在我们正在处理的语言中,专有名词前有一个(大多情况下)强制性的冠词,很像希腊语。也就是说,“亚伯拉罕和撒拉去了”需要写成:
O Abraam hai i Sara telǎrdǎs
The.M Abraam and the.F Sara went
如果我们输入带有 \k o Abraam\k* 的关键术语,这会破坏排序顺序。如果我手动将 o/i 输入在 \k...\k* 之外,情况会更糟,因为 Paratext 并没有像我期望的那样保持段落完整,而是将段落一分为二,冠词也随之丢失。我猜想一些使用前置敬称的语言也会遇到类似的问题。
目前我能想到 3 种解决方案,这些方案不需要在打印前手动编辑术语表:
- 将冠词包含在关键术语中,并在外部进行排序。这使得查找内容变得有些麻烦。
- 将冠词放在关键术语之后
\k Abraam\k*,o,然后编写某种更改脚本以识别并重新排序它们。
- 扩展
\k 的定义,使其允许一个默认的 USFM3.0 属性,从而解决脚本问题。
(a) 仅将冠词作为属性,它可以是:\k Abraam|o\k*
(b) 或者,可能更倾向于使用 \k Printed form |internal representation\k*,就像 \w 那样,其中内部表示形式用于指导链接和排序。
例如:\k o Abraam|Abraam\k*`
显然,3(b) 需要一些编程工作。我不知道选项 3(a) 会对 Paratext 的内部机制产生什么影响。它会破坏 \w 链接等吗?
我假设我不是第一个遇到这个问题的人。其他人是如何解决这个问题的?
In the language we’re working in, there’s a (mostly) obligatory article before proper names, much like in Greek. I.e, ‘Abraam and Sara went’ would need to be:
O Abraam hai i Sara telǎrdǎs
The.M Abraam and the.F Sara went
If we enter key terms with \k o Abraam\k* then this breaks the sort order. If I manually enter the o/i outside the \k...\k* this is worse, because rather than keeping the paragraph together as I hoped, paratext splits the paragraph in two and the article is lost. I imagine that some languages using prefixed honorifics suffer similar problems.
I can think of 3 solutions for the moment, that don’t require manual glossary editing before we get to printing:
- Include the article in the keyterm and sort externally. This makes finding things a bit of a pain.
- Include the article after the keyterm
\k Abraam\k*,o and then write some kind of changes script to recognise and reorder them.
- Extend the definition of
\k so that it allows a default USFM3.0 attribute, which solves the scripting issues.
(a) just having the article as an attribute, it could be: \k Abraam|o\k*
(b) Alternatively, it might be preferable to have \k Printed form |internal representation\k* like with \w, where the internal representation informs linking and the sort order.
E.g. \k o Abraam|Abraam\k*`
Clearly 3(b) would need some programming effort. I don’t know what option 3 (a) would do to paratext’s internal mechanisms. Would it break the \w linking, etc?
I assume I’m not the first to face this. How have others solved this issue?
机器翻译自 English