Is this meant to eventually support pre-existing HTML markup?

[dejan]

Is this meant to eventually support pre-existing HTML markup?

[subimage]

I believe it should work with existing HTML markup, or markdown — which is
why I made the changes and submitted the patches I did. If you're working
on anything like a CMS or blog software that's the only way to go. User
input in the wild is unpredictable stuff.

On Wed, Aug 13, 2014 at 6:39 PM, zrisher [email protected] wrote:

Hi all,

I notice that there are a lot of areas where this would break existing
HTML markup (i.e. any link where the src= one of the supported urls). I see
related pull requests @subimage https://github.com/subimage #116
#116, #114
#114 that would fix this in
certain situations. But if this is really meant to work with anyone's
existing markup, and you need URL filters to only target text, some
architectural changes (possibly involving parsing content into nodes via
Nokogiri and then only running filters on text) would be easier and more
DRY than adding the conditions onto every regex. That's a big change in
functionality.

The content of README.md seems to indicate that this is meant to be used
as either a complete filter on very simple content fields (without much
previous markup, which gets escaped), or as a simple filter called
individually when you know you need it (once you've isolated exactly the
text that needs changing).

I love the respository - it works very well for my project when used in
the latter way. But if it's meant to work on all prexisting markup it needs
some changes. If not, perhaps this could be indicated in the readme so
similar issues and pull requests aren't raised.

@dejan https://github.com/dejan, could you let us know if you see this
repository eventually fully supporting pre-existing markup? Or should we
plan on calling filters on isolated sections in those cases?

—
Reply to this email directly or view it on GitHub
#122.

seth @ subimage llc
http://subimage.com

[zrisher]

With the changes made today, this repo no longer meets any of my use cases for it (i.e. the various social links in Rails). So I won't be using it, and if I'm the only one who had issues with running this over existing HTML markup, this issue can be closed.