|
| 1 | +.. _rust-search-text-guide: |
| 2 | + |
| 3 | +=========== |
| 4 | +Search Text |
| 5 | +=========== |
| 6 | + |
| 7 | +.. facet:: |
| 8 | + :name: genre |
| 9 | + :values: reference |
| 10 | + |
| 11 | +.. meta:: |
| 12 | + :keywords: code example, full text, string search |
| 13 | + |
| 14 | +.. contents:: On this page |
| 15 | + :local: |
| 16 | + :backlinks: none |
| 17 | + :depth: 2 |
| 18 | + :class: singlecol |
| 19 | + |
| 20 | +Overview |
| 21 | +-------- |
| 22 | + |
| 23 | +In this guide, you can learn how to use the {+driver-short+} to run a text search. |
| 24 | +A text search allows you to efficiently query fields with string values. |
| 25 | + |
| 26 | +.. important:: |
| 27 | + |
| 28 | + MongoDB text search differs from the more powerful Atlas Search |
| 29 | + feature. To learn more, see the :atlas:`Atlas Search documentation |
| 30 | + </atlas-search/>`. |
| 31 | + |
| 32 | +This guide includes the following sections: |
| 33 | + |
| 34 | +- :ref:`Sample Data for Examples <rust-text-sample-data>` presents |
| 35 | + the sample data that is used by the text search examples |
| 36 | + |
| 37 | +- :ref:`Text Index <rust-text-index>` describes how to create a text |
| 38 | + index on a string-valued field |
| 39 | + |
| 40 | +- :ref:`Text Search <rust-text-search>` describes how to perform text |
| 41 | + searches with different search criteria |
| 42 | + |
| 43 | +- :ref:`Aggregation <rust-search-text-aggregation>` describes how to |
| 44 | + perform text searches by using aggregation pipelines |
| 45 | + |
| 46 | +- :ref:`Additional Information <rust-crud-text-addtl-info>` |
| 47 | + provides links to resources and API documentation for types |
| 48 | + and methods mentioned in this guide |
| 49 | + |
| 50 | +.. _rust-text-sample-data: |
| 51 | + |
| 52 | +Sample Data for Examples |
| 53 | +------------------------ |
| 54 | + |
| 55 | +The examples in this guide use the following ``Dish`` struct as a model |
| 56 | +for documents in the ``menu`` collection: |
| 57 | + |
| 58 | +.. literalinclude:: /includes/fundamentals/code-snippets/crud/text.rs |
| 59 | + :start-after: start-dish-struct |
| 60 | + :end-before: end-dish-struct |
| 61 | + :language: rust |
| 62 | + :dedent: |
| 63 | + |
| 64 | +The examples use the following sample documents that describe |
| 65 | +dishes you can order at a restaurant: |
| 66 | + |
| 67 | +.. code-block:: json |
| 68 | + :copyable: false |
| 69 | + |
| 70 | + { "name": "Shepherd’s Pie", "description": "A vegetarian take on the classic dish that uses lentils as a base. Serves 2." }, |
| 71 | + { "name": "Green Curry", "description": "A flavorful Thai curry, made vegetarian with tofu. Vegetarian and vegan friendly." }, |
| 72 | + { "name": "Herbed Branzino", "description": "Grilled whole fish stuffed with herbs and pomegranate seeds. Serves 3-4." }, |
| 73 | + { "name": "Kale Tabbouleh", "description": "A bright, herb-based salad. A perfect starter for vegetarians and vegans." }, |
| 74 | + { "name": "Garlic Butter Trout", "description": "Baked trout seasoned with garlic, lemon, dill, and, of course, butter. Serves 2." } |
| 75 | + |
| 76 | +.. _rust-text-index: |
| 77 | + |
| 78 | +Text Index |
| 79 | +---------- |
| 80 | + |
| 81 | +Before you perform a text search, you must create a **text index** on |
| 82 | +the collection. A text index specifies the string or string array field |
| 83 | +on which you can perform text searches. |
| 84 | + |
| 85 | +The examples in this guide perform text searches on the |
| 86 | +``description`` field of documents in the ``menu`` collection. To enable |
| 87 | +text searches on the ``description`` field, create a text index as shown |
| 88 | +in the following code: |
| 89 | + |
| 90 | +.. literalinclude:: /includes/fundamentals/code-snippets/crud/text.rs |
| 91 | + :language: rust |
| 92 | + :dedent: |
| 93 | + :start-after: begin-idx |
| 94 | + :end-before: end-idx |
| 95 | + |
| 96 | +.. _rust-text-search: |
| 97 | + |
| 98 | +Text Search |
| 99 | +----------- |
| 100 | + |
| 101 | +A text search retrieves documents that contain a specified **term** or |
| 102 | +**phrase** in the value of the indexed field. A term is a sequence of |
| 103 | +characters that excludes whitespace characters. A phrase is a sequence |
| 104 | +of terms with any number of whitespace characters. |
| 105 | + |
| 106 | +To perform a text search, include the ``$text`` evaluation query operator, |
| 107 | +followed by the ``$search`` field in your query filter. The ``$text`` operator |
| 108 | +specifies that you are performing a text search on the text-indexed |
| 109 | +fields. The ``$search`` field specifies the term or phrase to search for |
| 110 | +in the text-indexed field or fields. |
| 111 | + |
| 112 | +Query filters for text searches use the following format: |
| 113 | + |
| 114 | +.. code-block:: rust |
| 115 | + |
| 116 | + let filter = doc! { "$text": { "$search": "<search term or phrase>" } }; |
| 117 | + |
| 118 | +.. _rust-text-term: |
| 119 | + |
| 120 | +Search for a Term |
| 121 | +~~~~~~~~~~~~~~~~~ |
| 122 | + |
| 123 | +To search for a term, specify the term as a string in your query filter. |
| 124 | +To search for multiple terms, separate each term with a space. |
| 125 | + |
| 126 | +.. note:: |
| 127 | + |
| 128 | + When searching for multiple terms, the ``find()`` method returns any |
| 129 | + document in which the text-indexed field or fields contain at least |
| 130 | + one of the terms. |
| 131 | + |
| 132 | + For example, if your search terms are ``"one two |
| 133 | + three"``, MongoDB returns documents in which the indexed field |
| 134 | + contains ``"one"``, ``"two"``, ``"three"``, or more than one of these terms. |
| 135 | + |
| 136 | +Example |
| 137 | +``````` |
| 138 | + |
| 139 | +The following example performs a search for documents in which the |
| 140 | +``description`` field contains the term ``"herb"``: |
| 141 | + |
| 142 | +.. io-code-block:: |
| 143 | + :copyable: true |
| 144 | + |
| 145 | + .. input:: /includes/fundamentals/code-snippets/crud/text.rs |
| 146 | + :start-after: begin-by-term |
| 147 | + :end-before: end-by-term |
| 148 | + :language: rust |
| 149 | + :dedent: |
| 150 | + |
| 151 | + .. output:: |
| 152 | + :language: none |
| 153 | + :visible: false |
| 154 | + |
| 155 | + Dish { name: "Kale Tabbouleh", description: "A bright, herb-based salad. A perfect starter for vegetarians and vegans." } |
| 156 | + Dish { name: "Herbed Branzino", description: "Grilled whole fish stuffed with herbs and pomegranate seeds. Serves 3-4." } |
| 157 | + |
| 158 | +.. tip:: |
| 159 | + |
| 160 | + Even though the search term is ``"herb"``, the text search also matches |
| 161 | + documents in which the ``description`` field contains ``"herbs"``. |
| 162 | + This is because a MongoDB text index uses suffix stemming to match |
| 163 | + similar words. To learn more about how MongoDB matches terms, see |
| 164 | + :manual:`Index Entries </core/index-text/#index-entries>` in the |
| 165 | + Server manual. |
| 166 | + |
| 167 | +.. _rust-text-phrase: |
| 168 | + |
| 169 | +Search for a Phrase |
| 170 | +~~~~~~~~~~~~~~~~~~~ |
| 171 | + |
| 172 | +To search for a phrase, specify the phrase with escaped quotes in your |
| 173 | +query filter: |
| 174 | + |
| 175 | +.. code-block:: rust |
| 176 | + |
| 177 | + let filter = doc! { "$text": { "$search": "\"<some phrase>\"" } }; |
| 178 | + |
| 179 | +If you don't add escaped quotes around the phrase, the |
| 180 | +search performs a :ref:`term search <rust-text-term>`. |
| 181 | + |
| 182 | +Example |
| 183 | +``````` |
| 184 | + |
| 185 | +The following example performs a search for documents in which the |
| 186 | +``description`` field contains the phrase ``"serves 2"``: |
| 187 | + |
| 188 | +.. io-code-block:: |
| 189 | + :copyable: true |
| 190 | + |
| 191 | + .. input:: /includes/fundamentals/code-snippets/crud/text.rs |
| 192 | + :start-after: begin-by-phrase |
| 193 | + :end-before: end-by-phrase |
| 194 | + :language: rust |
| 195 | + :dedent: |
| 196 | + |
| 197 | + .. output:: |
| 198 | + :language: none |
| 199 | + :visible: false |
| 200 | + |
| 201 | + Dish { name: "Shepherd’s Pie", description: "A vegetarian take on the classic dish that uses lentils as a base. Serves 2." } |
| 202 | + Dish { name: "Garlic Butter Trout", description: "Baked trout seasoned with garlic, lemon, dill, and, of course, butter. Serves 2." } |
| 203 | + |
| 204 | +Exclude Terms from Search |
| 205 | +~~~~~~~~~~~~~~~~~~~~~~~~~ |
| 206 | + |
| 207 | +To specify a term or phrase you want to exclude from your text search, |
| 208 | +prefix it with a minus sign in your query filter: |
| 209 | + |
| 210 | +.. code-block:: rust |
| 211 | + |
| 212 | + let filter = doc! { "$text": { "$search": "<term> -<excluded term>" } }; |
| 213 | + |
| 214 | +.. important:: |
| 215 | + |
| 216 | + You must search for at least one term or phrase to exclude |
| 217 | + other terms from your search. If you only exclude terms, the |
| 218 | + search doesn't return any documents. |
| 219 | + |
| 220 | +Example |
| 221 | +``````` |
| 222 | + |
| 223 | +The following example performs a search for documents in which the |
| 224 | +``description`` field contains the term ``"vegan"``, but does not |
| 225 | +contain the term ``"tofu"``: |
| 226 | + |
| 227 | +.. io-code-block:: |
| 228 | + :copyable: true |
| 229 | + |
| 230 | + .. input:: /includes/fundamentals/code-snippets/crud/text.rs |
| 231 | + :start-after: begin-exclude-term |
| 232 | + :end-before: end-exclude-term |
| 233 | + :language: rust |
| 234 | + :dedent: |
| 235 | + |
| 236 | + .. output:: |
| 237 | + :language: none |
| 238 | + :visible: false |
| 239 | + |
| 240 | + Dish { name: "Kale Tabbouleh", description: "A bright, herb-based salad. A perfect starter for vegetarians and vegans." } |
| 241 | + |
| 242 | +Sort by Relevance |
| 243 | +~~~~~~~~~~~~~~~~~ |
| 244 | + |
| 245 | +A text search assigns a numerical text score to indicate how closely |
| 246 | +each result matches the string in your query filter. A higher text score |
| 247 | +indicates that the result is more relevant to your query. To reveal the text |
| 248 | +score in your output, use a projection to retrieve the ``textScore`` |
| 249 | +field from the metadata. You can sort the text score in descending order |
| 250 | +by specifying a sort on the ``textScore`` metadata field. |
| 251 | + |
| 252 | +Example |
| 253 | +``````` |
| 254 | + |
| 255 | +This example performs the following actions: |
| 256 | + |
| 257 | +- Performs a search for documents in which the ``description`` field contains the term ``"vegetarian"`` |
| 258 | +- Sorts the results in descending order on text score |
| 259 | +- Includes only the ``name`` and ``score`` fields in the output |
| 260 | + |
| 261 | +.. io-code-block:: |
| 262 | + :copyable: true |
| 263 | + |
| 264 | + .. input:: /includes/fundamentals/code-snippets/crud/text.rs |
| 265 | + :start-after: begin-sort-relevance |
| 266 | + :end-before: end-sort-relevance |
| 267 | + :language: rust |
| 268 | + :dedent: |
| 269 | + |
| 270 | + .. output:: |
| 271 | + :language: none |
| 272 | + :visible: false |
| 273 | + |
| 274 | + Document({"name": String("Green Curry"), "score": Double(0.9166666666666667)}) |
| 275 | + Document({"name": String("Kale Tabbouleh"), "score": Double(0.5625)}) |
| 276 | + Document({"name": String("Shepherd’s Pie"), "score": Double(0.5555555555555556)}) |
| 277 | + |
| 278 | +.. _rust-search-text-aggregation: |
| 279 | + |
| 280 | +Aggregation |
| 281 | +----------- |
| 282 | + |
| 283 | +You can include the ``$text`` evaluation query operator in a |
| 284 | +:manual:`$match </reference/operator/aggregation/match/>` aggregation |
| 285 | +stage to perform a text search in an aggregation pipeline. |
| 286 | + |
| 287 | +The following sections demonstrate how to perform text searches by using |
| 288 | +aggregation pipelines instead of the ``find()`` method. |
| 289 | + |
| 290 | +Match a Search Term |
| 291 | +~~~~~~~~~~~~~~~~~~~ |
| 292 | + |
| 293 | +The following example uses an aggregation to perform a search for |
| 294 | +documents in which the ``description`` field contains the term |
| 295 | +``"herb"``: |
| 296 | + |
| 297 | +.. io-code-block:: |
| 298 | + :copyable: true |
| 299 | + |
| 300 | + .. input:: /includes/fundamentals/code-snippets/crud/text.rs |
| 301 | + :start-after: begin-agg-term |
| 302 | + :end-before: end-agg-term |
| 303 | + :language: rust |
| 304 | + :dedent: |
| 305 | + |
| 306 | + .. output:: |
| 307 | + :language: none |
| 308 | + :visible: false |
| 309 | + |
| 310 | + Document({"_id": ObjectId("..."), "name": String("Kale Tabbouleh"), "description": String("A bright, herb-based salad. A perfect starter for vegetarians and vegans.")}) |
| 311 | + Document({"_id": ObjectId("..."), "name": String("Herbed Branzino"), "description": String("Grilled whole fish stuffed with herbs and pomegranate seeds. Serves 3-4.")}) |
| 312 | + |
| 313 | +Sort by Relevance |
| 314 | +~~~~~~~~~~~~~~~~~ |
| 315 | + |
| 316 | +This example uses an aggregation to perform the following actions: |
| 317 | + |
| 318 | +- Performs a search for documents in which the ``description`` field contains the term ``"vegetarian"`` |
| 319 | +- Sorts the results in descending order on text score |
| 320 | +- Includes only the ``name`` and ``score`` fields in the output |
| 321 | + |
| 322 | +.. io-code-block:: |
| 323 | + :copyable: true |
| 324 | + |
| 325 | + .. input:: /includes/fundamentals/code-snippets/crud/text.rs |
| 326 | + :start-after: begin-agg-relevance |
| 327 | + :end-before: end-agg-relevance |
| 328 | + :language: rust |
| 329 | + :dedent: |
| 330 | + |
| 331 | + .. output:: |
| 332 | + :language: none |
| 333 | + :visible: false |
| 334 | + |
| 335 | + Document({"name": String("Green Curry"), "score": Double(0.9166666666666667)}) |
| 336 | + Document({"name": String("Kale Tabbouleh"), "score": Double(0.5625)}) |
| 337 | + Document({"name": String("Shepherd’s Pie"), "score": Double(0.5555555555555556)}) |
| 338 | + |
| 339 | +.. _rust-crud-text-addtl-info: |
| 340 | + |
| 341 | +Additional Information |
| 342 | +---------------------- |
| 343 | + |
| 344 | +For a runnable example that uses the ``find()`` method, see the |
| 345 | +:ref:`rust-find-usage` usage example. |
| 346 | + |
| 347 | +To learn more about the operations in this guide, see the |
| 348 | +following documentation: |
| 349 | + |
| 350 | +- :ref:`rust-query-guide` |
| 351 | +- :ref:`rust-retrieve-guide` |
| 352 | +- :ref:`rust-aggregation` |
| 353 | +- :ref:`rust-indexes` |
| 354 | +- :manual:`Text Indexes </core/index-text/>` in the Server manual |
| 355 | +- :manual:`$text </reference/operator/query/text/>` in the Server manual |
| 356 | +- :manual:`$meta </reference/operator/aggregation/meta/>` in the Server manual |
| 357 | + |
| 358 | +API Documentation |
| 359 | +~~~~~~~~~~~~~~~~~ |
| 360 | + |
| 361 | +To learn more about the methods and types mentioned in this |
| 362 | +guide, see the following API documentation: |
| 363 | + |
| 364 | +- `find() <{+api+}/struct.Collection.html#method.find>`__ |
| 365 | +- `FindOptions <{+api+}/options/struct.FindOptions.html>`__ |
| 366 | +- `Cursor <{+api+}/struct.Cursor.html>`__ |
| 367 | +- `clone_with_type() <{+api+}/struct.Collection.html#method.clone_with_type>`__ |
| 368 | +- `aggregate() <{+api+}/struct.Collection.html#method.aggregate>`__ |
0 commit comments