Setting sim/replay/log-raw-speed logs raw speed to
/sim/replay/log-raw-speed-values/value[].
'scripts/python/recordreplay.py --test-motion' uses this to test for a bug
where the user aircraft simtime and signals information don't quite match in
continuous recordings.
Recording of extra properties is only supported in Continuous recordings.
Modified Continuous recording format to allow future forwards
compatibility. See docs-mini/README-recordings.md for details. This breaks
compatibility with previously-generated Continuous recordings, but this only
affects next.
To reduce overhead we record all extra property values in the first frame and
then later frames contain only extra property changes. When replaying, if the
user jumps backwards we replay all extra property changes since the start of
the recording. Similarly if the user jumps forwards, we replay any intervening
extra property changes.
Recording extra properties:
This is enabled by:
/sim/replay/record-extra-properties
The extra properties that are recorded are identified by the property
paths in the values of /sim/replay/record-extra-properties-paths/path[]
properties. We record the entire tree for each specified path.
Recording of main window position size:
We have specific support for record and replay of main window position/size.
This is enabled by:
/sim/replay/record-main-window
Recording of main window view:
We have specific support for recording the view type and direction/zoom
settings.
This is enabled by:
/sim/replay/record-main-view
We record the /sim/current-view/ property tree, excluding some subtrees
that continuously vary but are not required for replaying of the view.
When replaying, we allow separate control of what extra property changes are
replayed, with:
/sim/replay/replay-extra-property-changes
/sim/replay/replay-extra-property-removal
/sim/replay/replay-main-window-position
/sim/replay/replay-main-window-size
/sim/replay/replay-main-view
We work around some problems caused by the use of tied properties when
replaying changes to view-number.
Window position issue:
When replaying window position and size changes, things get a little tricky
because osgViewer::GraphicsWindow::setWindowRectangle() takes a position
for the entire window, but osgGA::GUIEventAdapter::RESIZE events contain
the position of the interior of the window; for example the y values will
differ by the height of the window's titlebar. This can cause windows to
move progressively further down each time they are positioned or resized.
There doesn't seem to be a way of finding the size of a window's
furniture directly. So instead this commit adds a new method
osgGA::GUIEventAdapter::setWindowRectangleInteriorWithCorrection() which
wraps osgViewer::GraphicsWindow::setWindowRectangle(). We listen for the
following osgGA::GUIEventAdapter::RESIZE event and calculate corrections
for x and y position that are used for subsequent calls.
docs-mini/README-recordings.md:
Updated to document new Continuous format.
scripts/python/recordreplay.py:
New script to test various aspects of record/replay.
Other:
We now create convenience softlink to most recent continuous recording, using
SGPath::makeLink(). Note that SGPath::makeLink() currently does nothing on
Windows.
Changed format of Continuous recordings to contain a single property tree
in header. This is much simpler than having separate Config and Meta trees.
- Instantiate an HTTPSConnection object when the URL scheme is 'https'.
- Clarify and simplify the initialization of HTTPSocketRequest and
HTTPDownloadRequest:
+ clarify initialization of their 'callback' attribute (it's the
method of the same name; make it clear that the base class
constructor, namely HTTPGetCallback.__init__(), doesn't modify the
'callback' attribute when an object of class HTTPSocketRequest or
HTTPDownloadRequest is initialized);
+ HTTPDownloadRequest doesn't need access to the TerraSync object
-> remove the corresponding instance attribute and constructor
argument.
- Don't use super() when initializing HTTPDownloadRequest objects
(see [1]).
[1] https://fuhm.net/super-harmful/
- Refuse to recursively delete a directory that does not contain a
.dirindex file. This will protect users against data loss in case they
inadvertently use the --remove-orphan option with the wrong target
directory.
- Correctly handle the case where we have a file on disk that is now
listed as a directory on the server: remove the file if we are in
'sync' mode, so that the directory can be created and sync'ed from the
server.
- only accept ASCII-encoded .dirindex files (this is guaranteed to work
fine "everywhere");
- reject .dirindex files with a 'path' entry that contains a backslash
or starts with a slash;
- reject .dirindex files with a 'path' entry that contains a '..'
component;
- reject .dirindex files with an 'f', 'd' or 't' entry whose name field
contains a slash or a backslash;
- reject .dirindex files with an 'f', 'd' or 't' entry whose name field
is '..';
- add comment lines (starting with '#') in the sample good .dirindex
file used by unit tests.
Remove the 'if __name__ == "__main__": unittest.main()'. Indeed, the
module can't be run this way due to its imports. Tests from this module
can be run with:
cd scripts/python/TerraSync
python3 -m unittest tests.test_virtual_path
In Python, common usage is not to define accessors, but to directly use
class or instance attributes (especially when the associated data is
constant after instance creation). If it later happens that a given
attribute needs getter or setter logic, this can always be done via the
@property decorator, and doesn't affect calling code at all. See for
instance:
https://docs.python.org/3/library/functions.html#propertyhttps://mail.python.org/pipermail/tutor/2012-December/thread.html#92990
Apply this to the DirIndex class and rename the following attributes for
better readability: f -> files, d -> directories, t -> tarballs.
The tests can be run from directory 'scripts/python/TerraSync' using:
python3 -m unittest tests.test_dirindex
(or just 'python3 -m unittest' to run all tests pertaining to
terrasync.py).
python3 has a default implementation for __ne__ when __eq__ is defined. The opposite is not true -- having only __ne__ does not have a default __eq__ implementation.
Also note that there are cases when eq/ne will both be True or both False, therefore, developers are encouraged to explicitly define these methods in pairs.
During the recent SourceForge migration, the TerraSync server hosted
there used to send .dirindex files with the following contents:
Project web is currently offline pending the final migration of its
data to our new datacenter.
Make sure terrasync.py aborts with a user-understandable error in such a
case.
Add more functions and properties to VirtualPath that directly
correspond to functions and properties of pathlib.PurePath, except that
types are adapted of course, and that for API consistency, VirtualPath
methods use mixedCaseStyle whereas those of pathlib.PurePath use
underscore_style.
Due to some misleading 'url' variable name, network error messages used
to contain things such as '/scenery/Airports/N/E/4/.dirindex' (i.e., the
path on the server) instead of the full URL. For the same reason, the
callback function of HTTPGetCallback objects was passed this
path-on-server instead of the URL. This should all be fixed now.
Option --only-subdir allows one to restrict terrasync.py processing[1]
to a chosen subdirectory of the TerraSync repository. Example:
terrasync.py --target=/your/TerraSync/repo --only-subdir="Airports/L/F/P"
[1] This works in both 'check' and 'sync' modes.
Add classes VirtualPath and MutableVirtualPath (the latter derived from
the former) to manipulate slash-separated paths where the root '/'
represents the TerraScenery root. This makes it clear what a function
expects when you see that one of its arguments is a VirtualPath
instance: you don't have to ask yourself whether it can start or end
with a slash, how to interpret it, etc. Operating on these paths is also
easy[1], be it to assemble URLs in order to retrieve files or to join
their relative part with a local directory path in order to obtain a
real (deeper) local path.
VirtualPath and MutableVirtualPath are essentially the same; the former
is hashable and therefore has to be immutable, whereas the latter can be
modified in-place with the /= operator (used to append path components),
and therefore can't be hashable. As a consequence, MutableVirtualPath
instances can't be used as dictionary keys, elements of a set or
frozenset, etc.
VirtualPath and MutableVirtualPath use the pathlib.PurePath API where
applicable (part of this API has been implemented in
[Mutable]VirtualPath; more can be added, of course). These classes have
no assumptions related to TerraSync and thus should be fit for use in
other projects.
To convert a [Mutable]VirtualPath instance to a string, just use str()
on it. The result is guaranteed to start with a '/' and not to end with
a '/', except for the virtual root '/'. Upon construction, the given
string is interpreted relatively to the virtual root, i.e.:
VirtualPath("") == VirtualPath("/")
VirtualPath("abc/def/ghi") == VirtualPath("/abc/def/ghi")
etc.
VirtualPath and MutableVirtualPath instances sort like the respective
strings str() converts them too. The __hash__() method of VirtualPath is
based on the type and this string representation, too. Such objects can
only compare equal (using ==) if they have the same type. If you want to
compare the underlying virtual paths inside a VirtualPath and a
MutableVirtualPath, use the samePath() method of either class.
For more info, see scripts/python/TerraSync/terrasync/virtual_path.py
and unit tests in scripts/python/TerraSync/tests/test_virtual_path.py.
[1] Most useful is the / operator, which works as for SGPath:
VirtualPath("/abc/def/ghi") == VirtualPath("/abc") / "def" / "ghi"
VirtualPath("/abc/def/ghi") == VirtualPath("/abc") / "def/ghi"
- New directory scripts/python/TerraSync/terrasync.
- Move scripts/python/terrasync.py to
scripts/python/TerraSync/terrasync/main.py (main module in the new
structure).
- Add empty __init__.py file to scripts/python/TerraSync/terrasync/ to
make this directory a Python package.
- Wrap the main code from previous terrasync.py in a main() function of
the terrasync.main module. Also move command-line arguments parsing to
a separate parseCommandLine() function.
- Add an executable script scripts/python/TerraSync/terrasync.py for end
users, that just calls terrasync.main.main().
For end users, the only difference is that they now have to use
scripts/python/TerraSync/terrasync.py instead of
scripts/python/terrasync.py (which doesn't exist anymore, since all this
lives under the scripts/python/TerraSync directory from now on).
This structure will allow to cleanly split the code into modules and to
add unit tests.
Using os.path.abspath() here in TerraSync.setTarget() adds a safety
layer in case the process later calls os.chdir() or similar[1], which
would change the meaning of the "." directory. Also remove the strip()
call which I don't consider useful here, see my message at:
https://sourceforge.net/p/flightgear/mailman/message/36208140/
[1] Not the case currently, but who knows what will happen in the
future...
You may now call terrasync.py with --mode=sync or --mode=check. 'sync'
mode is the default and corresponds to terrasync.py's usual behavior.
In 'check' mode, terrasync.py never writes to disk and aborts at the
first mismatch between local and remote data. The exit status in 'check'
mode is:
- 0 if the program terminated successfully and no mismatch was found
between the local and remote repositories;
- 1 in case an error was encountered;
- 2 if there was a mismatch between local and remote data.
In 'sync' mode, the exit status is:
- 0 if the program terminated successfully;
- 1 in case an error was encountered.
A mismatch in 'check' mode is *not* an error, it is just one of the two
expected results. An error is a worse condition (uncaught exception,
network retrieval aborted after retrying failed, stuff like that).
Additionally, calling terrasync.py with --report causes it to print
lists of:
- files and dirs that were missing or had mismatching hashes (this is
okay in 'sync' mode: these things have been "fixed" in the target
directory before the report was printed);
- files and dirs that have been found to be orphaned (i.e., found
under the target directory but not mentioned in the corresponding
.dirindex file). These are the ones removed in 'sync' mode when
--remove-orphan is passed.
- Add computeHash() utility function that can work with any file-like
object (e.g., a connected socket).
- Rename hash_of_file() to hashForFile(), and of course implement it
using our new computeHash().
- Add class HTTPSocketRequest derived from HTTPGetCallback. It allows
one to process data from the network without storing it to a file (it
uses the file-like interface provided by http.client.HTTPResponse).
The callback returns the http.client.HTTPResponse object, which can be
conveniently used in a 'with' statement.
- Simplify the API of TerraSync.updateDirectory(): its 'dirIndexHash'
argument must now be a hash (a string); the None object is not allowed
anymore (with the soon-to-come addition of --mode=check, having to
deal with this special case in updateDirectory() would make the logic
too difficult to follow, or we would have to really completely
separate check-only mode from update mode, which would entail code
duplication).
Since TerraSync.updateDirectory() must now always have a hash to work
with, compute the hash of the root '.dirindex' file from the server in
TerraSync.start(), using our new HTTPSocketRequest class---which was
written for this purpose, since that will have to work in check-only
mode (but not only), where we don't want to write any file to disk.
- TerraSync.updateFile(): correctly handle the case where a directory
inside the TerraSync repository is (now) a file according to the
server: the directory must be recursively removed before the file can
be downloaded in the place formerly occupied by the directory.
- Add stub class Report. Its methods do nothing for now, but are already
called in a couple of appropriate places. The class will be completed
in a future commit, of course.
The goal of removeDirectoryTree() is to provide a safety net around
recursive directory removal with shutil.rmtree(), in order to prevent
user or bug-caused catastrophic events such as /, /home /home/joeuser or
C:\ being recursively erased.
- Add method assembleUrl() to HTTPGetter.
- Raise a NetworkError exception with the particular URL and number of
retries when it has been exhausted.
- Number of retries is now trivial to expose as a parameter, and set to
5 in HTTPGetter.
- Sleep for one second between self.httpConnection.close() and
self.httpConnection.connect() when retrying a failed HTTP request.
- Apply DRY principle.
- New generic exception class TerraSyncPyException.
- Add subclass NetworkError of TerraSyncPyException.
- Raise a NetworkError exception when the HTTP return code is not 200.
- hash_of_file() does not silently ignore errors anymore; exceptions
should be dealt with wherever appropriate by the callers.
Whenever hash_of_file() returns, its return value is now the SHA-1
hash of the specified file. This is less error-prone IMHO than
returning None. Otherwise, calling code could erroneously conclude
that there is a matching hash when the file to check is actually
missing. For a concrete example, see the 'dirIndexHash' parameter of
TerraSync.updateDirectory(), which so far is used precisely with the
value None to express that "we are just starting the recursion and
have no hash from the server to compare to".
When called, the callback passed to HTTPGetter.get() is now explicitly
passed the URL and the http.client.HTTPResponse instance.
Remove the HTTPGetCallback.result attribute (not needed anymore, leaves
more freedom when implementating HTTPGetCallback subclasses...).
