Re: [JDBC] FW: Re: [PATCHES] Ant configuration
Just a thought. Why not separate postgres and postgres jdbc in that case? Jayesh -Original Message- From: Bruce Momjian [SMTP:[EMAIL PROTECTED]] Sent: Monday, October 22, 2001 5:56 PM To: Barry Lind Cc: [EMAIL PROTECTED] Subject: Re: [JDBC] FW: Re: [PATCHES] Ant configuration If we can include the ANT libraries in our CVS then my objection to ANT (requiring users to trackdown and download ANT) goes away, and I would then suggest we continue to use ANT for the other reasons you mention. Does anyone know if we could include ANT into our CVS and redistribute it in order to build the jdbc code? Are their license incompatibilities between the Apache License and the PostgreSQL license that would prevent this? I don't think we want to go there. Too much bloat to add it. Looks like 600k. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup.| Drexel Hill, Pennsylvania 19026 ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [JDBC] FW: Re: [PATCHES] Ant configuration
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 23-Oct-2001 Jayesh K. Parayali wrote: Just a thought. Why not separate postgres and postgres jdbc in that case? To be honest, this is the one thing I want to avoid if possible. Its important that the postgres build system builds the 'official' interfaces as well, since it keeps the two connect tightly. I prefer to use ant to build Java code, but to be honest, I really just want to be able to type either of the following, in the src/interfaces/jdbc directory: ant jar or make jar Both should give the same results. (Its kinda silly that the current technique has 'make' call 'ant', but anyways...) The only two issues are multiple build tools and ease for building the jdbc driver for non-UNIX users. And to be more honest, I think we've spent too much time on this as is. If we can't decide as a group, we should either a) vote on it (Least the CVS committers should) or b) leave it for now. Either way, I'm going to stay out of this discussion for now. Virtually, Ned Wolpert [EMAIL PROTECTED] D08C2F45: 28E7 56CB 58AC C622 5A51 3C42 8B2B 2739 D08C 2F45 -BEGIN PGP SIGNATURE- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE71bdxiysnOdCML0URAoXlAJ4w6Nd9pXuCoJAawEpxBaE/DADsRwCfS0dK tjPvtMsIWudhz641Ro12SgE= =/uGM -END PGP SIGNATURE- ---(end of broadcast)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly
Re: [JDBC] FW: Re: [PATCHES] Ant configuration
Added to TODO: o Move to using 'make' rather than 'ant'(?) At least we know it is an issue for final decision. --- -- Start of PGP signed section. On 23-Oct-2001 Jayesh K. Parayali wrote: Just a thought. Why not separate postgres and postgres jdbc in that case? To be honest, this is the one thing I want to avoid if possible. Its important that the postgres build system builds the 'official' interfaces as well, since it keeps the two connect tightly. I prefer to use ant to build Java code, but to be honest, I really just want to be able to type either of the following, in the src/interfaces/jdbc directory: ant jar or make jar Both should give the same results. (Its kinda silly that the current technique has 'make' call 'ant', but anyways...) The only two issues are multiple build tools and ease for building the jdbc driver for non-UNIX users. And to be more honest, I think we've spent too much time on this as is. If we can't decide as a group, we should either a) vote on it (Least the CVS committers should) or b) leave it for now. Either way, I'm going to stay out of this discussion for now. Virtually, Ned Wolpert [EMAIL PROTECTED] D08C2F45: 28E7 56CB 58AC C622 5A51 3C42 8B2B 2739 D08C 2F45 -- End of PGP signed section. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 853-3000 + If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup.| Drexel Hill, Pennsylvania 19026 ---(end of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/users-lounge/docs/faq.html
Re: [JDBC] FW: Re: [PATCHES] Ant configuration
Tom, My reasons for not wanting to depend on Ant are: Many binary distributions of Postgres are not including the JDBC code because of the added overhead of the jdbc build process (i.e. getting Ant installed). Users have complained to the mail lists in the past about the requirement for Ant (they expect make; make install to work for jdbc just like it does for the rest of the postgres). (However I realize there are equally good arguments on why depending on Ant is a good thing.) Thus I thought a good compromise was to include the Ant libraries in postgres for building, however after seeing the size of them, I agree that is clearly not an option. Your rationale with regards to since you require java to build, what is one more dependency is a little flawed. The java dependency is both a build time and runtime dependency. In order to use the result of the build you need java so the expectation that you will have java if you are dealing with jdbc is high (otherwise why would you bother). However the dependency on Ant is only for building. The one thing this thread has made abundantly clear for me it that we are far from agreement on what the ideal solution is (and there probably isn't an ideal solution), thus I feel we should stick with status quo for 7.2 and deal with all of this in a few months after 7.2 is production. thanks, --Barry Tom Lane wrote: Barry Lind [EMAIL PROTECTED] writes: If we can include the ANT libraries in our CVS then my objection to ANT (requiring users to trackdown and download ANT) goes away, and I would then suggest we continue to use ANT for the other reasons you mention. The sheer bulk of the ANT libraries rules that out, even if there weren't a management/synchronization issue: do you want a PG release to be using an older ANT than what you have already installed locally? On the other hand, I can see no reason why we shouldn't say that you *must* have ANT installed to build the JDBC driver. You've gotta have Java to build JDBC, no? Seems like ANT is just one more dependency, and hardly an unreasonable one if it's the standard for Java projects. What's wrong with saying we don't build the JDBC driver if ANT isn't installed? regards, tom lane ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [JDBC] FW: Re: [PATCHES] Ant configuration
Barry Lind writes: If we can include the ANT libraries in our CVS then my objection to ANT (requiring users to trackdown and download ANT) goes away, and I would then suggest we continue to use ANT for the other reasons you mention. My problem with Ant isn't primarily the availability or what's standard or not. My problem is simply that it doesn't work the way I want the build system to work (see recently listed issues such as exit status and multiple build dirs). If we can get these fixed then I have no problem, but otherwise Ant introduces bugs into the build system and those will be fixed, no matter what's the standard today. -- Peter Eisentraut [EMAIL PROTECTED] http://funkturm.homeip.net/~peter ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster
Re: [JDBC] FW: Re: [PATCHES] Ant configuration
If we can include the ANT libraries in our CVS then my objection to ANT (requiring users to trackdown and download ANT) goes away, and I would then suggest we continue to use ANT for the other reasons you mention. Does anyone know if we could include ANT into our CVS and redistribute it in order to build the jdbc code? Are their license incompatibilities between the Apache License and the PostgreSQL license that would prevent this? thanks, --Barry Ned Wolpert wrote: I meant to send this to the group... -FW: [EMAIL PROTECTED]- Date: Fri, 19 Oct 2001 10:45:43 -0700 (MST) Sender: [EMAIL PROTECTED] From: Ned Wolpert [EMAIL PROTECTED] To: Barry Lind [EMAIL PROTECTED] Subject: Re: [JDBC] [PATCHES] Ant configuration -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Two cents with of thought... :-) On 18-Oct-2001 Barry Lind wrote: * People with Java background probably know Ant better that 'make'. Ant seems to become 'build tool of choice' in Java world. Perhaps, but I'm not convinced that either of these arguments is true. I have seen make used by more projects than ant, and a developer building postgres from source, certainly needs to be familiar with make. Ant is great in a pure-java project. I have seen ant used as the primary build engine in many Java projects, and it works great for me. But, here are some (conflicting) thoughts. 1) To some degree, the make process of PostgreSQL should be consistant. Ant and Make are two seperate building systems. Its easier to put the JDBC driver into make than it is to put postgresql into ant. 2) Correct me if I'm wrong, but currently, Make is needed to run Ant in the JDBC driver anyways, right? (To grab the version numbers and set the properties when running ant) Yet... 3) Java is system independant, and the JDBC driver is a client piece. I should be able to compile the JDBC driver outside of PostgreSQL server if I wanted to, right? Example, I can't compile PostgreSQL on windozes 95. But I could compile the JDBC driver via ant if Java and ant are installed. (Or, rather, I should be able to) 4) Ant doesn't need to be 'installed' in other projects. Rather, they include the needed ant libs in the CVS, which is allowed by the license, and create a build script to call the ant process to build. (Usually a build.sh and a build.bat) We could do that to also set the version numbers rather than have make do it. I guess I can make an argument either way. The real way to answer this is to ask if one cares if you can build the JDBC driver on window's only platform without cygwin installed. -If the answer is no, that one wouldn't build the JDBC driver without building the rest of postgresql, then the build process should be tied to make, like the rest of postgresql, for simplisity in maintence. -However, if the JDBC driver should be able to be built without having to build postgresql, (on platforms without UNIX-compatiablity layer like cygwin) for client use only, then yes, use Ant to achive platform-independance. I'm lucky enough to only work with Linux, so I don't really care on the outcome of this :-) I can work with either. Virtually, Ned Wolpert [EMAIL PROTECTED] D08C2F45: 28E7 56CB 58AC C622 5A51 3C42 8B2B 2739 D08C 2F45 -BEGIN PGP SIGNATURE- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE70GbGiysnOdCML0URAnl5AJ9VkJs0QXr1GEzzFVxW5CzNLDRl1wCfbOmM O2tENKYPCsDAx6I42NoYh5U= =CSwI -END PGP SIGNATURE- --End of forwarded message- ---(end of broadcast)--- TIP 2: you can get off all lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED]) ---(end of broadcast)--- TIP 6: Have you searched our list archives? http://archives.postgresql.org
