Open Source Projects Europe

    if (!splitter.text_to_words(doc.text))

    if (!splitter.text_to_words(doc.text))

        LOGDEB(("Db::addOrUpdate: split failed for main text\n"));

        LOGDEB(("Db::addOrUpdate: split failed for main text\n"));

    ////// Special terms for other metadata. No positions for these.

    ////// Special terms for other metadata. No positions for these.

    // Mime type

    // Mime type

    newdocument.add_term(wrap_prefix(mimetype_prefix) + doc.mimetype);

    // Simple file name indexed unsplit for specific "file name"

    // Simple file name indexed unsplit for specific "file name"

    // searches. This is not the same as a filename: clause inside the

    // searches. This is not the same as a filename: clause inside the

    // query language.

    // query language.

    // We also add a term for the filename extension if any.

    // We also add a term for the filename extension if any.

        // a pathological case anyway

        // a pathological case anyway

        if (fn.size() > 230)

        if (fn.size() > 230)

        utf8truncate(fn, 230);

        utf8truncate(fn, 230);

        string::size_type pos = fn.rfind('.');

        string::size_type pos = fn.rfind('.');

        if (pos != string::npos && pos != fn.length() - 1) {

        if (pos != string::npos && pos != fn.length() - 1) {

        newdocument.add_term(wrap_prefix(fileext_prefix) + 

                     fn.substr(pos + 1));

        }

        }

        newdocument.add_term(wrap_prefix(unsplitfilename_prefix) + fn);

    }

    }

    }

    }

    // Udi unique term: this is used for file existence/uptodate

    // Udi unique term: this is used for file existence/uptodate

    // checks, and unique id for the replace_document() call.

    // checks, and unique id for the replace_document() call.

    string uniterm = make_uniterm(udi);

    string uniterm = make_uniterm(udi);

    newdocument.add_term(uniterm);

    // Parent term. This is used to find all descendents, mostly to delete them 

    // Parent term. This is used to find all descendents, mostly to delete them 

    // when the parent goes away

    // when the parent goes away

    if (!parent_udi.empty()) {

    if (!parent_udi.empty()) {

    newdocument.add_term(make_parentterm(parent_udi));

    }

    }

    // Dates etc.

    // Dates etc.

    time_t mtime = atoll(doc.dmtime.empty() ? doc.fmtime.c_str() : 

    time_t mtime = atoll(doc.dmtime.empty() ? doc.fmtime.c_str() : 

            doc.dmtime.c_str());

    struct tm *tm = localtime(&mtime);

    struct tm *tm = localtime(&mtime);

    char buf[9];

    char buf[9];

    snprintf(buf, 9, "%04d%02d%02d",

    snprintf(buf, 9, "%04d%02d%02d",

        tm->tm_year+1900, tm->tm_mon + 1, tm->tm_mday);

        tm->tm_year+1900, tm->tm_mon + 1, tm->tm_mday);

    // Date (YYYYMMDD)

    // Date (YYYYMMDD)

    newdocument.add_term(wrap_prefix(xapday_prefix) + string(buf)); 

    // Month (YYYYMM)

    // Month (YYYYMM)

    buf[6] = '\0';

    buf[6] = '\0';

    newdocument.add_term(wrap_prefix(xapmonth_prefix) + string(buf));

    // Year (YYYY)

    // Year (YYYY)

    buf[4] = '\0';

    buf[4] = '\0';

    newdocument.add_term(wrap_prefix(xapyear_prefix) + string(buf)); 

    //////////////////////////////////////////////////////////////////

    //////////////////////////////////////////////////////////////////

    // Document data record. omindex has the following nl separated fields:

    // Document data record. omindex has the following nl separated fields:

    // - url

    // - url