Start a new topic

Automatic change of numbers

ich hab doch noch eine Frage zu einem nervenden Detail, das ich nicht in den Griff bekommen.
Bei der Übersetzung von Kommazahlen aus dem Englischen oder Französischen ins Deutsche, soll sich ja der Punkt zum Komma ändern, also z. B. 7.5 zu 7,5.
Selbst wenn ich das beim Korrekturlesen manuell ändere und als „checked segment“ ins Memory gebe, steht danach wieder der Punkt da. Mit welcher Einstellung kann ich das unterbinden?


I still have a question about an annoying detail that I can't get to grips with.
When translating comma numbers from English or French into German, the point is supposed to change to the comma, e.g. 7.5 to 7.5.
Even if I change this manually during proofreading and put it into memory as a "checked segment", the dot is there again. With which setting can I prevent this?

I would check in Preferences > Auto-assembling (especially Transfer numbers to segments and Fuzzy match auto-correction, try with these OFF).


(At least some) Auto-assembling settings are also used for (fuzzy) match transferring.


Note to Igor: Especially, I don't think Fuzzy match auto-correction should be enabled by default…


Additionally, if these are repeated segments, also try disabling numbers and non-translatables from Auto-propagations settings.

From Anthony Rudd: 13. Reformat number In many cases, the number formatting differs between the source and target languages, for example, the significance of periods and commas is reversed between English and German numbers (123,456.78 → 123.456,78). In this example, three capturing groups are used on the Replace string. Because only the content of each “number block” is required, each must be matched as separate capturing group; for clarity, each “number block” is written within parentheses. Regex: \b(\d{1,3})(,(\d{3}))(\.(\d+))\b Replace string: $1.$3,$5 Test string: 12,345.6 Result string: 12.345,6 Because the “Replace string” does not permit any flexibility, the “Regex” / “Replace string” pair must be appropriate for the matched data; if necessary, multiple “Regex” / “Replace string” pairs for all possibilities. must be executed. Tip: To avoid programming multiple regexes to handle the various formats, it may be better to program a more general regex with optional fields, and then “clean up” the results, as shown below. General regex: \b(\d{1,3})(,(\d{3}))?(,(\d{3}))?(\.(\d+))?\b Replace string: $1.$3.$5,$7 Test string: 1,234,567.89 12,345.6 12 Result string: 1.234.567,89 12.345.,6 12.., Whereby, the erroneous separators “.,” and “..,” can be cleaned up with the following Find regex (note the use of the “null capturing group” that is required to reference an “empty” value): (?(\d*\.\d+)*)(\.+(?,\d+)|\.+,(?)) and Replace string: ${ho}${lo} Test string: 1.234.567,89 12.345.,6 12.., Result string: 1.234.567,89 12.345,6 12 Note: The “General regex” can be extended using the same scheme to handle additional leading “thousand blocK How about building this into CT as a QA
Not in the QA menu but in the Task menu. And the Format numbers should take the language direction from the project into account.

Here is Anthony's info in a better legible form. I'll also attach the great PDF with updated techniques. And I'll file to requests for enhancements.


image

image


pdf
Login to post a comment