gettext-0.14.5

[gerard@…]

Added

[markh@…]

New version.

[markh@…]

Having diffed between this and 0.10.38, 99% of it is changes in the autoconf stuff so that 2.13 and 2.50 are supported and documentation + locale changes. There is literally only one source code change so I don't think it'll cause a problem. It compiles and makes cleanly anyways.

[markh@…]

Done

[gerard@…]

updated in book

[gimli@…]

Bleh they keep releasing new versions :(

[highos@…]

Version increment. A copy was put in LFS FTP conglomeration.

[highos@…]

Version increment. A copy was put in LFS FTP conglomeration.

[highos@…]

Version increment.

[highos@…]

* Bug 392 has been marked as a duplicate of this bug. *

[highos@…]

Version increment (0.12)

[highos@…]

Version increment (0.12.1)

[tushar@…]

The book mentions gettext failures related to java in Ch 5. I have uploaded a patch to the patches site that implements a --without-java switch.

[greg@…]

The book mentions gettext failures related to java in Ch 5. I have uploaded a patch to the patches site that implements a --without-java switch.

The book also recommends not to run the gettext test suite in Ch 5. I assume because you've reopened this bug you're requesting the patch be added to the book? Even tho' you've already mentioned this on lfs-dev and didn't get any response? I object to this. There are already too many patches in core LFS. Only obvious critical stuff should be patched. This is not critical. Maybe we could make a mention of the patch and point to the patches project (like we do for some hints) but not include it in the book "proper".

PS - Have you submitted the patch upstream?

[greg@…]

Version increment (0.13)

[tushar@…]

The book also recommends not to run the gettext test suite in Ch 5.

Though some users may still be interested in running it.

I assume because you've reopened this bug you're requesting the patch be added to the book?

Yes. Mainly because the installed files are linked to libs on the host which does not convey a very clean feeling.

Even tho' you've already mentioned this on lfs-dev and didn't get any response? I object to this.

Does that mean that if something is posted on the dev list, it should not be added to the bugzilla! There is nothing to object, if as an editor you think the bug report is not appropriate, you have the liberty to mark it invalid/won't.

There are already too many patches in core LFS. Only obvious critical stuff should be patched. This is not critical.

There is no such thing as *too many/less* patches.

There are also lot of non critical patches/workarounds in the current book that may not be useful for most of the users (gawk-libexecdir, gcc-libiberty, kbd-more-prgs to name a few). Are you also planning on removing those?

Maybe we could make a mention of the patch and point to the patches project (like we do for some hints) but not include it in the book "proper".

Your choice. Though I don't think there is any need to mention it "improperly" since the patches project has been mentioned in the book for optional patches.

PS - Have you submitted the patch upstream?

Nope, coz the patch is for the configure script, don't have any autotools knowledge to patch configure.ac.

[greg@…]

Does that mean that if something is posted on the dev list, it should not be added to the bugzilla!

No, but I was commenting (ironically) on that fact that stuff in Bugzilla usually ends up back on dev list for discussion. Bit of a merry-go-round :-)

There is no such thing as *too many/less* patches.

In theory, yes. But as has already been proven, quite a few patches from LFS 5.0 were redundant and/or easily replaced by seds.

I still believe Patches in the book are a maintenance burden. And I still believe only obvious/critical/important stuff should be patched. This gettext patch doesn't meet the criteria for direct inclusion IMHO.

(gawk-libexecdir, gcc-libiberty, kbd-more-prgs to name a few). Are you also planning on removing those?

Yes to the gawk-libexecdir. gcc-libiberty is already replaced by a sed. Like I said, we need to draw a line somewhere as to what "official" patches we put into the book.

The danger of putting a pointer to an experimental, optional patch into the book is that it might open up the flood gates. Soon, every man and his dog will want their favourite hack mentioned in the book.

PS - do you know if the patch still works with latest gettext-0.13 ?

[greg@…]

Made note of availability of the disable java patch.

[Matthew Burgess]

Version increment (0.13.1)

[petri.koistinen@…]

Version increment (0.14)

[Matthew Burgess]

Thanks Petri - updating the summary.

[Matthew Burgess]

Version increment

[jim@…]

Version Update - 0.14.2

Updated in multi-arch

[Matthew Burgess]

Version increment (0.14.3). Minor release, this from NEWS:

• Usability improvements in gettextize and autopoint.
• Programming languages support:
  • Scheme: Use the GNU guile definition of format strings.

[Matthew Burgess]

Version increment (0.14.4). Another minor release, fixing autoconf macro compatibility with gcc-4.0.

[jim@…]

Upgrade completed to 0.14.4

[jim@…]

Version Update - 0.14.5