|
a/src/rcldb/stemdb.cpp |
|
b/src/rcldb/stemdb.cpp |
|
... |
|
... |
| 41 |
namespace Rcl {
|
41 |
namespace Rcl {
|
| 42 |
|
42 |
|
| 43 |
/**
|
43 |
/**
|
| 44 |
* Expand for one or several languages
|
44 |
* Expand for one or several languages
|
| 45 |
*/
|
45 |
*/
|
| 46 |
bool StemDb::stemExpand(const std::string& langs, const std::string& term,
|
46 |
bool StemDb::stemExpand(const std::string& langs, const std::string& _term,
|
| 47 |
vector<string>& result)
|
47 |
vector<string>& result)
|
| 48 |
{
|
48 |
{
|
| 49 |
vector<string> llangs;
|
49 |
vector<string> llangs;
|
| 50 |
stringToStrings(langs, llangs);
|
50 |
stringToStrings(langs, llangs);
|
|
|
51 |
|
|
|
52 |
// The stemdb keys may have kept their diacritics or not but they
|
|
|
53 |
// are always lower-case. It would be more logical for the term
|
|
|
54 |
// transformers to perform before doing the stemming, but this
|
|
|
55 |
// would be inefficient when there are several stemming languages
|
|
|
56 |
string term;
|
|
|
57 |
unacmaybefold(_term, term, "UTF-8", UNACOP_FOLD);
|
| 51 |
|
58 |
|
| 52 |
for (vector<string>::const_iterator it = llangs.begin();
|
59 |
for (vector<string>::const_iterator it = llangs.begin();
|
| 53 |
it != llangs.end(); it++) {
|
60 |
it != llangs.end(); it++) {
|
| 54 |
SynTermTransStem stemmer(*it);
|
61 |
SynTermTransStem stemmer(*it);
|
| 55 |
XapComputableSynFamMember expander(getdb(), synFamStem, *it, &stemmer);
|
62 |
XapComputableSynFamMember expander(getdb(), synFamStem, *it, &stemmer);
|