Finally no need to worry about parent-left conflicts and broken chain.
With :block/order, we only need to re-compute new orders for siblings
with same order (it can happens if there're bugs in our code, or
updates from rtc), but it doesn't break UI.
Another huge potential benefit after discussing with Zhiyuan is:
Ee might be able to simplify both RTC and undo/redo, currently, we
need to handle each new op for both of them, with recently
refactorings like properties being db attributes, :block/order
is a string instead of a ref, we can handle most property value
conflicts using last-write-wins, and others (e.g. :block/parent,
property with :default type) specifically.
I haven't fixed the issues of using :block/left in RTC and undo/redo,
because we might change both soon.
graph-parser is primarily a file graph dep that is maintained separately
from db graphs. We want to move as many namespaces out of it as
possible. Also moved differing count assertions to
their respective tests
outliner-tx/transact! now operates outliner op one by one instead of
batching processing, the benefit is that we can safely rely on the ops
order, it simplifies code a lot too.
Addressed fixmes for db-based-graph fns that were inlined to prevent
circular dependencies b/n db-property and sqlite-util. Moved multi
graph fns to db-property-util and existing db-property-util to
db-property-build to address this
This PR also remove supports for:
1. merge pages when renaming a page to existing page
2. namespaces such as a/b/c
3. nested page such as [[a [[nested page]]]]
Pages merge might be added back depends on RTC, but it should be
decoupled from renaming, otherwise it's too complex.
Namespaces and nested pages have been contributed some critical bugs
that lead data-loss, they're so complex together with page alias,
it's just impossible to have a good test coverage and ensure the app
is stable, especially when page rename and RTC.
