Cannot fetch using d3.text while following Intro to TF-IDF tutorial

Community Help
1

Howdy y’all!

Digging into Observable for the first time. Spent the morning trying to reproduce the Introduction to text analysis with TF-IDF tutorial by @kerryrodden. I’m trying to use this to analyze a collection of Church documents over time and consistently getting hung trying to fetch the text of each document.

I initially named the folder “documents”, but since “document” is reserved I switched it to “writings”. No matter what I try I’m getting these errors.

04
041499×646 73.4 KB

Does anyone know what I’m doing wrong?

1 Like
2

This is the repo I’m working with.

3

And this is the fork I just created hoping to fix the issue to no avail.

4

Hi @jameshahn2! Here’s a quick way for you to see what’s happening:

Promise.all(files.map(async f => [f, await d3.text(f).catch(error => error)]))

Or, to just show the broken URLs:

{
  const errors = [];
  for (const f of files) {
    try {
      await d3.text(f);
    } catch (error) {
      errors.push(f);
      yield errors;
    }
  }
}

For me this yields two broken links:

https://raw.githubusercontent.com/jameshahn2/synod-documents/master/writings/Justicia_in_Mundo_SYNOD.txt
https://raw.githubusercontent.com/jameshahn2/synod-documents/master/writings/Africa_Munus_SYNOD.txt

The first appears intended to refer to

https://raw.githubusercontent.com/jameshahn2/synod-documents/master/writings/Justicia-in-Mundo_SYNOD.txt

But the second referenced file doesn’t appear to exist in the repository.

If you want to ignore any failing files, you can do this:

writings = Promise
  .all(files.map(d => d3.text(d).catch(() => null)))
  .then(texts => texts.filter(t => t !== null))

Here is a suggestion you can merge with this change:

3 Likes
5

Awesome! I’ll get to the bottom of this.

Thank you, sir!!