Skip to content

Latest commit

 

History

History
70 lines (53 loc) · 2.69 KB

README.md

File metadata and controls

70 lines (53 loc) · 2.69 KB
Raw

Movies to DayOne

Problem: from 2003 to 2014, I kept track of movies I watched in an OrgMode file. By 2014, I wanted to do this with a more mobile-first approach, so I started tracking them in the Day One journal (which ended up getting a template managed by an iOS Workflow/Shortcut).

I captured the following attributes:

  • Date Watched
  • Title -- as shown by IMDb, for instance: 28 Days Later... (2002)
  • Where -- viewing location (theater name or perhaps media, like "VHS")
  • Score -- integer rating, 1-5
  • New -- Boolean -- was this first time I saw it?

When I translated this to DayOne, I wanted to expand to match the current Day One entry template, which is something like this:

28 Days Later... (2002)   # title
5/5                       # rating
IMDb summary
-- IMDb link

In Day One, I use tags for easier filtering: this entry would be tagged "5/5" and "Movies". If it was a first viewing, it also would be tagged "new".

Data (cleaning) Requirements

I wanted to look up the IMDb summaries and URLs based on the titles. I found a web service, the OMDb API, that made this substantially easier, although not perfect.

Day One can also store the geographic coordinates of the entry location. Since I had theater names, I wanted to find them (where possible: some theaters closed since I went).

I converted the Org Mode file to a CSV and walked through the data, looking up the IMDb and location information as I went. I stored the results in a SQLite database, with movies_to_db.py.

Outcome

I had roughly 1,300 entries in the Org Mode file (not all were unique), but the OMDb API limited free use to 1,000 queries a day, so I did my lookups over a weekend.

If a movie has a non-English original name, I like to store that. There were several instances where a title I stored (via copy/paste from IMDb into the file) couldn't be found. I could generally find the titles by doing other web searches and then find the IMDb pages.

There were also a handful of movies where the title search I did through OMDb API yielded a documentary segment about the movie instead of the actual movie, so I had to do some manual cleanup (clean_db.py).

Finally, I loaded the historic entries into a new Day One journal, using db_to_dayone.py. After that, I was frequently able to use the Day One map tool to zoom to the specific theater locations and add back in theater names, using Day One's affiliation with Foursquare.

Regrets

I should have been more careful to write out HTTPS URLs in the journal entries with movies_to_db.py. I must have looked at an old entry instead of a current one when I was building the template, and I didn't notice this until too late.