On Wed, Nov 24, 2010 at 01:33:21PM +0800, jida...@jidanni.org wrote: > OK, here are my changes to > ftp://alpha.gnu.org/gnu/units/units-1.88a.tar.gz to add the Taiwanese > units. The Japanese people I Cc'd might consider doing the same. > > Note I only added units that seem to me to be still in use.
There's no need to be so restrictive---units.dat is packed with historic and archaic units. But it's up to you if you want to go there. I've found that dealing with historic units can be difficult. > Dear units(1) author: Consider removing trailing whitespace from all > files. Date in units.man should be updated. Some files are not 644 mode > but 600! units --check "says Table 'ansicoated' lacks..." The date in units.man reflects, I believe, the last date I made nontrivial changes to the manual. Is there some reason it should be some other date? The situation with ansicoated is problematic. The difficulty is that the actual official definition for the ANSI grit size is noninvertible. There are two different scales, one for coarser grits and one for finer ones and the two scales overlap a bit and disagree where they overlap. I added group and other read permission to two files, configure.ac and texi2man to address your other observation. It appears that emacs is inserting a single space sometimes when I use autofill to write comments. Is this the white space you're suggesting that I remove? (I don't see that it presents a problem.) Or is there something more troublesome somewhere? Do you find that the utf8 support in the new version is working properly? One thing I don't understand is how chinese characters are supposed to interact with western characters. It seems like when they are mixed, columns don't line up properly. (An example would be when you type "?" at the "You have" prompt to get a list of conformable units.) I took a look at your patch and have two concerns. One is that I have utf8 stuff placed inside !utf8 / !endutf8 pairs so that the non-ASCII definitions can be ignored when utf8 support is not available. See the large block of utf8 definitions at the end of the file included in the alpha distribution of the code. (It's fine to use extended chars in comments outside of these blocks.) Either we can wrap each of the three new sections with this, or we can group the new sections into one section where the taiwanese character appear. I don't know which makes more sense. The other things was here: > +# http://en.wikipedia.org/wiki/Taiwanese_units_of_measurement > +坪 tsubo # http://zh.wikipedia.org/wiki/坪 > +甲 2934 坪 # http://zh.wikipedia.org/wiki/甲_(单位) > +分地 1|10 甲 # http://zh.wikipedia.org/wiki/分 #+地 to > disambiguate > + I format units.dat so that it displays in an 80 (western) character wide window. This last line is too long, and I don't understand what's being disambiguated, so I'm not sure of the best fix. Also, do you have any idea where 2934 came from? It's a strange number and for the most part, units definitions don't include strange numbers like that. (In other words, can this unit be defined in a way that is less strange?) -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
