Closed shortjared closed 8 years ago
Hi @shortjared, I'm unable to reproduce this bug. Here's what I do:
$ cd my-jekyll-site
$ cd _site/
$ mkdir 2015-12-15-talking-about-your-environments
$ touch index.html
$ cd ..
$ cd ..
$ s3_website push
[info] Deploying /tmp/_site/* to my-test-site.com
[succ] Created 2015-12-15-talking-about-your-environments/index.html (public, max-age=120 | text/html; charset=utf-8 | gzip)
[info] Summary: Created 1 file. Applied 1 redirect. Transferred 20 B, 0 B/s.
[info] Successfully pushed the website to http://my-test-site.s3-website-us-east-1.amazonaws.com
It seems to me that the code at https://github.com/laurilehmijoki/s3_website/blob/a17868b3bdd498d0331c48078dc9b67455e9264a/src/main/scala/s3/website/model/push.scala#L135 does not imply that it matches all the files that contain the string env
. The Scala REPL also seems to agree:
scala> Seq(".env") contains "2015-12-15-talking-about-your-environments/index.html"
res0: Boolean = false
Is there something in your exclude_from_upload
setting? Values in that setting are treated as regular expressions.
Interesting. Yep.
exclude_from_upload:
- .env
.env is treated as the regex so anything with env gets kicked out. Closing, thanks!
This caused me a couple hours of debugging trying to track down why only one specific file generated with Jekyll would not deploy no matter what I did.
It turns out that any file that contains the substring "env" in the file name gets excluded by s3_website. So for instance
2015-12-15-talking-about-your-environments/index.html
, gets excluded (silent by default).I believe the guilty line is: https://github.com/laurilehmijoki/s3_website/blob/a17868b3bdd498d0331c48078dc9b67455e9264a/src/main/scala/s3/website/model/push.scala#L135