More docs updates

[mathias]

Just some quick notes that I took while getting up to speed with the new version.

Related: I can call

(org.jsoup.Jsoup/parse "foo" "UTF-8")

just fine in a REPL in the project, but I get

(html->text (:body (first (inbox 1)) ))

IllegalArgumentException No matching method found: parse  clojure.lang.Reflector.invokeMatchingMethod (Reflector.java:80)

when trying to use html->text. Any clues as to how I can get around this / fix this? Thanks in advance!

[owainlewis]

Thanks Mathias. I'll have a look at the HTML parser issue. my first guess is that whatever gets returned isn't a string perhaps. The doc updates look good so I'll merge them now.

[owainlewis]

I think the html->text function is working but the message body being returned previously was a bit confusing. Now if the message is multi-part (i.e html and text parts) a vector is returned with each part i.e

[{:content-type "blah" :body "foo"} {:content-type "text/html" :body "bar"}]

So basically the body is a sequence rather than the text you might expect for multipart mesasges. The html->text will only work on a html string.

here is a working example using one of the test fixtures in /test/fixtures/25

(def message (read-mail-from-file "test/clojure_mail/fixtures/25"))
(def html-part (second (message-body message)))
(def result (html->text (:body html-part)))

;; => 

"Request to share You are the owner of ContractsBuilder. 
niuserre@gmail.com has asked that you share this item with: 
niuserre@gmail.com You can add these people in 
Sharing settings. Google Docs makes it easy to create, 
store and share online documents, spreadsheets 
and presentations."

[owainlewis]

To solve the problem in your example you would need to dig deeper into the body to get the actual HTML so something like

;; a vector containing [{:content-type "text/html" :body "foobar"}]

(def message-body-parts (:body (first inbox "user@gmail.com" "password" 1)))

(def html-part (second message-body-parts))

(html->text (:body (html-part))

Obviously that's a little confusing so would be happy to look at some way to improve that. : )