When converting with the Get format api. file https://calendar10.tripod.com/calendars/2022jun.pdf to text, the month day (1 - 31) is missing from the returned text.
What is the get api format for “Engine3”?
Hi. The format to use the new OCR Engine 3 is exactly the same as for engine1 or 2!
The only difference is that you use the parameter
ocrengine = 3
Just getting back to this and it’s working with OCR engine 1
Using get api with PHP Curl
‘https://api.ocr.space/parse/imageurl?apikey=XXXXXXXXXXXXX&OCREngine=1&scale=true&filetype=PDF&detectOrientation=true&url=https://calendar10.tripod.com/calendars/2022oct.pdf’
The PHP with get returned text data varies from the text data returned by the https://ocr.space page as follows:
The ocr.space version text returns the embedded newlines, x0A, and carriage returns, x0D.
The get version converts newlines and carriage returns to a single space.
The code I wrote expected the newlines and carriage returns. IS there any way to have this data returned in the text?
Answering my own question.
The newlines and carriage returns are in the returned data, but were apparently stripped out with the echo command I used to display the returned data. Never noticed this before.
Next problem
Using get api with PHP Curl
‘https://api.ocr.space/parse/imageurl?apikey=XXXXXXXXXXXXX&OCREngine=1&scale=true&filetype=PDF&detectOrientation=true&url=https://calendar10.tripod.com/calendars/2022oct.pdf’
Sometimes it works, but around 50% of the attempts, Curl times out after 30 seconds without a response from ocr.space. I’ve also encountered this issue using the online page at ocr.space.