Simultaneously writing to sqlite3 database is not possible. Process exclusively
locks the database for this. Another process will fail to perform any request
when database is locked. Alternatively it can wait. Handling this situation
properly requires complexity that is not really needed. Users are not expected
to run multiple openmw processes simultaneously using the same navmeshdb.
Before this change running multiple openmw processes using the same navmeshdb
can lead to a crash when first transaction fails to start because there is
exception thrown and not catched.
Remove use of explicit transactions from DbWorker. Handling all possible
transaction states due to different errors brings unnecessary complexity.
Initially they were introduced to increase time between flushes to disk. This
makes sense for navmeshtool because of massive number of writes but for the
engine this is not an issue.
Need to load the last present record from a sequence of loaded records. That
means reverse should be called before unique or unique should be applied for
a reversed range. Since unique keeps only the first element from a sub
sequence of equal elements.
Use forEachUnique with reversed range to avoid redundant container
modifications.
Use "pragma max_page_count" to define max allowed file size in combination with
"pragma page_size" based on a new setting "max navmeshdb file size".
* Stop navmeshtool on the first db error.
* Disable writes to db in the engine on first "database or disk is full"
SQLite3 error. There is no special error code for this error.
* Change default "write to navmeshdb" to true.
* Use time intervals for transaction duration instead of number of changes.
* Remove tiles outside processing range. Useful when new content profile map
has different bounds.
* Remove ignored tiles. For a case when content profile maps have intersection
but there is no more data for navmesh.
* Remove older tiles at the same worldspace position. If navmesh tile data has
changed with new content, the old ones unlikely to be used.
* Vacuum the database when there are modifications. SQLite leaves empty pages
in the file on database modification. Vacuum cleans up unused pages reducing
the file size.
If object is too big iteration over all tiles covering it can take too much
time. Limit bounds to a square around a player position to cover only tiles
that will be present in navmesh based on max tiles number option.
Each object is associated with a set of tiles its present in. Culling can
reduce this set but it has to be update when bounds change position. Do this
in TileCachedRecastMeshManager::setBounds updating the set and adding/removing
objects to the corresponding CachedRecastMeshManagers.
* Avoid inheritance.
* Define equality operators out of the class definition.
* Replace toString with toStringView where it doesn't make sense to create a string.
Choose a parent base on which node is used to iterate over children nodes.
This leads to duplicate handing of child nodes. A node will be handled so many
times how many parents it has.
For example:
p1 p2
\ /
c
Will be handled as:
p1 p2
| |
c c
If c has children they will be handled X times c is handled.
Perform all request to db in a single thread to avoid blocking navmesh
generator threads due to slow write operations.
Write to db navmesh for all changes except update as it done for memory cache.
Batch multiple db operations into a single transaction to speed up writing by
not executing fsync after each insert/update query. All reads are performed in
the same transaction so they see uncommited data.
Merge conflicts included:
* One setting being removed (branch had changed its type).
* One setting's description being changed (branch had changed its type).
* List of files in components/files was changed both upstream and on the
branch.
* Upstream had changed something in a file the branch deletes.
With this PR we refactor `StringUtils::replaceAll` to accept `string_view` as suggested in a code comment. In addition, while we are touching this rebuild happy file, we slim it down a bit by moving a few sparingly used functions elsewhere.
This PR aims to start addressing `ESM` design issues that have silenced errors we incorporated into groundcover `ESM` loading approaches.
- We move the resolution of `parentFileIndices` from `ESMStore` to `ESMReader` as suggested in a `TODO` comment.
- We improve a highly misleading comment which downplayed the significance of `parentFileIndices`.
- We document important preconditions.
- We move a user facing error message to the highest level and improve its context.
- We remove an inappropriate `setGlobalReaderList` method. We now pass this reader list into the method that requires it.
- We remove a thoroughly pointless optimisation of `Store<ESM::LandTexture>`'s construction that has unnecessarily depended on `getGlobalReaderList`.
There should be no functional changes for `master`, but this PR should remove an issue blocking PR #3208.
To construct serializer from given entities:
* Data source/destination - any value that has to be serialized/deserialized,
usually already existing type.
* Format - functional object to define high level serialization logic to
define specific format and data schema. Like order of fields, allocation.
* Visitor - functional object to define low level serialization logic to
operator on given data part.
* BinaryWriter - copies given value into provided buffer.
* BinaryReader - copies value into given destination from provided buffer.
* SizeAccumulator - calculates required buffer size for given data.
It can fail due to float arithmetic precision with error:
Expected equality of these values:
getAsString(lua, "moveAndScale")
Which is: "TransformM{ move(6, 22, 18) scale(0.5, 1, 0.5) rotation(angle=8.53284e-17, axis=(1, 0, 0)) }"
"TransformM{ move(6, 22, 18) scale(0.5, 1, 0.5) }"
A component to load ESM content files with limited support for record types and
selection which of them to load. Supported record types are:
ACTI, CELL, CONT, DOOR, GMST, LAND, STAT.