UPDATES PLANNED FOR TODAY
  • comment-white.png
    *
    /i/ieddit
    2y
    UPDATES PLANNED FOR TODAY (text)
  • 2y
    parent [-]
    [05/Oct/2019:14:14:08 +0000] "GET /r/OMITTED/undefined HTTP/1.1" 404 209 "https://ieddit.com/r/OMMITTED/" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/77.0.3865.90 Safari/537.36"

    Additionally, I keep seeing requests like these showing up in the logs. I believe this may be caused by browser addons, as I utilize a bare minimum of javascrip so 'undefined' REALLY should not being showing up in any requests. That being said, it is still a possibility this bug is being caused by something I have done.

    Any users who encounter this, please comment/message me as to the details of how you encountered these URLs with 'undefined' in them, and which browser/addons you may be using.
    3
    reply
  • edit text now stays even after captcha fail
    2
    reply
  • got an absolute ton done today

    inline images/text (by far the most impactful)
    edit now preserves your new edited text if you fail the captcha
    fixed a captcha bug
    when linking directly to comments, the comment-reply box is hidden by default

    i'm done for the night.
    2
    reply
  • inline image expansion added
    1
    reply
  • I would HIGHLY recommend against image expansion. What happens when this site gets flooded by illegal photos of children? Or live executions? Do you really want to serve time in prison for the sake of this project? Let imgur manage those risks

    Even now, if someone were to illegally dox a user or declare an actionable death threat, all we can do is mark it NSFW
    3
    reply
  • 2y
    parent [-]
    this is a VERY valid concern

    So, the way this would work, is the images will still be hosted on external sites.

    For example, say somebody posts an image, https://imgur.com/a/abcd - the actual image is

    The code I already have written for thumbnails: 1. first checks if the url has a image/x content type. if so, that image is used as thumbnail. 2. if it isn't a direct image, it looks for the og:image opengraph tag, and then fetches the images from the og:image url. 3. if that fails, it just trys its best to find the largest image in the page source.

    in case 1 or 2, if an image is found, I'd store that link in a "remote_image_url" column, for the post. the way expansion would work, is whenever a user clicks to expand an image, it actually will just add an element. I would not be hosting any full-size images on the site.
    3
    reply
  • 2y
    parent [-]
    Thanks
    3
    reply
  • Just wanna say, good on ya mate!
    2
    reply
  • You do an impressive amount of work for being only one person!
    2
    reply