >>>> > >>>> > >>>> > >>>> On Mon, Jan 17, 2022 at 5:06 AM James Turton <dz...@apache.org> > wrote: > >>>> > >>>>> Thank you Ted and Paul for the feedback. Since Java is compiled, > Maven > >>>>> is probably better fit than GitHub for distribution? If Drillbits > can > >>>>> write to their jars/3rdparty directory then I can imagine Drill > gaining > >>>>> the ability to fetch and install plugins itself without too much > >>>>> trouble, at least for Drill clusters with Internet access. > >>>>> "Sideloading" by downloading from Maven and copying manually would > >>>>> always remain possible. > >>>>> > >>>>> @Paul I'll try to get a little time with you to get some ideas about > >>>>> designing a plugin API. > >>>>> > >>>>> On 2022/01/14 23:20, Paul Rogers wrote: > >>>>>> Hi All, > >>>>>> > >>>>>> James raises an important issue, I've noticed that it used to be > easy > >>>>> to > >>>>>> build and test Drill, now it is a struggle, because of the many odd > >>>>>> external dependencies we have introduced. That acts as a big damper > on > >>>>>> contributions: none of us get paid enough to spend more time > fighting > >>>>>> builds than developing the code... > >>>>>> > >>>>>> Ted is right that we need a good way to install plugins. There are > two > >>>>>> parts. Ted is talking about the high-level part: make it easy to > >>>>> point to > >>>>>> some repo and use the plugin. Since Drill is Java, the Maven repo > >>>>> could be > >>>>>> a good mechanism. In-house stuff is often in an internal repo that > >>>>> does > >>>>>> whatever Maven needs. > >>>>>> > >>>>>> The reason that plugins are in the Drill project now is that Drill's > >>>>> "API" > >>>>>> is all of Drill. Plugins can (and some do) access all of Drill > though > >>>>> the > >>>>>> fragment context. The API to Calcite and other parts of Drill are > >>>>> wide, and > >>>>>> tend to be tightly coupled with Drill internals. By contrast, other > >>>>> tools, > >>>>>> such as Presto/Trino, have defined very clean APIs that extensions > >>>>> use. In > >>>>>> Druid, everything is integrated via Google Guice and an extension > can > >>>>>> replace any part of Druid (though, I'm not convinced that's actually > >>>>> a good > >>>>>> idea.) I'm sure there are others we can learn from. > >>>>>> > >>>>>> So, we need to define a plugin API for Drill. I started down that > >>>>> route a > >>>>>> while back: the first step was to refactor the plugin registry so it > >>>>> is > >>>>>> ready for extensions. The idea was to use the same mechanism for all > >>>>> kinds > >>>>>> of extensions (security, UDFs, metastore, etc.) The next step was to > >>>>> build > >>>>>> something that roughly followed Presto, but that kind of stalled > out. > >>>>>> > >>>>>> In terms of ordering, we'd first need to define the plugin API. > Then, > >>>>> we > >>>>>> can shift plugins to use that. Once that is done, we can move > plugins > >>>>> to > >>>>>> separate projects. (The metastore implementation can also move, if > we > >>>>>> want.) Finally, figure out a solution for Ted's suggestion to make > it > >>>>> easy > >>>>>> to grab new extensions. Drill is distributed, so adding a new plugin > >>>>> has to > >>>>>> happen on all nodes, which is a bit more complex than the typical > >>>>>> Julia/Python/R kind of extension. > >>>>>> > >>>>>> The reason we're where we're at is that it is the path of least > >>>>> resistance. > >>>>>> Creating a good extension mechanism is hard, but valuable, as Ted > >>>>> noted. > >>>>>> Thanks, > >>>>>> > >>>>>> - Paul > >>>>>> > >>>>>> On Thu, Jan 13, 2022 at 10:18 PM Ted Dunning<ted.dunn...@gmail.com> > >>>>> wrote: > >>>>>>> The bigger reason for a separate plug-in world is the enhancement > of > >>>>>>> community. > >>>>>>> > >>>>>>> I would recommend looking at the Julia community for examples of > >>>>>>> effective ways to drive plug in structure. > >>>>>>> > >>>>>>> At the core, for any pure julia package, you can simply add a > >>>>> package by > >>>>>>> referring to the github repository where the package is stored. For > >>>>>>> packages that are "registered" (i.e. a path and a checksum is > >>>>> recorded in a > >>>>>>> well known data store), you can add a package by simply naming it > >>>>> without > >>>>>>> knowing the path. All such plugins are tested by the authors and > the > >>>>>>> project records all dependencies with version constraints so that > >>>>> cascading > >>>>>>> additions are easy. The community leaders have made tooling > >>>>> available so > >>>>>>> that you can test your package against a range of versions of Julia > >>>>> by > >>>>>>> pretty simple (to use) Github actions. > >>>>>>> > >>>>>>> The result has been an absolute explosion in the number of pure > Julia > >>>>>>> packages. > >>>>>>> > >>>>>>> For packages that include C or Fortran (or whatever) code, there is > >>>>> some > >>>>>>> amazing tooling available that lets you record a build process on > >>>>> any of > >>>>>>> the supported platforms (Linux, LinuxArm, 32 or 64 bit, windows, > >>>>> BSD, OSX > >>>>>>> and so on). WHen you register such a package, it is automagically > >>>>> built on > >>>>>>> all the platforms you indicate and the binary results are checked > >>>>> into a > >>>>>>> central repository known as Yggdrasil. > >>>>>>> > >>>>>>> All of these registration events for different packages are > recorded > >>>>> in a > >>>>>>> central registry as I mentioned. That registry is recorded in > Github > >>>>> as > >>>>>>> well which makes it easy to propagate changes. > >>>>>>> > >>>>>>> > >>>>>>> > >>>>>>> On Thu, Jan 13, 2022 at 8:45 PM James Turton<dz...@apache.org> > >>>>> wrote: > >>>>>>>> Hello dev community > >>>>>>>> > >>>>>>>> Discussions about reorganising the Drill source code to better > >>>>> position > >>>>>>>> the project to support plug-ins for the "long tail" of weird and > >>>>>>>> wonderful systems and data formats have been coming up here and > >>>>> there > >>>>>>>> for a few months, e.g. inhttps:// > github.com/apache/drill/pull/2359. > >>>>>>>> > >>>>>>>> A view which I personally share is that adding too large a number > >>>>> and > >>>>>>>> variety of plug-ins to the main tree would create a lethal > >>>>> maintenance > >>>>>>>> burden for developers working there and lead down a road of > >>>>> accumulating > >>>>>>>> technical debt. The Maven tricks we must employ to harmonise the > >>>>>>>> growing set of dependencies of the main tree to keep it buildable > >>>>> are > >>>>>>>> already enough, as is the size of our distributable and the count > of > >>>>>>>> open bug reports. > >>>>>>>> > >>>>>>>> > >>>>>>>> Thus, the idea of splitting out "/contrib" into a new > >>>>>>>> apache/drill-contrib repo after selecting a subset of plugins to > >>>>> remain > >>>>>>>> in apache/drill. I'll now volunteer a set of criteria to decide > >>>>> whether > >>>>>>>> a plug-in should live in this notional apache/drill-contrib. > >>>>>>>> > >>>>>>>> 1. The plug-in queries an unstructured data format (even if it > >>>>> only > >>>>>>>> reads metadata fields) e.g. Image format plug-in. > >>>>>>>> 2. The plug-in queries a data format that was designed for > human > >>>>>>>> consumption e.g. Excel format plug-in. > >>>>>>>> 3. The plug-in cannot be expected to run with speed and > >>>>> reliability > >>>>>>>> comparable to querying structured data on the local network > >>>>> e.g. > >>>>>>>> Dropbox storage plugin. > >>>>>>>> 4. The plug-in queries an obscure system or format e.g. we > >>>>> receive a > >>>>>>>> plug-in for some data format used only on old Cray > >>>>> supercomputers. > >>>>>>>> 5. The plug-in can for some reason not be well supported by the > >>>>> Drill > >>>>>>>> devs e.g. it has a JNI dependency on some difficult native > >>>>> libs. > >>>>>>>> > >>>>>>>> Any one of those suggests that an apache/drill-contrib is the > better > >>>>>>>> home to me, but what is your view? Would we apply significantly > >>>>> more > >>>>>>>> relaxed standards when reviewing PRs to apache/drill-contrib? > >>>>> Would we > >>>>>>>> tag, build and test apache/drill-contrib with every release of > >>>>>>>> apache/drill, or would it run on its own schedule, perhaps with > >>>>> users > >>>>>>>> downloading builds made continuously from snapshots of HEAD? > >>>>>>>> > >>>>>>>> > >>>>>>>> Regards > >>>>>>>> James > >>>>>>>> > >>>>>>>> > >>>>>>>> > >>>>> > >
