W. Trevor King [Tue, 14 Dec 2010 21:14:55 +0000 (16:14 -0500)]
Merge Veeco -> Bruker and NumPy/SciPy doc guide patches.
W. Trevor King [Tue, 14 Dec 2010 21:13:36 +0000 (16:13 -0500)]
Update URL for NumPy/SciPy documentation guide.
W. Trevor King [Fri, 19 Nov 2010 18:08:07 +0000 (13:08 -0500)]
Veeco -> Bruker since sale of metrology wing this August.
W. Trevor King [Sat, 13 Nov 2010 13:31:53 +0000 (08:31 -0500)]
Better contact point detection now handles some initial vibrations.
W. Trevor King [Thu, 11 Nov 2010 13:27:15 +0000 (08:27 -0500)]
Actually, filtering the copy directly is *almost* ok.
W. Trevor King [Thu, 11 Nov 2010 13:15:54 +0000 (08:15 -0500)]
Use hooke.command_by_name[...] in hooke.plugin.*.
W. Trevor King [Thu, 11 Nov 2010 13:13:09 +0000 (08:13 -0500)]
Only shallow copy when filtering a playlist.
We should iterate through the base playlist, because iterating over a
playlist that's having items removed is poorly defined. However, if
were iterating through the input playlist and removing items from the
copy, we need a shallow copy so the item ids still match. Curves can
belong to several playlists, so shallow copies are ok.
Another problem with the old version was that the _index was being
updated in the copy, not the current curve, so the
current_curve_callback() bit was *not* working.
W. Trevor King [Thu, 11 Nov 2010 12:11:38 +0000 (07:11 -0500)]
.`_setup_item` -> `._setup_item` in hooke.playlist.Playlist.unload.__doc__.
W. Trevor King [Wed, 10 Nov 2010 18:10:42 +0000 (13:10 -0500)]
Adjust playlist curve setup to load curves after YAML restore.
W. Trevor King [Wed, 10 Nov 2010 17:56:07 +0000 (12:56 -0500)]
Add yaml load test to hooke.curve.Data.
W. Trevor King [Wed, 10 Nov 2010 16:13:32 +0000 (11:13 -0500)]
expanduser() in hooke.playlist.load().
W. Trevor King [Tue, 9 Nov 2010 21:18:06 +0000 (16:18 -0500)]
Don't save digests in the playlist file.
They create dependency loops:
digest_{i+1} = checksum(file with digest_i)
With the old way, exiting after
$ hk.py -c 'new_playlist --output_playlist mylist'
-c 'glob_curves_to_playlist ...*.curve'
-c 'save_playlist --output mylist' -p
would complain of unsaved playlists.
W. Trevor King [Tue, 9 Nov 2010 20:24:47 +0000 (15:24 -0500)]
data_logger.normalize_logdir -> normalize_log_dir with data_logger v0.4.
W. Trevor King [Tue, 9 Nov 2010 20:22:31 +0000 (15:22 -0500)]
Update WTK driver links with git.php -> gitweb.cgi.
W. Trevor King [Tue, 9 Nov 2010 20:05:03 +0000 (15:05 -0500)]
Update test/wtk_driver.py for data_logger version 0.4.
W. Trevor King [Wed, 3 Nov 2010 19:18:18 +0000 (15:18 -0400)]
Don't use questions in the README section headings.
W. Trevor King [Wed, 3 Nov 2010 19:14:48 +0000 (15:14 -0400)]
Fixup reStructuredText in README.
W. Trevor King [Wed, 3 Nov 2010 18:41:49 +0000 (14:41 -0400)]
Explain my Hooke ebuild in doc/install.txt.
W. Trevor King [Wed, 3 Nov 2010 15:37:50 +0000 (11:37 -0400)]
Ran update_copyright.py, updating hooke/ui/gui/dialog/string.py.
W. Trevor King [Wed, 3 Nov 2010 15:36:52 +0000 (11:36 -0400)]
Fix hg.py -> hk.py in setup.py.
W. Trevor King [Thu, 28 Oct 2010 16:43:35 +0000 (12:43 -0400)]
Merged `igorbinarywave.py --not-strict` and playlist expanduser() commits.
W. Trevor King [Thu, 28 Oct 2010 16:39:29 +0000 (12:39 -0400)]
Add -n/--not-strict to igorbinarywave.py (currently for IBW files with non-empty padding).
W. Trevor King [Wed, 29 Sep 2010 16:41:23 +0000 (12:41 -0400)]
Move some playlist-related expanduser() calls from hooke.plugin.* to hooke.playlist and hooke.curve.
W. Trevor King [Wed, 29 Sep 2010 14:27:48 +0000 (10:27 -0400)]
Update hooke.playlist.FilePlaylist doctests for explicit_user_call.
W. Trevor King [Wed, 29 Sep 2010 14:25:40 +0000 (10:25 -0400)]
Update hooke.command_stack.CommandStack doctests for explicit_user_call.
W. Trevor King [Wed, 29 Sep 2010 14:17:07 +0000 (10:17 -0400)]
Cleanups so hooke/ui/gui/ passes unittests.
W. Trevor King [Wed, 29 Sep 2010 14:08:00 +0000 (10:08 -0400)]
Log errors when instantiation fails in hooke.util.pluggable.construct_*().
W. Trevor King [Wed, 29 Sep 2010 13:53:30 +0000 (09:53 -0400)]
Relax doctest constraints on hooke.util.fit.ModelFitter.
I'm not sure of the reason behind the differences, but sometimes the
exit condition varies slightly depending on which test box I use.
W. Trevor King [Wed, 29 Sep 2010 13:47:50 +0000 (09:47 -0400)]
Add os.path.expanduser() wrappers to user-supplied paths.
This allows things like
[gui user interface]
perspective path = ~/.hooke/resources/gui/perspective
to work as expected.
W. Trevor King [Wed, 29 Sep 2010 12:59:25 +0000 (08:59 -0400)]
Zip files use slashes (not os.path.sep) for their path separator (JPK driver).
W. Trevor King [Wed, 29 Sep 2010 11:42:59 +0000 (07:42 -0400)]
Quote backslash-containing arguments in doc/tutorial.py.
W. Trevor King [Wed, 15 Sep 2010 19:22:38 +0000 (15:22 -0400)]
albedomanu -> Manuel Gensler based on an email from the latter.
W. Trevor King [Thu, 9 Sep 2010 12:26:49 +0000 (08:26 -0400)]
Use `with Closing(...) as f:` in JPKDriver._zip_segment().
W. Trevor King [Thu, 9 Sep 2010 12:24:58 +0000 (08:24 -0400)]
Extended test/jpk_driver.py to test .jpk-data1D file reading.
W. Trevor King [Thu, 9 Sep 2010 11:37:03 +0000 (07:37 -0400)]
Extent JPK driver to also read .jpk-data1D-file files.
Also:
* Lowercase most JPK exception messages.
* Decode integer types to floats as specified by 'encoder'.
* Give the path to the old file when suggeesting an `out2jpk-force` upgrade.
W. Trevor King [Wed, 8 Sep 2010 20:22:09 +0000 (16:22 -0400)]
Fix test/jpk_driver.py now that we can load each of the examples.
W. Trevor King [Wed, 8 Sep 2010 20:21:39 +0000 (16:21 -0400)]
Ensure unique segment names in the JPK driver.
W. Trevor King [Wed, 8 Sep 2010 20:13:18 +0000 (16:13 -0400)]
Fix distance_base_unit extraction in JPK driver for encoded data.
W. Trevor King [Wed, 8 Sep 2010 20:09:23 +0000 (16:09 -0400)]
Fix JPK list field splitting for empty lists.
W. Trevor King [Wed, 8 Sep 2010 20:02:24 +0000 (16:02 -0400)]
Only pass curve params out to segments if they exist.
W. Trevor King [Wed, 8 Sep 2010 19:48:22 +0000 (15:48 -0400)]
Skip some param translation in the JPK driver for channel-less segments.
W. Trevor King [Wed, 8 Sep 2010 19:45:20 +0000 (15:45 -0400)]
Improve handling of the base calibration slot units in JPK driver.
W. Trevor King [Wed, 8 Sep 2010 17:59:08 +0000 (13:59 -0400)]
I'm using the 0.12 specs for the JPK driver, so allow version 0.12 files.
The previous range(12) code only went up to version 0.11.
W. Trevor King [Wed, 8 Sep 2010 17:57:56 +0000 (13:57 -0400)]
Adjust unit extraction from the JPK file format.
I've emailed Michael Haggerty to track down the version that made the
change from 'data.unit=' to 'data.unit.unit='. For now, I'm assuming
it's just <= 0.2.
W. Trevor King [Wed, 8 Sep 2010 17:35:23 +0000 (13:35 -0400)]
Add support for raster and constant data to JPK driver.
W. Trevor King [Wed, 8 Sep 2010 17:18:19 +0000 (13:18 -0400)]
Add support for {,u}int{16,32} segments in JPK driver.
TODO: check that these are converted to float32 or similar when
converting from bits to meters. Potentially change from multi-column
Data blocks to lists of single column blocks to allow for different
dtypes in each block.
W. Trevor King [Wed, 8 Sep 2010 14:58:45 +0000 (10:58 -0400)]
Break JPK data block name extraction out into ._zip_segment_name().
Also removes restrictions on allowed names.
W. Trevor King [Wed, 8 Sep 2010 14:54:40 +0000 (10:54 -0400)]
Extract 'z piezo sensitivity (m/V)' from JPK files into Curve.info.
Not particularly important, since we convert from V to m inside the
driver, but it's an interesting property for getting a feel for how a
particular instrument works.
W. Trevor King [Wed, 8 Sep 2010 14:51:08 +0000 (10:51 -0400)]
Accept version 0.2 (and similar?) JPK files with channels (vs. channels.list).
I've emailed Michael Haggerty for clarification on this issue, but
this will work until I hear back.
W. Trevor King [Wed, 8 Sep 2010 14:41:03 +0000 (10:41 -0400)]
Mark block level spring constant as a hack in the JPK driver.
W. Trevor King [Wed, 8 Sep 2010 14:39:57 +0000 (10:39 -0400)]
Don't assume you know which sJPK egments will be created.
W. Trevor King [Wed, 8 Sep 2010 14:38:10 +0000 (10:38 -0400)]
Allow JPK file versions up to 0.12 (unfinished).
W. Trevor King [Wed, 8 Sep 2010 14:33:56 +0000 (10:33 -0400)]
Add version 0.2 and 0.12 JPK file examples from Michael Haggerty.
Also adjust test/jpk_driver.py to attempt to load each file in the JPK
playlist. Note that this currently fails for some of the new files
because I haven't fixed the JPK driver yet.
W. Trevor King [Wed, 8 Sep 2010 14:30:28 +0000 (10:30 -0400)]
Move hooke.driver.igorbinarywave to hooke.util.igorbinarywave.
It's a utility for the MFP3D driver, not a driver in its own right.
W. Trevor King [Tue, 7 Sep 2010 19:07:02 +0000 (15:07 -0400)]
Update test/ to pass after
8f1ebe298b48.
W. Trevor King [Tue, 7 Sep 2010 18:43:11 +0000 (14:43 -0400)]
Don't specify curve or data block types. See doc/standards.txt.
W. Trevor King [Mon, 6 Sep 2010 00:26:03 +0000 (20:26 -0400)]
Add information about JPKForceSpec.txt to hooke.driver.jpk.__doc__.
Update this if you adjust the driver based on a newer spec, or if JPK
changes their licensing or contact email address.
W. Trevor King [Mon, 6 Sep 2010 00:02:09 +0000 (20:02 -0400)]
Improve error message for old (ASCII) JPK file formats.
Michael Haggerty gave me the name of JPK's upgrade script so we can
point users in the right direction.
W. Trevor King [Thu, 2 Sep 2010 19:26:38 +0000 (15:26 -0400)]
Remove 'results' panel from GUI. Use 'block info' instead.
W. Trevor King [Thu, 2 Sep 2010 19:23:07 +0000 (15:23 -0400)]
Added --config option to bin/hk.py to override built in config file path.
W. Trevor King [Thu, 2 Sep 2010 14:19:35 +0000 (10:19 -0400)]
Remove hooke/plugin/export.py.
You can achieve the similar effects with
do_fits -> 'block info'
do_force_curve -> 'export block'
do_notes -> 'save playlist'
do_overlay -> 'export block' or future 'superimpose' + 'export block'
do_results -> 'block info' or 'export block'
W. Trevor King [Thu, 2 Sep 2010 14:12:59 +0000 (10:12 -0400)]
Remove hooke/plugin/results.py.
I'm not sure what it was for in the first place. Probably not a great
reason to remove it ;), but if it turns out to have been important, we
can always bring it back later.
W. Trevor King [Thu, 2 Sep 2010 14:10:18 +0000 (10:10 -0400)]
Remove hooke/plugin/procplots.py.
Nothing substantially different from hooke/plugin/plotmanip.py.
W. Trevor King [Thu, 2 Sep 2010 14:06:54 +0000 (10:06 -0400)]
Remove hooke/plugin/generalvclamp.py.
The only functions without analogs in the current framework are
already listed at the OldCruft class at the end of
hooke/plugin/curve.py.
W. Trevor King [Thu, 2 Sep 2010 13:58:14 +0000 (09:58 -0400)]
Remove hooke/plugin/flatfilts*
Their functionality is now distributed among
hooke/plugin/flatfilt.py
hooke/plugin/convfilt.py
hooke/util/peak.py
W. Trevor King [Thu, 2 Sep 2010 10:37:55 +0000 (06:37 -0400)]
Update test/ with new 'block names' in 'curve info' (from
3371b5652dfe).
W. Trevor King [Thu, 2 Sep 2010 10:26:00 +0000 (06:26 -0400)]
Add 'block names' argument to 'curve info'.
Many commands key off the block name, so there should be a way for the
user to determine what the block names are.
W. Trevor King [Thu, 2 Sep 2010 10:23:05 +0000 (06:23 -0400)]
Add list of possible keys to 'block info' error for key discovery.
W. Trevor King [Thu, 2 Sep 2010 02:37:17 +0000 (22:37 -0400)]
Updated test/ to match new pformated command line output.
W. Trevor King [Thu, 2 Sep 2010 02:22:03 +0000 (22:22 -0400)]
pprint.pformat output dicts in ui.commandline.
Also more str() -> unicode() (see
2f57e930cd52).
W. Trevor King [Thu, 2 Sep 2010 02:06:59 +0000 (22:06 -0400)]
Added 'block info' to retrieve and export portions of Curve.data[*].info.
W. Trevor King [Wed, 1 Sep 2010 22:28:13 +0000 (18:28 -0400)]
Note that unloaded curves will not have any Data or DataInfo portion saved in playlist files.
W. Trevor King [Wed, 1 Sep 2010 20:31:08 +0000 (16:31 -0400)]
Ran update_copyright.py.
W. Trevor King [Wed, 1 Sep 2010 14:01:35 +0000 (10:01 -0400)]
Add 'hg update wtk' to Mercurial installation directions.
W. Trevor King [Tue, 31 Aug 2010 12:44:49 +0000 (08:44 -0400)]
Check data copying in Curve.__copy__ and .__deepcopy__ doctests.
W. Trevor King [Mon, 30 Aug 2010 17:56:49 +0000 (13:56 -0400)]
Store a copy of params in CurveCommand._add_to_command_stack.
We may modify the parameter dict later during the command's execution,
and we want future runs via the command stack to replicate the first
run from the raw initial parameters. If you don't do this, you'll get
errors like
$ ./bin/hk.py
-c 'load_playlist test/data/vclamp_picoforce/playlist'
-c 'start_command_capture'
-c 'zero_surface_contact_point --block approach'
-c 'zero_surface_contact_point --block retract'
-c 'stop_command_capture'
-c 'apply_command_stack_to_playlist --enable-evaluate'
-c 'flat_filter_playlist --block retract --min_points 1'
Success
...
Failure
Expected a list of Peaks, not <class 'hooke.command.Failure'>:
surface distance (m) not in retract (['z piezo (m)', 'deflection (m)',
'surface distance (m) (m)', 'surface deflection (m) (m)']):...
where the units '(m)' were appended twice, once during the initial
command run, and again during the reload's command stack exectution.
W. Trevor King [Mon, 30 Aug 2010 17:22:27 +0000 (13:22 -0400)]
Maintain Curve._hooke setting during copy and deepcopy.
W. Trevor King [Sun, 29 Aug 2010 00:47:43 +0000 (20:47 -0400)]
Flesh out command line help messages.
W. Trevor King [Sat, 28 Aug 2010 23:35:49 +0000 (19:35 -0400)]
.__setup_params() -> ._setup_params() in hooke/plugin/*.py.
Methods starting with '__' get special name mangling.
http://docs.python.org/reference/expressions.html#atom-identifiers
We don't want any of that.
W. Trevor King [Sat, 28 Aug 2010 22:58:38 +0000 (18:58 -0400)]
Don't distinguish between ASCII unicode and strings in YAML output.
W. Trevor King [Sat, 28 Aug 2010 12:42:19 +0000 (08:42 -0400)]
Document fit parameter extraction via PyYAML.
W. Trevor King [Sat, 28 Aug 2010 11:38:47 +0000 (07:38 -0400)]
Document --ui in doc/gui.txt.
W. Trevor King [Sat, 28 Aug 2010 11:34:27 +0000 (07:34 -0400)]
Added convenience --ui option to hooke.hooke.main().
W. Trevor King [Fri, 27 Aug 2010 23:00:21 +0000 (19:00 -0400)]
Trivial fixes to test/command_stack.py. Now test/ passes again.
W. Trevor King [Fri, 27 Aug 2010 22:44:01 +0000 (18:44 -0400)]
Don't load curves becore clearing their command stack.
Loading a curve (which would happen if it had been unloaded before
being summoned by current_curve_callback) executes its command stack.
Since 'clear curve command stack' is only trying to remove the stack,
that stack execution is a waste of time. With the unloaded callback,
we can remove the command stack without ever loading the curve.
W. Trevor King [Fri, 27 Aug 2010 22:39:44 +0000 (18:39 -0400)]
Fix 'flat filter peaks' CommandMessage declaration in test/apply_command_stack_to_playlist.py
W. Trevor King [Fri, 27 Aug 2010 21:36:35 +0000 (17:36 -0400)]
Don't fail if Playlist.unload() is called on an unloaded Curve.
W. Trevor King [Fri, 27 Aug 2010 21:31:39 +0000 (17:31 -0400)]
Double initial guess for the surface kink position.
The previous algorithm would fail for curves if
1) They didn't have many tension features to stretch out the range.
2) The contact slope "softened" a bit in the kink region.
For example, try it on most of the test/vclamp_picoforce data ;).
The new algorithm works perfectly on the ideal, boring curve:
_______
/
/
W. Trevor King [Fri, 27 Aug 2010 20:14:25 +0000 (16:14 -0400)]
Keep Playlist._loaded up to date on Curve.unload() calls.
W. Trevor King [Fri, 27 Aug 2010 19:46:21 +0000 (15:46 -0400)]
Add 'command stack' to list of ignored commands in propertyeditor.py
W. Trevor King [Fri, 27 Aug 2010 19:25:01 +0000 (15:25 -0400)]
Added logging to hooke.util.callback's callback routines.
W. Trevor King [Fri, 27 Aug 2010 19:20:22 +0000 (15:20 -0400)]
Oops, the CommandMessage.explicit_user_call, not Command.explicit_user_call.
Fixes bug from
daf1c9f9c8e1.
W. Trevor King [Fri, 27 Aug 2010 15:29:30 +0000 (11:29 -0400)]
Added hooke.util.caller.caller_names() for minimal stack traces.
W. Trevor King [Fri, 27 Aug 2010 11:03:36 +0000 (07:03 -0400)]
Added CommandStack.explicit_user_call to distinguish user- from UI-generated calls.
W. Trevor King [Tue, 24 Aug 2010 11:32:35 +0000 (07:32 -0400)]
More robust guess for `zero surface contact point` + fit check arguments
W. Trevor King [Mon, 23 Aug 2010 15:46:26 +0000 (11:46 -0400)]
Add PyYAML to doc/install.txt and doc/hacking.txt dependency lists.
W. Trevor King [Mon, 23 Aug 2010 14:29:14 +0000 (10:29 -0400)]
Extend JPK v0.5 parsing back to 0.3 and 0.4 file formats.
On Mon, Aug 23, 2010 at 02:15:08PM +0200, Michael Haggerty wrote:
> The fundamentals of our file format has not changed much since it was
> introduced. Most of the changes involve metadata, such as how we
> describe force spectroscopy scanning patterns, that are not even
> described in the document that I sent you.
W. Trevor King [Mon, 23 Aug 2010 13:19:51 +0000 (09:19 -0400)]
Remove ._on_notebook_page_close binding for the removed notebook GUI panel.
W. Trevor King [Mon, 23 Aug 2010 13:14:30 +0000 (09:14 -0400)]
Restore HookeFrame.select_plugin (was removed in
29866287a21e).
W. Trevor King [Sun, 22 Aug 2010 19:26:07 +0000 (15:26 -0400)]
Cleaned non-folders-panel cruft methods out of hooke.ui.gui.HookeFrame