Question

I’m currently working on a function to find all images referenced in an html file, currently I am trying to to find these substrings within the file: ".bmp" ".gif" ".jpg" ".png" and also want to find their roots eg: /images/foo/ and then use these two substrings to make a new string: /images/foo/bar.jpg I know how I am going to concatenate the strings but I have no idea how I am going to locate the actual substrings, I feel quite overwhelmed right now and would really appreciate some help.

Answer 1

The “right” answer to this question ought to urge you to use tools that were built for the job. Smart people write stuff like libxml for a reason. Re-inventing the wheel will only make things more difficult. With libxml, for example, you easily traverse an XML tree like so:

for (cur_node = a_node; cur_node; cur_node = cur_node->next) {
    if (cur_node->type == XML_ELEMENT_NODE) {
        printf("node type: Element, name: %s\n", cur_node->name);
}

The “wrong” answer is to come up with some “trick” for finding the beginning of an image string, either by looking for the beginning of the image tag (<img) or a quote " as Doug mentions in the comments.

You’ll notice that I put right and wrong in quotations. I’m somewhat of a purist and would strongly suggest an XML-oriented solution because it’s wholly generalizable and easily extendible (tomorrow you may say: oh I also need the anchor text). A DOM parser makes every subsequent problem a breeze to solve.

But if you’re working on a proof of concept or prototype (or maybe even homework) where everything’s well-formed and you don’t release your code in the wild, the “wrong” approach may be sufficient.