Initial feedback on the refactored frontend

[aarppe]

Things to consider and improve:

1. Headings
   • [x] (center-justify the headings)
   • [x] left-justify the linguistic word-content, i.e. the Cree words in the non-row/column-header cells
   • [ ] in panes, separate components of multipart heading with – (n-dash: –) or pipe (|, e.g. Prs+Ind -> present tense | independent or something is happening now | ni-/ki- word (BACKEND)
2. Column/row heading
   • [x] left-justify the non-heading row and column labels
3. Cell content
   • [x] show multiple word-forms for any cell vertically stacked
   • [x] left-justify the non-heading row and column labels
   • [ ] indicate the source of the audio with emojis, with human emoji for human-pronounced audio, and robot-head emoji for synthesized audio (BACKEND)
   • [ ] indicate word-forms that have been observed in a corpus in bold-face (as previously). (BACKEND)
4. General Settings / Other
   • [x] In the drop-down settings, separate the orthography options from the labelings options with a horizontal (hair)line
   • [ ] explore if the font in the paradigms could be more similar/slender, as in the production version

[aarppe]

@Eluviian @nienna73 Here's a list of lesser issues with the refactored frontend that would benefit from attention.

[nienna73]

Of the tasks above, the following are actually backend tasks and should be moved to the backend repo:

• Separate components of multipart headings with a dash or a pipe. The relabelling happens on the backend, so only the backend knows about the individual parts. As such, only the backend can add the divider

• Indicate the source of the audio with emojis As of right now, the backend only looks for the recording URL. It will need to add/return the speaker as well as the recording URL in order to implement this. The backend should decide on the emoji, not the frontend. The backend should look at the recording speaker, see if it's SDOL (I believe that's "Synthesized Dolores"), and add a new field called "recording_emoji" with the correct value based on the speaker value. The frontend should only have to display the emoji.

• Indicate wordforms that have been observed in the corpus I'm not actually too sure how this is happening right now, so whoever implements this will have to look at the original/legacy morphodict code so find the logic. Once again, this will need to be added to the paradigm as a new field in the JSON called "observed" with a value of "true" or "false" (or something like that). Once the backend does this, the frontend will decide on styles based on the value of "observed". The backend portion needs to be done first, though.

[nienna73]

The wording of left-justify the linguistic word-content needs clarification. What is "linguistic word-content"? Which page is this on? Where is this information located? One or multiple screenshots would be helpful.

[nienna73]

I didn't mean to close the issue, I had a keyboard/mouse disagreement that posted my earlier comment before I was ready. This issue should have never been closed.

[aarppe]

The wording of left-justify the linguistic word-content needs clarification. What is "linguistic word-content"? Which page is this on? Where is this information located? One or multiple screenshots would be helpful.

Doing this like the current production version (which left-justifies everything except the pane headings), which we could generally adhere to: