public static bool Intersect(List <PersonMorphCollection> list) { bool ret = false; while (true) { bool ch = false; for (int i = 0; i < (list.Count - 1); i++) { for (int j = i + 1; j < list.Count; j++) { if (PersonMorphCollection.Intersect2(list[i], list[j])) { ch = true; } if (PersonMorphCollection.IsEquals(list[i], list[j])) { list.RemoveAt(j); j--; ch = true; } } } if (ch) { ret = true; } else { break; } } return(ret); }
static bool Intersect2(PersonMorphCollection col1, PersonMorphCollection col2) { if (col1.Head != col2.Head) { return(false); } bool ret = false; List <string> vals1 = col1.Values; List <string> vals2 = col2.Values; List <string> uni = new List <string>(); foreach (string v in vals1) { if (vals2.Contains(v)) { uni.Add(v); continue; } } foreach (string v in vals1) { if (!uni.Contains(v)) { col1.Remove(v, Pullenti.Morph.MorphGender.Undefined); ret = true; } } foreach (string v in vals2) { if (!uni.Contains(v)) { col2.Remove(v, Pullenti.Morph.MorphGender.Undefined); ret = true; } } if (col1.Gender != Pullenti.Morph.MorphGender.Undefined) { if (col2.Remove(null, (col1.Gender == Pullenti.Morph.MorphGender.Feminie ? Pullenti.Morph.MorphGender.Masculine : Pullenti.Morph.MorphGender.Feminie))) { ret = true; } } if (col2.Gender != Pullenti.Morph.MorphGender.Undefined) { if (col1.Remove(null, (col2.Gender == Pullenti.Morph.MorphGender.Feminie ? Pullenti.Morph.MorphGender.Masculine : Pullenti.Morph.MorphGender.Feminie))) { ret = true; } } return(ret); }
public static bool IsEquals(PersonMorphCollection col1, PersonMorphCollection col2) { if (col1.Head != col2.Head) { return(false); } foreach (PersonMorphVariant v in col1.Items) { if (!col2.ContainsItem(v.Value, v.Gender)) { return(false); } } foreach (PersonMorphVariant v in col2.Items) { if (!col1.ContainsItem(v.Value, v.Gender)) { return(false); } } return(true); }
public static PersonMorphCollection AddPrefix(PersonMorphCollection prefix, PersonMorphCollection body) { PersonMorphCollection res = new PersonMorphCollection(); res.Head = string.Format("{0}-{1}", prefix.Head, body.Head); foreach (PersonMorphVariant pv in prefix.Items) { foreach (PersonMorphVariant bv in body.Items) { Pullenti.Morph.MorphGender g = bv.Gender; if (g == Pullenti.Morph.MorphGender.Undefined) { g = pv.Gender; } else if (pv.Gender != Pullenti.Morph.MorphGender.Undefined && pv.Gender != g) { g = Pullenti.Morph.MorphGender.Undefined; } res.Add(string.Format("{0}-{1}", pv.Value, bv.Value), null, g, false); } } return(res); }
internal static Pullenti.Ner.ReferentToken CreateReferentToken(Pullenti.Ner.Person.PersonReferent p, Pullenti.Ner.Token begin, Pullenti.Ner.Token end, Pullenti.Ner.MorphCollection morph, List <PersonAttrToken> attrs, Pullenti.Ner.Person.PersonAnalyzer.PersonAnalyzerData ad, bool forAttribute, bool afterBePredicate) { if (p == null) { return(null); } bool hasPrefix = false; if (attrs != null) { foreach (PersonAttrToken a in attrs) { if (a.Typ == PersonAttrTerminType.BestRegards) { hasPrefix = true; } else { if (a.BeginChar < begin.BeginChar) { begin = a.BeginToken; if ((a.EndToken.Next != null && a.EndToken.Next.IsChar(')') && begin.Previous != null) && begin.Previous.IsChar('(')) { begin = begin.Previous; } } if (a.Typ != PersonAttrTerminType.Prefix) { if (a.Age != null) { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_AGE, a.Age, false, 0); } if (a.PropRef == null) { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_ATTR, a.Value, false, 0); } else { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_ATTR, a, false, 0); } } else if (a.Gender == Pullenti.Morph.MorphGender.Feminie && !p.IsFemale) { p.IsFemale = true; } else if (a.Gender == Pullenti.Morph.MorphGender.Masculine && !p.IsMale) { p.IsMale = true; } } } } else if ((begin.Previous is Pullenti.Ner.TextToken) && (begin.WhitespacesBeforeCount < 3)) { if ((begin.Previous as Pullenti.Ner.TextToken).Term == "ИП") { PersonAttrToken a = new PersonAttrToken(begin.Previous, begin.Previous); a.PropRef = new Pullenti.Ner.Person.PersonPropertyReferent(); a.PropRef.Name = "индивидуальный предприниматель"; p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_ATTR, a, false, 0); begin = begin.Previous; } } Pullenti.Ner.MorphCollection m0 = new Pullenti.Ner.MorphCollection(); foreach (Pullenti.Morph.MorphBaseInfo it in morph.Items) { Pullenti.Morph.MorphBaseInfo bi = new Pullenti.Morph.MorphBaseInfo(); bi.CopyFrom(it); bi.Number = Pullenti.Morph.MorphNumber.Singular; if (bi.Gender == Pullenti.Morph.MorphGender.Undefined) { if (p.IsMale && !p.IsFemale) { bi.Gender = Pullenti.Morph.MorphGender.Masculine; } if (!p.IsMale && p.IsFemale) { bi.Gender = Pullenti.Morph.MorphGender.Feminie; } } m0.AddItem(bi); } morph = m0; if ((attrs != null && attrs.Count > 0 && !attrs[0].Morph.Case.IsUndefined) && morph.Case.IsUndefined) { morph.Case = attrs[0].Morph.Case; if (attrs[0].Morph.Number == Pullenti.Morph.MorphNumber.Singular) { morph.Number = Pullenti.Morph.MorphNumber.Singular; } if (p.IsMale && !p.IsFemale) { morph.Gender = Pullenti.Morph.MorphGender.Masculine; } else if (p.IsFemale) { morph.Gender = Pullenti.Morph.MorphGender.Feminie; } } if (begin.Previous != null) { Pullenti.Ner.Token ttt = begin.Previous; if (ttt.IsValue("ИМЕНИ", "ІМЕНІ")) { forAttribute = true; } else { if (ttt.IsChar('.') && ttt.Previous != null) { ttt = ttt.Previous; } if (ttt.WhitespacesAfterCount < 3) { if (ttt.IsValue("ИМ", "ІМ")) { forAttribute = true; } } } } if (forAttribute) { return new Pullenti.Ner.ReferentToken(p, begin, end) { Morph = morph, MiscAttrs = (int)p.m_PersonIdentityTyp } } ; if ((begin.Previous != null && begin.Previous.IsCommaAnd && (begin.Previous.Previous is Pullenti.Ner.ReferentToken)) && (begin.Previous.Previous.GetReferent() is Pullenti.Ner.Person.PersonReferent)) { Pullenti.Ner.ReferentToken rt00 = begin.Previous.Previous as Pullenti.Ner.ReferentToken; for (Pullenti.Ner.Token ttt = (Pullenti.Ner.Token)rt00; ttt != null;) { if (ttt.Previous == null || !(ttt.Previous.Previous is Pullenti.Ner.ReferentToken)) { break; } if (!ttt.Previous.IsCommaAnd || !(ttt.Previous.Previous.GetReferent() is Pullenti.Ner.Person.PersonReferent)) { break; } rt00 = ttt.Previous.Previous as Pullenti.Ner.ReferentToken; ttt = rt00; } if (rt00.BeginToken.GetReferent() is Pullenti.Ner.Person.PersonPropertyReferent) { bool ok = false; if ((rt00.BeginToken as Pullenti.Ner.ReferentToken).EndToken.Next != null && (rt00.BeginToken as Pullenti.Ner.ReferentToken).EndToken.Next.IsChar(':')) { ok = true; } else if (rt00.BeginToken.Morph.Number == Pullenti.Morph.MorphNumber.Plural) { ok = true; } if (ok) { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_ATTR, rt00.BeginToken.GetReferent(), false, 0); } } } if (ad != null) { if (ad.OverflowLevel > 10) { return new Pullenti.Ner.ReferentToken(p, begin, end) { Morph = morph, MiscAttrs = (int)p.m_PersonIdentityTyp } } ; ad.OverflowLevel++; } List <PersonAttrToken> attrs1 = null; bool hasPosition = false; bool openBr = false; for (Pullenti.Ner.Token t = end.Next; t != null; t = t.Next) { if (t.IsTableControlChar) { break; } if (t.IsNewlineBefore) { if (t.NewlinesBeforeCount > 2) { break; } if (attrs1 != null && attrs1.Count > 0) { break; } Pullenti.Ner.Mail.Internal.MailLine ml = Pullenti.Ner.Mail.Internal.MailLine.Parse(t, 0, 0); if (ml != null && ml.Typ == Pullenti.Ner.Mail.Internal.MailLine.Types.From) { break; } if (t.Chars.IsCapitalUpper) { PersonAttrToken attr1 = PersonAttrToken.TryAttach(t, (ad == null ? null : ad.LocalOntology), PersonAttrToken.PersonAttrAttachAttrs.No); bool ok1 = false; if (attr1 != null) { if (hasPrefix || attr1.IsNewlineAfter || ((attr1.EndToken.Next != null && attr1.EndToken.Next.IsTableControlChar))) { ok1 = true; } else { for (Pullenti.Ner.Token tt2 = t.Next; tt2 != null && tt2.EndChar <= attr1.EndChar; tt2 = tt2.Next) { if (tt2.IsWhitespaceBefore) { ok1 = true; } } } } else { Pullenti.Ner.Token ttt = CorrectTailAttributes(p, t); if (ttt != null && ttt != t) { end = (t = ttt); continue; } } if (!ok1) { break; } } } if (t.IsHiphen || t.IsCharOf("_>|")) { continue; } if (t.IsValue("МОДЕЛЬ", null)) { break; } Pullenti.Ner.Token tt = CorrectTailAttributes(p, t); if (tt != t && tt != null) { end = (t = tt); continue; } bool isBe = false; if (t.IsChar('(') && t == end.Next) { openBr = true; t = t.Next; if (t == null) { break; } PersonItemToken pit1 = PersonItemToken.TryAttach(t, null, PersonItemToken.ParseAttr.No, null); if ((pit1 != null && t.Chars.IsCapitalUpper && pit1.EndToken.Next != null) && (t is Pullenti.Ner.TextToken) && pit1.EndToken.Next.IsChar(')')) { if (pit1.Lastname != null) { Pullenti.Morph.MorphBaseInfo inf = new Pullenti.Morph.MorphBaseInfo() { Case = Pullenti.Morph.MorphCase.Nominative }; if (p.IsMale) { inf.Gender |= Pullenti.Morph.MorphGender.Masculine; } if (p.IsFemale) { inf.Gender |= Pullenti.Morph.MorphGender.Feminie; } PersonMorphCollection sur = PersonIdentityToken.CreateLastname(pit1, inf); if (sur != null) { p.AddFioIdentity(sur, null, null); end = (t = pit1.EndToken.Next); continue; } } } if ((t is Pullenti.Ner.TextToken) && t.Chars.IsLatinLetter) { List <PersonItemToken> pits = PersonItemToken.TryAttachList(t, null, PersonItemToken.ParseAttr.CanBeLatin, 10); if (((pits != null && pits.Count >= 2 && pits.Count <= 3) && pits[0].Chars.IsLatinLetter && pits[1].Chars.IsLatinLetter) && pits[pits.Count - 1].EndToken.Next != null && pits[pits.Count - 1].EndToken.Next.IsChar(')')) { Pullenti.Ner.Person.PersonReferent pr2 = new Pullenti.Ner.Person.PersonReferent(); int cou = 0; foreach (PersonItemToken pi in pits) { foreach (Pullenti.Ner.Slot si in p.Slots) { if (si.TypeName == Pullenti.Ner.Person.PersonReferent.ATTR_FIRSTNAME || si.TypeName == Pullenti.Ner.Person.PersonReferent.ATTR_MIDDLENAME || si.TypeName == Pullenti.Ner.Person.PersonReferent.ATTR_LASTNAME) { if (Pullenti.Ner.Core.MiscHelper.CanBeEqualCyrAndLatSS(si.Value.ToString(), pi.Value)) { cou++; pr2.AddSlot(si.TypeName, pi.Value, false, 0); break; } } } } if (cou == pits.Count) { foreach (Pullenti.Ner.Slot si in pr2.Slots) { p.AddSlot(si.TypeName, si.Value, false, 0); } end = (t = pits[pits.Count - 1].EndToken.Next); continue; } } } } else if (t.IsComma) { t = t.Next; if ((t is Pullenti.Ner.TextToken) && (t as Pullenti.Ner.TextToken).IsValue("WHO", null)) { continue; } if ((t is Pullenti.Ner.TextToken) && t.Chars.IsLatinLetter) { List <PersonItemToken> pits = PersonItemToken.TryAttachList(t, null, PersonItemToken.ParseAttr.CanBeLatin, 10); if ((pits != null && pits.Count >= 2 && pits.Count <= 3) && pits[0].Chars.IsLatinLetter && pits[1].Chars.IsLatinLetter) { Pullenti.Ner.Person.PersonReferent pr2 = new Pullenti.Ner.Person.PersonReferent(); int cou = 0; foreach (PersonItemToken pi in pits) { foreach (Pullenti.Ner.Slot si in p.Slots) { if (si.TypeName == Pullenti.Ner.Person.PersonReferent.ATTR_FIRSTNAME || si.TypeName == Pullenti.Ner.Person.PersonReferent.ATTR_MIDDLENAME || si.TypeName == Pullenti.Ner.Person.PersonReferent.ATTR_LASTNAME) { if (Pullenti.Ner.Core.MiscHelper.CanBeEqualCyrAndLatSS(si.Value.ToString(), pi.Value)) { cou++; pr2.AddSlot(si.TypeName, pi.Value, false, 0); break; } } } } if (cou == pits.Count) { foreach (Pullenti.Ner.Slot si in pr2.Slots) { p.AddSlot(si.TypeName, si.Value, false, 0); } end = (t = pits[pits.Count - 1].EndToken); continue; } } } } else if ((t is Pullenti.Ner.TextToken) && (t as Pullenti.Ner.TextToken).IsVerbBe) { t = t.Next; } else if (t.IsAnd && t.IsWhitespaceAfter && !t.IsNewlineAfter) { if (t == end.Next) { break; } t = t.Next; } else if (t.IsHiphen && t == end.Next) { t = t.Next; } else if (t.IsChar('.') && t == end.Next && hasPrefix) { t = t.Next; } Pullenti.Ner.Token ttt2 = CreateNickname(p, t); if (ttt2 != null) { t = (end = ttt2); continue; } if (t == null) { break; } PersonAttrToken attr = null; attr = PersonAttrToken.TryAttach(t, (ad == null ? null : ad.LocalOntology), PersonAttrToken.PersonAttrAttachAttrs.No); if (attr == null) { if ((t != null && t.GetReferent() != null && t.GetReferent().TypeName == "GEO") && attrs1 != null && openBr) { continue; } if ((t.Chars.IsCapitalUpper && openBr && t.Next != null) && t.Next.IsChar(')')) { if (p.FindSlot(Pullenti.Ner.Person.PersonReferent.ATTR_LASTNAME, null, true) == null) { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_LASTNAME, t.GetSourceText().ToUpper(), false, 0); t = t.Next; end = t; } } if (t != null && t.IsValue("КОТОРЫЙ", null) && t.Morph.Number == Pullenti.Morph.MorphNumber.Singular) { if (!p.IsFemale && t.Morph.Gender == Pullenti.Morph.MorphGender.Feminie) { p.IsFemale = true; p.CorrectData(); } else if (!p.IsMale && t.Morph.Gender == Pullenti.Morph.MorphGender.Masculine) { p.IsMale = true; p.CorrectData(); } } break; } if (attr.Morph.Number == Pullenti.Morph.MorphNumber.Plural) { break; } if (attr.Typ == PersonAttrTerminType.BestRegards) { break; } if (attr.IsDoubt) { if (hasPrefix) { } else if (t.IsNewlineBefore && attr.IsNewlineAfter) { } else if (t.Previous != null && ((t.Previous.IsHiphen || t.Previous.IsChar(':')))) { } else { break; } } if (!morph.Case.IsUndefined && !attr.Morph.Case.IsUndefined) { if (((morph.Case & attr.Morph.Case)).IsUndefined && !isBe) { break; } } if (openBr) { if (Pullenti.Ner.Person.PersonAnalyzer.TryAttachPerson(t, ad, false, 0, true) != null) { break; } } if (attrs1 == null) { if (t.Previous.IsComma && t.Previous == end.Next) { Pullenti.Ner.Token ttt = attr.EndToken.Next; if (ttt != null) { if (ttt.Morph.Class.IsVerb) { if (Pullenti.Ner.Core.MiscHelper.CanBeStartOfSentence(begin)) { } else { break; } } } } attrs1 = new List <PersonAttrToken>(); } attrs1.Add(attr); if (attr.Typ == PersonAttrTerminType.Position || attr.Typ == PersonAttrTerminType.King) { if (!isBe) { hasPosition = true; } } else if (attr.Typ != PersonAttrTerminType.Prefix) { if (attr.Typ == PersonAttrTerminType.Other && attr.Age != null) { } else { attrs1 = null; break; } } t = attr.EndToken; } if (attrs1 != null && hasPosition && attrs != null) { Pullenti.Ner.Token te1 = attrs[attrs.Count - 1].EndToken.Next; Pullenti.Ner.Token te2 = attrs1[0].BeginToken; if (te1.WhitespacesAfterCount > te2.WhitespacesBeforeCount && (te2.WhitespacesBeforeCount < 2)) { } else if (attrs1[0].Age != null) { } else if (((te1.IsHiphen || te1.IsChar(':'))) && !attrs1[0].IsNewlineBefore && ((te2.Previous.IsComma || te2.Previous == end))) { } else { foreach (PersonAttrToken a in attrs) { if (a.Typ == PersonAttrTerminType.Position) { Pullenti.Ner.Token te = attrs1[attrs1.Count - 1].EndToken; if (te.Next != null) { if (!te.Next.IsChar('.')) { attrs1 = null; break; } } } } } } if (attrs1 != null && !hasPrefix) { PersonAttrToken attr = attrs1[attrs1.Count - 1]; bool ok = false; if (attr.EndToken.Next != null && attr.EndToken.Next.Chars.IsCapitalUpper) { ok = true; } else { Pullenti.Ner.ReferentToken rt = Pullenti.Ner.Person.PersonAnalyzer.TryAttachPerson(attr.BeginToken, ad, false, -1, false); if (rt != null && (rt.Referent is Pullenti.Ner.Person.PersonReferent)) { ok = true; } } if (ok) { if (attr.BeginToken.WhitespacesBeforeCount > attr.EndToken.WhitespacesAfterCount) { attrs1 = null; } else if (attr.BeginToken.WhitespacesBeforeCount == attr.EndToken.WhitespacesAfterCount) { Pullenti.Ner.ReferentToken rt1 = Pullenti.Ner.Person.PersonAnalyzer.TryAttachPerson(attr.BeginToken, ad, false, -1, false); if (rt1 != null) { attrs1 = null; } } } } if (attrs1 != null) { foreach (PersonAttrToken a in attrs1) { if (a.Typ != PersonAttrTerminType.Prefix) { if (a.Age != null) { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_AGE, a.Age, true, 0); } else if (a.PropRef == null) { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_ATTR, a.Value, false, 0); } else { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_ATTR, a, false, 0); } end = a.EndToken; if (a.Gender != Pullenti.Morph.MorphGender.Undefined && !p.IsFemale && !p.IsMale) { if (a.Gender == Pullenti.Morph.MorphGender.Masculine && !p.IsMale) { p.IsMale = true; p.CorrectData(); } else if (a.Gender == Pullenti.Morph.MorphGender.Feminie && !p.IsFemale) { p.IsFemale = true; p.CorrectData(); } } } } if (openBr) { if (end.Next != null && end.Next.IsChar(')')) { end = end.Next; } } } int crlfCou = 0; for (Pullenti.Ner.Token t = end.Next; t != null; t = t.Next) { if (t.IsTableControlChar) { break; } if (t.IsNewlineBefore) { Pullenti.Ner.Mail.Internal.MailLine ml = Pullenti.Ner.Mail.Internal.MailLine.Parse(t, 0, 0); if (ml != null && ml.Typ == Pullenti.Ner.Mail.Internal.MailLine.Types.From) { break; } crlfCou++; } if (t.IsCharOf(":,(") || t.IsHiphen) { continue; } if (t.IsChar('.') && t == end.Next) { continue; } Pullenti.Ner.Referent r = t.GetReferent(); if (r != null) { if (r.TypeName == "PHONE" || r.TypeName == "URI" || r.TypeName == "ADDRESS") { string ty = r.GetStringValue("SCHEME"); if (r.TypeName == "URI") { if ((ty != "mailto" && ty != "skype" && ty != "ICQ") && ty != "http") { break; } } p.AddContact(r); end = t; crlfCou = 0; continue; } } if (r is Pullenti.Ner.Person.PersonIdentityReferent) { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_IDDOC, r, false, 0); end = t; crlfCou = 0; continue; } if (r != null && r.TypeName == "ORGANIZATION") { if (t.Next != null && t.Next.Morph.Class.IsVerb) { break; } if (begin.Previous != null && begin.Previous.Morph.Class.IsVerb) { break; } if (t.WhitespacesAfterCount == 1) { break; } bool exist = false; foreach (Pullenti.Ner.Slot s in p.Slots) { if (s.TypeName == Pullenti.Ner.Person.PersonReferent.ATTR_ATTR && (s.Value is Pullenti.Ner.Person.PersonPropertyReferent)) { Pullenti.Ner.Person.PersonPropertyReferent pr = s.Value as Pullenti.Ner.Person.PersonPropertyReferent; if (pr.FindSlot(Pullenti.Ner.Person.PersonPropertyReferent.ATTR_REF, r, true) != null) { exist = true; break; } } else if (s.TypeName == Pullenti.Ner.Person.PersonReferent.ATTR_ATTR && (s.Value is PersonAttrToken)) { PersonAttrToken pr = s.Value as PersonAttrToken; if (pr.Referent.FindSlot(Pullenti.Ner.Person.PersonPropertyReferent.ATTR_REF, r, true) != null) { exist = true; break; } } } if (!exist) { PersonAttrToken pat = new PersonAttrToken(t, t); pat.PropRef = new Pullenti.Ner.Person.PersonPropertyReferent() { Name = "сотрудник" }; pat.PropRef.AddSlot(Pullenti.Ner.Person.PersonPropertyReferent.ATTR_REF, r, false, 0); p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_ATTR, pat, false, 0); } continue; } if (r != null) { break; } if (!hasPrefix || crlfCou >= 2) { break; } Pullenti.Ner.ReferentToken rt = t.Kit.ProcessReferent("PERSON", t); if (rt != null) { break; } } if (ad != null) { ad.OverflowLevel--; } if (begin.IsValue("НА", null) && begin.Next != null && begin.Next.IsValue("ИМЯ", null)) { Pullenti.Ner.Token t0 = begin.Previous; if (t0 != null && t0.IsComma) { t0 = t0.Previous; } if (t0 != null && (t0.GetReferent() is Pullenti.Ner.Person.PersonIdentityReferent)) { p.AddSlot(Pullenti.Ner.Person.PersonReferent.ATTR_IDDOC, t0.GetReferent(), false, 0); } } return(new Pullenti.Ner.ReferentToken(p, begin, end) { Morph = morph, MiscAttrs = (int)p.m_PersonIdentityTyp }); }