PF2 JSON data project H1 problem

[lillo84]

if you refer to h1 inside the id ctl00_MainContent_DetailedOutput

I also found the error that derives from an incorrect closure of the html code.

So.. i change the code from this
main = soup.find("span", {'id': 'ctl00_MainContent_DetailedOutput'}) #print (main) pfsLegal = main.find("img", {'title': 'PFS Legal'}) if(pfsLegal): pfsLegal = True else: pfsLegal = False name = main.find("h1", {'class': 'title'}).text

to this:
main = soup.find("div", {'id': 'main'}) #print (main) pfsLegal = main.find("img", {'title': 'PFS Standard'}) if(pfsLegal): pfsLegal = True else: pfsLegal = False for finder in main.find_all("a", {'href': eachEntry}): # temporary for error into html source name = finder.text

[jimbarnesrtp]

Thanks saw it this week ut a note in the readme about it, asked aon to fix their html as it breaks many things

[lillo84]

Nice thing Sir, I hope everything goes well..

[jimbarnesrtp]

I am struggling with this one, in that philsophically I was trying to aim for an event driven parsing kind of like the java stax xml parsing pattern, instead of hard coding everything that I was looking for, but this breaks most things and instead of being event driven having to rewrite the code to now look for specific things, thoughts? I understand this way is of course more fragile, just more frustrating as bad html breaks it but html should be right

[lillo84]

I know, the human error is frustrating, but the base idea is right.
in a few years, when the contents become a lot, it will take less time to modify the script than to copy them manually.
I would understand if instead there is a way to get the data directly from the database d20pfsrd.

Read also the other post I made on the differences between nethys and d20pfsrd. Because I believe there is a problem License Content in nethys than d20pfsrd

[jimbarnesrtp]

sorry it has been busy, and yes the stuff in d20 is the more open version of the names for many things, so is something to considering

[jimbarnesrtp]

working on some rewrites for some of the things, also switching to the html5lib for the codec is making it handle parsing better, adding testing to the code as well, and also pulling from the datasets coming from archives of nethys