tchildrencount count()
{
if(next)
return 1 + next->count();
else
return 1;
}
int cmp(const Lemma * lt)
{
assert(Type != 0);
assert(lt->Type != 0);
int strdif = strcmp(Type,lt->Type);
if(strdif)
return strdif;
assert(BaseForm != 0);
assert(lt->BaseForm != 0);
strdif = strcmp(BaseForm,lt->BaseForm);
if(strdif)
return strdif;
if(S.Offset > lt->S.Offset)
return 1;
else if(S.Offset < lt->S.Offset)
return -1;
if(S.frequency > lt->S.frequency)
return 1;
else if(S.frequency < lt->S.frequency)
return -1;
if(next)
{
assert(lt->next);
return next->cmp(lt->next);
}
else
return 0;
}
int print(FILE * fp)
{
fprintf(fp,"%d %s %s %d",S.Offset,BaseForm,Type,S.frequency);
if(next)
{
fprintf(fp,",");
return 1 + next->print(fp);
}
return 1;
}
tcount strcnt()
{
tcount ret = 0;
if(*Type)
ret = 1;
if(*BaseForm)
ret += 1;
if(next)
ret += next->strcnt();
return ret;
}
void move(tcount pos)
{
LEMMAS[pos].Type = Type;
LEMMAS[pos].BaseForm = BaseForm;
LEMMAS[pos].S = S;
if(next)
{
LEMMAS[pos].next = LEMMAS + pos + 1;
next->move(pos + 1);
}
}
bool addFreq(char * lextype,int n,char * bf,bool relaxed,unsigned int offset)
{
bool added = false;
int off = offset - S.Offset;
if(!strcmp(lextype,Type) && (!bf || /*S.Offset == offset*/ off >= 0))
{
if(relaxed)
{
assert(bf);
char * bfoff = bf + S.Offset;
int len = strlen(bfoff);
if(!strncmp(BaseForm,bfoff,len) && BaseForm[len] == ',')
{
S.frequency += n;
added = true;
//return true;
}
}
else
{
if(!bf || !strcmp(bf + S.Offset,BaseForm))
{
S.frequency += n;
added = true;
//return true;
}
}
}
/*else*/ // Full form may have several (equal or different) base forms
// all having the same type!
if(next)
// TODO Bart 20060829. Shouldn't the next line be
// return next->addFreq(lextype,n,bf,relaxed,offset) || added;
// ?
// Bart 20061114 added || added
return next->addFreq(lextype,n,bf,relaxed,offset) || added;
else
{
if(added)
{
return true;
}
else
{
notypematch++;
notypematchcnt += n;
return false;
}
}
}
tcount Lemma::printLemmas(tcount pos,DictNode * parent,FILE * fp)
{
if(parent)
{
fprintf(fp,PERCLD ": ",pos);
}
fprintf(fp,"%d %s %s %d",S.Offset,BaseForm,Type,S.frequency);
if(next)
{
fprintf(fp,",");
return next->printLemmas(pos + 1,NULL,fp);
}
else
return pos + 1;
}
bool add(char * lextype,char * baseform,int offset)
{
if( strcmp(lextype,Type)
|| strcmp(baseform,BaseForm)
|| (unsigned int)offset != S.Offset
)
{
if(next)
return next->add(lextype,baseform,offset);
else
{
next = new Lemma(lextype,baseform,offset);
return true;
}
}
return false;
}
bool addFreq(char * lextype,int n,char * bf,bool relaxed,ptrdiff_t offset)/*20120709 unsigned int -> ptrdiff_t*/
{
bool added = false;
ptrdiff_t off = offset - S.Offset;/*20120709 int -> ptrdiff_t*/
if(!strcmp(lextype,Type) && (!bf || off >= 0))
{
if(relaxed)
{
assert(bf);
char * bfoff = bf + S.Offset;
size_t len = strlen(bfoff);
if(!strncmp(BaseForm,bfoff,len) && BaseForm[len] == ',')
{
S.frequency += n;
added = true;
//return true;
}
}
else
{
if(!bf || !strcmp(bf + S.Offset,BaseForm))
{
S.frequency += n;
added = true;
//return true;
}
}
}
/*else*/ // Full form may have several (equal or different) base forms
// all having the same type!
if(next)
return next->addFreq(lextype,n,bf,relaxed,offset) || added;
else
{
if(added)
{
return true;
}
else
{
notypematch++;
notypematchcnt += n;
return false;
}
}
}
void getStrings()
{
if(*Type)
pstrings[istrings++] = &Type;
else
{
delete [] Type;
Type = nul;
}
if(*BaseForm)
pstrings[istrings++] = &BaseForm;
else
{
delete [] BaseForm;
BaseForm = nul;
}
if(next)
next->getStrings();
}
int cmp(const Lemma * lt)
{
int d = Type - lt->Type;
if(d)
return d;
d = BaseForm - lt->BaseForm;
if(d)
return d;
d = S.Offset - lt->S.Offset;
if(d)
return d;
d = S.frequency - lt->S.frequency;
if(d)
return d;
if(next)
{
assert(lt->next);
return next->cmp(lt->next);
}
else
return 0;
}
void Lemmacnt(tcount * nLemmas)
{
++*nLemmas;
if(next)
next->Lemmacnt(nLemmas);
}
