# Searching for individual fields in a single search vector

**URL:** <https://forum.djangoproject.com/t/searching-for-individual-fields-in-a-single-search-vector/12522>\
**Category:** Using Django\
**Created:** [March 9, 2022, 5:37pm UTC](https://forum.djangoproject.com/t/searching-for-individual-fields-in-a-single-search-vector/12522 "2022-03-09T17:37:23Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![mabersold](https://sea2.discourse-cdn.com/flex026/user_avatar/forum.djangoproject.com/mabersold/32/6497_2.png) [@mabersold](https://forum.djangoproject.com/u/mabersold)\
**Post date:** [March 9, 2022, 5:37pm UTC](https://forum.djangoproject.com/t/searching-for-individual-fields-in-a-single-search-vector/12522/1 "2022-03-09T17:37:23Z")

</div>

I have a model with multiple fields that I want to be searchable, and I want to use a search vector in the database for searching. Let’s say my model is Song, and a song has several searchable fields: a title, a filename, and comment text. In addition, there is also a SearchVector field called search\_document that is meant to be used in the search query. Here’s a simplified version of the class:

```auto
class Song(models.Model):
    filename=models.CharField()
    title=models.CharField()
    comment_text=models.TextField()
    search_document=SearchVectorField()

```

The search vector is populated via a signal. Whenever an object is saved, the signal creates a search vector using the following weighting:

```auto
def index_components(self):
        return {
            'A': self.title,
            'B': self.filename,
            'C': self.comment_text
        }

```

The method called in the signal looks like this:

```auto
def make_updater(instance):
    components = instance.index_components()
    pk = instance.pk

    def on_commit():
        search_vectors = []
        for weight, text in components.items():
            search_vectors.append(
                SearchVector(Value(text, output_field=TextField()), weight=weight)
            )
        instance. __class__.objects.filter(pk=pk).update(
            search_document=reduce(operator.add, search_vectors)
        )

    return on_commit

```

And finally, a search query on the Song class looks like this:

```auto
Song.objects.annotate(
    type=Value('song', output_field=CharField()),
        rank=rank_annotation
    ).filter(
        search_document=query
    ).values('pk', 'type', 'rank')

```

This works, but it searches for all three fields together. I would like to also be able to search only for a particular field. In particular, I would like the search for comment\_text to be optional, and to be able to search for title and filename without comment\_text.

Is it possible to modify my query to only search for title and filename without comment\_text while still keeping all three within the same SearchVectorField? Or will I need to store comment\_text into a different SearchVectorField?