Fixed number rulesets for French, Portuguese, etc.

This collection fixes the flaw in the auto-translation rules included with the current memoQ installations (as of version 10.2) and called "Numbers-FrenchGroup" ruleset. The rules provided by memoQ fail to use non-breaking spaces as digit grouping characters for billions, millions, thousands, etc. in large numbers, which may lead to these breaking across lines in a document. This doesn't look good in most publications :-)

A special set of rules ("Numbers NBSP groups decimal comma with exclusions.mqres") excludes from the conversion certain numbers associated with Figures, etc. The example provided can be extended to suit your particular needs.

Two improved number rulesets for English, DGT (EU) styleguide-compliant numbers included

The number ruleset included with memoQ was overhauled and re-organized to recognize some additional characters used to represent negative numbers and to handle section numbering. These rules are to be selected where the target language is English.

The EU DGT has different specifications for how to write English numbers in their texts. Instead of a comma for digit grouping (thousands, millions, billions, etc.) they use a non-breating space (Unicode value \u00a0). Use this ruleset when doing English translations for the DGT or others with the same style specifications. Unlike the rules to be used typically for French and Portuguese texts, these rules write numbers with a period (".") as the decimal delimiter. This ruleset also includes XML comments as documentation to facilitate re-use for projects such as auto-translation rules for currency expressions. These comments enable such rules to be developed with minimal confusion and greater accuracy.

I generally recommend that auto-translation rules be maintained and documented in this manner. Notepad++ is free and Open Source, and there are many other code editors that can be used for this purpose as well.

Numbers like 1-626-555-1212 may break at the end of a line,

so to prevent that, I created a rule that takes up to five blocks of numbers separated by hyphens and makes the same string with non-breaking hyphens substituted. A technique like this may be useful with phone numbers and some catalog numbers, for example.