)]}'
{
  "commit": "efbbf31023483691a9467124a38bf8103f41d3ae",
  "tree": "8dcf431e76203ef0b01590ec4dfaa2154b4c01c5",
  "parents": [
    "8f743ef97916f32e2666832f41ccfdbf5f1f9d47"
  ],
  "author": {
    "name": "Piotr Sikora",
    "email": "piotrsikora@google.com",
    "time": "Wed Mar 15 15:55:35 2017 -0700"
  },
  "committer": {
    "name": "Piotr Sikora",
    "email": "piotrsikora@google.com",
    "time": "Wed Mar 15 15:55:35 2017 -0700"
  },
  "message": "Proxy: split configured header names and values.\n\nPreviously, each configured header was represented in one of two ways,\ndepending on whether or not its value included any variables.\n\nIf the value didn\u0027t include any variables, then it would be represented\nas as a single script that contained complete header line with HTTP/1.1\ndelimiters, i.e.:\n\n     \"Header: value\\r\\n\"\n\nBut if the value included any variables, then it would be represented\nas a series of three scripts: first contained header name and the \": \"\ndelimiter, second evaluated to header value, and third contained only\n\"\\r\\n\", i.e.:\n\n     \"Header: \"\n     \"$value\"\n     \"\\r\\n\"\n\nThis commit changes that, so that each configured header is represented\nas a series of two scripts: first contains only header name, and second\ncontains (or evaluates to) only header value, i.e.:\n\n    \"Header\"\n    \"$value\"\n\nor\n\n    \"Header\"\n    \"value\"\n\nThis not only makes things more consistent, but also allows header name\nand value to be accessed separately.\n\nSigned-off-by: Piotr Sikora \u003cpiotrsikora@google.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "839d479eaf6fe178b57d6871dbd7f93bdb299b27",
      "old_mode": 33188,
      "old_path": "src/http/modules/ngx_http_proxy_module.c",
      "new_id": "61bf55c157cb398c8294f3cb000fbe1631b28513",
      "new_mode": 33188,
      "new_path": "src/http/modules/ngx_http_proxy_module.c"
    }
  ]
}