public static WordList CreateFromWords(IEnumerable <string> words, AffixConfig affix)
{
var wordListBuilder = new Builder(affix ?? new AffixConfig.Builder().MoveToImmutable());
if (words is IList <string> wordsAsList)
{
wordListBuilder.InitializeEntriesByRoot(wordsAsList.Count);
}
else
{
wordListBuilder.InitializeEntriesByRoot(0);
}
foreach (var word in words)
{
var wordEntry = new WordEntry(word, FlagSet.Empty, MorphSet.Empty, WordEntryOptions.None);
WordEntrySet entryList = !wordListBuilder.EntriesByRoot.TryGetValue(word, out entryList)
? WordEntrySet.Create(wordEntry)
: WordEntrySet.CopyWithItemAdded(entryList, wordEntry);
wordListBuilder.EntriesByRoot.Add(word, entryList);
}
return(wordListBuilder.MoveToImmutable());
}
/// <summary>
/// Forbid compoundings when there are special patterns at word bound.
/// </summary>
public bool Check(string word, int pos, WordEntry r1, WordEntry r2, bool affixed)
{
var wordAfterPos = word.Substring(pos);
foreach (var patternEntry in items)
{
int len;
if (
StringEx.IsSubset(patternEntry.Pattern2, wordAfterPos)
&&
(
r1 == null
||
patternEntry.Condition.IsZero
||
r1.ContainsFlag(patternEntry.Condition)
)
&&
(
r2 == null
||
patternEntry.Condition2.IsZero
||
r2.ContainsFlag(patternEntry.Condition2)
)
&&
// zero length pattern => only TESTAFF
// zero pattern (0/flag) => unmodified stem (zero affixes allowed)
(
string.IsNullOrEmpty(patternEntry.Pattern)
||
(
(
patternEntry.Pattern.StartsWith('0') &&
r1.Word.Length <= pos &&
StringEx.EqualsOffset(word, pos - r1.Word.Length, r1.Word, 0, r1.Word.Length)
)
||
(
!patternEntry.Pattern.StartsWith('0')
&&
(
(
len = patternEntry.Pattern.Length
) != 0
)
&&
StringEx.EqualsOffset(word, pos - len, patternEntry.Pattern, 0, len)
)
)
)
)
{
return(true);
}
}
return(false);
}
/// <summary>
/// Forbid compoundings when there are special patterns at word bound.
/// </summary>
internal bool Check(string word, int pos, WordEntry r1, WordEntry r2, bool affixed)
{
#if DEBUG
if (word == null)
{
throw new ArgumentNullException(nameof(word));
}
if (r1 == null)
{
throw new ArgumentNullException(nameof(r1));
}
if (r2 == null)
{
throw new ArgumentNullException(nameof(r2));
}
#endif
var wordAfterPos = word.Subslice(pos);
foreach (var patternEntry in items)
{
if (
HunspellTextFunctions.IsSubset(patternEntry.Pattern2, wordAfterPos)
&&
(
patternEntry.Condition.IsZero
||
r1.ContainsFlag(patternEntry.Condition)
)
&&
(
patternEntry.Condition2.IsZero
||
r2.ContainsFlag(patternEntry.Condition2)
)
&&
// zero length pattern => only TESTAFF
// zero pattern (0/flag) => unmodified stem (zero affixes allowed)
(
string.IsNullOrEmpty(patternEntry.Pattern)
||
PatternWordCheck(word, pos, patternEntry.Pattern.StartsWith('0') ? r1.Word : patternEntry.Pattern)
)
)
{
return(true);
}
}
return(false);
}
public bool EntryContainsRuleFlags(WordEntry rv)
{
foreach (var rule in items)
{
foreach (var flag in rule)
{
if (!flag.Equals('*') && !flag.Equals('?') && rv.ContainsFlag(flag))
{
return(true);
}
}
}
return(false);
}
private bool IsKeepCase(WordEntry rv) => rv.ContainsFlag(Affix.KeepCase);
public SpellCheckResult CheckDetails()
{
var word = WordToCheck;
if (string.IsNullOrEmpty(word) || word.Length >= MaxWordUtf8Len || !WordList.HasEntries)
{
return(new SpellCheckResult(false));
}
if (word == DefaultXmlToken)
{
// Hunspell supports XML input of the simplified API (see manual)
return(new SpellCheckResult(true));
}
// input conversion
if (!Affix.InputConversions.HasReplacements || !Affix.InputConversions.TryConvert(word, out string convertedWord))
{
convertedWord = word;
}
if (!CleanWord2(out string scw, convertedWord, out CapitalizationType capType, out int abbv))
{
return(new SpellCheckResult(false));
}
if (HunspellTextFunctions.IsNumericWord(word))
{
// allow numbers with dots, dashes and commas (but forbid double separators: "..", "--" etc.)
return(new SpellCheckResult(true));
}
var resultType = SpellCheckResultType.None;
string root = null;
WordEntry rv = null;
if (capType == CapitalizationType.Huh || capType == CapitalizationType.HuhInit || capType == CapitalizationType.None)
{
if (capType == CapitalizationType.HuhInit)
{
resultType |= SpellCheckResultType.OrigCap;
}
rv = CheckWord(scw, ref resultType, out root);
if (abbv != 0 && rv == null)
{
rv = CheckWord(scw + ".", ref resultType, out root);
}
}
else if (capType == CapitalizationType.All)
{
rv = CheckDetailsAllCap(abbv, ref scw, ref resultType, out root);
}
if (capType == CapitalizationType.Init || (capType == CapitalizationType.All && rv == null))
{
rv = CheckDetailsInitCap(abbv, capType, ref scw, ref resultType, out root);
}
if (rv != null)
{
if (rv.ContainsFlag(Affix.Warn))
{
resultType |= SpellCheckResultType.Warn;
if (Affix.ForbidWarn)
{
return(new SpellCheckResult(root, resultType, false));
}
}
return(new SpellCheckResult(root, resultType, true));
}
// recursive breaking at break points
if (Affix.BreakPoints.HasItems && !EnumEx.HasFlag(resultType, SpellCheckResultType.Forbidden))
{
// calculate break points for recursion limit
if (Affix.BreakPoints.FindRecursionLimit(scw) >= 10)
{
return(new SpellCheckResult(root, resultType, false));
}
// check boundary patterns (^begin and end$)
foreach (var breakEntry in Affix.BreakPoints)
{
if (breakEntry.Length == 1 || breakEntry.Length > scw.Length)
{
continue;
}
var pLastIndex = breakEntry.Length - 1;
if (
breakEntry.StartsWith('^') &&
StringEx.EqualsOffset(scw, 0, breakEntry, 1, pLastIndex) &&
Check(scw.Substring(pLastIndex))
)
{
return(new SpellCheckResult(root, resultType, true));
}
if (breakEntry.EndsWith('$'))
{
var wlLessBreakIndex = scw.Length - breakEntry.Length + 1;
if (
StringEx.EqualsOffset(scw, wlLessBreakIndex, breakEntry, 0, pLastIndex) &&
Check(scw.Substring(0, wlLessBreakIndex))
)
{
return(new SpellCheckResult(root, resultType, true));
}
}
}
// other patterns
foreach (var breakEntry in Affix.BreakPoints)
{
var found = scw.IndexOfOrdinal(breakEntry);
var remainingLength = scw.Length - breakEntry.Length;
if (found > 0 && found < remainingLength)
{
var found2 = scw.IndexOfOrdinal(breakEntry, found + 1);
// try to break at the second occurance
// to recognize dictionary words with wordbreak
if (found2 > 0 && (found2 < remainingLength))
{
found = found2;
}
if (!Check(scw.Substring(found + breakEntry.Length)))
{
continue;
}
// examine 2 sides of the break point
if (Check(scw.Substring(0, found)))
{
return(new SpellCheckResult(root, resultType, true));
}
// LANG_hu: spec. dash rule
if (Affix.IsHungarian && "-".Equals(breakEntry, StringComparison.Ordinal))
{
if (Check(scw.Substring(0, found + 1)))
{
return(new SpellCheckResult(root, resultType, true));
}
}
}
}
// other patterns (break at first break point)
foreach (var breakEntry in Affix.BreakPoints)
{
var found = scw.IndexOfOrdinal(breakEntry);
var remainingLength = scw.Length - breakEntry.Length;
if (found > 0 && found < remainingLength)
{
if (!Check(scw.Substring(found + breakEntry.Length)))
{
continue;
}
// examine 2 sides of the break point
if (Check(scw.Substring(0, found)))
{
return(new SpellCheckResult(root, resultType, true));
}
// LANG_hu: spec. dash rule
if (Affix.IsHungarian && "-".Equals(breakEntry, StringComparison.Ordinal))
{
if (Check(scw.Substring(0, found + 1)))
{
return(new SpellCheckResult(root, resultType, true));
}
}
}
}
}
return(new SpellCheckResult(root, resultType, false));
}
private bool AddWord(string word, FlagSet flags, MorphSet morphs, bool onlyUpperCase)
{
if (Affix.IgnoredChars.HasItems)
{
word = word.RemoveChars(Affix.IgnoredChars);
}
if (Affix.ComplexPrefixes)
{
word = word.Reverse();
if (morphs.HasItems && !Affix.IsAliasM)
{
var newMorphs = new string[morphs.Count];
for (int i = 0; i < morphs.Count; i++)
{
newMorphs[i] = morphs[morphs.Count - i - 1].Reverse();
}
morphs = MorphSet.TakeArray(newMorphs);
}
}
WordEntryOptions options;
if (morphs.HasItems)
{
if (Affix.IsAliasM)
{
options = WordEntryOptions.AliasM;
var morphBuilder = new List <string>();
foreach (var originalValue in morphs)
{
if (IntEx.TryParseInvariant(originalValue, out int morphNumber) && Affix.TryGetAliasM(morphNumber, out MorphSet aliasedMorph))
{
morphBuilder.AddRange(aliasedMorph);
}
else
{
morphBuilder.Add(originalValue);
}
}
morphs = MorphSet.Create(morphBuilder);
}
else
{
options = WordEntryOptions.None;
}
if (morphs.AnyStartsWith(MorphologicalTags.Phon))
{
options |= WordEntryOptions.Phon;
}
}
else
{
options = WordEntryOptions.None;
}
bool saveEntryList = false;
word = Builder.Dedup(word);
if (!Builder.EntriesByRoot.TryGetValue(word, out WordEntrySet entryList))
{
saveEntryList = true;
entryList = WordEntrySet.Empty;
}
var upperCaseHomonym = false;
for (var i = 0; i < entryList.Count; i++)
{
var existingEntry = entryList[i];
if (!onlyUpperCase)
{
if (existingEntry.ContainsFlag(SpecialFlags.OnlyUpcaseFlag))
{
existingEntry = new WordEntry(
existingEntry.Word,
flags,
existingEntry.Morphs,
existingEntry.Options);
entryList.DestructiveReplace(i, existingEntry);
return(false);
}
}
else
{
upperCaseHomonym = true;
}
}
if (!upperCaseHomonym)
{
saveEntryList = true;
entryList = WordEntrySet.CopyWithItemAdded(entryList, new WordEntry(
word,
flags,
Builder.Dedup(morphs),
options));
}
if (saveEntryList)
{
Builder.EntriesByRoot[word] = entryList;
}
return(false);
}
