The WDF is the "within-document frequency" value for a particular
term. It's intended to provide an indication of how frequent a term is
within a document, (for use in computing relevance). Xapian's term
generator already computes WDF values when we use that, (which we do
for indexing all mail content).
We don't use the term generator when adding single terms for things
that don't actually appear in the mail document, (such as tags, the
filename, etc.). In this case, the WDF value for these terms doesn't
matter much.
But Xapian's flint backend can be more efficient with changes to terms
that don't affect the document "length". So there's a performance
advantage for manipulating tags (with the flint backend) if the WDF of
these terms is 0.
All notmuch searches currently sort by value (either date or message
ID) so it's just wasted effort for Xapian to compute relevance values
for each result. We now explicitly tell Xapian that we're uninterested
in the relevance values.
The first phase copies data from the old format to the new format
without deleting anything. This allows an old notmuch to still use the
database if the upgrade process gets interrupted. The second phase
performs the deletion (after updating the database version number). If
the second phase is interrupted, there will be some unused data in the
database, but it shouldn't cause any actual harm.
Our signal handler is designed to quickly flush out changes and then
exit. But if a database upgrade is in progress when the user
interrupts, then we just want to immediately abort. We could do
something fancy like add a return value to our progress_notify
function to allow it to tell the upgrade process to abort. But it's
actually much cleaner and robust to delay the installation of our
signal handler so that the default abort happens on SIGINT.
This takes advantage of the recently added library support to detect
if the database needs to be upgraded and then automatically performs
that upgrade, (with a nice progress report).
The recent support for renames in the database is our first time
(since notmuch has had more than a single user) that we have a
database format change. To support smooth upgrades we now encode a
database format version number in the Xapian metadata.
Going forward notmuch will emit a warning if used to read from a
database with a newer version than it natively supports, and will
refuse to write to a database with a newer version.
The library also provides functions to query the database format
version:
notmuch_database_get_version
to ask if notmuch wants a newer version than that:
notmuch_database_needs_upgrade
and a function to actually perform that upgrade:
notmuch_database_upgrade
Previously, when notmuch detected that a directory had been deleted it
was only removing files immediately in that directory. We now
correctly recurse to also remove any directories (and files, etc.)
within sub-directories, etc.
Previously we had NOTMUCH_DATABASE_MODE_READ_ONLY but
NOTMUCH_STATUS_READONLY_DATABASE which was ugly and confusing. Rename
the latter to NOTMUCH_STATUS_READ_ONLY_DATABASE for consistency.
Previously, many checks were deep in the library just before a cast
operation. These have now been replaced with internal errors and new
checks have instead been added at the beginning of all top-levelentry
points requiring a read-write database.
The new checks now also use a single function for checking and
printing the error message. This will give us a convenient location to
extend the check, (such as based on database version as well).
The original wording made it sound like this function was just doing
some string manipulation. But this function actually creates new
directory documents as a side effect. So make that explicit in its
documentation.
When a notmuch database is upgraded to the new database format, (to
support file rename and deletion), any message documents corresponding
to deleted files will not currently be upgraded. This means that a
search matching these documents will find no filenames in the expected
place.
Go ahead and return the filename as originally stored, (rather than
aborting with an internal error), in this case.
When we know that we are adding a new directory to the database, (and
we therefore are using inode rather than strcmp-based sorting of the
filenames), then we *never* want to see any names from the
database. If we get any names that could only make us inadvertently
remove files that we just added.
Since it's not obvious from the Xapian documentation whether new terms
being added as part of new documents will appear in the in-progress
all-terms iteration we are using, (and this might differ based on
Xapian backend and also might differ based on how many new directories
are added and whether a flush threshold is reached).
For all of these reasons, we play it safe and use NULL rather than a
real notmuch_filenames_t iterator in this case to avoid any problem.
The bug here was that we would see that the database did not know
anything about a directory so would get results from the filesystem in
inode rather than strcmp order.
However, we wouldn't actually ask for the list of files from the
database until after recursing into the sub-directories. So by the
time we traverse the filenames looking for deletions, the database
*does* have entries and we end up detecting erroneous deletions
because our filename list from the filesystem isn't in strcmp order.
So ask for the list of names from the database before doing any
additions to avoid this problem.
Previously we only scanned the list of filenames in the filesystem and
detected a deletion whenever that scan skipped a name that existed in
the database. That much was fine, but we *also* need to continue
walking the list of names from the database when the filesystem list
is exhausted.
Without this, removing the last file or directory within any
particular directory would go undetected.
As described in the previous commit message, we introduced multiple
symlink-based regressions in commit
3df737bc4addfce71c647792ee668725e5221a98
Here, we fix the case of symlinks to regular files by doing an extra
stat of any DT_LNK files to determine if they do, in fact, link to
regular files.
In commit 3df737bc4addfce71c647792ee668725e5221a98 we switched from
using stat() to using the d_type field in the result of scandir() to
determine whether a filename is a regular file or a directory. This
change introduced a regression in that the recursion would no longer
traverse through a symlink to a directory. (Since stat() would resolve
the symlink but with scandir() we see a distinct DT_LNK value in
d_type).
We fix this for directories by allowing both DT_DIR and DT_LNK values
to recurse, and then downgrading the existing not-a-directory check
within the recursion to not be an error. We also add a new
not-a-directory check outside the recursion that is an error.
Similar to the return value of notmuch_database_add_message, we now
enhance the return value of notmuch_database_remove_message to
indicate whether the message document was entirely removed (SUCCESS)
or whether only this filename was removed and the document exists
under other filenamed (DUPLICATE_MESSAGE_ID).
Previously, adding a filename with the same message ID as an existing
message would do nothing. But we recently fixed this to instead add
the new filename to the existing message document. So update the
documentation to match now.
In the presentation we often omit citations and signatures, but this
is not content that should be omitted from the index, (especially
when the citation detection is wrong---see cases where a line
beginning with "From" is corrupted to ">From" by mail processing
tools).
The "notmuch new" command will now efficiently notice if any files or
directories have been removed from the mail store and will
appropriately update its database.
Any given mail message (as determined by the message ID) may have
multiple corresponding filenames, and notmuch will return one of
them. When a filen is deleted, the corresponding filename will be
removed from the message in the database. When the last filename is
removed from a message, that message will be entirely removed from the
database.
All file additions are handled before any file removals so that rename
is supported properly.
It is essential to defer the actual removal of any filenames from the
database until we are entirely done adding any new files. This is to
avoid any information loss from the database in the case of a renamed
file or directory.
Note that we're *still* not actually doing any removal---still just
printing messages indicating the filenames that were detected as
removed. But we're at least now printing those messages at a time when
we actually *can* do the actual removal.
This takes advantage of the notmuch_directory_t interfaces added
recently (with cooresponding storage of directory documents in the
database) to detect when files or entire directories are deleted or
renamed within the mail store.
This also fixes the recent regression where *all* files would be
processed by every run of "notmuch new", (now only new files are
processed once again).
The deleted files and directories are only detected so far. They
aren't properly removed from the database.
Previously, we were re-scanning the entire list of entries for every
directory entry. Instead, we can simply check if the entries look like
a maildir once, up-front.
We now do two scans over the entries returned from scandir. The first
scan is looking for directories (and making the recursive call). The
second scan is looking for new files to add to the database.
This is easier to read than the previous code which had a single loop
and some if statements with ridiculously long bodies. It also has the
advantage that once the directory scan is complete we can do a single
comparison of the filesystem and database mtimes and entirely skip the
second scan if it's not needed.
Previously we had an array named "namelist" and its count named
"num_entries". We now use an array name of "fs_entries" and a count
named "num_fs_entries" to try to preserve sanity.
We were previousl using the stat for two reasons. One was to obtain
the mtime of the file. This usage was removed in the previous commit,
(since the mtime is unreliable in the case of a file being moved into
the mail store).
The second reason was to identify regular and directory file
types. But this information is already available in the result we get
from scandir.
What's left is simply a stat for each directory in the mailstore,
(which we are still using to compare filesystem mtime with the mtime
stored in the database).
This check was buggy in that moving a pre-existing file into the mail
store, (where the file existed before the last run of "notmuch new"),
does not update the mtime of the file. So the message would never be
added to the database.
The fix here is not practical in the long run, (since it causes *all*
files in the mail store to be processed in every run of "notmuch new"
(!)). But this change will let us drop a stat() call that we don't
otherwise need and will help move us toward proper database-backed
detection of new files, (which will fix the bug without the
performance impact of the current fix).
The previous name of "path_mtime" was very ambiguous. The new names
are much more obvious (fs_mtime is the mtime from the filesystem and
db_mtime is the mtime from the database).
This was a very dangerous bug. An interrupted "notmuch new" session
would still update the timestamp for the directory in the
database. This would result in mail files that were not processed due
to the original interruption *never* being picked up by future runs of
"notmuch new". Yikes!
This new directory ojbect provides all the infrastructure needed to
detect when files or directories are deleted or renamed. There's still
code needed on top of this (within "notmuch new") to actually do that
detection.
This commit contains my changes to the API proposed by Keith. Nothing
is dramatically different. There are minor things like changing
notmuch_files_t to notmuch_filenames_t and then various things needed
for completeness as noticed while implementing this, (such as
notmuch_directory_destroy and notmuch_directory_set_mtime).
This will allow applications to support the removal of messages, (such
as when a file is deleted from the mail store). No removal support is
provided yet in commands such as "notmuch new".
The existing find_doc_ids function is convenient when the caller
doesn't want to be bothered constructing a term. But when the caller
*does* have the term already, that interface is just wasteful. So we
export a lower-level interface that maps a pre-constructed term to a
document-ID iterators.
The code to map a filename to a direntry is something that we're going
to want in a future _remove_message function, so put it in a new
function _notmuch_database_filename_to_direntry .
The library interface is unchanged so far, (still just
notmuch_database_add_message), but internally, the old
_set_filename function is now _add_filename instead.
Instead of storing the complete message filename in the data portion
of a mail document we now store a 'direntry' term that contains the
document ID of a directory document and also the basename of the
message filename within that directory. This will allow us to easily
store multple filenames for a single message, and will also allow us
to find mail documents for files that previously existed in a
directory but that have since been deleted.
Some pending commits want the _split_path functionality separate from
mapping a directory to a document ID. The split_path function now
returns the basename as well as the directory name.
