SubtitleEdit/Dictionaries/ita_OCRFixReplaceList.xml

68 lines
2.3 KiB
XML
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<?xml version="1.0" encoding="utf-8"?>
<OCRFixReplaceList>
<WholeWords>
<Word from="lN" to="IN" />
<Word from="lL" to="IL" />
<Word from="l" to="I" />
</WholeWords>
<PartialWordsAlways>
<!-- Will be replaced always -->
<WordPart from="¤" to="o" />
<WordPart from="lVI" to="M" />
<WordPart from="IVl" to="M" />
<WordPart from="lVl" to="M" />
<WordPart from="I\/I" to="M" />
<WordPart from="l\/I" to="M" />
<WordPart from="I\/l" to="M" />
<WordPart from="l\/l" to="M" />
<WordPart from="IVIa" to="Ma" />
<WordPart from="IVIe" to="Me" />
<WordPart from="IVIi" to="Mi" />
<WordPart from="IVIo" to="Mo" />
<WordPart from="IVIu" to="Mu" />
<WordPart from="IVIy" to="My" />
</PartialWordsAlways>
<PartialWords>
<!-- Will be used to check words not in dictionary.
If new word(s) and longer than 5 chars and exists in spelling dictionary, it is (or they are) accepted -->
<WordPart from="IVI" to="M" />
<WordPart from="/" to="l" />
<WordPart from="|" to="I" />
<WordPart from="vv" to="w" />
<WordPart from="m" to="rn" />
<WordPart from="l" to="i" />
<WordPart from="€" to="e" />
<WordPart from="I" to="l" />
<WordPart from="c" to="o" />
<WordPart from="i" to="t" />
<WordPart from="cc" to="oo" />
<WordPart from="ii" to="tt" />
<WordPart from="ii" to="ll" />
<!-- "f " will be two words -->
<WordPart from="f" to="f " />
<WordPart from="c" to="e" />
<WordPart from="I" to="t" />
<WordPart from="n" to="o" />
<WordPart from="s" to="e" />
<WordPart from="l-l" to="H" />
<WordPart from="l_" to="L" />
<WordPart from="°" to="o" />
<WordPart from=")/" to="y" />
<WordPart from=")'" to="y" />
<WordPart from="a" to="d" />
<WordPart from="|" to="l" />
<WordPart from="¤" to="o" />
<WordPart from="vx/" to="w" />
</PartialWords>
<WholeLines />
<PartialLinesAlways />
<PartialLines />
<BeginLines />
<EndLines />
<RegularExpressions>
<RegEx find="\b\|\b" replaceWith="I" />
<RegEx find="([\p{Ll},] )(I)([oai][, \.])" replaceWith="$1l$3" />
<RegEx find="\b(I)([àì]|'[ \r\n])\b" replaceWith="l$2" />
<RegEx find="([\p{Ll},] )(II)\b" replaceWith="$1Il" />
</RegularExpressions>
</OCRFixReplaceList>