Accessing Google Photos

≡

I like having a blog, but honestly it’s faster to write it in google photos as that has a better WYSWIG viewer. But, I don’t get features like search, or previews, which matters to me. I also don’t trust Google not to break photos in the future. Of course exporting is a PITA, here are some hacks to export and make it into my blog format. NOTE: There is an API to extrat photos, but it doesn’t include the text (which I like to have, as recall WYSWIG editor)

Download the album page

Shells are weird, might want to shortcut it. Use follow if shortcutting.

export ALBUM_URL="https://shorturl.at/hjLX0"
http --follow $ALBUM_URL >  album.html

Extract the text content

Oddly, the page is pretty dynamic, got to find the text in a script block

cat album.html | pup 'script:contains("AF_initDataCallback(")' | grep AF_init > data1

Bleah, need to figure out a clean way to paarse this … Yuk. Then need to parse the json slug with the dictionary that has keys like: 99218341.

      99218341: [
        [
          1,
          [
            "Last but not least there was a risk of snow, which made the family uncomfortable to drive on the highway. At first that had me grumpy... But then I remembered the most important lesson of all.\n\nPut the needs of your family ahead of your own, you can always do something with them later when they're comfortable\n\nHave a great weekend - and we'll see you next week!",
          ],
        ],
      ],
      101428965: [0, "vdhvk94a06vv00000000004a"],

Thinking, maybe the best way to do that is non greedy grep … Lets think through that ..

https://github.com/idvorkin/idvorkin.github.io/blob/master/dump_google_album.py?plain=1#L25

Extract the photos

prettier album.html > pretty.album.html
cat pretty.album.html | grep 'https.*usercontet'

Extract the title

Slopp doppy

cat pretty.album.html| grep ' - Google Photos'

References