Image to Text API

The Image to Text API detects and extracts text from images using state-of-the-art optical character recognition (OCR) algorithms. It can detect texts of different sizes, fonts, and even handwriting.

This function sends an HTTP request to an image-to-text API, then returns the server’s response body when the request succeeds. In practice, it looks like a small wrapper around a GET call that passes an API key in a header and returns either the extracted text or an error string.

const
URL_APILAY='https://api.apilayer.com/image_to_text/url?url=%s'; 

In short, this is an OCR/image-to-text API client function written in Delphi-style code for maXbox.

function Image_to_text_API3(AURL, url_imgpath, aApikey: string): string;
var httpq: THttpConnectionWinInet;
    rets: TStringStream;  
    heads: TStrings; iht: IHttpConnection; //losthost:THTTPConnectionLostEvent;
begin
  httpq:= THttpConnectionWinInet.Create(true); 
  rets:= TStringStream.create('');
  heads:= TStringlist.create;     
  try 
    heads.add('apikey='+aAPIkey);
    iht:= httpq.setHeaders(heads);
    httpq.Get(Format(AURL,[url_imgpath]), rets);
    if httpq.getresponsecode=200 Then result:= rets.datastring
      else result:='Failed:'+
             itoa(Httpq.getresponsecode)+Httpq.GetResponseHeader('message');
  except  
    writeln('EWI_HTTP: '+ExceptiontoString(exceptiontype,exceptionparam));
  finally
    httpq:= Nil;
    heads.Free;
    rets.Free;
  end;                  
end;                 //}

What it does

• Creates an HTTP client object and two helper containers: one for the response text and one for headers.
• Adds a header named apikey with the value aApikey.
• Calls GET on Format(AURL, [url_imgpath]), so AURL is treated as a format string and url_imgpath is inserted into it.
• If the HTTP status code is 200, it returns the raw response content from rets.DataString.

Otherwise, it returns a string starting with Failed: plus the status code and the server’s message response header.

That matches the general pattern of image-to-text/OCR API calls, where the service returns text in the response after being given an image reference and an API key.

1176_APILayer_Demo64_latinumtranslate.txt

The function says: “Call this web API with my image URL and API key, then give me back whatever text the API extracted from the image.” If the call fails, it gives you a failure message instead of crashing.

Important notes

• The request uses GET, which implies the image path is likely passed in the URL, not uploaded as a file.
• The header name apikey is custom; some APIs use different auth headers such as Authorization: Bearer ..., so this code is tailored to one specific service or wrapper.
• The except block only writes an error message; it does not set a fallback return value, so the function could end up returning an empty string on exceptions.
• httpq:= Nil; in finally is not the same as freeing the object; normally you would call Free to release it properly.

• If the image path may contain special characters, URL-encode it before inserting it into Format(AURL, [UrlImgPath]).

function GetApiKey: string;
begin
  Result:= GetEnvironmentVariable('IMAGE_TO_TEXT_API_KEY');
  if Result = '' then
    raise Exception.Create('Missing IMAGE_TO_TEXT_API_KEY');
end;

• Store API keys or tokens outside code, such as environment variables, config files with restricted access, or platform secrets storage.

{"lang":"en","all_text
":"ウォーターリリーここに生まれてウォーターリリーここがどこだかまだわから
Waterlilies were born here -
Waterlilies still don't know where here -
あの川に兄が浮か"}

Comments

[Max Kleiner]

Haiku (俳句) is a form of short poetry that originated in Japan. It is known for its strict structure, which traditionally consists of three lines with a total of 17 syllables: 5 syllables in the first line, 7 in the second, and 5 in the third. This concise format allows poets to capture a moment in nature or an emotional experience in a few words