Question

I store the source code from a website to a string, and I successfully removed all the html tags. However, there are random whitespaces between paragraphs. Sometimes it will only be one line, other times it will be 4 or 5 lines.

Here is what I did

- (NSString *)parseHTMLText:(NSString *)text {

    NSString *startingPt = @"<!-- (START) Pagination Content Wrapper -->";
    NSString *endingPt = @"<!-- (END) Pagination Content Wrapper -->";

    //isolate body text from entire source code
    NSString *leftTrimmed = [text substringFromIndex:NSMaxRange([text rangeOfString:startingPt])] ;
    NSString *completeTrimmed = [leftTrimmed substringToIndex:[leftTrimmed rangeOfString:endingPt].location];
    completeTrimmed = [completeTrimmed stringByTrimmingCharactersInSet:[NSCharacterSet whitespaceAndNewlineCharacterSet]];

    completeTrimmed =  [self removeHTMlTagsFromString:completeTrimmed];
    completeTrimmed = [completeTrimmed stringByTrimmingCharactersInSet:[NSCharacterSet whitespaceAndNewlineCharacterSet]];

    return completeTrimmed; 

}

- (NSString *)removeHTMlTagsFromString:(NSString *)text {

    //check if there are any html tags

    if ([text rangeOfString:@"<"].location != NSNotFound && [text rangeOfString:@">"].location != NSNotFound) {
        //find first index of "<"
        int startIndex = [text rangeOfString:@"<"].location;
        NSString *startOfTag =  [text substringFromIndex:startIndex];
        // find length to ">"
        int length = [startOfTag rangeOfString:@">"].location + 1;
        text = [text stringByReplacingCharactersInRange:NSMakeRange(startIndex, length) withString:@""];

        text = [self removeHTMlTagsFromString:text];
    }

    return text;

}

I tried this, but it doesn’t work

completeTrimmed = [completeTrimmed stringByReplacingOccurrencesOfString:@"     " withString:@""];

Answer 1

If the original HTML looked like:

<tag1>
<tag2>
<tag3>
text text text
</tag3>
</tag2>
</tag1>

Then when you remove all the tags you’ll still have the newlines that separated them.

Use a DOM parsing library rather than primitive string functions, and your problem should be solved.