PDF Find Table

Explore Samples

AI powered document analysis can scan your document for tables and return the array of tables on pages with coordinates and information about columns detected in these tables.

Available Methods

[POST] /pdf/find/table (AI powered)
[POST] /pdf/find/table (legacy table finder)

[POST] /pdf/find/table (AI powered)

This function finds tables in documents using an AI-powered table detection engine.

This endpoint locates tables in an input PDF document and returns JSON with:

The array of tables objects;
X, Y, Width, and Height coordinates for every table found;
Rect param for every table that you can re-use with pdf/convert/to/json, pdf/convert/to/csv, pdf/convert/to/csv, and other endpoints to extract a selected table only;
PageIndex page index for a page with a table. The very first page is 0 (zero);
Columns array with the set of X coordinates for every column inside the table that was found;

To extract the table into CSV, JSON, or XML please use pdf/convert/to/csv, pdf/convert/to/json2, and pdf/convert/to/xml endpoints with rect parameter value from rect output param for this table accordingly.

Attributes
url required URL to the source file. Supports links from Google Drive, Dropbox and from built-in PDF.co files storage. For uploading files via API please check Files Upload section. If you are randomly getting `Too Many Requests` or `Access Denied` error for your input URL, Please try to add `cache:` to enable built-in URL caching. You can also `encrypt data` for output files and `decrypt data` input files with user-controlled data encryption to learn more.
httpusername optional HTTP auth user name if required to access source `url`.
httppassword optional HTTP auth password if required to access source `url`.
pages optional Comma-separated list of page indices (or ranges) to process. IMPORTANT: the very first page starts at `0` (zero). To set a range use the dash `-`, for example: `0,2-5,7-`. To set a range from the index to the last page use range like this: `2-` (from page #3 as the index starts at zero and till the of the document). For ALL pages just leave this param empty. Example: `0,2-5,7-` means first page, then 3rd page to 6th page, and then the range from 8th (index = `7`) page till the end of the document, The input must be in string format.
inline optional Must be one of: `true`, or `false`. When `false`, the endpoint returns a link to the `.json` file with the output.
password optional Password of PDF file, The input must be in string format.
async optional Runs processing asynchronously and returns `JobId` that you may use with `/job/check` to check the state of the processing (possible states: `working`, `failed`, `aborted` and `success`). Must be one of: `true`, or `false`.
name optional File name for the generated output, The input must be in string format.
expiration optional Output link expiration in minutes. The default is `60` (i.e. 60 minutes or 1 hour). After this delay generated output file(s) (if any) will be auto-removed from PDF.co temporary files storage. Max allowed expiration period depends on your current subscription plan. To store permanent input files (e.g. re-usable images, pdf, documents), please use PDF.co built-in Files Storage instead.
profiles optional Use this parameter to set additional configurations for fine-tuning and extra options. Explore PDF.co knowledgebase for profile examples, The input must be in string format.

Method: POST
URL: /v1/pdf/find/table

Query parameters

No query parameters accepted.

Body payload

{
    "url": "pdfco-test-files.s3.us-west-2.amazonaws.compdf-to-text/sample.pdf",
    "async": "false",
    "inline": "true",
    "password": ""
}

Example responses

/pdf/find/table

{
    "body": {
        "tables": [
            {
                "PageIndex": 0,
                "X": 36,
                "Y": 34.4400024,
                "Width": 523.44,
                "Height": 160.82,
                "Columns": [
                    357.675
                ],
                "rect": "36, 34.4400024, 523.44, 160.82"
            },
            {
                "PageIndex": 0,
                "X": 36,
                "Y": 316.249969,
                "Width": 523.44,
                "Height": 120.620026,
                "Columns": [
                    157.117,
                    340.68,
                    475.84
                ],
                "rect": "36, 316.249969, 523.44, 120.620026"
            }
        ]
    },
    "pageCount": 1,
    "error": false,
    "status": 200,
    "name": "sample.json",
    "remainingCredits": 98892697,
    "credits": 21
}

Code Snippet

CURL

curl --location --request POST 'https://api.pdf.co/v1/pdf/find/table' \
--header 'x-api-key: ' \
--header 'Content-Type: application/json' \
--data-raw '{
    "url": "pdfco-test-files.s3.us-west-2.amazonaws.compdf-to-text/sample.pdf",
    "async": "false",
    "inline": "true",
    "password": ""
}'

[POST] /pdf/find/table (legacy table finder)

This function finds tables in documents using AI-powered table detection engine but you can enable legacy table finder mode as well via profiles parameter (see below).

This endpoint locates tables in input PDF document and returns JSON with:

array of tables objects;
X, Y, Width, Height coordinates for every table found;
rect param for every table that you can re-use with pdf/convert/to/json, pdf/convert/to/csv, pdf/convert/to/csv and other endpoints to extract a selected table only;
PageIndex page index for a page with table. The very first page is 0 (zero);
Columns array with the set of X coordinates for every column inside table that was found;

To extract table into CSV, JSON or XML please use pdf/convert/to/csv, pdf/convert/to/json2, pdf/convert/to/xml endpoints with rect parameter value from rect output param for this table accordingly.

Attributes
url required URL to the source file. Supports links from Google Drive, Dropbox and from built-in PDF.co files storage. For uploading files via API please check Files Upload section. If you are randomly getting a `Too Many Requests` or `Access Denied` error for your input URL, Please try to add `cache:` to enable built-in URL caching. You can also `encrypt data` for output files and `decrypt data` input files with user-controlled data encryption to learn more.
httpusername optional HTTP auth user name if required to access source `url`.
httppassword optional HTTP auth password if required to access source `url`.
pages optional Comma-separated list of page indices (or ranges) to process. IMPORTANT: the very first page starts at `0` (zero). To set a range use the dash `-`, for example: `0,2-5,7-`. To set a range from the index to the last page use range like this: `2-` (from page #3 as the index starts at zero and till the of the document). For ALL pages just leave this param empty. Example: `0,2-5,7-` means first page, then 3rd page to 6th page, and then the range from 8th (index = `7`) page till the end of the document, The input must be in string format.
inline optional Must be one of: `true`, or `false`. When `false`, the endpoint returns a link to the `.json` file with the output.
password optional Password of PDF file, The input must be in string format.
async optional Runs processing asynchronously and returns `JobId` that you may use with `/job/check` to check the state of the processing (possible states: `working`, `failed`, `aborted` and `success`). Must be one of: `true`, or `false`.
name optional File name for the generated output, The input must be in string format.
expiration optional Output link expiration in minutes. The default is `60` (i.e. 60 minutes or 1 hour). After this delay generated output file(s) (if any) will be auto-removed from PDF.co temporary files storage. Max allowed expiration period depends on your current subscription plan. To store permanent input files (e.g. re-usable images, pdf, documents), please use PDF.co built-in Files Storage instead.
profiles optional Use this parameter to set additional configurations for fine-tuning and extra options. Explore PDF.co knowledgebase for profile examples, The input must be in string format.

Legacy mode can be enabled like this:

"profiles": "{ 'Mode': 'Legacy'}"

or with a more detailed config telling min required rows, min columns, and column detection mode.

"profiles": "{ 'Mode': 'Legacy', 'ColumnDetectionMode': 'BorderedTables', 'DetectionMinNumberOfRows': 1, 'DetectionMinNumberOfColumns': 1, 'DetectionMaxNumberOfInvalidSubsequentRowsAllowed': 0, 'DetectionMinNumberOfLineBreaksBetweenTables': 0, 'EnhanceTableBorders': false }"

Method: POST
URL: /v1/pdf/find/table