A single-page application to let users know more about their Whatsapp usage. This app take a file an analyze the content to extract messages, organize them and print interesting charts with metrics. It also restructure the conversation in a like-Whatsapp layout. It include keywords and relationship analysis.
In its first version will only admit insert files extracted from Whatsapp. For future version it will has an owner email server so the user can send the conversation directly from its mobile phone.
Its main role is to parse a text and to give it correct format so the app can create a Conversation object that the rest of modules can manage. Singleton pattern.
This class user a graph as data structure to abstract the interactions between users. This way we can get metrics as followers, followings or low-interaction users.
Instance of KeywordTree to analyze the conversation and find the topics and the most common word of each user.
Conversation class models a chat between different persons. This class will be responsible for sort the message and format them for analysis. This class stores the messages in two different data structures:
-
Queue: Each message has a reference to the next. This class has pointers to the first one and the last one. This way we can have direct access to the full conversation.
-
Array of days: In this array we store instances of Day, days have 24 properties, one for each hour in a day. This way, each index in the array represents a day. Index 0 represent the day of the first message. With this data structure we can have direct access to each day-hour like this: Array[day - firstDay][12]
This class models each message in the conversation. This class includes a pointer to the next message so it can work as a node in a queue.
This class models each user in the conversation. This class includes main metrics related to the user.
- Abstract functions in main.js file.
- Configure SnapChart charts to print correctly in the main container.
- Print main metrics in html. Animate the process.
- Add icons to buttons.
- Implement a scroll-loader to keywords list to make html lighter.
- Develop email server to analyzer conversation sent by email.
- Abstract system to admit any kind of conversation or interaction between users.