| 169 | |
| 170 | # Splits each line of the file to create lines and conversations |
| 171 | def loadLinesAndConversations(fileName): |
| 172 | lines = {} |
| 173 | conversations = {} |
| 174 | with open(fileName, 'r', encoding='iso-8859-1') as f: |
| 175 | for line in f: |
| 176 | lineJson = json.loads(line) |
| 177 | # Extract fields for line object |
| 178 | lineObj = {} |
| 179 | lineObj["lineID"] = lineJson["id"] |
| 180 | lineObj["characterID"] = lineJson["speaker"] |
| 181 | lineObj["text"] = lineJson["text"] |
| 182 | lines[lineObj['lineID']] = lineObj |
| 183 | |
| 184 | # Extract fields for conversation object |
| 185 | if lineJson["conversation_id"] not in conversations: |
| 186 | convObj = {} |
| 187 | convObj["conversationID"] = lineJson["conversation_id"] |
| 188 | convObj["movieID"] = lineJson["meta"]["movie_id"] |
| 189 | convObj["lines"] = [lineObj] |
| 190 | else: |
| 191 | convObj = conversations[lineJson["conversation_id"]] |
| 192 | convObj["lines"].insert(0, lineObj) |
| 193 | conversations[convObj["conversationID"]] = convObj |
| 194 | |
| 195 | return lines, conversations |
| 196 | |
| 197 | |
| 198 | # Extracts pairs of sentences from conversations |