Add the ability to parse a line with a link and a description

[jletey]

Here's a rough draft ... tell me what you think

[nikitavoloboev]

I tried it with:

func TestParseLinkWithDescription(t *testing.T) {
	assert := assert.New(t)
	v := "- [Effective Go](https://golang.org/doc/effective_go.html)"
	result := ParseLinkWithDescription(v)
	assert.Equal(result["Title"], "Effective Go", "Check to see if the title is correct")
	assert.Equal(result["Link"], "https://golang.org/doc/effective_go.html", "Check to see if the link is correct")
	assert.Equal(result["Description"], "", "Check to see if the description is correct")
}

And go test failed. Can we test what happens if there is no description too?

[jletey]

Sure ... working on it

[nikitavoloboev]

Speaking of using this PR to solve nikitavoloboev/alfred-learn-anything#9, are you on mac or windows. And if on mac, do you have Alfred? :)

[jletey]

I'm on Mac ... but I don't have Alfred

[jletey]

Also, can you take a look at nikitavoloboev/license-up#28

[jletey]

When we don't have a description, the regex completely changes ... how exactly do you want to go about checking if there is a description.

Regex for link without description: (?m)^- \[([^]]+)\]\(([^)]+)\)$
Regex for link with description: (?m)^- \[([^]]+)\]\(([^)]+)\) - (.+)$

[nikitavoloboev]

It has to be all encompassing.

The use case if to parse a file like this:

https://github.com/learn-anything/books

And extract all the links (books + # Other) inside the file to some data structure.

[nikitavoloboev]

If there is no description, it should only return the title and URL (those are bound to be there).

[nikitavoloboev]

but I don't have Alfred

That's a shame. It's a pretty amazing tool. Would definitely recommend. 🚀

There's lots of workflows for it too https://github.com/learn-anything/alfred-workflows#readme

[jletey]

That's a shame. It's a pretty amazing tool. Would definitely recommend. 🚀

I know it's an amazing tool ... but the problem is that they charge for workflows

[nikitavoloboev]

but the problem is that they charge for workflows

Can send your PayPal, I can sponsor a license for you. 🙂

[jletey]

Can send your PayPal, I can sponsor a license for you. 🙂

I don't have a PayPal account ... thank you though, that's very kind of you!

[jletey]

Should we change the function name from ParseLinkWithDescription to just ParseLink?

[nikitavoloboev]

ParseLinkWithDescription to just ParseLink

yes

[nikitavoloboev]

from ParseLinkWithDescription to just ParseLink

And do the error handling there in case there is no description. I believe there is none currently.

[jletey]

And do the error handling there in case there is no description. I believe there is none currently.

There should be a test case called TestParseLinkWithoutDescription

[nikitavoloboev]

There should be a test case called TestParseLinkWithoutDescription

Or maybe just TestParseLink and inside it test a link with description and one without and check for outputs. As well as some other text like blargh to check if handles this case properly too.

[jletey]

Doing that now

[jletey]

@nikitavoloboev How does it look now?

[nikitavoloboev]

Works great. Awesome!

[jletey]

Great! So is it now usable for your knowledge and other projects?

[nikitavoloboev]

Will have to check. Will need few minutes/hours. 😅

[jletey]

Got it

[nikitavoloboev]

So is it now usable for your knowledge and other projects?

Actually for nikitavoloboev/knowledge#25 it won't work as Imgur link embeds are of different format. Will have to write code specifically for parsing image embeds.

![](https://i.imgur.com/ZVPjkzh.png)

^ Is image embed.

[jletey]

I see ... you can probably just copy and paste the code for ParseLink into ParseImgurLink and then change up the regex

I can try to do this if you'd like

[nikitavoloboev]

Opened #15