On 16/07/07, Lübbe Onken <email@example.com> wrote:
> Hi Simon,
> You wrote:
> > Thanks for sorting this out, Stan.
> Yes, well done Stan :)
> > One more request, which is not
> > trivial enough for me to do myself. I want to exclude tags of the form
> > <tagname condition="pot"> for example <entry condition="pot">, which
> > are excluded from the translation because they are parameter or
> > switch names.
> Simon, The idea to use the condition="pot" attribute to exlude strings from spell checking is nice, but I'm afraid it breaks the translation, because the entire tag gets removed when the .pot file is created.
> Example from tsvn_app_automation.xml:
> <option condition="pot">/closeonend:0</option> don't close the dialog automatically
> Gets reduced to:
> don't close the dialog automatically
> by the .pot profiler. It's the way docbook profiling works, sorry. It drops entire tags. For the translation the full string including the <option> tag is used, which means that xml2po.py doesn't find a match.
Are you saying that condition="pot" only works properly when used
within a table where we can exclude the entire paragraph/entry?
My idea was that "/closeonend:0" should be excluded from the
translation because that must never be translated. But I guess that is
obvious enough to the translators. Getting it excluded from the
spellcheck too was a bonus.
> 1) A way to overcome this would be to tell xml2po.py that <option> (and maybe others) is a tag that has to be extracted separately. This would allow us to use the condition="pot" attribute
> 2) Another solution is to either use Stans tag filter and add more tags
> 3) or to use a different attribute for sections that should not be spell checked and teach the filter to exclude those attributed tags. In the example above:
> <option spellcheck="false">/closeonend:0</option> don't close the dialog automatically
> would still include the entire paragraph in the .pot file. The translator has to copy & paste the <option> tag, but at least he knows what he is translating.
> Three ways to solve the problem. I think the effort is about the same for each direction. Which path should we follow? My personal favourite is 3) because I think that some tags should go into the .pot file, but not be spell checked.
We represent command line options and property names in several
different and inconsistent ways in the docs:
For most app-automation we use a table with condition="pot" for the
option name entry.
For other app-automation parts we put it in an <option> tag.
For some properties we describe them in a <variablelist> and put the
property name inside a <term> tag
For other properties we just include the property name in the text in
a <literal> tag.
Are we doing this wrong, or do we just need to handle all those
different ways in our spellchecking?
We still need to exclude tags condition="pot" from spellchecking,
because most of the app automation switches use those, unless we also
put them inside a <literal> tag.
A relatively simple solution is to exclude the <option> tag from the
spellcheck, and use a <literal> tag inside the <term> tag for the
oo // \\ "De Chelonian Mobile"
(_,\/ \_/ \ TortoiseSVN
\ \_/_\_/> The coolest Interface to (Sub)Version Control
/_/ \_\ http://tortoisesvn.net
To unsubscribe, e-mail: firstname.lastname@example.org
For additional commands, e-mail: email@example.com
Received on Mon Jul 16 12:34:09 2007