7.4. Caveats¶
7.4.1. [% %] and [%+ %]¶
Generally, you should follow the English model, but if you need to change the order of [% %] in a line, you should be aware of those rules: when two [% %] members are following each other, the second member will not be separated with a space from the first one, even if you separate them with a space:
[% ELSIF message_tag == "bug_duplicate_of" %]
This [% terms.bug %] has been marked as a duplicate of [% terms.bug %] [% dupe_of FILTER html %]
For instance, the previous line of code will display in the browser as:
This bug has been marked as duplicate of bug12345
Instead you should add a "+" sign in the second member:
[% ELSIF message_tag == "bug_duplicate_of" %]
This [% terms.bug %] has been marked as a duplicate of [% terms.bug %] [%+ dupe_of FILTER html %]
Will then be displayed as:
This bug has been marked as duplicate of bug 12345
This is the same when a [% %] member is at the beginning of a new line:
[% IF groups_added_to.size %]
<li>
The account has been added to the
[% groups_added_to.join(', ') FILTER html %]
group[% 's' IF groups_added_to.size > 1 %].
</li>
[% END %]
The previous code will be shown as:
The account has been added to thebz_sudo_protect group.
You should instead put a "+" sign:
[% IF groups_added_to.size %]
<li>
The account has been added to the
[%+ groups_added_to.join(', ') FILTER html %]
group[% 's' IF groups_added_to.size > 1 %].
</li>
[% END %]
So that the sentence is displayed as:
The account has been added to the bz_sudo_protect group.
If one wants to localize the previous example into French for instance, the words order is not correct and the members have to me moved. The member [% groups_added_to.join(', ') FILTER html %] will actually display the name of one group or several group names separated with a comma and a space (', '). The member [% 's' IF groups_added_to.size > 1 %] will add an "s" to the word "group" if there is more than one. In French, the group name should be put after the term "group" and words need to be declensed when plural is used. That would give for instance the following:
[% IF groups_added_to.size %]
<li>
Le compte a été ajouté
[% IF groups_added_to.size > 1 %]
aux groupes[% ELSE %]au groupe[% END %][%+ groups_added_to.join(', ') FILTER html %].
</li>
[% END %]
The browser would then display for one group:
Le compte a été ajouté au groupe bz_sudo_protectAnd for several groups:
Le compte a été ajouté aux groupes bz_sudo_protect, canconfirm, editbugs7.4.2. Double quotes and single quotes¶
You should take care of the usage of the quotes in certain cases. You shouldn't put quotes inside a pair of quotes. The following example will break the user interface:
[% ELSIF message_tag == "buglist_adding_field" %]
[% title = "Adding field to search page..." %]
[% link = "Click here if the page "does not" redisplay automatically." %]
Instead, you can escape them with a backslash (""):
[% ELSIF message_tag == "buglist_adding_field" %]
[% title = "Adding field to search page..." %]
[% link = "Click here if the page \"does not\" redisplay automatically." %]
Or you can substitute the surrounding double quotes with single quotes:
[% ELSIF message_tag == "buglist_adding_field" %]
[% title = "Adding field to search page..." %]
[% link = 'Click here if the page "does not" redisplay automatically.' %]
7.4.3. Declensions¶
English only deals with one plural form and has no declension. Your locale might need to implement declensions and reorder words inside a sentence.
Warning
this is a bit tricky and making mistake here will break your Bugzilla localization (i.e. some features might not function when your locale package will be installed).
Let's say we have the following:
[% IF !Param("allowbugdeletion") %]
<p>
Sorry, there
[% IF comp.bug_count > 1 %]
are [% comp.bug_count %] [%+ terms.bugs %]
[% ELSE %]
is [% comp.bug_count %] [%+ terms.bug %]
[% END %]
pending for this component. You should reassign
[% IF comp.bug_count > 1 %]
these [% terms.bugs %]
[% ELSE %]
this [% terms.bug %]
[% END %]
to another component before deleting this component.
</p>
[% ELSE %]
Here, the following expression comp.bug_count obviously gives the count number of bugs for a component. IF comp.bug_count > 1 means "if there are more than one bug".
Let's say your language has to deal with three plural forms and that the terms "bug" and "pending" should be declensed as well.
First, you'll have to populate the /template/en/default/global/variables.none.tmp file with the declensions for "bug", which would give something like:
[% terms = {
"bug0" => "declension for zero bug",
"bug" => "declension for one bug",
"bug2" => "declension for two bugs",
"bug3" => "declension for three bugs",
"bugs" => "declension for more than three bugs",
Then, the previous code should look like:
[% IF !Param("allowbugdeletion") %]
<p>
Sorry, there
[% IF comp.bug_count > 3 %]
are [% comp.bug_count %] pending [% terms.bugs %]
[% ELSE %]
[% IF comp.bug_count == 0 %]
is [% comp.bug_count %] pending [% terms.bug0 %]
[% ELSE %]
[% IF comp.bug_count == 1 %]
is [% comp.bug_count %] pending [% terms.bug %]
[% ELSE %]
[% IF comp.bug_count == 2 %]
are [% comp.bug_count %] pending [% terms.bug2 %]
[% ELSE %]
[% IF comp.bug_count == 3 %]
are [% comp.bug_count %] pending [% terms.bug3 %]
[% END %]
for this component. You should reassign
[% IF comp.bug_count > 1 %]
these [% terms.bugs %]
[% ELSE %]
this [% terms.bug %]
[% END %]
to another component before deleting this component.
</p>
[% ELSE %]
7.4.4. $terms.foo¶
As seen previously, term substitutions are made across all templates files with help of *.none.tmpl files, which are:
- template/en/default/global/field-descs.none.tmpl
- template/en/default/global/variables.none.tmpl
- template/en/default/global/value-descs.none.tmpl
- template/en/default/global/reason-descs.none.tmpl
- template/en/default/global/setting-descs.none.tmpl
- template/en/default/bug/field-help.none.tmpl
These variables appear in the template files under three different forms: $terms.foo, ${terms.foo} and [% terms.foo %]
Note
If someone could explain me briefly why and when, I would add it there ☺.
During your localizing contribution, you will have to reorganize sentences, and sometimes a variable of the form $terms.foo would end at the end of a sentence which is ended with a dot, like this:
defaultplatform => "Plateforme qui est pré-sélectionnée dans le formulaire de soumission " _
"de $terms.bug.<br> " _
"Vous pouvez laisser ce champ vide : " _
"Bugzilla utilisera alors la plateforme indiquée par le navigateur.",
If you let it that way, the substitution would not take place and the result in the user interface might be quite weird. Instead, change the form $terms.foo, for the form ${terms.foo}, like this:
defaultplatform => "Plateforme qui est pré-sélectionnée dans le formulaire de soumission " _
"de ${terms.bug}.<br> " _
"Vous pouvez laisser ce champ vide : " _
"Bugzilla utilisera alors la plateforme indiquée par le navigateur.",
7.4.5. b[% %]ug¶
Once and a while you would find something like:
A b[% %]ug on b[% %]ugs.debian.org.
You remember that the file variables.none.tmpl holds the substitution for different terms. It is useful if you want to replace all over the user interface for instance the term "bug" with the term "ticket". The release process of Bugzilla has a script that parses all the templates to check if you wrote the bare word "foo" instead of "$terms.foo".
In the example above, we really want to write the term "bug" and we neither want it to be substituted afterwards nor being warned by the checking.
As far as I know, for the moment this checking is working only for the English terms --bug, bugs, Bugzilla, …-- so you can safely ignore the [% %] and localize b[% %]ug but you would keep b[% %]ugs.debian.org unchanged.
This documentation undoubtedly has bugs; if you find some, please file them here.
