Enable programmatic analysis and visualization of iMessages which are commonly used on Apple devices. Examples of use cases:
- how many people have you corresponded with in the last month or year?
- who is the quickest, or slowest, to text back?
- who have you messaged with most during a particular time?
- have you lost touch with anyone?
Target audience: any iMessage user, current or past. Project requires direct read access to the relevant database file, so currently requires minor technical skills.
- All relevant iMessage personal data is stored in
chat.db. The project requiresreadaccess to this file, and for now, it must be in the same directory asanalysis.py. For now, this must be done manually. - Further, for this file to contain up-to-date data, you must have sync enabled between your iPhone and your Mac laptop/computer.
[HOME_FOLDER/]Library/Messages/chat.db- actual chat data.[HOME_FOLDER/]Library/Messages/Attachments- message attachments[HOME_FOLDER/]Library/Messages/chat.db-shm- SQLite-specific file, irrelevant to analyis.[HOME_FOLDER/]Library/Messages/chat.db-wal- "write-ahead log", relevant only operationally, irrelevant to analysis.
- ✅ connect to
chat.db - ✅ read table stats e.g. tables, rows, row counts
- 🕗 read 10 latest messages
- ☐ if possible, get date when overall db last-updated
- add
Attachmentsfolder analysis - if possible, get date when overall db last-updated
- rewrite SQL queries to accomplish more within query
- (medium priority) use
logging - (low priority) allow exports of texts or analysis to
.pdf/spreadsheet
- ✅ Separate file for SQL queries
- 📝 add "
TABLE CREATE" queries for each table to new.sqlfile - 📝 make
chat.dbfile location configurable (config.py)
- OOP versus functions. Which is suited where.
- Typing: consistent type annotations. Where do annotations make sense and where not.
- Add init.py file to make project a package?
- Add docstrings for each function? Which convention to follow. Potentially autogenerate docs.
- Percentage breakdown of you-versus-other(s) message count per chat(group)
- per chat between you and another (and aggregate up to overall), mean/median/skew on both message length and frequency ** for all your 1-1 chats, plot 2D frequency and length of overall messages
- your most actively messages sent during 24hrs
I used the following for inspiration: https://stmorse.github.io/journal/iMessage.html
- Analysis and query ideas. https://spin.atomicobject.com/2020/05/22/search-imessage-sql/
- Analysis and query ideas. https://linuxsleuthing.blogspot.com/2015/01/getting-attached-apple-messaging.html
- Description of figuring out Message attachments. And the inspiration to add full
CREATE TABLE.sqlfile to the repo(a TODO as of this writing). https://github.com/dsouzarc/iMessageAnalyzer - Existing but older analysis app for Mac. Used for feature and visualization ideas.