Skip to content

Commit

Permalink
indexer: make lazy check even lazier
Browse files Browse the repository at this point in the history
In lazy-mode, we were skipping directories that did not change; however,
this didn't help for the case were users received new messages in big
maildirs.

So, add another check where we compare the ctime of message files with
the time of the last indexing operation. If it's smaller, ignore the
message-file. This is faster than having to consult the Xapian database
for each message.

Note that this requires in mu4e:
      (setq mu4e-index-lazy-check t)
or
   --lazy-check
as a parameter for 'mu index'.
  • Loading branch information
djcb committed Nov 10, 2024
1 parent fa59040 commit b0d8d42
Show file tree
Hide file tree
Showing 3 changed files with 29 additions and 14 deletions.
14 changes: 12 additions & 2 deletions lib/mu-indexer.cc
Original file line number Diff line number Diff line change
Expand Up @@ -145,6 +145,8 @@ struct Indexer::Private {
std::mutex lock_, w_lock_;
std::atomic<time_t> completed_{};
bool was_empty_{};

uint64_t last_index_{};
};

bool
Expand Down Expand Up @@ -206,12 +208,16 @@ Indexer::Private::handler(const std::string& fullpath, struct stat* statbuf,

case Scanner::HandleType::File: {
++progress_.checked;
if (conf_.lazy_check && static_cast<uint64_t>(statbuf->st_ctime) < last_index_) {
// in lazy mode, ignore the file if it has not changed
// since the last indexing op.
return false;
}

if ((size_t)statbuf->st_size > max_message_size_) {
if (static_cast<size_t>(statbuf->st_size) > max_message_size_) {
mu_debug("skip {} (too big: {} bytes)", fullpath, statbuf->st_size);
return false;
}

// if the message is not in the db yet, or not up-to-date, queue
// it for updating/inserting.
if (statbuf->st_ctime <= dirstamp_ && store_.contains_message(fullpath))
Expand Down Expand Up @@ -414,6 +420,10 @@ Indexer::Private::start(const Indexer::Config& conf, bool block)
mu_debug("indexing: {}; clean-up: {}", conf_.scan ? "yes" : "no",
conf_.cleanup ? "yes" : "no");

// remember the _previous_ indexing, so in lazy mode we can skip
// those files.
last_index_ = store_.config().get<Mu::Config::Id::LastIndex>();

state_.change_to(IndexState::Scanning);
/* kick off the first worker, which will spawn more if needed. */
workers_.emplace_back(std::thread([this] { item_worker(); }));
Expand Down
4 changes: 2 additions & 2 deletions lib/mu-indexer.hh
Original file line number Diff line number Diff line change
Expand Up @@ -54,8 +54,8 @@ public:
bool ignore_noupdate{};
/**< ignore .noupdate files */
bool lazy_check{};
/**< whether to skip directories that don't have a changed
* mtime */
/**< whether to skip directories or message files that haven't changed since the
* previous indexing operation, based on their ctime */
};

/**
Expand Down
25 changes: 15 additions & 10 deletions man/mu-index.1.org
Original file line number Diff line number Diff line change
Expand Up @@ -41,7 +41,7 @@ If there is a file called _.noupdate_ in a directory, the contents of that
directory and all of its subdirectories will be ignored. This can be useful to
speed up things you have some maildirs that never change.

_.noupdate_ does not affect already-indexed message: you can still search for
_.noupdate_ does not affect already-indexed messages: you can still search for
them. _.noupdate_ is ignored when you start indexing with an empty database (such
as directly after *mu init*).

Expand All @@ -58,7 +58,7 @@ the database for which there is no longer a corresponding file in the Maildir.
If you do not want this, you can use *-n*, *--nocleanup*.

When *mu index* catches one of the signals *SIGINT*, *SIGHUP* or *SIGTERM* (e.g., when
you press Ctrl-C during the indexing process), it attempts to shutdown
you press *Ctrl-C* during the indexing process), it attempts to shutdown
gracefully; it tries to save and commit data, and close the database etc. If it
receives another signal (e.g., when pressing Ctrl-C once more), *mu index* will
terminate immediately.
Expand All @@ -67,12 +67,17 @@ terminate immediately.

** --lazy-check
In lazy-check mode, *mu* does not consider messages for which the time-stamp
(ctime) of the directory they reside in has not changed since the previous
indexing run. This is much faster than the non-lazy check, but won't update
messages that have change (rather than having been added or removed), since
merely editing a message does not update the directory time-stamp. Of course,
you can run *mu-index* occasionally without *--lazy-check*, to pick up such
messages.
(*ctime*) of the directory in which they reside, has not changed since the
previous time this directory was checked.

This is much faster than the non-lazy check, but won't update messages that have
changed (rather than having been added or removed), since merely editing a
message does not update the directory time-stamp. Of course, you can run
*mu-index* occasionally without *--lazy-check*, to pick up such messages.

Furthermore, in lazy-check mode, files which have a *ctime* smaller than the time
the previous indexing operation was completed, are ignored. This helps for the
use-case where new messages can appear in big maildirs.

** --nocleanup
Disable the database cleanup that *mu* does by default after indexing.
Expand Down Expand Up @@ -172,7 +177,7 @@ The instructions are a little different since we have a proper repeatable
benchmark now. After building,

#+begin_example
$ sudo sh -c 'sync && echo 3 > /proc/sys/vm/drop_caches'
$ sudo sh -c 'sync && echo 3 > /proc/sys/vm/drop_caches'
% THREAD_NUM=4 build/lib/tests/bench-indexer -m perf
# random seed: R02Sf5c50e4851ec51adaf301e0e054bd52b
1..1
Expand All @@ -185,7 +190,7 @@ ok 1 /bench/indexer/4-cores
#+end_example

Things are again a little faster, even though the index does a lot more now
(text-normalizatian, and pre-generating message-sexps). A faster machine helps,
(text-normalization, and pre-generating message-sexps). A faster machine helps,
too!

** recent releases
Expand Down

0 comments on commit b0d8d42

Please sign in to comment.