sqlite-comprehend by simonw

64 downloads this week        Star

README source code


PyPI Changelog Tests License

Tools for running data in a SQLite database through AWS Comprehend

See sqlite-comprehend: run AWS entity extraction against content in a SQLite database for background on this project.


Install this tool using pip:

pip install sqlite-comprehend


You can see examples of tables generated using this command here:


You will need AWS credentials with the comprehend:BatchDetectEntities IAM permission.

You can configure credentials using these instructions. You can also save them to a JSON or INI configuration file and pass them to the command using -a credentials.ini, or pass them using the --access-key and --secret-key options.

Entity extraction

The sqlite-comprehend entities command runs entity extraction against every row in the specified table and saves the results to your database.

Specify the database, the table and one or more columns containing text in that table. The following runs against the text column in the pages table of the sfms.db SQLite database:

sqlite-comprehend sfms.db pages text

Results will be written into a pages_comprehend_entities table. Change the name of the output table by passing -o other_table_name.

You can run against a subset of rows by adding a --where clause:

sqlite-comprehend sfms.db pages text --where 'id < 10'

You can also used named parameters in your --where clause:

sqlite-comprehend sfms.db pages text --where 'id < :maxid' -p maxid 10

Only the first 5,000 characters for each row will be considered. Be sure to review Comprehend's pricing - which starts at $0.0001 per hundred characters.

If your context includes HTML tags, you can strip them out before extracting entities by adding --strip-tags:

sqlite-comprehend sfms.db pages text --strip-tags

Rows that have been processed are recorded in the pages_comprehend_entities_done table. If you run the command more than once it will only process rows that have been newly added.

You can delete records from that _done table to run them again.

sqlite-comprehend entities --help

Usage: sqlite-comprehend entities [OPTIONS] DATABASE TABLE COLUMNS...

  Detect entities in columns in a table

  To extract entities from columns text1 and text2 in mytable:

      sqlite-comprehend entities my.db mytable text1 text2

  To run against just a subset of the rows in the table, add:

      --where "id < :max_id" -p max_id 50

  Results will be written to a table called mytable_comprehend_entities

  To specify a different output table, use -o custom_table_name

  --where TEXT                WHERE clause to filter table
  -p, --param <TEXT TEXT>...  Named :parameters for SQL query
  -o, --output TEXT           Custom output table
  -r, --reset                 Start from scratch, deleting previous results
  --strip-tags                Strip HTML tags before extracting entities
  --access-key TEXT           AWS access key ID
  --secret-key TEXT           AWS secret access key
  --session-token TEXT        AWS session token
  --endpoint-url TEXT         Custom endpoint URL
  -a, --auth FILENAME         Path to JSON/INI file containing credentials
  --help                      Show this message and exit.


Assuming an input table called pages the tables created by this tool will have the following schema:

CREATE TABLE [pages] (
   [text] TEXT
CREATE TABLE [comprehend_entity_types] (
   [value] TEXT
CREATE TABLE [comprehend_entities] (
   [name] TEXT,
   [type] INTEGER REFERENCES [comprehend_entity_types]([id])
CREATE TABLE [pages_comprehend_entities] (
   [id] INTEGER REFERENCES [pages]([id]),
   [score] FLOAT,
   [entity] INTEGER REFERENCES [comprehend_entities]([id]),
   [begin_offset] INTEGER,
   [end_offset] INTEGER
CREATE UNIQUE INDEX [idx_comprehend_entity_types_value]
    ON [comprehend_entity_types] ([value]);
CREATE UNIQUE INDEX [idx_comprehend_entities_type_name]
    ON [comprehend_entities] ([type], [name]);
CREATE TABLE [pages_comprehend_entities_done] (


To contribute to this tool, first checkout the code. Then create a new virtual environment:

cd sqlite-comprehend
python -m venv venv
source venv/bin/activate

Now install the dependencies and test dependencies:

pip install -e '.[test]'

To run the tests: