I have managed to get some HTML from a website using HTMLDomParser (http://simplehtmldom.sourceforge.net/):
$html = HtmlDomParser::file_get_html('https://example.com/id=1');
This results in the following HTML:
<p> ANDE 2, the Atmospheric Neutral Density Experiment 2, is a pair of microsatellites (Castor and Pollux) launched from Cape Canaveral on STS 127 on 15 July 2009 at 22:03 UT and deployed from the payload bay of the shuttle on 30 July 2009 at 17:22 UT.</p>
<p><strong>Launch Date:</strong> 2009-07-15<br/><strong>Launch Vehicle:</strong> Shuttle<br/><strong>Launch Site:</strong> Cape Canaveral, United States<br/></p>
What I want to attempt to do is to take this HTML and save it to a mySQL database. So, in the database I will have four columns:
- Description
- Launch Date
- Launch Vehicle
- Launch Site
The HTML layout will always stay the same. My other database contains the id's for the URL's. I am able to replace the id=1 with a variable, so I would be able to manage to save multiple webpage data.
The only thing I am worrying about if the id does not exist. Would it be possible just to skip that page if it cannot find the HTML it is looking for?
I have never done this before, so I am a bit overwhelmed by this. Would it even be possible to do this?
Basically, I just want to save data from a DomParser into my database.
EDIT: Since there has not been an answer, I think we need to tackle this systematically - first we should try creating the function that saves the data to the database.
EDIT 2: I was just thinking - maybe the function that needs to save the HTML can split the result from HTMLDomParser by tag ('strong', 'p', etc.) and create a variable for each column and then save it. I guess the question here first is how do you split the HTML?
EDIT 3: Since I did not receive any answers / suggestions, I am guessing it is very complex to implement this. I still have not managed to come up with anything that saves the HTML successfully to my database.