{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "ename": "ERROR",
     "evalue": "repl.go:6:5: error loading package \"dataframe\" metadata, maybe you need to download (go get), compile (go build) and install (go install) it? can't find import: \"dataframe\"",
     "output_type": "error",
     "traceback": [
      "repl.go:6:5: error loading package \"dataframe\" metadata, maybe you need to download (go get), compile (go build) and install (go install) it? can't find import: \"dataframe\""
     ]
    }
   ],
   "source": [
    "import (\n",
    "    \"io/ioutil\"\n",
    "    \"fmt\"\n",
    "    \"strings\"\n",
    "    \"strconv\"\n",
    "    \"dataframe\"\n",
    ")\n",
    "\n",
    "const kitchenReviews = \"./datasets/words/processed_acl/kitchen\"\n",
    "\n",
    "type Pair struct {\n",
    "    Phrase string\n",
    "    Frequency int\n",
    "}\n",
    "\n",
    "func pairsAndFilters(splitPairs []string) ([]Pair, map[string]bool) {\n",
    "    var (\n",
    "    pairs []Pair\n",
    "    m map[string]bool\n",
    "    )\n",
    "    m = make(map[string]bool)\n",
    "    for _, pair := range splitPairs{\n",
    "        p := strings.Split(pair, \":\")\n",
    "        phrase := p[0]\n",
    "        m[phrase] = true\n",
    "        if len(p) < 2 {\n",
    "            continue\n",
    "        }\n",
    "        \n",
    "        freq, err := strconv.Atoi(p[1])\n",
    "        if err != nil {\n",
    "            fmt.Println(\"Fuck \", err)\n",
    "        }\n",
    "        pairs = append(pairs, Pair{\n",
    "            Phrase: phrase, \n",
    "            Frequency: freq,\n",
    "        })\n",
    "    }\n",
    "    return pairs, m\n",
    "}\n",
    "\n",
    "func exclude(pairs []Pair, exclusions map[string]bool) []Pair {\n",
    "    var ret []Pair\n",
    "    for _, pair := range pairs{\n",
    "        if !exclusions[pair.Phrase] {\n",
    "            ret = append(ret, pair)\n",
    "        }\n",
    "    }\n",
    "    return ret\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "ename": "ERROR",
     "evalue": "repl.go:21:10: undefined \"dataframe\" in dataframe.LoadStructs <*ast.SelectorExpr>",
     "output_type": "error",
     "traceback": [
      "repl.go:21:10: undefined \"dataframe\" in dataframe.LoadStructs <*ast.SelectorExpr>"
     ]
    }
   ],
   "source": [
    "\n",
    "positives, err := ioutil.ReadFile(kitchenReviews + \"/positive.review\")\n",
    "if err != nil {\n",
    "    fmt.Println(\"could not read positive reviews: \", err)\n",
    "}\n",
    "\n",
    "negatives, err := ioutil.ReadFile(kitchenReviews + \"/negative.review\")\n",
    "if err != nil {\n",
    "    fmt.Println(\"could not read negative reviews: \", err)\n",
    "}\n",
    "\n",
    "pairsPositives := strings.Fields(string(positives))\n",
    "pairsNegatives := strings.Fields(string(negatives))\n",
    "\n",
    "parsedPositives, posPhrases := pairsAndFilters(pairsPositives)\n",
    "parsedNegatives, negPhrases := pairsAndFilters(pairsNegatives)\n",
    "parsedPositives = exclude(parsedPositives, negPhrases)\n",
    "parsedNegatives = exclude(parsedNegatives, posPhrases)\n",
    "\n",
    "\n",
    "dfPos := dataframe.LoadStructs(parsedPositives)\n",
    "dfNeg := dataframe.LoadStructs(parsedNegatives)\n",
    "\n",
    "fmt.Println(dfPos)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Go",
   "language": "go",
   "name": "gophernotes"
  },
  "language_info": {
   "codemirror_mode": "",
   "file_extension": ".go",
   "mimetype": "",
   "name": "go",
   "nbconvert_exporter": "",
   "pygments_lexer": "",
   "version": "go1.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}