V7 SpellCheck Incorrectly Decodes Some Linefeeds and Spaces During Form UrlEncoded Requests

Question

While using the "proof" mode to check a document. The results are coming back with what appears to be errors decoding urlencoded data.

Attached test document "cancerTestSmall.txt"

When the request is sent as a POST with formUrlEncoded data, I see random mis-decoded spaces and linefeeds in the results.

Examples...

{
"offset": 941,
"suggestions": [
{
"suggestion": "20 tumor",
"score": 0.6754141504061596
}
]
},
{
"offset": 953,
"suggestions": [
{
"suggestion": "0A our",
"score": 0.6754141504061596
}
]
},

Also attaching the results json file and http request details.

172392-encodingissueshttprequest.txt

172359-encodingissuesjson.txt

172411-cancertestsmall.txt