browser example does not work

[under-score]

interesting project as I need the x,y location of the textareas but unfortunately I get

    Uncaught TypeError: Cannot read property '5' of null
    at parseOcradResultsFile (ocrad.js:96142)
    at postprocess (ocrad.js:96223)
    at OCRAD (ocrad.js:96242)
    at OCRImage (location.html:20)
    at HTMLImageElement.onload (location.html:43)

[Fonsori]

Have you managed to find a solution to that? That error only shows when setting the "verbose" flag to true right?
Looks like setting {raw: true} works the same but returns pure text with no object access to .letters and the like.

Actually I may of found a working edit to ocrad.js, a one line insert that stops a for loop. Must test it further though

[under-score]

sorry, don't know the answer

[rajivraman]

I also came across the same issue. Any luck with a solution?

[rajivraman]

For anyone that runs across the problem, the issue is that the out.txt file generated by verbose mode is missing the tail end of its content. So parseOcradResultsFile() fails while processing it. The trick is to ensure that the underlying filesystem flushes all file streams before attempting to read out.txt. The easiest way I've found to do this is to first expose the flushing method to the API:

API.delete_file            = function(filename){ return FS.unlink(filename) };
API.quitFS                 = function() { return FS.quit() }; // NEW

Then call it from this block of code in _simple():

if(opt.raw){
    API.quitFS(); // NEW
    ret = API.read_text('/out.txt');
    API.delete_file('/out.txt');

[sdkfz181tiger]

Wow!!
It works very well!!
Thank you sooooo much!!
I added 2 lines berow!!
96057: API.quitFS(); // NEW
96084: API.quitFS = function() { return FS.quit() }; // NEW