Skip to content

Commit

Permalink
Support locking of directory entries
Browse files Browse the repository at this point in the history
Moves the locking algorithm from file entry to file system entry.

Fixes #137
  • Loading branch information
Nathan Memmott committed Dec 9, 2023
1 parent bd6ef5a commit a0294b7
Showing 1 changed file with 46 additions and 24 deletions.
70 changes: 46 additions & 24 deletions index.bs
Original file line number Diff line number Diff line change
Expand Up @@ -110,7 +110,9 @@ specifications do not need to bother implementing [=/file system entry=]'s

Issue(101): Make access check algorithms associated with a FileSystemHandle.

Each [=/file system entry=] has an associated <dfn for="file system entry" id=entry-name>name</dfn> (a [=string=]).
Each [=/file system entry=] has an associated <dfn for="file system entry" id=entry-name>name</dfn> (a [=string=]),
a <dfn for="file system entry">lock</dfn> (a string that may exclusively be "`open`", "`taken-exclusive`" or "`taken-shared`")
and a <dfn for="file system entry">shared lock count</dfn> (a number representing the number of shared locks that are taken at a given point in time).

A <dfn>valid file name</dfn> is a [=string=] that is not an empty string, is not equal to "." or "..",
and does not contain '/' or any other character used as path separator on the underlying platform.
Expand All @@ -124,22 +126,20 @@ Issue: We should consider having further normative restrictions on file names th
never be allowed using this API, rather than leaving it entirely up to underlying file
systems.

A <dfn export id=file>file entry</dfn> additionally consists of
<dfn for="file entry" export>binary data</dfn> (a [=byte sequence=]), a
<dfn for="file entry">modification timestamp</dfn> (a number representing the number of milliseconds since the <a spec=FileAPI>Unix Epoch</a>),
a <dfn for="file entry">lock</dfn> (a string that may exclusively be "`open`", "`taken-exclusive`" or "`taken-shared`")
and a <dfn for="file entry">shared lock count</dfn> (a number representing the number shared locks that are taken at a given point in time).

A user agent has an associated <dfn>file system queue</dfn> which is the
result of [=starting a new parallel queue=]. This queue is to be used for all
file system operations.

<div algorithm>
To <dfn for="file entry/lock">take</dfn> a [=file entry/lock=] with a |value| of
"`exclusive`" or "`shared`" on a given [=file entry=] |file|:

1. Let |lock| be the |file|'s [=file entry/lock=].
1. Let |count| be the |file|'s [=file entry/shared lock count=].
To <dfn for="file system entry" id=file-entry-lock-take>take a lock</dfn> with a |value| of
"`exclusive`" or "`shared`" on a given [=/file system entry=] |entry|:

1. Let |lock| be the |entry|'s [=file system entry/lock=].
1. Let |count| be the |entry|'s [=file system entry/shared lock count=].
1. If |entry| is a [=directory entry=]:
1. Let |descendantLockStatus| be the result of [=file system entry/checking for a descendant lock=] on |entry|.
1. If |descendantLockStatus| is "`taken`":
1. Return "`failure`".
1. If |value| is "`exclusive`":
1. If |lock| is "`open`":
1. Set lock to "`taken-exclusive`".
Expand All @@ -159,11 +159,29 @@ Note: These steps have to be run on the [=file system queue=].
</div>

<div algorithm>
To <dfn for="file entry/lock">release</dfn> a [=file entry/lock=] on a given
[=file entry=] |file|:
To <dfn for="file system entry">check for a descendant lock</dfn> on a given
[=directory entry=] |directory|:

1. [=set/For each=] |child| of |directory|'s [=directory entry/children=]:
1. Let |lock| be the |child|'s [=file system entry/lock=].
1. If |lock| is not "`open`":
1. Return "`taken`".
1. If |child| is a [=directory entry=]:
1. Let |descendantLockStatus| be the result of [=file system entry/checking for a descendant lock=] on |child|.
1. If |descendantLockStatus| is "`taken`":
1. Return "`taken`".
1. Return "`open`".

1. Let |lock| be the |file|'s associated [=file entry/lock=].
1. Let |count| be the |file|'s [=file entry/shared lock count=].
Note: These steps have to be run on the [=file system queue=].

</div>

<div algorithm>
To <dfn for="file system entry/lock">release</dfn> a [=file system entry/lock=] on a given
[=/file system entry=] |entry|:

1. Let |lock| be the |entry|'s associated [=file system entry/lock=].
1. Let |count| be the |entry|'s [=file system entry/shared lock count=].
1. If |lock| is "`taken-shared`":
1. Decrease |count| by 1.
1. If |count| is 0, set |lock| to "`open`".
Expand All @@ -176,6 +194,10 @@ Note: These steps have to be run on the [=file system queue=].
Note: Locks help prevent concurrent modifications to a file. A {{FileSystemWritableFileStream}}
requires a shared lock, while a {{FileSystemSyncAccessHandle}} requires an exclusive one.

A <dfn export id=file>file entry</dfn> additionally consists of
<dfn for="file entry" export>binary data</dfn> (a [=byte sequence=]) and a
<dfn for="file entry">modification timestamp</dfn> (a number representing the number of milliseconds since the <a spec=FileAPI>Unix Epoch</a>).

A <dfn export id=directory>directory entry</dfn> additionally consists of a [=/set=] of
<dfn for="directory entry">children</dfn>, which are themselves [=/file system entries=].
Each member is either a [=/file entry=] or a [=/directory entry=].
Expand Down Expand Up @@ -538,7 +560,7 @@ The <dfn method for=FileSystemFileHandle>getFile()</dfn> method steps are:
the temporary file starts out empty,
otherwise the existing file is first copied to this temporary file.

Creating a {{FileSystemWritableFileStream}} [=file entry/lock/take|takes a shared lock=] on the
Creating a {{FileSystemWritableFileStream}} [=file system entry/take a lock|takes a shared lock=] on the
[=file entry=] [=locate an entry|locatable=] with |fileHandle|'s [=FileSystemHandle/locator=].
This prevents the creation of {{FileSystemSyncAccessHandle|FileSystemSyncAccessHandles}}
for the entry, until the stream is closed.
Expand Down Expand Up @@ -575,7 +597,7 @@ The <dfn method for=FileSystemFileHandle>createWritable(|options|)</dfn> method
|result| with a "{{NotFoundError}}" {{DOMException}} and abort these steps.
1. [=Assert=]: |entry| is a [=file entry=].

1. Let |lockResult| be the result of [=file entry/lock/take|taking a lock=]
1. Let |lockResult| be the result of [=file system entry/take a lock|taking a lock=]
with "`shared`" on |entry|.

1. [=Queue a storage task=] with |global| to run these steps:
Expand Down Expand Up @@ -603,7 +625,7 @@ The <dfn method for=FileSystemFileHandle>createWritable(|options|)</dfn> method
[=file entry=] [=locate an entry|locatable=] by |fileHandle|'s [=FileSystemHandle/locator=].
To ensure the changes are reflected in this file, the handle can be flushed.

Creating a {{FileSystemSyncAccessHandle}} [=file entry/lock/take|takes an exclusive lock=] on the
Creating a {{FileSystemSyncAccessHandle}} [=file system entry/take a lock|takes an exclusive lock=] on the
[=file entry=] [=locate an entry|locatable=] with |fileHandle|'s [=FileSystemHandle/locator=].
This prevents the creation of further {{FileSystemSyncAccessHandle|FileSystemSyncAccessHandles}}
or {{FileSystemWritableFileStream|FileSystemWritableFileStreams}}
Expand Down Expand Up @@ -645,7 +667,7 @@ The <dfn method for=FileSystemFileHandle>createSyncAccessHandle()</dfn> method s
|result| with a "{{NotFoundError}}" {{DOMException}} and abort these steps.
1. [=Assert=]: |entry| is a [=file entry=].

1. Let |lockResult| be the result of [=file entry/lock/take|taking a lock=]
1. Let |lockResult| be the result of [=file system entry/take a lock|taking a lock=]
with "`exclusive`" on |entry|.

1. [=Queue a storage task=] with |global| to run these steps:
Expand Down Expand Up @@ -1181,15 +1203,15 @@ given a [=file entry=] |file| in a [=/Realm=] |realm|:
file on disk being written to.

1. [=Enqueue the following steps=] to the [=file system queue=]:
1. [=file entry/lock/release|Release the lock=] on
1. [=file system entry/lock/release|Release the lock=] on
|stream|'s [=FileSystemWritableFileStream/[[file]]=].
1. [=Queue a storage task=] with |file|'s [=relevant global object=]
to [=/resolve=] |closeResult| with `undefined`.

1. Return |closeResult|.
1. Let |abortAlgorithm| be these steps:
1. [=enqueue steps|Enqueue this step=] to the [=file system queue=]:
1. [=file entry/lock/release|Release the lock=] on
1. [=file system entry/lock/release|Release the lock=] on
|stream|'s [=FileSystemWritableFileStream/[[file]]=].
1. Let |highWaterMark| be 1.
1. Let |sizeAlgorithm| be an algorithm that returns `1`.
Expand Down Expand Up @@ -1649,7 +1671,7 @@ The <dfn method for=FileSystemSyncAccessHandle>flush()</dfn> method steps are:
: |handle| . {{FileSystemSyncAccessHandle/close()}}
:: Closes the access handle or no-ops if the access handle is already closed.
This disables any further operations on it and
[=file entry/lock/release|releases the lock=] on the
[=file system entry/lock/release|releases the lock=] on the
[=FileSystemSyncAccessHandle/[[file]]=] associated with |handle|.
</div>

Expand All @@ -1661,7 +1683,7 @@ The <dfn method for=FileSystemSyncAccessHandle>close()</dfn> method steps are:
1. Set |lockReleased| to false.
1. Let |file| be [=this=]'s [=FileSystemSyncAccessHandle/[[file]]=].
1. [=Enqueue the following steps=] to the [=file system queue=]:
1. [=file entry/lock/release|Release the lock=] on |file|.
1. [=file system entry/lock/release|Release the lock=] on |file|.
1. Set |lockReleased| to true.
1. [=Pause=] until |lockReleased| is true.

Expand Down

0 comments on commit a0294b7

Please sign in to comment.