[EMAIL PROTECTED] writes: > First, I did/do not mean to offend/attack you,
Oh, I know that. My little rant was not directed toward you, it aimed at world in general. :-) > okok, granted, to dissolve > <a href=foo.html target="_top"><img src="pic.htm.jpg" name=index.html > alt="<oops>>"><br><-fool.htm-></a> > for example, you'd really have a hard time, I suppose. There are less convoluted examples than that. Imagine someone forgetting to close a tag, either by forgetting to close the double quote, or by forgetting to close the ">". Alternatively, imagine someone leaving a bogus "<" in the file. So you could have something like: bogus-char<wget now thinks its in a tag but heres a real tag it will miss <img src="foo">. Because of one bogus "<" character Wget could lose the entire valid <img> tag. So being "strict" isn't in all cases such a bad idea -- it allows the parser to weed out syntactically invalid tags at the benefit of being able to back out and capture the valid stuff. The ability to recover from an error condition is what makes grokking broken HTML relatively easy. In programming languages such a serious syntax error might invalidate the entire input stream. Compared to that, Wget's error recovery seems acceptable to me. >> Some amount of shoveling is necessary and is performed by all small >> programs to protect their users, but there has to be a point where you >> draw the line. There is only so much shit Wget can shovel. > Unfortunately, the amount of shit on the web will not decrease. > I fear that the opposite may be true. > no, wait, I am pretty sure... I know that, but what I wrote still stands. I trust there are many people with higher nausea threshold, but it doesn't mean they don't have one. >> I'm not saying Ian's example is where the line has to be drawn. (Your >> example is equivalent to Ian's -- Wget would only choke on the last >> "going" part). But I'm sure that the line exists and that it is not >> far from those two examples. > > Ok, but I understand you correctly that these two examples (mine was > intended to be equivalent, but without JS) should be on the "parse > and retrieve" side of this line, not the "ignore and blame > Frontpage" side? I intentionally refrained from specifying which side of the line those examples are. Which means it's up to the others to decide whether they care about that anomalous case enough to get knee-deep into html-parse.c.
