Transliteration consistency

Question

In one project we use \tl … \tl* to mark words translaterated from English. These are quite common and include terms such as “incense”, “temple” and “donkey”. We mark them because their spelling differs considerably from vernacular words.

Now I’m wondering if there is a way to check if these words have been marked consistently. As far as I’m aware, the only way is manual: make a list of all words marked with \tl, then search each of them twice:

• search temple

• search \tl temple\tl* and remove the hits from the list.

Any remaining hits (except for those occurring in figure captions, filenames etc.) are probably wrong.

Would there be an easier way to check this?

I don’t know if this use of \tl is common – if so, it might be nice to have this kind of check available ready-made in Paratext. But if this concerns just a few users, building an extra check may not be worth the effort…

Paulus+Kieviet

Answer 1

Hi Paulus+Kieviet,

I’d do a RegexPal Find/Replace something like this:

Find: (?<!\\tl )(incense|temple|donkey)
Replace: \tl \1\tl*

This will find any occurrence of the words ‘incense’, ’temple’, or ‘donkey’ which are not preceded by ‘\tl’.
The replace will add ‘\tl…\tl*’ around whichever word is found.
This simple search may encounter problems if two transliterated words occur in sequence, e.g. ’temple incense’.

LivingField

Comments

Hi LivingField,

Thank you for your reply. This would indeed be a speedier way to find instances without \tl. Thanks for your help!

The line does need some tweaking; after some trial and error I came up with:

Find: (?<!\tl )\b(incense|temple|donkey)

Replace: \tl \1\tl*

The \b is to prevent it from finding “\tl frankincense”.

Inevitably there will be some false positives (like the word “temple” in picture captions, filenames and picture descriptions, and cases in footnotes that are surrounded by +tl … +tl*), but that’s OK as long as I don’t hit “Yes to all”. Of course one could also use RegEx Pal just to find instances, without replacing them straightaway.

Paulus+Kieviet

---

Sorry, I didn’t paste the correct version of the Find and Replace lines in Regex Pal. Eventually I came up with:

Find: (?<!tl )\b(word1|word2|etc)\b

Replace with: \tl \1\tl*

(In the Find box, I’m leaving out the backslash in front of “tl” – this way it will also ignore “\+tl incense” in footnotes. A lot of the transliterated words happen to be explained in footnotes: ephah, talent, denarius etc.etc.

This will work in any language which doesn’t have “tl” word-finally. Few languages allow word-final “tl”, I’d think…)

Paulus+Kieviet

Answer 2

FYI, Regex Pal is in the Main menu, not the Project menu: Paratext menu > Advanced > Regex Pal.

You may want do your own “Mark point in history” before/after making changes in RegexPal. RegexPal DOES mark a point in project history, but it just says “Before Regex Pal Modifications”, which doesn’t tell you what changes were made, so may be difficult to locate in the history if you need to later.

james_post