If there was a zip file that included an open source compiler and linker and a well commented makefile along with SQLite's source code so that anyone (at least under Windows) can generate the version SQLite dll and command shell that they want without having to search for tools, I think more people would be more likely to experiment with some of the non standard features.
Vance Jay A. Kreibich wrote: > Grrr.... I didn't meant to send this just yet. But since I did, I > guess I need to finish it. > > On Thu, Jul 23, 2009 at 10:50:37AM -0500, Jay A. Kreibich scratched on the > wall: >> On Thu, Jul 23, 2009 at 09:46:24AM -0400, Wilson, Ron P scratched on the >> wall: >> >>> I think the OP just has the wrong expectations. >> Yes, and no. While SQLite doesn't have a one-click-to-install >> download, I have to agree that the current build and distribution >> state of SQLite is... let's just say "less than ideal." I've been >> writing a lot of documentation on just this issue, and unless you >> want a perfect vanilla install, there are definitely a lot of hoops >> you have to jump through compared to most open-source projects of >> similar design. >> >> A few versions ago we transitioned from a traditional UNIX style >> project, complete with "configure" script, to having the amalgamation >> be the "standard" distribution. I've always felt like that >> transition is incomplete, and we've never gotten back to where we >> were before. >> >> The amalgamation works well enough if what you want is mostly >> defaults. The issue is that, while you can change a few of the >> #defines for numeric defaults, most of the more interesting build >> options won't work with the amalgamation. Only that's it. As the >> website clearly states, there is no other supported option. > > The "by the file" distribution is bad enough, but you're totally out > of luck if you need to go to the tree for some of the really complex > build options. Of course, the "by the file" distribution is there > and available for download because a lot of people still need it, but > apparently not enough to justify keeping it updated. That's a bit of > a contradiction... for a piece of software that prides itself on its > testing systems, the end-users sees a whole lot of "there but not > supported; it might work it might not; you're on your own" stuff. > That's normally a big red flag in my book. Stuff should be there, be > supported, and be documented, or it shouldn't. The current situation > is only easily understandable if you've been following SQLite for a few > years. > > And the OP is right... the build docs suck. There are no build docs > for most downloads, just a archive file with source. No Makefiles, no > nothing. There is no "how to build" on the documentation page (just > docs on build options) and the Wiki pages on building and the > amalgamation are so out of date that they have negative value, doing > little more than confusing people. > > I realize that you might take the argument that anyone smart enough > need to build the SQLite engine into an application should be able to > figure that out, and with enough time, maybe that's right a fair > percentage of the time. On the other hand, the whole point of moving > to the amalgamation was to simplify putting SQLite into an > application. I think overall it does this, but only if you're > working from the default everything. The amalgamation might be > easier or better (for some definition of those terms) if you > understand it, but it is radically different from nearly every other > open source project out there. A bit of hand holding, even for > experienced developers, is not out of place. > > SQLite is a great product, but there is still a lot of room for > improvement on the packaging, distribution, and "productization" > of the code. > > > And even if his post was a bit negative, I'm a little disappointed at > the community response. We're here to help, not criticize. If that > was my introduction to the product and its users, you can be sure I > wouldn't be coming back. If you're not going to try to add something > to the conversation, there is no obligation to hit reply. > > -j > > _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
