Colon code isn’t parsed correctly by Facebook

[TJScalzo]

The newest blog post was shared on Facebook but the metadata that it pulled from the page was interpreted literally. I guess the solution would be to define the title some other way. Perhaps using a different designation for the colon character would work?

[gisellegk]

Yeah I saw that... Liquid didn't like having a colon in the metadata table (front matter? Or something like that?) So I just slapped on a quick fix. I wonder if there's a way to escape the character instead of encoding it like that?? Or choosing a different encoding.

Also I think the page/tab title pulls from the same field so we should look to see if the issue persists there too

On Aug 7, 2017 8:04 PM, "Tim Scalzo" notifications@github.com wrote:

The newest blog post was shared on Facebook but the metadata that it pulled from the page was interpreted literally. [image: img_4299] https://user-images.githubusercontent.com/15825214/29050626-309bb5a0-7bab-11e7-88a4-be64d114f722.PNG I guess the solution would be to define the title some other way. Perhaps using a different designation for the colon character would work?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/LetsBuildRockets/letsbuildrockets.github.io/issues/36, or mute the thread https://github.com/notifications/unsubscribe-auth/AC4tvxg2b1-8zWonhORF6SaGrQy8GWJhks5sV6YXgaJpZM4OwFtL .

[TJScalzo]

It seems like it's not an issue with the page title in Safari and in Chrome.

It seems like it's something specific to Facebook's method of pulling front matter/page metadata. I'll check if there's some documentation on how their link previews work.

[gisellegk]

After @TJScalzo 's fix, I did the following steps to fix this issue

1. Forced fb to re-scrape the info from the website using this tool
2. Went to the facebook post and as admin, refreshed the share attachment.

Not sure if both steps are necessary but it's fixed now.