Keywords - Replace whitespaces with CamelCase

[dderom]

I use the Better BibTeX to import refs into Obsidian. I update my Zotero on a regular basis and add keywords as I go along. However, keywords with a whitespace (eg Natural Language Processing) are not allowed in Obsidian. One either has to use CamelCase, underscore or dash (aka NaturalLanguageProcessing or Natural_Language Processing or Natural-Language-Processing).

I would like to convert on export the tags with Camelcase tags. I assume the Postscript on the export tab would be of use. However, can't seem to get it to work.

Any pointers on where I could find some info on how to get this going?

[retorquere]

with a postscript:

if (Translator.BetterTeX && zotero.tags) {
  const tags = zotero.tags
    .map(tag => (typeof tag === 'string' ? tag : tag.tag))
    .map(tag => tag.replace(/\b\w/g, letter => letter.toUpperCase()).replace(/\s/g, ''))
  tex.add({ name: 'keywords', value: tags, enc: 'tags' })
}

[dderom]

Brilliant, that did it. Thanks for the swift reply ánd the plugin.

[Gladbjerg]

Pardon my question, but how do you get the keywords to export from Zotero? (I can't get it to work).

[Gladbjerg]

this postscript will add the keywords to CSL exports, but I don't know how you'd use them.

if (Translator.BetterCSL && zotero.tags) {
  csl.keywords = zotero.tags.map(tag => (typeof tag === 'string' ? tag : tag.tag))
}

Thank you. Would that also remove the whitespace? (I prefer an underscore).

[Gladbjerg]

And right now I am stuck on not being able to import them to Obsidian (at least not using the mentioned plugin).

[retorquere]

Thank you. Would that also remove the whitespace? (I prefer an underscore).

No, for that you'd need

if (Translator.BetterCSL && zotero.tags) {
  csl.keywords = zotero.tags.map(tag => (typeof tag === 'string' ? tag : tag.tag)).map(tag => tag.replace(/\s+/g, '_'))
}

And right now I am stuck on not being able to import them to Obsidian (at least not using the mentioned plugin).

I can't help with that, I'm not a regular Obsidian user.

[dderom]

As soon as you have the zotero postscript, you need to add a template to import the keywords. In other words, go to the settings/options of the plugin (cf. Citations) of Obsidian and add the following in the 'literature note content template':

---
type: reference
authors: {{authorString}}
year: {{year}}
DOI: {{DOI}}
URL: {{URL}}
year: {{year}}
citekey: {{citekey}}
---
# {{title}} ({{year}}_

## Abstract
{{abstract}}

## Tags
{{#each entry.data.fields.keywords}}{{#if @last}}#{{this}} {{else}}#{{this}} {{/if}}{{/each}}

Simply giving you a template I use, where the first part is front matter, the second imports the keywords of the zotero reference. So essentially, you need only the last part to import the keywords:

## Tags
{{#each entry.data.fields.keywords}}{{#if @last}}#{{this}} {{else}}#{{this}} {{/if}}{{/each}}

[Gladbjerg]

As soon as you have the zotero postscript, you need to add a template to import the keywords. In other words, go to the settings/options of the plugin (cf. Citations) of Obsidian and add the following in the 'literature note content template':

---
type: reference
authors: {{authorString}}
year: {{year}}
DOI: {{DOI}}
URL: {{URL}}
year: {{year}}
citekey: {{citekey}}
---
# {{title}} ({{year}}_

## Abstract
{{abstract}}

## Tags
{{#each entry.data.fields.keywords}}{{#if @last}}#{{this}} {{else}}#{{this}} {{/if}}{{/each}}

Simply giving you a template I use, where the first part is front matter, the second imports the keywords of the zotero reference. So essentially, you need only the last part to import the keywords:

## Tags
{{#each entry.data.fields.keywords}}{{#if @last}}#{{this}} {{else}}#{{this}} {{/if}}{{/each}}

Where did you find the documentation regarding importing the tags? I can see, that you import every tag as a tag with # before the tha (e.g. #tag1, #tag2). It is possbile to change the import, so it will be a comma e.g. Tag1, Tag2 etc.

[Gladbjerg]

So I have had a chance to play around with it.. I am now able to export the Tags to the BBT file. And also import it into Obsidian. But how do I get the tags to be seperated with a "," and not a "#" in front of a tag (because I don't want to use them with Obsidian keywords, but rather as YAML). eg. tags: tag1, tag2, tag3 and not tags: #tag1 #tag2 #tag3. If I do export via CamelCase, then it capitalises the letters after æ,ø,å eg. LæRebøGer where it should be Lærebøger (which it is in the Zotero Tag)

[retorquere]

I need to know what you mean by "BBT file" here, because the discussion above switches between BibLaTeX, CSL-JSON, and BBT-JSON. I am not adding # marks anywhere, so I don't know where those come from, or at what stage of the process you are following they appear.

The camelcase issue is because \w only understands ascii, so assuming you're exporting to BibTeX, try this:

if (Translator.BetterTeX && zotero.tags) {
  const tags = zotero.tags
    .map(tag => (typeof tag === 'string' ? tag : tag.tag))
    .map(tag => tag.replace(/(^|\s+)./g, letter => letter.toUpperCase()).replace(/\s/g, ''))
  tex.add({ name: 'keywords', value: tags, enc: 'tags' })
}

[Gladbjerg]

Sorry for not explaining it better. By BBT I mean the Better BibLaTeX file.
NO I found out, that the adding of the # is done at import into Obsidian. And the problem with the formatting is also in the import from the Better BibLaTeX file..

My problem is when I import from the Better BibLaTeX file with the Obsidian Citation plugin..

Sorry for the confusion.

[retorquere]

If the # characters are added after the .bib file is produced, there's nothing that BBT can do. You'll have to ask the author of the Obsidian Citation plugin.