feat(search): Improve search functionality: update documentation, refine Lucene query transformation, and add advanced search tests

This commit is contained in:
2025-04-21 15:27:55 +00:00
parent 0834ec5c91
commit 23b499b3a8
7 changed files with 373 additions and 76 deletions

View File

@@ -343,22 +343,72 @@ export class SmartDataDbDoc<T extends TImplements, TImplements, TManager extends
if (searchableFields.length === 0) {
throw new Error(`No searchable fields defined for class ${className}`);
}
// field:value exact match (case-sensitive for non-regex fields)
const fv = query.match(/^(\w+):(.+)$/);
if (fv) {
const field = fv[1];
const value = fv[2];
// empty query -> return all
const q = query.trim();
if (!q) {
return await (this as any).getInstances({});
}
// simple exact field:value (no spaces, no wildcards, no quotes)
const simpleExact = q.match(/^(\w+):([^"'\*\?\s]+)$/);
if (simpleExact) {
const field = simpleExact[1];
const value = simpleExact[2];
if (!searchableFields.includes(field)) {
throw new Error(`Field '${field}' is not searchable for class ${className}`);
}
return await (this as any).getInstances({ [field]: value });
}
// safe regex across all searchable fields (case-insensitive)
const escaped = escapeForRegex(query);
const orConditions = searchableFields.map((field) => ({
[field]: { $regex: escaped, $options: 'i' },
}));
return await (this as any).getInstances({ $or: orConditions });
// quoted phrase across all searchable fields: exact match of phrase
const quoted = q.match(/^"(.+)"$|^'(.+)'$/);
if (quoted) {
const phrase = quoted[1] || quoted[2] || '';
// build regex that matches the exact phrase (allowing flexible whitespace)
const parts = phrase.split(/\s+/).map((t) => escapeForRegex(t));
const pattern = parts.join('\\s+');
const orConds = searchableFields.map((f) => ({ [f]: { $regex: pattern, $options: 'i' } }));
return await (this as any).getInstances({ $or: orConds });
}
// wildcard field:value (supports * and ?) -> direct regex on that field
const wildcardField = q.match(/^(\w+):(.+[*?].*)$/);
if (wildcardField) {
const field = wildcardField[1];
const pattern = wildcardField[2];
if (!searchableFields.includes(field)) {
throw new Error(`Field '${field}' is not searchable for class ${className}`);
}
// escape regex special chars except * and ?, then convert wildcards
const escaped = pattern.replace(/([.+^${}()|[\\]\\])/g, '\\$1');
const regexPattern = escaped.replace(/\*/g, '.*').replace(/\?/g, '.');
return await (this as any).getInstances({ [field]: { $regex: regexPattern, $options: 'i' } });
}
// wildcard plain term across all fields (supports * and ?)
if (!q.includes(':') && (q.includes('*') || q.includes('?'))) {
// build wildcard regex pattern: escape all except * and ? then convert
const escaped = q.replace(/([.+^${}()|[\\]\\])/g, '\\$1');
const pattern = escaped.replace(/\*/g, '.*').replace(/\?/g, '.');
const orConds = searchableFields.map((f) => ({ [f]: { $regex: pattern, $options: 'i' } }));
return await (this as any).getInstances({ $or: orConds });
}
// detect advanced Lucene syntax: field:value, wildcards, boolean, grouping
const luceneSyntax = /(\w+:[^\s]+)|\*|\?|\bAND\b|\bOR\b|\bNOT\b|\(|\)/;
if (luceneSyntax.test(q)) {
const filter = (this as any).createSearchFilter(q);
return await (this as any).getInstances(filter);
}
// multi-term unquoted -> AND of regex across fields for each term
const terms = q.split(/\s+/);
if (terms.length > 1) {
const andConds = terms.map((term) => {
const esc = escapeForRegex(term);
const ors = searchableFields.map((f) => ({ [f]: { $regex: esc, $options: 'i' } }));
return { $or: ors };
});
return await (this as any).getInstances({ $and: andConds });
}
// single term -> regex across all searchable fields
const esc = escapeForRegex(q);
const orConds = searchableFields.map((f) => ({ [f]: { $regex: esc, $options: 'i' } }));
return await (this as any).getInstances({ $or: orConds });
}