Process and strip markdown/HTML in OPF meta tags #583
Conversation
As described by @jaycolmvar in #560, this strips markdown/HTML from meta tags in the OPF file. That markup may be present in meta.yml since we do allow it, so that it's possible to include, say, italics in a title or project name.
|
I wasn't checking email for a while, so just saw this. Two comments: First, in looking at the Dublin Core metadata definitions (http://dublincore.org/specifications/dublin-core/dcmi-terms/) it looks as if some properties might allow for more than plain text. For example, the "description" properties is defined as follows: "Description may include but is not limited to: an abstract, a table of contents, a graphical representation, or a free-text account of the resource." Maybe I'm missing something, but this seems to allow for the inclusion of say, an IMG tag to reference a "graphical representation". Second, what about things like HTML character entities? Although I can't find a reference, I presume that the DC properties like "creator", "publisher", etc., can contain non-ASCII7 characters. What would happen if you put a line in _data/meta.yml like title: All About © and ® Would the entities get replaced with the corresponding UTF-8 strings? |
|
@jaycolmvar Thanks so much for your input here. Sorry I haven't had a chance to test what you suggest. I'll get there! |
As described by @jaycolmvar in #560, this strips markdown/HTML from meta tags in the OPF file. That markup may be present
in
meta.ymlsince we do allow it, so that it's possible to include, say, italics in a title or project name.I haven't tested this very thoroughly – only on one simple epub so far.
@LouiseSteward @bertuss @jaycolmvar can you think of any cases or reasons this might be a bad idea? Essentially, we need plain text that is valid XML in the
package.opffile we build with theepub-packageinclude.