MCP-2240 Un-parsable date fix

[jbregstone]

What was the problem?

If the code to parse a date failed then the date we show on the PDF was set to an empty string.

Associated tickets or Slack threads:

• MCP-2240
• https://amida.slack.com/archives/C01Q7979Z7D/p1672172009573269

How does this fix it?¹

In the case of an un-parsable date it is passed directly to the PDF without being parsed

How to test this PR

• Step 1
• Step 2²

Footnotes

1. Pull-Requests guidelines. If PR is significant, update Current Software State wiki page. ↩

2. To check if a PR will succeed in the SecRel workflow, test PRs in the SecRel pipeline. ↩

[anthonyvelazquez]

So we will need to change the date to something like unknown-{date}. Passing the date as it is, I won't know if I need to fill in the table cell with No associated date / Non-confident date {date} or if I should just display the date. Maybe adding a prefix/suffix or passing an additional field that lets me know its a bad date?

[anthonyvelazquez]

Wait, I can do the parse on my side if needed or should we keep that logic in your service?

[jbregstone]

Passing the date as it is, I won't know if I need to fill in the table cell with No associated date / Non-confident date {date} or if I should just display the date.

I see that could be an issue. If it simplifies things, then we could move the parsing to the PDF side. I thought it would be better to have it in here since the code is already going through and reading each date. I could change dateformatted to start with with the string "unparsed". In the pdf service it could check if the dateformatted starts with that string and then handle the case accordingly. Keep in mind this is an edge case so I want to minimize designing around it.

[anthonyvelazquez]

Since you already have the logic on your side, I think just passing unparsed is fine. I have helper functions for each code so I'll add that logic on my side in there so it doesnt affect future PDFs

[jbregstone]

The four cases to account for in the pdf helper logic:

1. If there is no date at all -> dateFormatted is an empty string
2. If there is well formatted date -> dateFormatted looks like 04/06/1950
3. If there is a date but its empty -> dateFormatted is 'unparsed ()`
4. if there is a poorly formatted date (like "1999/m/d")-> dateFormatted is 'unparsed (1999/m/d)'