Well, if you're interested in building a database index, this might be a good article to read.
I'm looking at ways to implement our tag indices.
Finally, a chance to use all of this Computer Science stuff I didn't learn in college.
Regular expression matching if the pattern is in the beginning of a sentence, the end of a sentence is is surrounded by white space
We were finding "tags" in the URL anchor fragments. Hopefully this should fix it.
We should also have a list of all tags. Good for inspiring ideas.
It should also list the number of entries for each.
Show the results in REVERSE chronological order. This is a big deal because it would've been a PITA with plain text files. #notetaking #todo
#notetaking system #todo:
- Linkify tags
- Show the entry hash and link before each entry (especially in search results)
- Can't input tab characters in Firefox
- Obviously our tag indexing system is completely pessimal
- Individual entry pages should perhaps include a "what links here" list
- I'm not convinced we're showing search results in order (relying on undefined behavior)
- It kind of sucks writing new entries in a blank window instead of below the previous tags (put an entry field at the bottom of the search results, and pre-fill it with the search tags?)
And lots more, I'm sure...
- It sucks having to click a button instead of just being able to type some dashes to finish the entry and start a new one
It's even harder than I thought, because we do need to concatenate entries in real time for our search results.
Our current system of piping the template header and then the entry body at read-time doesn't work with GZIP compression, among other failings. We need to create static HTML files ahead-of-time, which means we should do escaping and fancy formatting then too.
RethinkDB: An open-source distributed database built with love over three years (rethinkdb.com)
Instead of "NoSQL", we need a "NoDB" movement.
I think our new #notetaking system is approaching working status.