stewartmcgown / grammarly-api Goto Github PK

Normally, queries take at least 3 seconds to process. Is this Grammarly's fault? Is there any way to improve this?

Hi,

I am facing the issue of inconsistent highlightBegin/highlightEnd values:

Sample sentence:

It is good that she have come.

Response from stewartmcgown@grammarly-api (one request):

https://gist.github.com/lexx888/11c83c06395866ce7c3c4183d3621297

has:

 "highlightBegin": 19,
"highlightEnd": 23,

Response from Grammarly demo document (pasted as one sentence, not typed):

https://gist.github.com/lexx888/a115013c9acbc126dbeab2ea0cf41824

has:

"highlightBegin": 20,
"highlightEnd": 24,

see:

And it should be 20 to 24:

"It is good that she have come.\n".substring(20,24)  
> "have"

This problem consistently arises in almost any texts with 2 or more mistakes, and it seems that the highlight region is shifted differently in different cases.

Any ideas why this might happen?

A nice feature would be the addition of an async generator API, where consumers could read corrections as soon as they are processed.

This would require a generator wrapper around the web socket.

According to the response object, the default dialect is British English. Is there a way to change it to American, and if so, how?

As the title, the detailed error is as follows

throw new Error('Unable to create a session with these credentials.');      
                  ^

Error: Unable to create a session with these credentials.
    at Object.<anonymous> (E:\***\node_modules\@stewartmcgown\grammarly-api\build\lib\auth.js:138:19)
    at Generator.next (<anonymous>)
    at fulfilled (E:\***\node_modules\@stewartmcgown\grammarly-api\build\lib\auth.js:4:58)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)  

The reason for this error is that too many websockets have been established, has it been officially disabled by Grammarly?

Cheers for the excellent work here. Has made interfacing with Grammarly a breeze.

In the README, it indicates that dialect is an optional param to analyze, but I can't seem to get different results when I pass in "american".

Interacting with the web browser, I have no corrections for the word yogurt. But through the API, where it seems the dialect is "british", I get corrections for yogurt to be yoghurt instead.

Any chance I'm doing something wrong with the API? My call looks like this.

results = await premium.analyse(sentence, 5000, "american");

You should be able to simply pass in credentials for Premium and get access to premium alerts.

I am a seasoned PHP developer and have absolutely no idea how node projects run. Can you please provide steps/instructions about how I can run this project after cloning it. I did npm run build and I see

nps is executing `build` : nps build.prebuild && tsc -p tsconfig.build.json
nps is executing `build.prebuild` : node node_modules/rimraf/bin.js ./build 

What's after this? How do I actually see the errors or the error object?
Also, does this project still run as tests are failing

StringObject interface from utils.d.ts cannot be found when installing this API through npm.

Error Message when building the first usage example from the readme file.

node_modules/@stewartmcgown/grammarly-api/build/lib/connection.d.ts:4:34 - error TS2304: Cannot find name 'StringObject'.
 
 4 export interface Headers extends StringObject {

node_modules/@stewartmcgown/grammarly-api/build/lib/responses.d.ts:46:23 - error TS2304: Cannot find name 'StringObject'.
 
 46     extra_properties: StringObject;