Fix error when reading first page from WAL commit in SQLite3

[PimSanders]

This PR fixes an issue where a InvalidPageType would be thrown when attempting to read data from page 1 of a WAL commit, e.g. Error while executing 'notifications.wpndatabase': Unknown page type: 83.

This issue was discussed while working on #9, see: #9 (comment).

Back then I was not able to create a database with data on page 1, and page 1 being in WAL. I have found that the Windows notifications database on both my VM's does this, which allowed me to work on a fix. It is fixed by just moving the head past the header. This fix works on the notifications database.

I have not yet been able to achieve this in Python's SQLite3 library, so there is no new test data.

Required for: fox-it/dissect.target#1454

[codecov]

Codecov Report

❌ Patch coverage is 90.00000% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 82.05%. Comparing base (cba61a9) to head (6ab5bd3).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
dissect/database/sqlite3/sqlite3.py	90.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main      #15   +/-   ##
=======================================
  Coverage   82.05%   82.05%           
=======================================
  Files          31       31           
  Lines        2346     2352    +6     
=======================================
+ Hits         1925     1930    +5     
- Misses        421      422    +1     

Flag	Coverage Δ
unittests	82.05% <90.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Schamper]

Pushed a slightly changed suggestion, let me know if that works.

Question: does the header actually change? I.e. is it worth to read the database header from the WAL instead of always from fh like we do now?

[PimSanders]

I don't think that is necessary. It is possible for the header to change, for example when a PRAGMA command is executed, but that seems to be written directly to the database file.

[Schamper]

Cool, if you can confirm my change is fine, this LGTM.

[PimSanders]

Yes, it still works as expected.