On Thu, Nov 04, 2004 at 01:18:26AM +0000, Yves Rutschle wrote: Hello, The story of the HTML module is that it is the very first module which were written by someone else than me. At one point, Laurent wrote in the documentation that it was almost done. But later, when I reviewed the code I discovered that the way it did split the sentences makes it very hard to use. "I <b>like</b> it" is splitted in 3 msgid which have to be translated separately ("I" ; "like" and "it"). See http://po4a.alioth.debian.org/en/po4a.7.html#Why_not_to_split_on_ and http://graal.ens-lyon.fr/~mquinson/l10n.html#l2.2 for my point on this. I should have explained my point to Laurent before... That's why I added the comment about the useless state and decided not to distribute it. Nowadays, this module should probably be reimplemented using Jordi's great work on XML-like formats. done by Denis, so no comment (beside "thanks"). I guess you speak of: https://savannah.nongnu.org/projects/po4a The only description of the package is: This project moved to http://alioth.debian.org/projects/po4a/ I agree that we should turn the project completely off. I should contact savannah's admin about this. (I used a bug on savannah to remove all developpers of the project, which thus cannot be modified anymore without admin intervention. What a lamer). The letters PO in `.po' files means Portable Object, to distinguish it from `.mo' files, where MO stands for Machine Object. (info gettext) consider this as a [meaningless] legacy of the eighties or even before. the T of pot stands for template. Have fun around, Mt.

On Sun, 7 Nov 2004, Martin Quinson wrote: The first result on Google points to http://www.nongnu.org/po4a/, which seems to be the same as the "Homepage" link at the top of the savannah project (which points to http://www.freesoftware.fsf.org/po4a but has the same content). There's some old po4a documentation. There should also be a link to the alioth project. Regards, Jordi Vilalta

On Sun, Nov 07, 2004 at 12:37:57AM +0100, Martin Quinson wrote: That's not actually the reason I found it useless as is: the current CVS version slices paragraphs randomly (well, on 512 bytes boundaries or something like that), which means that irrelevant changes in formatting anywhere in the file fuzzify the entire file. Yes, I have actually run into a couple of those problems myself. While the splitting in 3 as in your above example is, indeed, a bit confusing, I don't find that makes it useless, and more to the point, I just don't think there is any other good solution: the bottom line is, you want to specify that something in that sentence is important, which will need to be in a different msgid. The solution, I find, is to have the translator understand the structure of the original text so she'll know to translate: "It's a <b>blue</b> car" => ("It's a", "blue", "car" ) into: "C'est une voiture <b>bleue</b>" => ( "C'est une voiture", "bleue", " ") I guess an alternative would be to have a list of "small formatting tags" ( bold, italics etc) that do not actually split at all, and appear in the msgid with the onus on the translator to know enough HTML to know what to do with them (so you'd have something like: msgid "It's a <b>blue</b> car" msgstr "C'est une voiture <b>bleue</b>" That would have the advantage of providing the translator with context information. In fact that goes a long way towards your point of splitting at paragraph level :-) That's actually fairly easily achievable: the list of paragraph-marking tags is fairly small (<p>, <div>, <h1,2,3,4,...>) and XHTML makes it mandatory for text to be included in a block-level element of some sort. I know next to nothing about XML; last time I saw some, I thought it looked quite different from HTML. A quick read of Jordi's module makes me think it's mostly an XML parser: Html.pm relies on Gisle Aas' HTML parser, and it doesn't seem to be very beneficial to change parsers just for fun; besides, Gisle's parser is supposed to be quite good at handling broken HTML, which I doubt XML is very good at (then again, helping bad HTML spread probably isn't good :) Y.

On Sun, Nov 07, 2004 at 01:05:12AM +0100, Jordi Vilalta wrote: There is no developper of the po4a project anymore (I removed everybody, including myself). I asked the savannah admin to kill the project for us, since they are the only ones who can help us now. https://savannah.gnu.org/support/index.php?func=detailitem&item_id=10... Bye, Mt.

On Mon, Nov 08, 2004 at 02:22:42PM +0100, Martin Quinson wrote: Thank you for sharing your experience; I'm getting convinced now. I read that article a long time ago, printed in a real paper version of TPJ... I think that's actually the most single interesting article I read in all TPJs :) [splitting in HTML blocs] Ok. Well, I'm afraid that means I'm gonna have to ditch the current Html.pm and redo one from scratch (bar a couple of routines that may be recued). So, we'll now be cutting the html along blocks and display formatting tags inline (at first sight, it looks like cutting along tags that have a 'display: block' property, while keeping those that have a 'display: inline' property). While thinking about it there is at least one thing I'd like feedback on: I'd personally rather not expose "complicated" tags to the translator, i.e. while I think it's acceptable to present them with <b> and <i> and so on, I don't think something like: This is a <a href="blahblah.com/this/that/blah.html">link</a> to <img src="blahblah.com/this/that/blah.png" alt="blah" title="Blah"> belongs in a PO. So I'd propose to collapse the inside of long inline tags, so as to simply state there is a tag (e.g. "you're in a link") without detailing what the tag contains. Thus, the example line would appear, in the PO, as: This is a <a>link</a> to <img>blah</img> (Meanwhile we also output the title field of the img as a separate msgid; the alt field is a replacement for the image for text browsers, and therefore belongs interpolated in the rest of the text). One argument to expose the full tag would be that it allows the translator to update links (change a link to blah.html into a link to blah.fr.html for example), allowing the complete translation of a Web site. I'm not fond of the idea though, as: - The tranlator doesn't necessarily know how the translation would be implemented - The burden of maintaining the Web site should not be on the translator - A small script can easily take care of that (I'll be happy to provide what I've written later on, but I'm not sure it belongs in po4a). Any comments on this? [HTML::Parser vs Jordi's XML parser] Me too, but I hate reimplementation of code (reinventing the wheel) more. Besides HTML::Parser is also quite widely spread, and only one apt-get away at worse (or emerge, or whatever -- if it's not one command away, you need a better distribution :) ) Y.