bahman/scylla
1
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

Typos: fix typos in comments

Browse Source 
Fixes some typos as found by codespell run on the code.
In this commit, I was hoping to fix only comments, not user-visible alerts, output, etc.
Follow-up commits will take care of them.

Refs: https://github.com/scylladb/scylladb/issues/16255
Signed-off-by: Yaniv Kaul <yaniv.kaul@scylladb.com>
This commit is contained in:
Yaniv Kaul
2023-12-02 22:37:22 +02:00
parent bae6f3387f
commit c658bdb150
242 changed files with 466 additions and 466 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
db/commitlog/commitlog.cc
View File

@@ -204,7 +204,7 @@ struct db::commitlog::entry_writer {
* Should return the total, exact, size for all entries + overhead (i.e. schema)
* for this segment.
*
* Can be called more than once, if segment switch is neccesary (because race)
* Can be called more than once, if segment switch is necessary (because race)
*/
virtual size_t size(segment&) = 0;
/**
@@ -627,7 +627,7 @@ std::enable_if_t<std::is_fundamental<T>::value, T> read(Input& in) {
/*
* A single commit log file on disk. Manages creation of the file and writing mutations to disk,
* as well as tracking the last mutation position of any "dirty" CFs covered by the segment file. Segment
* files are initially allocated to a fixed size and can grow to accomidate a larger value if necessary.
* files are initially allocated to a fixed size and can grow to accommodate a larger value if necessary.
*
* The IO flow is somewhat convoluted and goes something like this:
*
@@ -654,7 +654,7 @@ std::enable_if_t<std::is_fundamental<T>::value, T> read(Input& in) {
* Note that we do not care which order segment chunks finish writing
* to disk, other than all below a flush point must finish before flushing.
*
* We currently do not wait for flushes to finish before issueing the next
* We currently do not wait for flushes to finish before issuing the next
* cycle call ("after" flush point in the file). This might not be optimal.
*
* To close and finish a segment, we first close the gate object that guards
@@ -1108,7 +1108,7 @@ public:
} catch (...) {
// If we get an IO exception (which we assume this is)
// we should close the segment.
// TODO: should we also trunctate away any partial write
// TODO: should we also truncate away any partial write
// we did?
me->_closed = true; // just mark segment as closed, no writes will be done.
throw;
@@ -2216,7 +2216,7 @@ future<> db::commitlog::segment_manager::do_pending_deletes() {
}
// #8376 - if we had an error in recycling (disk rename?), and no elements
// are available, we could have waiters hoping they will get segements.
// are available, we could have waiters hoping they will get segments.
// abort the queue (wakes up any existing waiters - futures), and let them
// retry. Since we did deletions instead, disk footprint should allow
// for new allocs at least. Or more likely, everything is broken, but