Get a comic page
const url = 'http://localhost:8080/api/v1/publications/1/pages/1';const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};
try { const response = await fetch(url, options); const data = await response.json(); console.log(data);} catch (error) { console.error(error);}curl --request GET \ --url http://localhost:8080/api/v1/publications/1/pages/1 \ --header 'Authorization: Bearer <token>'The page’s original image bytes, read straight from the archive or fetched through the library’s source. An extension chapter’s pages have no ETag.
Authorizations
Section titled “Authorizations”Parameters
Section titled “Parameters”Path Parameters
Section titled “Path Parameters”Zero-based position in reading order.
Responses
Section titled “Responses”The page image.
Headers
Section titled “Headers”Changes whenever the file’s content changes. Send it back in If-None-Match.
The client’s copy is current.
The request has no valid session token.
object
Explains the problem to a developer.
Example
{ "code": "bad_request"}The resource does not exist.
object
Explains the problem to a developer.
Example
{ "code": "bad_request"}The file changed on disk since it was scanned. A rescan has started; retry after it finishes.
object
Explains the problem to a developer.
Example
{ "code": "bad_request"}The server failed; details are in its log.
object
Explains the problem to a developer.
Example
{ "code": "bad_request"}The publication is from a source library, and its source couldn’t be
read (source_unavailable) or its extension failed (source_failed);
the message says why.
object
Explains the problem to a developer.
Example
{ "code": "bad_request"}