File
In Alfred, a file is an individual document or data unit undergoing specialized operations tailored for document analysis and management. It is processed by the platform for tasks such as classification, optical character recognition (OCR), and data extraction. Each file, as a distinct entity, transitions through various statuses, which mark its progress in the automated workflow.
For example, a file could be a digital version of a business contract, initially uploaded for processing, then analyzed for key content, classified according to its nature, and finally integrated into a database for efficient access and utilization. This process underscores the file's role in Alfred's robust document handling ecosystem.
Roles and Responsibilities
Data Carrier and Information Source: At its core, a file in Alfred serves as a carrier of data and a source of information. From the moment it is uploaded, through various processing stages, to its final state, it retains and conveys valuable information that is key to document processing tasks.
Subject of Automated Processes: Throughout its lifecycle, the file acts as the subject of various automated processes like OCR, classification, extraction, and indexing. Its role is to be efficiently processed by these automated systems, ensuring accurate and effective handling of the document's content.
Facilitator of Workflow Progression: As the file moves through different statuses, it facilitates the progression of the overall workflow in Alfred. Each change in status reflects a step forward in the document processing pipeline, contributing to the workflow's momentum.
Intermediary for User Interaction: In stages requiring user input or confirmation, the file acts as an intermediary between the automated system and the user. It presents processed information for user review, validation, or further input, bridging the gap between automation and human oversight.
Indicator of System Health and Efficiency: The file's progression through various statuses also serves as an indicator of the system's health and efficiency. Stages like error handling or re-queuing provide insights into potential issues or bottlenecks in the workflow, prompting necessary adjustments or improvements.
End-Point of Processing and Analysis: Finally, in its concluded states, whether successfully processed, tagged, or in error and deletion stages, the file represents the end-point of document processing and analysis. It signifies the completion of a cycle within the Alfred platform, either culminating in successful integration into the system or identifying areas for system refinement.
File Retrieval EndPoints
Get all files from your company
GET https://tagshelf.host/api/file/all
Get a list of all files within your company context.
Headers
X-TagshelfAPI-Key
String
Application API Key
Authorization
String
Bearer <access_token> or amx <hmac_token>
[
{
"id": "eada91d5-ea19-4a80-afbe-166c12649d52",
"creation_date": "2020-08-12T00:00:00.000",
"update_date": "2020-08-12T00:00:00.000",
"file_name": "SAMPLE.pdf",
"file_name_without_extension": "SAMPLE",
"blob_name": "Tag/SAMPLE.pdf",
"blob_url": "https://demo.bucket.com/Tag/SAMPLE.pdf",
"user_name": null,
"md5_hash": "KTuG8VB6fviFQF6sshyGvQ==",
"content_type": "application/pdf",
"channel": "web",
"should_be_classified": false,
"classifier": "bayes",
"classification_score": 0.98,
"status": "waiting_to_be_sent_to_third_party",
"is_duplicate": false,
"duplicate_origin_id": null,
"tag_id": "49973d90-03c3-4f2a-b556-fd1a2f5635b9",
"is_parent": false,
"parent_id": null,
"deferred_session_id": null,
"tag_name": "Sale Invoices",
"company_id": "c46109f2-a507-47da-ae05-f839be543855",
"file_size": 222222,
"proposed_tag_id": "1c40a594-4b0f-4f8c-b534-73f64e61b2ea",
"proposed_tag_variance": 0.000632407575208874,
"classification_score_above_deviation": false,
"confirmed_tag_id": "1c40a594-4b0f-4f8c-b534-73f64e61b2ea",
"confirmed_by": "[email protected]",
"manual_classification": false
},
...
]Get file details by ID
GET https://tagshelf.host/api/file/detail/:id
Get all file details by File ID.
Path Parameters
id*
String
File ID
Headers
X-TagshelfAPI-Key
String
Application API Key
Authorization
String
Bearer <access_token> or amx <hmac_token>
{
"id": "eada91d5-ea19-4a80-afbe-166c12649d52",
"creation_date": "2020-08-12T00:00:00.000",
"update_date": "2020-08-12T00:00:00.000",
"file_name": "SAMPLE.pdf",
"file_name_without_extension": "SAMPLE",
"blob_name": "Tag/SAMPLE.pdf",
"blob_url": "https://demo.bucket.com/Tag/SAMPLE.pdf",
"user_name": null,
"md5_hash": "KTuG8VB6fviFQF6sshyGvQ==",
"content_type": "application/pdf",
"channel": "web",
"should_be_classified": false,
"classifier": "bayes",
"classification_score": 0.98,
"status": "waiting_to_be_sent_to_third_party",
"is_duplicate": false,
"duplicate_origin_id": null,
"tag_id": "49973d90-03c3-4f2a-b556-fd1a2f5635b9",
"is_parent": false,
"parent_id": null,
"deferred_session_id": null,
"tag_name": "Sale Invoices",
"company_id": "c46109f2-a507-47da-ae05-f839be543855",
"file_size": 222222,
"proposed_tag_id": "1c40a594-4b0f-4f8c-b534-73f64e61b2ea",
"proposed_tag_variance": 0.000632407575208874,
"classification_score_above_deviation": false,
"confirmed_tag_id": "1c40a594-4b0f-4f8c-b534-73f64e61b2ea",
"confirmed_by": "[email protected]",
"manual_classification": false
}
Get file events by ID
GET https://tagshelf.host/api/file/events/:id
Get a list of events triggered on the specified File.
Path Parameters
id*
String
File ID
Headers
X-TagshelfAPI-Key
String
Application API Key
Authorization
String
Bearer <access_token> or amx <hmac_token>
[
{
"id": "de0e8920-bc18-405d-89c3-4e741bd2d7a0",
"creation_date": "2022-11-17T00:00:00.000",
"update_date": "2022-11-17T00:00:00.000",
"file_id": "9fb3b05e-36f2-4528-a7fc-67d13d3c4e83",
"event_name": "alfred_event_file_done",
"event_envelope": "{}",
"sent_to_event_broker": true,
"sent_to_web_hook": true
},
...
]Get file versions by ID
GET https://tagshelf.host/api/file/versions/:id
Get a list of versions from the specified File.
Path Parameters
id*
String
File ID
Headers
X-TagshelfAPI-Key
String
Application API Key
Authorization
String
Bearer <access_token> or amx <hmac_token>
[
{
"id": "d854a6d3-652e-4d00-9cd9-563b94e0c6ab",
"update_date": "2020-08-12T00:00:00.000",
"creation_date": "2020-08-12T00:00:00.000",
"file_log_id": "9ddb12be-2375-4891-9edc-ec457adc252c",
"content": ""
},
...
]Download file by ID
GET https://tagshelf.host/api/file/download/:id
Returns a byte stream representing the data for a given file.
Headers
X-TagshelfAPI-Key
String
Application API Key
Authorization
String
Bearer <access_token> or amx <hmac_token>
Return the File as a stream.
Generate shareable link
POST https://tagshelf.host/api/file/share/:id
Generate a shareable link based on your desired configuration on expiration time, URL format (short or long), etc.
Headers
Content-Type
String
application/json
Authorization
String
Bearer <access_token> or amx <hmac_token>
X-TagshelfAPI-Key
String
Application API Key
Request Body
length
Integer
Shareable link duration in seconds. After the defined time elapsed, the link will no longer works.
short_url
Boolean
Whether or not you would like use a short link instead of the original link.
always_use_last_version
Boolean
Whether or not you would like to use the latest version of the File or simply use the last version available during the shareable link creation.
{
"file_id": "6e5ef3ed-71c5-442d-97b2-d4992edb14ce",
"share_url": "https://staging.tagshelf.com//api/file/public/01debd6b-31d8-4a95-9b83-9d0d5e2f8ed6"
}File Upload EndPoints
Upload Files from Remote Sources
POST https://tagshelf.host/api/file/upload
This endpoint facilitates the uploading of files from remote sources, accommodating various types of external storage like URLs, or Blob storage from cloud providers including AWS, GCP, or DigitalOcean.
This endpoint is specifically tailored for integrating files stored in diverse remote locations into the Alfred platform for processing.
In Alfred, uploading a file via this endpoint contributes to the initiation of a Job, which encompasses a series of processes over several files. Each file uploaded becomes a processing unit within this Job. The individual processing of each file, involving tasks like classification, OCR, and data extraction, starts only after the Job is initiated. The Job is considered complete when all its associated files have finished their respective processing cycles.
This design ensures that while each file undergoes its unique processing workflow, it remains a cohesive part of the larger Job, aligning with Alfred's approach to handling multiple files in a coordinated and efficient manner.
Headers
Content-Type
string
application/json
Authorization
string
Bearer <access_token> or amx <hmac_token>
X-TagshelfAPI-Key
string
Application API Key
Request Body
url or urls
string or array of strings
Use url, when you have an URl to single remote file. Use urls, when you have URl's for multiple remote files. the current limit for this parameter is 100 elements.
source
string
Configured object storage source name. Ideal for referring to files hosted in existing cloud containers. When used, file_name are container are required.
container
string
Virtual container where the referenced remote file is located. When used, source and file_name are required.
file_name or file_names
string or array of strings
Unique identifier or identifiers of file within an object storage source. When used, source and container are required.
merge
boolean
Boolean value [true/false] - When set to true, will merge all of the remote files into a single PDF file. All of the remote files MUST be images. By default this field is set to false.
metadata
string
JSON object or JSON array of objects containing metadata fields for a given remote file. When merge field is set to false: When using the urls field this should be a JSON object array that matches the urls field array length. When using the url field the metadata field should be a JSON object. When the merge field is set to true: The metadata field should be a JSON object.
priority
string
Sets the Job's priority in the processing queue. Valid values are: low, normal or high.
propagate_metadata
boolean
This parameter enables the specification of a single metadata object to be applied across multiple files from remote URLs or remote sources. When used, propagate_metadata ensures that the defined metadata is consistently attached to all the specified files during their upload and processing. This feature is particularly useful for maintaining uniform metadata across a batch of files, streamlining data organization and retrieval.
parent_file_prefix
string
The parent_file_prefix parameter is used to specify a virtual folder destination for the uploaded files, diverging from the default 'Inbox' folder. By setting this parameter, users can organize files into specific virtual directories, enhancing file management and accessibility within Alfred's system.
page_rotation
integer
This parameter allows for the manual rotation of remote files upon upload. page_rotation is applied to ensure the correct orientation of the files for processing, which is crucial for tasks like OCR and document analysis. Represents rotation in degrees.
{
"job_id": "559167cb-4c76-4e28-bf15-34cbf614119c"
}When retrieving media from HTTP/HTTPS URLs, be aware that the request might not succeed if the host blocks the request (for instance, as a part of request throttling or DOS attack prevention) or if the host's response is delayed. For critical applications, it is advisable to avoid reliance on externally hosted media to guarantee the success of your requests.
For optimal processing efficiency in Alfred, ensure that files uploaded via the /api/file/upload endpoint from remote sources include accurate content type information. Relying solely on Alfred's automatic file type detection may lead to processing inefficiencies or errors.
Upload file by stream
POST https://tagshelf.host/api/file/uploadfile
Uploads a single remote file to the system's pipeline. A successful response will be returned in JSON with a file's unique identifier for the submitted workflow. If the request fails, the response will contain a message to help understand what went wrong. File processing is asynchronous, there fore a polling policy (within the API rate limits) should be implemented that queries the /api/file/detail end-point and evaluates its response body and the file statuses. Anatomy of the Request
Headers
Content-Type
string
multipart/form-data
Authorization
string
Bearer <access_token> or amx <hmac_token>
X-TagshelfAPI-Key
string
Application API Key
Request Body
metadata
string
String which represents a JSON object or JSON array of objects containing metadata fields for a given remote file.
session_id
string
Session ID to link multiple files to a job.
file*
file or string
Allows the upload of either a local file object or a base64-encoded string of the file.
{
"file_id": "559167cb-4c76-4e28-bf15-34cbf614119c"
}Last updated
Was this helpful?