message
stringlengths 13
484
| diff
stringlengths 38
4.63k
|
|---|---|
Disable SignUp
* Disable SignUp
Disabled signup from setting only hides signup link but users can signup anyway. Disables the sign_up function.
* Fix translate String | @@ -13,6 +13,7 @@ import frappe.permissions
import frappe.share
import re
from frappe.limits import get_limits
+from frappe.website.utils import is_signup_enabled
STANDARD_USERS = ("Guest", "Administrator")
@@ -602,6 +603,9 @@ def verify_password(password):
@frappe.whitelist(allow_guest=True)
def sign_up(email, full_name, redirect_to):
+ if not is_signup_enabled():
+ frappe.throw(_('Sign Up is disabled'), title='Not Allowed')
+
user = frappe.db.get("User", {"email": email})
if user:
if user.disabled:
|
Fix backwards compatibility for pip VcsSupport
VcsSupport classes need to be instantiated on pip<19.2
This provides backward compatibility | @@ -61,15 +61,17 @@ class VCSRepository(object):
def obtain(self):
# type: () -> None
- lte_pip_19 = (
- pip_shims.parsed_pip_version.parsed_version < pip_shims.parse_version("19.0")
+ lt_pip_19_2 = (
+ pip_shims.parsed_pip_version.parsed_version < pip_shims.parse_version("19.2")
)
+ if lt_pip_19_2:
+ self.repo_backend = self.repo_backend(self.url)
if os.path.exists(
self.checkout_directory
) and not self.repo_backend.is_repository_directory(self.checkout_directory):
self.repo_backend.unpack(self.checkout_directory)
elif not os.path.exists(self.checkout_directory):
- if lte_pip_19:
+ if lt_pip_19_2:
self.repo_backend.obtain(self.checkout_directory)
else:
self.repo_backend.obtain(self.checkout_directory, self.parsed_url)
|
Update NZBGetPostProcess.py
fail more gracefully | @@ -236,14 +236,14 @@ if 'NZBOP_SCRIPTDIR' in os.environ and not os.environ['NZBOP_VERSION'][0:5] < '1
if success:
sys.exit(POSTPROCESS_SUCCESS)
else:
- sys.exit(POSTPROCESS_ERROR)
+ sys.exit(POSTPROCESS_NONE)
elif (radarrcat.startswith(category)):
#DEBUG#print "Radarr Processing Activated"
success = radarr.processMovie(path, settings, True, pathMapping=path_mapping)
if success:
sys.exit(POSTPROCESS_SUCCESS)
else:
- sys.exit(POSTPROCESS_ERROR)
+ sys.exit(POSTPROCESS_NONE)
elif (sickragecat.startswith(category)):
#DEBUG#print "Sickrage Processing Activated"
autoProcessTVSR.processEpisode(path, settings, nzb, pathMapping=path_mapping)
|
[cleanup] Deprecate pagegenerators.UnconnectedPageGenerator
Use Site.unconnected_pages() instead | @@ -950,8 +950,7 @@ class GeneratorFactory(object):
total=intNone(value),
site=self.site)
elif arg == '-unconnectedpages':
- gen = UnconnectedPageGenerator(total=intNone(value),
- site=self.site)
+ gen = self.site.unconnected_pages(total=intNone(value))
elif arg == '-imagesused':
if not value:
value = pywikibot.input(
@@ -1309,6 +1308,7 @@ def RecentChangesPageGenerator(start=None, end=None, reverse=False,
return gen
+@deprecated('site.unconnected_pages()')
@deprecated_args(step=None)
def UnconnectedPageGenerator(site=None, total=None):
"""
@@ -1323,8 +1323,7 @@ def UnconnectedPageGenerator(site=None, total=None):
site = pywikibot.Site()
if not site.data_repository():
raise ValueError('The given site does not have Wikibase repository.')
- for page in site.unconnected_pages(total=total):
- yield page
+ return site.unconnected_pages(total=total)
@deprecated_args(referredImagePage='referredFilePage', step=None)
|
[varLib] Allow sparse masters in HVAR
Part of
Part of | @@ -420,42 +420,46 @@ def _merge_TTHinting(font, masterModel, master_ttfs, tolerance=0.5):
var = TupleVariation(support, delta)
cvar.variations.append(var)
-def _add_HVAR(font, model, master_ttfs, axisTags):
+def _add_HVAR(font, masterModel, master_ttfs, axisTags):
log.info("Generating HVAR")
- hAdvanceDeltas = {}
+ glyphOrder = font.getGlyphOrder()
+
+ hAdvanceDeltasAndSupports = {}
metricses = [m["hmtx"].metrics for m in master_ttfs]
- for glyph in font.getGlyphOrder():
- hAdvances = [metrics[glyph][0] for metrics in metricses]
+ for glyph in glyphOrder:
+ hAdvances = [metrics[glyph][0] if glyph in metrics else None for metrics in metricses]
# TODO move round somewhere else?
- hAdvanceDeltas[glyph] = tuple(otRound(d) for d in model.getDeltas(hAdvances)[1:])
+ hAdvanceDeltasAndSupports[glyph] = masterModel.getDeltasAndSupports(hAdvances)
- # Direct mapping
- supports = model.supports[1:]
+ singleModel = models.allSame(id(v[1]) for v in hAdvanceDeltasAndSupports.values())
+
+ directStore = None
+ if singleModel:
+ # Build direct mapping
+ supports = hAdvanceDeltasAndSupports[0][1][1:]
varTupleList = builder.buildVarRegionList(supports, axisTags)
varTupleIndexes = list(range(len(supports)))
- n = len(supports)
- items = []
- for glyphName in font.getGlyphOrder():
- items.append(hAdvanceDeltas[glyphName])
-
- # Build indirect mapping to save on duplicates, compare both sizes
- uniq = list(set(items))
- mapper = {v:i for i,v in enumerate(uniq)}
- mapping = [mapper[item] for item in items]
- advanceMapping = builder.buildVarIdxMap(mapping, font.getGlyphOrder())
-
- # Direct
- varData = builder.buildVarData(varTupleIndexes, items)
+ varData = builder.buildVarData(varTupleIndexes, [])
+ for glyphName in glyphOrder:
+ varData.add_item(hAdvanceDeltasAndSupports[glyphName][0])
directStore = builder.buildVarStore(varTupleList, [varData])
- # Indirect
- varData = builder.buildVarData(varTupleIndexes, uniq)
- indirectStore = builder.buildVarStore(varTupleList, [varData])
- mapping = indirectStore.optimize()
- advanceMapping.mapping = {k:mapping[v] for k,v in advanceMapping.mapping.items()}
-
+ # Build optimized indirect mapping
+ storeBuilder = varStore.OnlineVarStoreBuilder(axisTags)
+ mapping = {}
+ for glyphName in glyphOrder:
+ deltas,supports = hAdvanceDeltasAndSupports[glyphName]
+ storeBuilder.setSupports(supports)
+ mapping[glyphName] = storeBuilder.storeDeltas(deltas)
+ indirectStore = storeBuilder.finish()
+ mapping2 = indirectStore.optimize()
+ mapping = [mapping2[mapping[g]] for g in glyphOrder]
+ advanceMapping = builder.buildVarIdxMap(mapping, glyphOrder)
+
+ use_direct = False
+ if directStore:
# Compile both, see which is more compact
writer = OTTableWriter()
|
(fix) updated poll interval lengths
updated SHORT_POLL_INTERVAL and LONG_POLL_INTERVAL both to 1.0 so
_status_poll_loop can actively query acc balances. | @@ -54,10 +54,10 @@ class BitmartExchange(ExchangeBase):
trading functionality.
"""
API_CALL_TIMEOUT = 10.0
- SHORT_POLL_INTERVAL = 5.0
+ SHORT_POLL_INTERVAL = 1.0
UPDATE_ORDER_STATUS_MIN_INTERVAL = 10.0
UPDATE_TRADE_STATUS_MIN_INTERVAL = 10.0
- LONG_POLL_INTERVAL = 120.0
+ LONG_POLL_INTERVAL = 1.0
@classmethod
def logger(cls) -> HummingbotLogger:
|
TST: updated repr test for files
Updated the repr test for files. | @@ -172,23 +172,23 @@ class TestBasics():
def setup(self):
"""Runs before every method to create a clean testing setup."""
self.out = ''
+
# Use a two-year as default. Some tests will use custom ranges.
self.start = dt.datetime(2008, 1, 1)
self.stop = dt.datetime(2009, 12, 31)
- # store current pysat directory
+ # Store current pysat directory
self.data_paths = pysat.params['data_dirs']
- # create temporary directory
+ # Create temporary directory
self.tempdir = tempfile.TemporaryDirectory()
pysat.params['data_dirs'] = [self.tempdir.name]
- self.testInst = \
- pysat.Instrument(inst_module=pysat.instruments.pysat_testing,
- clean_level='clean',
- temporary_file_list=self.temporary_file_list,
- update_files=True)
- # create instrument directories in tempdir
+ self.testInst = pysat.Instrument(
+ inst_module=pysat.instruments.pysat_testing, clean_level='clean',
+ temporary_file_list=self.temporary_file_list, update_files=True)
+
+ # Create instrument directories in tempdir
create_dir(self.testInst)
def teardown(self):
@@ -201,7 +201,7 @@ class TestBasics():
"""The repr output will match the str output"""
self.out = self.testInst.files.__repr__()
assert isinstance(self.out, str)
- assert self.out.find("Local files") > 0
+ assert self.out.find("pysat.Files(") >= 0
def test_basic_str(self):
"""Check for lines from each decision point in str"""
|
[Docs] Changed output of model server
When I run the model server the output is [value] not {"predictions": [value]} - perhaps this is something I have done but if not the docs should probably reflect this. I haven't tested the R version so that may need updating too. | @@ -306,7 +306,7 @@ in MLflow saved the model as an artifact within the run.
the server should respond with output similar to::
- {"predictions": [6.379428821398614]}
+ [6.379428821398614]
.. container:: R
|
Tests: Run compile library test in a per version directory.
* This is to avoid clashes between different versions run at the
same time and still easily find them. | @@ -109,12 +109,11 @@ def action(stage_dir, root, path):
)
)
-
+from nuitka.PythonVersions import python_version
compileLibraryTest(
search_mode = search_mode,
- stage_dir = os.path.join(tmp_dir, "compile_library"),
+ stage_dir = os.path.join(tmp_dir, "compile_library_%s" % python_version ),
decide = decide,
action = action
)
-
|
Developer Manual: Added description of how context managers are named.
* This is a very useful coding rule in my mind, as it makes
it really easy to recognize them. | @@ -293,6 +293,21 @@ There is no code in packages themselves. For programs, we use
Names of modules should be plurals if they contain classes. Example is
that a ``Nodes`` module that contains a ``Node`` class.
+Names for context manages start with ``with``
+=============================================
+
+In order to easily recognize that something is to be used as a context
+manager, we follow a pattern of naming them ``withSomething``, to make
+that easily recognized.
+
+.. code:: python
+
+ with withEnvironmentPathAdded(os.path.join(sys.prefix, "bin")):
+ with withDirectoryChange(self.qt_datadir):
+ ...
+
+This makes these easy to recognize even in their definition.
+
Prefer list contractions over built-ins
=======================================
|
TST: added unit tests
Added unit tests to cover newly uncovered lines. | @@ -271,6 +271,28 @@ class TestConstellationBasics(object):
del self.testConst, self.load_date, self.custom_args
return
+ @pytest.mark.parametrize("apply_inst", [False, True])
+ def test_bad_set_custom(self, apply_inst):
+ """Test ValueError raised when not setting custom functions correctly.
+
+ Parameters
+ ----------
+ apply_inst : bool
+ Apply custom function at Instrument level (True) or Constellation
+ level (False)
+
+ """
+
+ with pytest.raises(ValueError) as verr:
+ pysat.Constellation(
+ instruments=[pysat.Instrument(
+ 'pysat', 'testing', num_samples=10, clean_level='clean')
+ for i in range(5)],
+ custom=[{'apply_inst': apply_inst}])
+
+ assert str(verr).find("Input dict to custom is missing the") >= 0
+ return
+
@pytest.mark.parametrize("apply_inst", [False, True])
def test_repr(self, apply_inst):
"""Test `__repr__` with custom method.
@@ -305,12 +327,17 @@ class TestConstellationBasics(object):
"""
self.testConst.custom_attach(mult_data, apply_inst=apply_inst,
- args=self.custom_args)
+ args=self.custom_args,
+ kwargs={'dkey': 'mlt'})
self.out = self.testConst.__str__()
assert isinstance(self.out, str)
assert self.out.find("Constellation-level Data Processing") >= 0
assert self.out.find(
"Custom Functions: {:d} applied".format(num_func)) >= 0
+
+ if num_func > 0:
+ assert self.out.find("Args=") >= 0
+ assert self.out.find("Kwargs=") >= 0
return
def test_single_custom_function_error(self):
|
UI - FM - Alarm Diagnostic: Wrong field types fixed
HG--
branch : ashapovalov/ui-fm-alarm-diagnostic-wrong-field-typ-1490087280651 | @@ -142,7 +142,7 @@ Ext.define("NOC.fm.alarmdiagnosticconfig.Application", {
},
{
name: "periodic_delay",
- xtype: "main.ref.script.LookupField",
+ xtype: "numberfield",
fieldLabel: __("Delay"),
min: 0,
allowBlank: true
@@ -155,7 +155,7 @@ Ext.define("NOC.fm.alarmdiagnosticconfig.Application", {
},
{
name: "periodic_script",
- xtype: "textfield",
+ xtype: "main.ref.script.LookupField",
fieldLabel: __("Script"),
allowBlank: true
},
|
CI: Removed test_converters_compatibility
TT 1.x uses inkscape as converter only. Hence, this test is not required anymore. | @@ -200,14 +200,6 @@ def is_current_version_compatible(test_id,
or not os.path.isfile(mod_args["preamble-file"]):
mod_args["preamble-file"] = os.path.join(EXTENSION_DIR, "default_packages.tex")
- if converter == "pstoedit":
- textext.CONVERTERS = {textext.PstoeditPlotSvg.get_pdf_converter_name(): textext.PstoeditPlotSvg}
- elif converter == "pdf2svg":
- textext.CONVERTERS = {textext.Pdf2SvgPlotSvg.get_pdf_converter_name(): textext.Pdf2SvgPlotSvg}
- else:
- textext.CONVERTERS = {textext.PstoeditPlotSvg.get_pdf_converter_name(): textext.PstoeditPlotSvg,
- textext.Pdf2SvgPlotSvg.get_pdf_converter_name(): textext.Pdf2SvgPlotSvg}
-
# run TexText
tt = textext.TexText()
tt.affect([
@@ -250,31 +242,3 @@ def test_compatibility(root, inkscape_version, textext_version, converter, test_
)
sys.stderr.write(message + "\n")
assert result, message
-
-
-def test_converters_compatibility(root, inkscape_version, textext_version, converter, test_case):
- if inkscape_version.startswith("_") or textext_version.startswith("_") or converter.startswith(
- "_") or test_case.startswith("_"):
- pytest.skip("skip %s (remove underscore to enable)" % os.path.join(inkscape_version, textext_version, converter,
- test_case))
-
- assert converter in ["pdf2svg", "pstoedit"]
- # switch converters
- if converter == "pdf2svg":
- replaced_converter = "pstoedit"
- elif converter == "pstoedit":
- replaced_converter = "pdf2svg"
-
- test_id = "%s-%s-%s-%s-%s" % (inkscape_version, textext_version, converter, replaced_converter, test_case)
- result, message = is_current_version_compatible(
- test_id,
- svg_original=os.path.join(root, inkscape_version, textext_version, converter, test_case, "original.svg"),
- svg_modified=os.path.join(root, inkscape_version, textext_version, converter, test_case, "modified.svg"),
- json_config=os.path.join(root, inkscape_version, textext_version, converter, test_case, "config.json"),
- converter=replaced_converter,
- fuzz="50%",
- pixel_diff_abs_tol=150,
- pixel_diff_rel_tol=0.005
- )
- sys.stderr.write(message+"\n")
- assert result, message
|
Allow pipelines to be opened withe the JSON widget
Pipieline files can currently be opened in either the pipeline
editor or the file editor. This uodate also allows users to open
pipeline files with the JSON viewer.
Fixes | @@ -108,12 +108,15 @@ const extension: JupyterFrontEndPlugin<void> = {
});
// Add the default behavior of opening the widget for .pipeline files
- app.docRegistry.addFileType({
+ app.docRegistry.addFileType(
+ {
name: PIPELINE,
displayName: 'Pipeline',
extensions: ['.pipeline'],
icon: pipelineIcon
- });
+ },
+ ['JSON']
+ );
app.docRegistry.addWidgetFactory(pipelineEditorFactory);
const tracker = new WidgetTracker<DocumentWidget>({
|
Catch RemoteError if scanner has crashed
If you restart the scanner (because it has crashed for e.g.)
but don't restart the web script it won't show any data
This way only the Worker status will be missing. | @@ -236,7 +236,10 @@ def get_pokemarkers():
if config.MAP_WORKERS:
# Worker stats
+ try:
markers.extend(get_worker_markers())
+ except RemoteError:
+ print('Unable to connect to manager for worker data.')
return markers
def get_spawnpointsmarkers():
|
fix: Use correct path of built assets
Get path from bundled_assets | @@ -576,13 +576,15 @@ def get_server_messages(app):
def get_messages_from_include_files(app_name=None):
"""Returns messages from js files included at time of boot like desk.min.js for desk and web"""
+ from frappe.utils.jinja_globals import bundled_asset
messages = []
app_include_js = frappe.get_hooks("app_include_js", app_name=app_name) or []
web_include_js = frappe.get_hooks("web_include_js", app_name=app_name) or []
include_js = app_include_js + web_include_js
for js_path in include_js:
- relative_path = os.path.join(frappe.local.sites_path, js_path.lstrip('/'))
+ file_path = bundled_asset(js_path)
+ relative_path = os.path.join(frappe.local.sites_path, file_path.lstrip('/'))
messages_from_file = get_messages_from_file(relative_path)
messages.extend(messages_from_file)
|
Standalone: Fix Qt plugins in subfolders.
* Also scan original files properly by proving the correct
path.
* This should e.g. fix the JPEG plugin. | @@ -105,12 +105,12 @@ if os.path.exists(guess_path):
return [
(
- os.path.join(plugin_dir, os.path.basename(filename)),
filename,
+ os.path.join(target_plugin_dir, os.path.relpath(filename, plugin_dir)),
full_name
)
for filename in
- getFileList(target_plugin_dir)
+ getFileList(plugin_dir)
if not filename.endswith(".qml")
]
|
Remove share directory
API documentation is installed by CMake in the doc directory. | @@ -84,6 +84,8 @@ class LibpqxxRecipe(ConanFile):
cmake = self._configure_cmake()
cmake.install()
+ tools.rmdir(os.path.join(self.package_folder, "share"))
+
def package_info(self):
pqxx_with_suffix = "pqxx-%s.%s" % tuple(self.version.split(".")[0:2])
is_package_with_suffix = self.settings.os != "Windows" and self.options.shared
|
Fix auth of mailboxlayer and vatlayer
Fix auth of apilayer mailboxlayer and vatlayer
resolve | @@ -404,7 +404,7 @@ API | Description | Auth | HTTPS | CORS |
API | Description | Auth | HTTPS | CORS |
|---|---|---|---|---|
| [Abstract Email Validation](https://www.abstractapi.com/email-verification-validation-api) | Validate email addresses for deliverability and spam | `apiKey` | Yes | Yes |
-| [apilayer mailboxlayer](https://mailboxlayer.com) | Email address validation | No | Yes | Unknown |
+| [apilayer mailboxlayer](https://mailboxlayer.com) | Email address validation | `apiKey` | Yes | Unknown |
| [Cloudmersive Validate](https://cloudmersive.com/validate-api) | Validate email addresses, phone numbers, VAT numbers and domain names | `apiKey` | Yes | Yes |
| [EVA](https://eva.pingutil.com/) | Validate email addresses | No | Yes | Yes |
| [Lob.com](https://lob.com/) | US Address Verification | `apiKey` | Yes | Unknown |
@@ -413,7 +413,7 @@ API | Description | Auth | HTTPS | CORS |
| [US Autocomplete](https://smartystreets.com/docs/cloud/us-autocomplete-api) | Enter address data quickly with real-time address suggestions | `apiKey` | Yes | Yes |
| [US Extract](https://smartystreets.com/products/apis/us-extract-api) | Extract postal addresses from any text including emails | `apiKey` | Yes | Yes |
| [US Street Address](https://smartystreets.com/docs/cloud/us-street-api) | Validate and append data for any US postal address | `apiKey` | Yes | Yes |
-| [vatlayer](https://vatlayer.com/documentation) | VAT number validation | No | Yes | Unknown |
+| [vatlayer](https://vatlayer.com/documentation) | VAT number validation | `apiKey` | Yes | Unknown |
| [Verifier](https://verifier.meetchopra.com/docs#/) | Verifies that a given email is real | `apiKey` | Yes | Yes |
| [Veriphone](https://veriphone.io) | Phone number validation & carrier lookup | `apiKey` | Yes | Yes |
|
Cleanup
Ran flake8 to clean up some unneeded whitespaces. | @@ -133,7 +133,6 @@ class Folder(CanvasObject):
)
return Folder(self._requester, response.json())
-
def upload(self, file, **kwargs):
"""
Upload a file to this folder.
@@ -155,7 +154,6 @@ class Folder(CanvasObject):
**kwargs
).start()
-
def update(self, **kwargs):
"""
Updates a folder.
|
Update problem.py
Should use "oneshot" for remote challenges. | @@ -246,6 +246,13 @@ class Remote(Service):
return output
+ def service(self):
+ """
+ Unlike the parent class, these are executables and should be restarted each time
+ """
+ return {"Type":"oneshot",
+ "ExecStart":"/bin/bash -c \"{}\"".format(self.start_cmd)
+ }
class FlaskApp(Service):
"""
|
Codacy
Still trying to clear the codacy flag. | @@ -1245,7 +1245,7 @@ def get_tts_engine(profile):
try:
flite_cmd = ['flite', '-lv']
voices = subprocess.check_output(
- flite_cmd,
+ ['flite','-lv'],
shell=False
).decode('utf-8').split(" ")[2:-1]
print(
|
Update README.md
Links in table now open the notebooks in colab | |-----------|----|----|
|Overview of Python ML/DL software ecosystem| Various | [markdown](software.md)|
|List of Python tutorials | Various | [markdown](python.md)|
-|Brief intro to colab| Colab | [notebook](colab_intro.ipynb)|
-|Brief intro to data analysis |Matplotlib, Pandas, Xarray | [notebook](pandas_intro.ipynb)|
-|Brief intro to sklearn | Sklearn | [notebook](sklearn_intro.ipynb)|
-|Brief intro to JAX | JAX | [notebook](jax_intro.ipynb)|
-|Brief survey of common datasets| Sklearn, TFDS| [notebook](datasets.ipynb)|
+|Brief intro to colab| Colab | [notebook](https://colab.sandbox.google.com/github/probml/pyprobml/blob/master/book1/intro/colab_intro.ipynb)|
+|Brief intro to data analysis |Matplotlib, Pandas, Xarray | [notebook](https://colab.sandbox.google.com/github/probml/pyprobml/blob/master/book1/intro/pandas_intro.ipynb)|
+|Brief intro to sklearn | Sklearn | [notebook](https://colab.sandbox.google.com/github/probml/pyprobml/blob/master/book1/intro/sklearn_intro.ipynb)|
+|Brief intro to JAX| Colab | [notebook](https://colab.sandbox.google.com/github/probml/pyprobml/blob/master/book1/intro/jax_intro.ipynb)
+|Brief survey of common datasets| Sklearn, TFDS| [notebook](https://colab.sandbox.google.com/github/probml/pyprobml/blob/master/book1/intro/datasets.ipynb)|
|
Fix master
Summary: The `warnOnSpreadAttributes` config option is failing on `yarn build-for-python`, remove it since it's not surfacing any issues right now anyway.
Test Plan: `yarn build-for-python`, verify no error.
Reviewers: yuhan | @@ -53,7 +53,7 @@ module.exports = {
],
},
],
- 'react/jsx-no-target-blank': ['error', {warnOnSpreadAttributes: true}],
+ 'react/jsx-no-target-blank': 'error',
'react/prefer-stateless-function': 'error',
'react/prop-types': 'off',
'react/display-name': 'off',
|
update analyze_spectral
There are specific datatypes that are allowed with writing a dictionary from the Outputs observations class to a json text file. Converting to string was useful for avoiding float32 but we don't want wavelengths or reflectance frequencies to be strings so instead update the datatype transformation steps. | @@ -38,7 +38,6 @@ def analyze_spectral(array, header_dict, mask, histplot=True):
wavelength_data = array[np.where(mask > 0)]
wavelength_freq = wavelength_data.mean(axis=0)
- #
min_wavelength = int(np.ceil(float(header_dict["wavelength"][0])))
max_wavelength = int(np.ceil(float(header_dict["wavelength"][-1])))
@@ -47,13 +46,12 @@ def analyze_spectral(array, header_dict, mask, histplot=True):
for i, wavelength in enumerate(header_dict["wavelength"]):
new_wavelengths.append(float(wavelength))
- new_freq.append(str(wavelength_freq[i]))
+ new_freq.append((wavelength_freq[i]).astype(np.float))
maxreflectance = np.amax(wavelength_data)
minreflectance = np.amin(wavelength_data)
avgreflectance = np.average(wavelength_data)
medianreflectance = np.median(wavelength_data)
- print("max")
# Store data into outputs class
outputs.add_observation(variable='max_reflectance', trait='maximum reflectance',
@@ -70,7 +68,7 @@ def analyze_spectral(array, header_dict, mask, histplot=True):
value=float(medianreflectance), label='reflectance')
outputs.add_observation(variable='spectral_frequencies', trait='thermal spectral_frequencies',
method='plantcv.plantcv.hyperspectral.analyze_spectral', scale='frequency', datatype=list,
- value=new_freq, label=header_dict["wavelength"])
+ value=new_freq, label=new_wavelengths)
params.debug = debug
analysis_img = None
|
Remove protractor-add-test-answer css class from modify training data
button | answer-group-editor .oppia-add-rule-button:active,
answer-group-editor .oppia-add-rule-button:focus,
- answer-group-editor .oppia-add-rule-button:hover {
+ answer-group-editor .oppia-add-rule-button:hover,
+ answer-group-editor .oppia-modify-training-data-button:active,
+ answer-group-editor .oppia-modify-training-data-button:focus,
+ answer-group-editor .oppia-modify-training-data-button:hover {
background-color: rgba(165,165,165,1);
color: white;
opacity: 1;
}
answer-group-editor .oppia-modify-training-data-button {
+ background-color: rgba(165,165,165,0.9);
+ border: 0;
+ border-radius: 0;
+ color: white;
+ opacity: 0.9;
+ padding: 7px;
+ width: 100%;
margin-top: 4%;
}
</style>
+ Add Another Possible Answer
</button>
- <div ng-if="isCurrentInteractionTrainable()">
- <button type="button" class="btn btn-default btn-lg oppia-add-rule-button protractor-test-add-answer oppia-modify-training-data-button" ng-click="openTrainingDataEditor()">
+ <div ng-if="isEditable && isCurrentInteractionTrainable()">
+ <button type="button" class="btn btn-default btn-lg oppia-modify-training-data-button" ng-click="openTrainingDataEditor()">
<i class="material-icons md-24">playlist_add</i> Modify Training Data
</button>
</div>
|
Force dtype=float for array returned by inf_like
The scalar case already explicity returned np.inf. | @@ -3182,19 +3182,24 @@ def vectorize_if_needed(func, *x):
def inf_like(x):
- """Return the shape of x with value infinity.
+ """Return the shape of x with value infinity and dtype='float'.
Preserves 'shape' for both array and scalar inputs.
+ But always returns a float array, even if x is of integer type.
- >>> inf_like(0.)
+ >>> inf_like(0.) # float scalar
inf
- >>> inf_like([0., 1., 2., 3.])
+ >>> inf_like(1) # integer scalar should give float output
+ inf
+ >>> inf_like([0., 1., 2., 3.]) # float list
+ array([inf, inf, inf, inf])
+ >>> inf_like([0, 1, 2, 3]) # integer list should give float output
array([inf, inf, inf, inf])
"""
if np.isscalar(x):
return np.inf
else:
- return np.full_like(x, np.inf)
+ return np.full_like(x, np.inf, dtype='float')
# Pre-defined cosmologies. This loops over the parameter sets in the
|
Tests: install python dependencies in pyright tests
Otherwise pyright shows 100+ warnings related to imports. | @@ -91,8 +91,9 @@ jobs:
- uses: actions/checkout@v3
- name: Install dependencies
run: |
- sudo apt-get install -y npm
+ sudo apt-get install -y npm libkrb5-dev libxmlsec1-dev
npm install --global pyright
+ python -m pip --no-cache-dir install --upgrade -r requirements.txt
- name: Make pyright report of current commit
run: |
tools/run_pyright.sh generate ${{ env.PYRIGHT_CURRENT_REPORT }}
|
DeleteUndefined: handle MaterializeAll
MaterializeAll has to be checked in llvm 4. Do so. | #include "llvm/Transforms/Utils/BasicBlockUtils.h"
#include <llvm/IR/DebugInfoMetadata.h>
+#if LLVM_VERSION_MAJOR >= 4
+#include <llvm/Support/Error.h>
+#endif
+
using namespace llvm;
class DeleteUndefined : public ModulePass {
@@ -106,7 +110,16 @@ static bool array_match(const StringRef &name, const char **array)
}
bool DeleteUndefined::runOnModule(Module& M) {
+#if LLVM_VERSION_MAJOR >= 4
+ if (llvm::Error err = M.materializeAll()) {
+ std::error_code ec = errorToErrorCode(std::move(err));
+ llvm::errs() << __PRETTY_FUNCTION__ << ": cannot load module: " <<
+ ec.message();
+ return false;
+ }
+#else
M.materializeAll();
+#endif
// delete/replace the calls in the rest of functions
bool modified = false;
|
Field.doc: automatically append precise list of types
TN: | @@ -1489,6 +1489,25 @@ class Field(BaseField):
:type: int
"""
+ @property
+ def doc(self):
+ result = super(Field, self).doc
+
+ # If parsers build this field, add a precise list of types it can
+ # contain: the field type might be too generic.
+ if not self.struct.synthetic:
+ precise_types = self.types_from_parser.minimal_matched_types
+ if len(precise_types) > 1:
+ type_descr = '\n'.join([
+ 'This field can contain one of the following nodes:'
+ ] + list(sorted('* {}'.format(t.dsl_name)
+ for t in precise_types)))
+ result = ('{}\n\n{}'.format(type_descr, type_descr)
+ if result else
+ type_descr)
+
+ return result
+
@property
def overriding(self):
"""
|
fix tostring deprecation
Fixes: bloscpack/test_cmdline/mktestarray.py:30: DeprecationWarning: tostring() is deprecated. Use tobytes() instead. | @@ -27,7 +27,7 @@ def exists(filename):
if not exists(DATA_FILE) and not exists(META_FILE):
a = numpy.linspace(0, 100, int(2e7))
with open(DATA_FILE, 'wb') as f:
- f.write(a.tostring())
+ f.write(a.tobytes())
with open(META_FILE, 'w') as m:
meta = dict(sorted(_ndarray_meta(a).items()))
m.write(json.dumps(meta))
|
Refine CosineAnnealingWarmRestarts doc for issue
Summary:
Fixes
Pull Request resolved: | @@ -697,18 +697,32 @@ class CosineAnnealingWarmRestarts(_LRScheduler):
for base_lr in self.base_lrs]
def step(self, epoch=None):
- """Step could be called after every update, i.e. if one epoch has 10 iterations
- (number_of_train_examples / batch_size), we should call SGDR.step(0.1), SGDR.step(0.2), etc.
+ """Step could be called after every batch update
+
+ Example:
+ >>> scheduler = CosineAnnealingWarmRestarts(optimizer, T_0, T_mult)
+ >>> iters = len(dataloader)
+ >>> for epoch in range(20):
+ >>> for i, sample in enumerate(dataloader):
+ >>> inputs, labels = sample['inputs'], sample['labels']
+ >>> scheduler.step(epoch + i / iters)
+ >>> optimizer.zero_grad()
+ >>> outputs = net(inputs)
+ >>> loss = criterion(outputs, labels)
+ >>> loss.backward()
+ >>> optimizer.step()
This function can be called in an interleaved way.
Example:
- >>> scheduler = SGDR(optimizer, T_0, T_mult)
+ >>> scheduler = ConsineAnnealingWarmRestarts(optimizer, T_0, T_mult)
>>> for epoch in range(20):
>>> scheduler.step()
>>> scheduler.step(26)
>>> scheduler.step() # scheduler.step(27), instead of scheduler(20)
"""
+ if epoch < 0:
+ raise ValueError("Expected non-negative epoch, but got {}".format(epoch))
if epoch is None:
epoch = self.last_epoch + 1
self.T_cur = self.T_cur + 1
|
Added print out of ignore_forNorm chromosomes
Number of chromosomes to be ignore might be more informative | @@ -20,6 +20,7 @@ outdir_ATACqc = 'MACS2_qc/'
# do workflow specific stuff now
include: os.path.join(workflow.basedir, "internals.snakefile")
+
### include modules of other snakefiles ########################################
################################################################################
@@ -57,6 +58,7 @@ onstart:
print("Gene annotation BED:", genes_bed)
print("Gene annotation GTF:", genes_gtf)
print("Blacklist regions BED:", blacklist_bed)
+ print("Ignore for normalization (bigwigs):", ignore_forNorm)
print("-" * 80, "\n")
### main rule ##################################################################
|
export: Fix an unnecessary Any.
This was introduced a few weeks ago in
"Import script: Check and add system bots after every import." | @@ -1390,7 +1390,7 @@ def import_uploads(import_dir: Path, processing_avatars: bool=False) -> None:
# Because the Python object => JSON conversion process is not fully
# faithful, we have to use a set of fixers (e.g. on DateTime objects
# and Foreign Keys) to do the import correctly.
-def do_import_realm(import_dir: Path) -> Any:
+def do_import_realm(import_dir: Path) -> Realm:
logging.info("Importing realm dump %s" % (import_dir,))
if not os.path.exists(import_dir):
raise Exception("Missing import directory!")
|
Interpret entity docs as Mako templates
This will allow us to have different doc renderings depending on the
context. For instance, having different casings depending on the
generated API.
TN: | @@ -1148,16 +1148,19 @@ def create_doc_printer(lang, formatter):
template_ctx['TODO'] = todo_markers[lang]
if isinstance(entity, str):
- doc = ctx.documentations[entity].render(
+ doc_template = ctx.documentations[entity]
+ elif entity.doc:
+ doc_template = Template(entity.doc)
+ else:
+ return ''
+
+ doc = doc_template.render(
ctx=get_context(),
capi=ctx.c_api_settings,
lang=lang,
null=null_names[lang],
TODO=todo_markers[lang]
)
- else:
- doc = entity.doc or ''
-
return formatter(doc, column, **kwargs)
func.__name__ = '{}_doc'.format(lang)
|
Process docs: added version note for "advanced argument tweaking" feature
related: openEOPlatform/documentation#41 | @@ -105,6 +105,8 @@ but you can call the corresponding client method in multiple equivalent ways::
Advanced argument tweaking
---------------------------
+.. versionadded:: 0.10.0
+
In some situations, you may want to finetune what the (convenience) methods generate.
For example, you want to play with non-standard, experimental arguments,
or there is a problem with a automatic argument handling/conversion feature.
|
Match: make input value casts unsafe
TN: | @@ -1015,10 +1015,13 @@ class Match(AbstractExpression):
# bound and initialized.
self.matchers = []
for m in matchers:
- # Initialize match_var...
+ # Initialize match_var. Note that assuming the code generation
+ # is bug-free, this cast cannot fail, so don't generate type
+ # check boilerplate.
let_expr = Let.Expr(
[m.match_var],
- [Cast.Expr(self.prefix_var.ref_expr, m.match_var.type)],
+ [Cast.Expr(self.prefix_var.ref_expr, m.match_var.type,
+ unsafe=True)],
# ... and cast this matcher's result to the Match result's
# type, as required by OOP with access types in Ada.
|
Remove unneeded config entries
Since channels that mods can't read are now implicitly ignored, there is no need to explicitly ignore them. | @@ -248,15 +248,13 @@ guild:
- *ADMIN_SPAM
- *MODS
- # Modlog cog ignores events which occur in these channels
+ # Modlog cog explicitly ignores events which occur in these channels.
+ # This is on top of implicitly ignoring events in channels that the mod team cannot view.
modlog_blacklist:
- - *ADMINS
- - *ADMINS_VOICE
- *ATTACH_LOG
- *MESSAGE_LOG
- *MOD_LOG
- *STAFF_VOICE
- - *DEV_CORE_VOTING
reminder_whitelist:
- *BOT_CMD
|
updated: reduction in overhead if memory is already accurate
updated: using contextlib instead of home-grown class for "with" context. | @@ -2,6 +2,7 @@ import Queue
import logging
import traceback
import threading
+import contextlib
import collections
import envi
@@ -230,18 +231,7 @@ class VivWorkspaceCore(object,viv_impapi.ImportApi):
self.reloc_by_va[rva] = rtype
self.relocations.append(einfo)
- if rtype == RTYPE_BASERELOC:
- # FIXME: we can't rebase something and expect the pointer in memory to be just the offset...
- # consider deprecating this reloc type in favor of BASEOFF
- ptr = self.readMemoryPtr(rva)
- ptr += imgbase
- if ptr != (ptr & e_bits.u_maxes[self.psize]):
- logger.warn('RTYPE_BASERELOC calculated a bad pointer: 0x%x (imgbase: 0x%x)', ptr, imgbase)
-
- with SupervisorMode(self):
- self.writeMemoryPtr(rva, ptr)
-
- logger.info('_handleADDRELOC: %x -> %x (map: 0x%x)', rva, ptr, imgbase)
+ # RTYPE_BASERELOC assumes the memory is already accurate (eg. PE's unless rebased)
if rtype == RTYPE_BASEOFF:
# add imgbase and offset to pointer in memory
@@ -250,10 +240,20 @@ class VivWorkspaceCore(object,viv_impapi.ImportApi):
if ptr != (ptr & e_bits.u_maxes[self.psize]):
logger.warn('RTYPE_BASEOFF calculated a bad pointer: 0x%x (imgbase: 0x%x)', ptr, imgbase)
- with SupervisorMode(self):
+ # writes are costly, especially on larger binaries
+ if ptr == self.readMemoryPtr(rva):
+ return
+
+ with self.getAdminRights():
self.writeMemoryPtr(rva, ptr)
- logger.info('_handleADDRELOC: %x -> %x (map: 0x%x)', rva, ptr, imgbase)
+ #logger.info('_handleADDRELOC: %x -> %x (map: 0x%x)', rva, ptr, imgbase)
+
+ @contextlib.contextmanager
+ def getAdminRights(self):
+ self._supervisor = True
+ yield
+ self._supervisor = False
def _handleADDMODULE(self, einfo):
print('DEPRICATED (ADDMODULE) ignored: %s' % einfo)
@@ -659,14 +659,6 @@ class VivWorkspaceCore(object,viv_impapi.ImportApi):
fva,spdelta,symtype,syminfo = locsym
self.localsyms[fva][spdelta] = locsym
-class SupervisorMode:
- def __init__(self, vw):
- self.vw = vw
- def __enter__(self):
- self.vw._supervisor = 1
- def __exit__(self, type, value, traceback):
- self.vw._supervisor = 0
-
def trackDynBranches(cfctx, op, vw, bflags, branches):
'''
track dynamic branches
|
[tests] Show additional informations with "urlshortener-blocked" APIError
add site and user to the result['error'] dict in case of T244062
if not logged in, site.user() is None; show the IP in that case
print "other" information in separate lines with APIError | @@ -120,10 +120,10 @@ class APIError(Error):
def __str__(self):
"""Return a string representation."""
if self.other:
- return '{0}: {1} [{2}]'.format(
+ return '{0}: {1}\n[{2}]'.format(
self.code,
self.info,
- '; '.join(
+ ';\n '.join(
'{0}: {1}'.format(key, val)
for key, val in self.other.items()))
@@ -2069,6 +2069,15 @@ class Request(MutableMapping):
self.wait()
continue
+ if code == 'urlshortener-blocked': # T244062
+ # add additional informations to result['error']
+ result['error']['current site'] = self.site
+ if self.site.user():
+ result['error']['current user'] = self.site.user()
+ else: # not logged in; show the IP
+ si = self.site.siteinfo
+ result['error']['current user'] = si['name']
+
# raise error
try:
# Due to bug T66958, Page's repr may return non ASCII bytes
|
Fix import ssl may fail under some Python installs
It's only required for certain proxy configurations, so we
don't want it to raise ImportError while the user imports
our library. | import abc
import asyncio
import socket
-import ssl as ssl_mod
import sys
+try:
+ import ssl as ssl_mod
+except ImportError:
+ ssl_mod = None
+
from ...errors import InvalidChecksumError
from ... import helpers
@@ -68,6 +72,12 @@ class Connection(abc.ABC):
loop=self._loop
)
if ssl:
+ if ssl_mod is None:
+ raise RuntimeError(
+ 'Cannot use proxy that requires SSL'
+ 'without the SSL module being available'
+ )
+
s.settimeout(timeout)
s = ssl_mod.wrap_socket(
s,
|
[air/xgboost] Resolve xgboost benchmark failure
After batch_size in predictor is applied correctly, we can reduce flakiness of our release test by using larger batch size for xgboost, since default 4096 is too small. This reduces runtime from ~310secs to ~200secs.
Full context and debugging history see attached issue. | @@ -102,7 +102,12 @@ def run_xgboost_prediction(model_path: str, data_path: str):
ds = data.read_parquet(data_path)
ckpt = XGBoostCheckpoint.from_model(booster=model)
batch_predictor = BatchPredictor.from_checkpoint(ckpt, XGBoostPredictor)
- result = batch_predictor.predict(ds.drop_columns(["labels"]))
+ result = batch_predictor.predict(
+ ds.drop_columns(["labels"]),
+ # Improve prediction throughput for xgboost with larger
+ # batch size than default 4096
+ batch_size=8192,
+ )
return result
|
Fix non-json response of container commit
The RESP BODY should be of json format, this patch fixes it.
Closes-Bug: | @@ -617,7 +617,7 @@ class Manager(periodic_task.PeriodicTasks):
repository, tag)
utils.spawn_n(do_container_commit)
- return snapshot_image.id
+ return {"uuid": snapshot_image.id}
def _do_container_image_upload(self, context, snapshot_image, data, tag):
try:
|
add SSH ingress on default sg for all addrs
Also fixes the ip_ranges on the other rules to match the CidrBlock
given when creating the vpc. | @@ -106,10 +106,9 @@ class EC2Provider(ExecutionProvider):
# Required: False
# Default: t2.small },
- "imageId" : #{"Description: String to append to the #SBATCH blocks
- # in the submit script to the scheduler
+ "imageId" : #{"Description: The ID of the AMI
# Type: String,
- # Required: False },
+ # Required: True },
"region" : #{"Description: AWS region to launch machines in
# in the submit script to the scheduler
@@ -377,7 +376,10 @@ class EC2Provider(ExecutionProvider):
return vpc
def security_group(self, vpc):
- ''' Create and configure security group.
+ """Create and configure a new security group.
+
+
+
Allows all ICMP in, all TCP and UDP in within VPC
This security group is very open. It allows all
@@ -386,13 +388,13 @@ class EC2Provider(ExecutionProvider):
by changing the allowed port ranges.
:param vpc - VPC in which to set up security group
- '''
+ """
sg = vpc.create_security_group(
GroupName="private-subnet", Description="security group for remote executors"
)
- ip_ranges = [{'CidrIp': '172.32.0.0/16'}]
+ ip_ranges = [{'CidrIp': '10.0.0.0/16'}]
# Allows all ICMP in, all TCP and UDP in within VPC
@@ -414,8 +416,16 @@ class EC2Provider(ExecutionProvider):
'IpRanges': [{
'CidrIp': '0.0.0.0/0'
}],
+ }, {
+ 'IpProtocol': 'TCP',
+ 'FromPort': 22,
+ 'ToPort': 22,
+ 'IpRanges': [{
+ 'CidrIp': '0.0.0.0/0'
+ }],
}
]
+
# Allows all TCP out, all TCP and UDP out within VPC
outPerms = [
{
|
Fix bugs that prevented the form handlers from working on firefox.
Now, the handlers directly interact with the input or textarea tag
when inputting text. | @@ -45,6 +45,14 @@ class BaseFormHandler:
def selenium(self):
return BuiltIn().get_library_instance("SeleniumLibrary")
+ @property
+ def input_element(self):
+ """Returns the actual <input> or <textarea> element inside the element"""
+ elements = self.element.find_elements_by_xpath(
+ ".//*[self::input or self::textarea]"
+ )
+ return elements[0] if elements else None
+
# these should be implemented by each of the handlers
def set(self, value):
pass
@@ -82,7 +90,7 @@ class LightningComboboxHandler(BaseFormHandler):
value_locator = f'//lightning-base-combobox-item[.="{value}"]'
wait = 5
try:
- self.element.click()
+ self.input_element.click()
self.selenium.wait_until_element_is_visible(value_locator, wait)
self.selenium.click_element(value_locator)
except Exception as e:
@@ -99,20 +107,22 @@ class LightningInputHandler(BaseFormHandler):
def set(self, value):
self.focus()
- if self.element.get_attribute("type") == "checkbox":
+ if self.input_element.get_attribute("type") == "checkbox":
# lightning-input elements are used for checkboxes
# as well as free text input.
checked = self.element.get_attribute("checked")
if (checked and value != "checked") or (not checked and value == "checked"):
- self.element.send_keys(" ")
+ self.input_element.send_keys(" ")
else:
self.clear()
- self.element.send_keys(value)
+ self.input_element.send_keys(value)
def clear(self):
"""Remove the value in the element"""
# oddly, element.clear() doesn't always work.
- self.selenium.driver.execute_script("arguments[0].value = '';", self.element)
+ self.selenium.driver.execute_script(
+ "arguments[0].value = '';", self.input_element
+ )
def focus(self):
"""Set focus to the element
@@ -134,7 +144,7 @@ class LightningLookupHandler(BaseFormHandler):
# instead of searching the whole document?
value_locator = f'//div[@role="listbox"]//*[.="{value}"]'
self.element.click()
- self.element.send_keys(value)
+ self.input_element.send_keys(value)
try:
self.selenium.wait_until_element_is_visible(value_locator, wait)
self.selenium.click_element(value_locator)
|
Fixed typo in requisites.rst
Glog -> Glob | @@ -68,7 +68,7 @@ first line in the stanza) or the ``- name`` parameter.
- require:
- pkg: vim
-Glog matching in requisites
+Glob matching in requisites
~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. versionadded:: 0.9.8
|
swarming: fix ts_mon_metrics_test
This is to fix | @@ -12,6 +12,7 @@ import swarming_test_env
swarming_test_env.setup_test_env()
from google.appengine.ext import ndb
+import webapp2
import gae_ts_mon
from test_support import test_case
@@ -83,7 +84,8 @@ class TestMetrics(test_case.TestCase):
def setUp(self):
super(TestMetrics, self).setUp()
gae_ts_mon.reset_for_unittest()
- gae_ts_mon.initialize()
+ self.app = webapp2.WSGIApplication(None, debug=True)
+ gae_ts_mon.initialize(self.app)
self.now = datetime.datetime(2016, 4, 7)
self.mock_now(self.now)
|
Update rogue_dns.txt
Root ```upgrinfo.com``` is going to ```domain.txt``` | @@ -223,4 +223,12 @@ ns2.whatnexthost.com
# Reference: https://www.virustotal.com/gui/domain/ns1.fakesemoiin23.com/relations
# Reference: https://www.virustotal.com/gui/domain/ns2.fakesemoiin23.com/relations
+ns1.fakesemoiin23.com
+ns2.fakesemoiin23.com
139.59.80.101:53
+
+# Reference: https://www.virustotal.com/gui/domain/upgrinfo.com/relations
+
+ns1.upgrinfo.com
+ns2.upgrinfo.com
+58.158.177.102:53
|
Config2: limit the number of retries to 15
This limits the waiting time until timeout from 43 min to 23. | @@ -649,7 +649,7 @@ special_page_limit = 500
step = -1
# Maximum number of times to retry an API request before quitting.
-max_retries = 25
+max_retries = 15
# Minimum time to wait before resubmitting a failed API request.
retry_wait = 5
|
Environment variable fallbacks should work for manually created configs
This gives LocalConfig.find() and LocalConfig() the same defaults.
The Code is also more easily understandable with the chain
of 'or' fallbacks in one place. | @@ -52,56 +52,68 @@ class LocalConfig(object):
current user.
"""
- def __init__(self, config, environment=None, files_loaded=None):
- self._config = config
+ def __init__(self, config, files_loaded=None, env=None, driver=None):
+ self._config = config # type: compat.configparser.ConfigParser
self.files_loaded = []
if files_loaded:
- self.files_loaded = files_loaded
+ self.files_loaded = files_loaded # type: list[str]
- self.environment = environment
+ # The user may specify these when running, otherwise they are loaded from config.
+ self._specified_environment = env # type: str
+ self._specified_driver = driver # type: str
+
+ if not config.has_section(self.environment):
+ raise ValueError('No config section found for environment %r' % (env,))
@classmethod
- def find(cls, paths=DEFAULT_CONF_PATHS, env=None):
+ def find(cls, paths=DEFAULT_CONF_PATHS, env=None, driver=None):
"""
Find config from possible filesystem locations.
'env' is which environment to use from the config: it corresponds to the name of a config section
+ `driver` is a specific driver to use rather than the defaults.
+
:type paths: list[str]
:type env: str
+ :type driver: str
:rtype: LocalConfig
"""
config = compat.read_config(_DEFAULT_CONF)
files_loaded = config.read(p for p in paths if p)
- env = env or os.environ.get('DATACUBE_ENVIRONMENT') or config.get('user', 'default_environment')
- if not config.has_section(env):
- raise ValueError('No config section found for environment %r' % (env,))
-
return LocalConfig(
config,
- environment=env,
- files_loaded=files_loaded
+ files_loaded=files_loaded,
+ env=env,
+ driver=driver
)
- def _prop(self, key):
+ def _environment_prop(self, key):
+ # Get the property for the current instance.
try:
return self._config.get(self.environment, key)
except compat.NoOptionError:
return None
+ @property
+ def environment(self):
+ return self._specified_environment or \
+ os.environ.get('DATACUBE_ENVIRONMENT') or \
+ self._config.get('user', 'default_environment')
+
@property
def db_hostname(self):
- return self._prop('db_hostname')
+ return self._environment_prop('db_hostname')
@property
def db_database(self):
- return self._prop('db_database')
+ return self._environment_prop('db_database')
@property
def db_connection_timeout(self):
- return int(self._prop('db_connection_timeout'))
+ return int(self._environment_prop('db_connection_timeout'))
@property
def db_username(self):
@@ -112,19 +124,24 @@ class LocalConfig(object):
# No default on Windows
default_username = None
- return self._prop('db_username') or default_username
+ return self._environment_prop('db_username') or default_username
@property
def default_driver(self):
- return self._prop('default_driver')
+ return (
+ self._specified_driver or
+ os.environ.get('DATACUBE_DEFAULT_DRIVER') or
+ # The default driver for the current environment
+ self._environment_prop('default_driver')
+ )
@property
def db_password(self):
- return self._prop('db_password')
+ return self._environment_prop('db_password')
@property
def db_port(self):
- return self._prop('db_port') or '5432'
+ return self._environment_prop('db_port') or '5432'
def __str__(self):
return "LocalConfig<loaded_from={}, config={}, environment={})".format(
|
Switched to using Message.from_bytes in RtMidi backend.
Since RtMidi always returns a complete message there's no need to use
the full parser. | @@ -7,6 +7,7 @@ import threading
import rtmidi
from .. import ports
+from ..messages import Message
from ._parser_queue import ParserQueue
def _get_api_lookup():
@@ -203,10 +204,13 @@ class Input(PortCommon, ports.BaseInput):
self._rt.set_callback(self._callback_wrapper)
def _callback_wrapper(self, msg_data, data):
- self._queue.put_bytes(msg_data[0])
- if self._callback:
- for msg in self._queue.iterpoll():
- self._callback(msg)
+ try:
+ msg = Message.from_bytes(msg_data[0])
+ except ValueError:
+ # Ignore invalid message.
+ return
+
+ (self._callback or self._queue.put)(msg)
class Output(PortCommon, ports.BaseOutput):
|
Add an annotation for time completed to show its non-None type
This is part of DFK static typing, issue | @@ -110,7 +110,7 @@ class DataFlowKernel(object):
self.hub_interchange_port = self.monitoring.start(self.run_id, self.run_dir)
self.time_began = datetime.datetime.now()
- self.time_completed = None
+ self.time_completed: Optional[datetime.datetime] = None
logger.info("Run id is: " + self.run_id)
|
node tests: Add test_trigger_submit_compose_form().
(Steve Howell also contributed to this.) | @@ -822,6 +822,26 @@ function test_with_mock_socket(test_params) {
assert(update_faded_messages_checked);
}());
+(function test_trigger_submit_compose_form() {
+ var prevent_default_checked = false;
+ var compose_finish_checked = false;
+ var e = {
+ preventDefault: function () {
+ prevent_default_checked = true;
+ },
+ };
+ compose.finish = function () {
+ compose_finish_checked = true;
+ };
+
+ var submit_handler = $('#compose form').get_on_handler('submit');
+
+ submit_handler(e);
+
+ assert(prevent_default_checked);
+ assert(compose_finish_checked);
+}());
+
(function test_set_focused_recipient() {
var sub = {
stream_id: 101,
|
fix "git clone" on windows
env replaces, it does not update, and windows really needs its env | @@ -273,13 +273,20 @@ def run_cmd(cwd, cmd, env=None):
if len(cmd) == 0:
raise dbt.exceptions.CommandError(cwd, cmd)
+ # the env argument replaces the environment entirely, which has exciting
+ # consequences on Windows! Do an update instead.
+ full_env = env
+ if env is not None:
+ full_env = os.environ.copy()
+ full_env.update(env)
+
try:
proc = subprocess.Popen(
cmd,
cwd=cwd,
stdout=subprocess.PIPE,
stderr=subprocess.PIPE,
- env=env)
+ env=full_env)
out, err = proc.communicate()
except OSError as exc:
|
[gcs/ha] Enable HA flags by default
PR to enable all three flags for GCS HA:
RAY_bootstrap_with_gcs=1
RAY_gcs_grpc_based_pubsub=1
RAY_gcs_storage=memory | @@ -294,9 +294,9 @@ RAY_CONFIG(bool, grpc_based_resource_broadcast, true)
// Feature flag to enable grpc based pubsub in GCS.
RAY_CONFIG(bool, gcs_grpc_based_pubsub, true)
// The storage backend to use for the GCS. It can be either 'redis' or 'memory'.
-RAY_CONFIG(std::string, gcs_storage, "redis")
+RAY_CONFIG(std::string, gcs_storage, "memory")
// Feature flag to enable GCS based bootstrapping.
-RAY_CONFIG(bool, bootstrap_with_gcs, false)
+RAY_CONFIG(bool, bootstrap_with_gcs, true)
/// Duration to sleep after failing to put an object in plasma because it is full.
RAY_CONFIG(uint32_t, object_store_full_delay_ms, 10)
|
Tweak language in PR template
This has irked me for a while.
Declaring certainty that a change won't produce a regression seems impossibly hard, and I think this is a more achievable reflection of the intent. | - [ ] All migrations are backwards compatible and won't block deploy
- [ ] The set of people pinged as reviewers is appropriate for the level of risk of the change
- [ ] If QA is part of the safety story, the "Awaiting QA" label is used
-- [ ] I am certain that this PR will not introduce a regression for the reasons below
+- [ ] I have confidence that this PR will not introduce a regression for the reasons below
### Automated test coverage
|
adding fields attribute to audio.Audio
docstring update | @@ -11,6 +11,7 @@ class Audio(PlexPartialObject):
Attributes:
addedAt (datetime): Datetime this item was added to the library.
+ fields (list): List of :class:`~plexapi.media.Field`.
index (sting): Index Number (often the track number).
key (str): API URL (/library/metadata/<ratingkey>).
lastViewedAt (datetime): Datetime item was last accessed.
@@ -33,6 +34,7 @@ class Audio(PlexPartialObject):
self._data = data
self.listType = 'audio'
self.addedAt = utils.toDatetime(data.attrib.get('addedAt'))
+ self.fields = self.findItems(data, etag='Field')
self.index = data.attrib.get('index')
self.key = data.attrib.get('key')
self.lastViewedAt = utils.toDatetime(data.attrib.get('lastViewedAt'))
|
Removes container cleanup test
This is tested elsewhere, eg, test_service_start_cleanup
Closes | from pathlib import Path
-import docker
import pytest
from django.core.exceptions import ValidationError
from django.test import TestCase
@@ -26,12 +25,7 @@ def test_submission_evaluation(
settings.task_always_eager = (True,)
# Upload a submission and create an evaluation
- dockerclient = docker.DockerClient(
- base_url=settings.COMPONENTS_DOCKER_BASE_URL
- )
-
eval_container, sha256 = evaluation_image
-
method = MethodFactory(
image__from_path=eval_container, image_sha256=sha256, ready=True
)
@@ -40,18 +34,11 @@ def test_submission_evaluation(
with pytest.raises(NotImplementedError):
_ = method.image.url
- num_containers_before = len(dockerclient.containers.list())
- num_volumes_before = len(dockerclient.volumes.list())
-
# This will create an evaluation, and we'll wait for it to be executed
submission = SubmissionFactory(
predictions_file__from_path=submission_file, phase=method.phase
)
- # The evaluation method should clean up after itself
- assert len(dockerclient.volumes.list()) == num_volumes_before
- assert len(dockerclient.containers.list()) == num_containers_before
-
# The evaluation method should return the correct answer
assert len(submission.evaluation_set.all()) == 1
assert (
|
OptionalParameter in ExternalPythonProgramTask
Updated ExternalPythonProgramTask parameters type to OptionalParameter | @@ -260,13 +260,13 @@ class ExternalPythonProgramTask(ExternalProgramTask):
:py:class:`luigi.parameter.Parameter` s for setting a virtualenv and for
extending the ``PYTHONPATH``.
"""
- virtualenv = luigi.Parameter(
+ virtualenv = luigi.OptionalParameter(
default=None,
positional=False,
description='path to the virtualenv directory to use. It should point to '
'the directory containing the ``bin/activate`` file used for '
'enabling the virtualenv.')
- extra_pythonpath = luigi.Parameter(
+ extra_pythonpath = luigi.OptionalParameter(
default=None,
positional=False,
description='extend the search path for modules by prepending this '
|
rng warning
Changed settings for generating the RNG warning, which were
happening too often. | @@ -198,7 +198,7 @@ def sample_rwalk(args):
nfail += 1
# Check if we're stuck generating bad numbers.
- if fail > 50 * walks:
+ if fail > 100 * walks:
warnings.warn("Random number generation appears to be "
"extremely inefficient. Adjusting the "
"scale-factor accordingly.")
|
TST: improve tox.ini
[CHANGED] invoke mpi4py futures correctly. To do this required ensuring
I correctly specify the tox venv environment python interpreter.
Other miscellaneous cleanup. | @@ -14,9 +14,7 @@ deps = py{36,37,38}: numba>0.48.0
scitrack
pandas
pytest-cov
- py37mpi: mpi4py
- py38mpi: mpi4py
- py39mpi: mpi4py
+ py{37mpi,38mpi,39mpi}: mpi4py
[testenv:py39]
changedir = tests
@@ -41,21 +39,21 @@ changedir = tests
basepython = python3.7
whitelist_externals = mpiexec
commands =
- mpiexec -n 1 pytest --junitxml=junit-{envname}.xml --cov-report xml --cov=cogent3 test_app/test_app_mpi.py
+ mpiexec -n 1 {envpython} -m mpi4py.futures -m pytest --junitxml=junit-{envname}.xml --cov-report xml --cov=cogent3 test_app/test_app_mpi.py
[testenv:py38mpi]
changedir = tests
basepython = python3.8
whitelist_externals = mpiexec
commands =
- mpiexec -n 1 pytest --junitxml=junit-{envname}.xml --cov-report xml --cov=cogent3 test_app/test_app_mpi.py
+ mpiexec -n 1 {envpython} -m mpi4py.futures -m pytest --junitxml=junit-{envname}.xml --cov-report xml --cov=cogent3 test_app/test_app_mpi.py
[testenv:py39mpi]
changedir = tests
basepython = python3.9
whitelist_externals = mpiexec
commands =
- mpiexec -n 1 pytest --junitxml=junit-{envname}.xml --cov-report xml --cov=cogent3 test_app/test_app_mpi.py
+ mpiexec -n 1 {envpython} -m mpi4py.futures -m pytest --junitxml=junit-{envname}.xml --cov-report xml --cov=cogent3 test_app/test_app_mpi.py
[testenv:py36]
changedir = tests
|
Decouple the cmd_restart from the actual restart
This is particularly useful if we want to restart without triggering the
configuration validation, for example after a xrandr. | @@ -261,6 +261,20 @@ class Qtile(CommandObject):
logger.debug('Stopping qtile')
self._stopped_event.set()
+ def restart(self):
+ argv = [sys.executable] + sys.argv
+ if '--no-spawn' not in argv:
+ argv.append('--no-spawn')
+ buf = io.BytesIO()
+ try:
+ pickle.dump(QtileState(self), buf, protocol=0)
+ except: # noqa: E722
+ logger.error("Unable to pickle qtile state")
+ argv = [s for s in argv if not s.startswith('--with-state')]
+ argv.append('--with-state=' + buf.getvalue().decode())
+ self._restart = (sys.executable, argv)
+ self.stop()
+
async def finalize(self):
self._eventloop.remove_signal_handler(signal.SIGINT)
self._eventloop.remove_signal_handler(signal.SIGTERM)
@@ -1098,19 +1112,7 @@ class Qtile(CommandObject):
logger.error("Preventing restart because of a configuration error: {}".format(error))
send_notification("Configuration error", str(error.__context__))
return
-
- argv = [sys.executable] + sys.argv
- if '--no-spawn' not in argv:
- argv.append('--no-spawn')
- buf = io.BytesIO()
- try:
- pickle.dump(QtileState(self), buf, protocol=0)
- except: # noqa: E722
- logger.error("Unable to pickle qtile state")
- argv = [s for s in argv if not s.startswith('--with-state')]
- argv.append('--with-state=' + buf.getvalue().decode())
- self._restart = (sys.executable, argv)
- self.stop()
+ self.restart()
def cmd_spawn(self, cmd):
"""Run cmd in a shell.
|
Make LaunchBar font configurable
When an icon can't be found, `LaunchBar` defaults to displaying
text. The text is formatted according to `base._TextBox`'s
defaults. This will pick up values set by `widget_defaults` but
doesn't allow individual customisation for this widget.
This PR fixes this issue by allowing users to configure the
appearance of text. | @@ -51,7 +51,10 @@ from libqtile.widget import base
class LaunchBar(base._Widget):
- """A widget that display icons to launch the associated command
+ """
+ A widget that display icons to launch the associated command.
+
+ Text will displayed when no icon is found.
Widget requirements: pyxdg_.
@@ -72,9 +75,13 @@ class LaunchBar(base._Widget):
("padding", 2, "Padding between icons"),
(
"default_icon",
- "/usr/share/icons/oxygen/256x256/mimetypes/" "application-x-executable.png",
+ "/usr/share/icons/oxygen/256x256/mimetypes/application-x-executable.png",
"Default icon not found",
),
+ ("font", "sans", "Text font"),
+ ("fontsize", None, "Font pixel size. Calculated if None."),
+ ("fontshadow", None, "Font shadow color, default is None (no shadow)"),
+ ("foreground", "#ffffff", "Text colour."),
]
def __init__(self, progs=None, width=bar.CALCULATED, **config):
@@ -124,10 +131,10 @@ class LaunchBar(base._Widget):
textbox._configure(self.qtile, self.bar)
textbox.layout = self.drawer.textlayout(
textbox.text,
- textbox.foreground,
- textbox.font,
- textbox.fontsize,
- textbox.fontshadow,
+ self.foreground,
+ self.font,
+ self.fontsize,
+ self.fontshadow,
markup=textbox.markup,
)
# the name will be displayed
|
BUG: Add missing DECREF in Py2 int() cast
The Long number is downcast to int if possible, so the old version
has to be DECREF'ed.
Thanks to Johannes Barthelmes for bisecting the offending commit.
Closes | @@ -1424,7 +1424,11 @@ static PyObject *
#ifndef NPY_PY3K
/* Invoke long.__int__ to try to downcast */
+ {
+ PyObject *before_downcast = long_result;
long_result = Py_TYPE(long_result)->tp_as_number->nb_int(long_result);
+ Py_DECREF(before_downcast);
+ }
#endif
return long_result;
|
Fix ctrl-\ behavior
This commit fixes two issues with ctrl-\ when using IPython on linux:
Previously, pressing ctrl-\ would make IPython exit without resetting the terminal configuration
IPython users could not override the behavior of ctrl-\ using `signal.signal(signal.SIGQUIT, ...)` as they would in other terminal apps | import signal
import sys
import re
+import os
from typing import Callable
@@ -56,7 +57,7 @@ def reformat_and_execute(event):
& insert_mode
))(reformat_and_execute)
- kb.add('c-\\')(force_exit)
+ kb.add("c-\\")(quit)
kb.add('c-p', filter=(vi_insert_mode & has_focus(DEFAULT_BUFFER))
)(previous_history_or_previous_completion)
@@ -458,10 +459,15 @@ def reset_search_buffer(event):
def suspend_to_bg(event):
event.app.suspend_to_background()
-def force_exit(event):
+def quit(event):
"""
- Force exit (with a non-zero return value)
+ On platforms that support SIGQUIT, send SIGQUIT to the current process.
+ On other platforms, just exit the process with a message.
"""
+ sigquit = getattr(signal, "SIGQUIT", None)
+ if sigquit is not None:
+ os.kill(0, signal.SIGQUIT)
+ else:
sys.exit("Quit")
def indent_buffer(event):
|
Fix missing call to logger converter.
If logger was not set and left to the default string 'bar', then it would have crashed when it tried to call `logger` as a function. See VideoClip.py:376 for correct equivalent. | @@ -6,6 +6,7 @@ out of VideoClips
import subprocess as sp
import os
import numpy as np
+from proglog import proglog
from moviepy.compat import PY3, DEVNULL
from moviepy.config import get_setting
@@ -200,6 +201,8 @@ def ffmpeg_write_video(clip, filename, fps, codec="libx264", bitrate=None,
""" Write the clip to a videofile. See VideoClip.write_videofile for details
on the parameters.
"""
+ logger = proglog.default_bar_logger(logger)
+
if write_logfile:
logfile = open(filename + ".log", 'w+')
else:
|
Bump cudf/cuml to 21.12
This is the newest version that supports both CUDA 11.X & Python 3.7 | @@ -89,7 +89,7 @@ RUN conda config --add channels nvidia && \
# b/232247930: uninstall pyarrow to avoid double installation with the GPU specific version.
RUN pip uninstall -y pyarrow && \
- conda install cudf=21.10 cuml=21.10 cudatoolkit=$CUDA_MAJOR_VERSION.$CUDA_MINOR_VERSION && \
+ conda install cudf=21.12 cuml=21.12 cudatoolkit=$CUDA_MAJOR_VERSION.$CUDA_MINOR_VERSION && \
/tmp/clean-layer.sh
{{ end }}
|
core: fix Dispatcher race introduced in
It must be constructed before are messages pumped. | @@ -2214,6 +2214,7 @@ class ExternalContext(object):
if self.config.get('setup_stdio', True):
self._setup_stdio()
+ self.dispatcher = Dispatcher(self)
self.router.register(self.parent, self.stream)
self.log_handler.uncork()
@@ -2222,7 +2223,6 @@ class ExternalContext(object):
self.parent, mitogen.context_id, os.getpid())
_v and LOG.debug('Recovered sys.executable: %r', sys.executable)
- self.dispatcher = Dispatcher(self)
self.dispatcher.run()
_v and LOG.debug('ExternalContext.main() normal exit')
except KeyboardInterrupt:
|
Update nasa-soho-comet-challenge-on-aws.yaml
added publications to dataset listing | @@ -18,3 +18,12 @@ DataAtWork:
Tutorials:
Tools & Applications:
Publications:
+ - Title: Topcoder NASA Comet Discovery: A Recap
+ URL: https://www.youtube.com/watch?v=E4OxTaqTP6E
+ AuthorName: TopCoder
+ - Title: Winners Selected for the NASA SOHO Comet Search with Artificial Intelligence Open-Science Challenge
+ URL: https://science.nasa.gov/science-news/winners-selected-for-the-nasa-soho-comet-search-with-artificial-intelligence-open-science-challenge
+ AuthorName: Denise Hill
+ - Title: Topcoder Challenge Finds Two New Comets For NASA
+ URL: https://www.topcoder.com/blog/topcoder-challenge-finds-two-new-comets-for-nasa/
+ AuthorName: Annika Nagy
|
Fix OneAccess.TDRE.get_metrics script
HG--
branch : feature/microservices | @@ -70,7 +70,7 @@ class Script(GetMetricsScript):
):
self.set_metric(
name=self.SLA_ICMP_RTT,
- value=delay,
+ value=int(delay * 1000000),
ts=ts,
tags={"probe": name}
)
@@ -83,7 +83,7 @@ class Script(GetMetricsScript):
):
self.set_metric(
name=self.SLA_UDP_RTT,
- value=delay,
+ value=int(delay * 1000000),
ts=ts,
tags={"probe": name}
)
@@ -95,7 +95,7 @@ class Script(GetMetricsScript):
):
self.set_metric(
name=self.SLA_JITTER,
- value=jitter,
+ value=int(jitter * 1000000),
ts=ts,
tags={"probe": name}
)
|
help_docs: Update `add-or-remove-users-from-a-stream` help doc.
Uses new `select-stream-view-subscribers.md` in instructions.
Also, adds a tip to bulk add users to a stream. | @@ -17,6 +17,8 @@ to a stream][configure-invites].
1. Select a stream.
+{!select-stream-view-subscribers.md!}
+
1. Under **Add subscribers**, enter a name or email address. The typeahead
will only include users who aren't already subscribed to the stream.
@@ -24,6 +26,11 @@ to a stream][configure-invites].
{end_tabs}
+!!! tip ""
+
+ To add users in bulk, you can copy members from an
+ existing stream or [user group](/help/user-groups).
+
## Remove users from a stream
{!admin-only.md!}
@@ -39,6 +46,8 @@ including streams the admin is not subscribed to.
1. Select a stream.
+{!select-stream-view-subscribers.md!}
+
1. Under **Subscribers**, find the user you would like to remove.
1. Click the **Unsubscribe** button in that row.
|
Add st_birthtime to struct stat
This is available on OS X, FreeBSD and NetBSD. On Linux, the definition
itself will not result in any errors. However, code that does attempt to
st_birthtime under Linux will now fail at C compile time rather than
Cython compile time. | @@ -18,6 +18,11 @@ cdef extern from "<sys/stat.h>" nogil:
time_t st_mtime
time_t st_ctime
+ # st_birthtime exists on *BSD and OS X.
+ # Under Linux, defining it here does not hurt. Compilation under Linux
+ # will only (and rightfully) fail when attempting to use the field.
+ time_t st_birthtime
+
# POSIX prescribes including both <sys/stat.h> and <unistd.h> for these
cdef extern from "<unistd.h>" nogil:
int fchmod(int, mode_t)
|
snow update link
* snow update link
* Update README.md
added more cmd example | @@ -5,7 +5,7 @@ IT service management. Cortex XSOAR interfaces with ServiceNow to help streamlin
- Query ServiceNow data with the ServiceNow query syntax.
- Manage Security Incident Response (SIR) tickets with Cortex XSOAR, update tickets and enrich them with data.
-Please refer to ServiceNow documentation for additional information. We especially recommend the [Operators available for filters and queries](https://docs.servicenow.com/bundle/istanbul-servicenow-platform/page/use/common-ui-elements/reference/r_OpAvailableFiltersQueries.html) page.
+Please refer to ServiceNow documentation for additional information. We especially recommend the [Operators available for filters and queries](https://docs.servicenow.com/bundle/sandiego-platform-user-interface/page/use/common-ui-elements/reference/r_OpAvailableFiltersQueries.html?cshalt=yes) page.
This integration was integrated and tested with the Orlando version of ServiceNow.
@@ -986,6 +986,7 @@ Queries the specified table in ServiceNow.
#### Command Example
```!servicenow-query-table table_name=alm_asset fields=asset_tag,sys_updated_by,display_name query=display_nameCONTAINSMacBook limit=4```
+```!servicenow-query-table table_name=sys_journal_field query=element_id=<SYS_ID>^ORDERBYsys_created_on limit=10 fields=value,name,element,sys_created_by,sys_created_on```
#### Context Example
```
|
hw DeviceMgr: speed-up client_for_keystore() for common-case
This method is often called when there is already an existing paired
client for the keystore, in which case we can avoid scan_devices() -
which would needlessly take several seconds. | @@ -511,10 +511,16 @@ class DeviceMgr(ThreadJob):
if handler is None:
raise Exception(_("Handler not found for") + ' ' + plugin.name + '\n' + _("A library is probably missing."))
handler.update_status(False)
+ pcode = keystore.pairing_code()
+ client = None
+ # search existing clients first (fast-path)
+ if not devices:
+ client = self.client_by_pairing_code(plugin=plugin, pairing_code=pcode, handler=handler, devices=[])
+ # search clients again, now allowing a (slow) scan
+ if client is None:
if devices is None:
devices = self.scan_devices()
- client = self.client_by_pairing_code(
- plugin=plugin, pairing_code=keystore.pairing_code(), handler=handler, devices=devices)
+ client = self.client_by_pairing_code(plugin=plugin, pairing_code=pcode, handler=handler, devices=devices)
if client is None and force_pair:
try:
info = self.select_device(plugin, handler, keystore, devices,
|
pull all services in parallel
This is done by default now | @@ -241,7 +241,7 @@ case $CMD in
esac
if [ "$CMD" == "test" ]; then
- docker-compose pull --parallel couch postgres redis elasticsearch kafka riakcs
+ docker-compose pull
docker-compose run --rm web run_tests "${TEST:-python}" "$@"
elif [ "$CMD" == "shell" ]; then
docker-compose run --rm web ./manage.py $CMD "$@"
|
Put get_questions on a long timeout
It varies by form source, so this should be safe to do. I tested this out
locally and it does invalidate properly, and interestingly, if you change the
form and then change it back, it'll hit the original cache, since the source is
the same. | @@ -1212,7 +1212,8 @@ class FormBase(DocumentSchema):
return xform.render()
@time_method()
- @quickcache(['self.source', 'langs', 'include_triggers', 'include_groups', 'include_translations'])
+ @quickcache(['self.source', 'langs', 'include_triggers', 'include_groups', 'include_translations'],
+ timeout=24 * 60 * 60)
def get_questions(self, langs, include_triggers=False,
include_groups=False, include_translations=False):
try:
|
BALD's isnan check is too late
Per-entropy isnan check. The current check is too late. | @@ -208,14 +208,21 @@ def get_bald_scores(logits, masks):
log_probs = jax.nn.log_softmax(logits)
probs = jax.nn.softmax(logits)
+ weighted_nats = -probs * log_probs
+ weighted_nats = jnp.where(jnp.isnan(weighted_nats), 0, weighted_nats)
+
+ marginal_entropy = jnp.mean(jnp.sum(weighted_nats, axis=-1), axis=1)
+
marginal_log_probs = jax.nn.logsumexp(log_probs, axis=1) - jnp.log(ens_size)
marginal_probs = jnp.mean(probs, axis=1)
- entropy_marginal = jnp.sum(-marginal_probs * marginal_log_probs, axis=-1)
- marginal_entropy = jnp.mean(jnp.sum(-probs * log_probs, axis=-1), axis=1)
- bald = entropy_marginal - marginal_entropy
+ weighted_marginal_nats = -marginal_probs * marginal_log_probs
+ weighted_marginal_nats = jnp.where(jnp.isnan(weighted_marginal_nats), 0, weighted_marginal_nats)
+
+ entropy_marginal = jnp.sum(weighted_marginal_nats, axis=-1)
- bald = jnp.where(jnp.isnan(bald), 0, bald)
+ # Mask results.
+ bald = entropy_marginal - marginal_entropy
bald = jnp.where(masks, bald, NINF_SCORE)
return bald
|
TST: set up TestMetaLabels
Cleaned up the TestMetaLabels class. | """Tests the pysat MetaLabels object."""
import logging
+import numpy as np
import pytest
import pysat
class TestMetaLabels(object):
- """Basic unit tests for the MetaLabels class."""
-
+ """Unit and integration tests for the MetaLabels class."""
def setup(self):
"""Set up the unit test environment for each method."""
- self.testInst = pysat.Instrument('pysat', 'testing')
- self.stime = pysat.instruments.pysat_testing._test_dates['']['']
- self.meta_labels = self.testInst.meta.labels
-
- self.label_dict = {'units': ('Units', str),
- 'name': ('Long_Name', str)}
- self.dval = None
- self.default_name = ['long_name']
- self.default_nan = ['fill', 'value_min', 'value_max']
- self.default_val = {'notes': '', 'units': '', 'desc': ''}
- self.frame_list = ['dummy_frame1', 'dummy_frame2']
+ testInst = pysat.Instrument('pysat', 'testing')
+ self.meta_labels = testInst.meta.labels
+ self.meta = pysat.Meta()
+
return
def teardown(self):
"""Clean up the unit test environment after each method."""
- del self.testInst, self.meta, self.stime, self.meta_labels
- del self.default_name, self.default_nan, self.default_val, self.dval
- del self.frame_list
+ del self.meta, self.meta_labels
return
# -----------------------
|
api/pupdevices/Light: drop blink, animate
External single-color lights won't have this for now. | @@ -556,7 +556,3 @@ Light
.. automethod:: pybricks.pupdevices.Light.on
.. automethod:: pybricks.pupdevices.Light.off
-
- .. automethod:: pybricks.pupdevices.Light.blink
-
- .. automethod:: pybricks.pupdevices.Light.animate
|
Fix padding type mismatch in gshard builder
Add num_packed_examples and batch_utilized_ratio eval metrics for gshard models | @@ -2359,7 +2359,8 @@ class UniTransformer(base_model.BaseTask):
def _ComputeNonPadding(self, input_batch):
if 'paddings' in input_batch.tgt:
- return 1.0 - input_batch.tgt.paddings
+ return tf.cast(1.0 - input_batch.tgt.paddings,
+ py_utils.FPropDtype(self.params))
non_padding = tf.cast(
tf.not_equal(input_batch.tgt.segment_ids, 0),
@@ -2455,9 +2456,11 @@ class UniTransformer(base_model.BaseTask):
tf.reduce_sum(non_padding))
avg_loss += p.aux_loss_coef * aux_loss
- non_padding_nor_eos = tf.where(
- tf.equal(input_batch.tgt.labels, 1),
- tf.zeros_like(non_padding, dtype=non_padding.dtype), non_padding)
+ num_items_in_batch = tf.reduce_sum(
+ tf.reduce_max(input_batch.tgt.segment_ids, axis=1))
+ num_nonpadding = tf.reduce_sum(
+ _ToInt32(tf.not_equal(input_batch.tgt.segment_ids, 0)))
+ batch_capacity = tf.size(input_batch.tgt.labels)
whole_tgt_correct = tf.cast(
tf.equal(
@@ -2471,12 +2474,11 @@ class UniTransformer(base_model.BaseTask):
}
eval_metrics = {
+ 'num_packed_examples': (num_items_in_batch, 1.0),
+ 'batch_utilized_ratio': (num_nonpadding / batch_capacity, 1.0),
'acc1':
(tf.reduce_sum(acc1 * non_padding) / tf.reduce_sum(non_padding),
tf.reduce_sum(non_padding)),
- 'acc1_excluding_eos': (tf.reduce_sum(acc1 * non_padding_nor_eos) /
- tf.reduce_sum(non_padding_nor_eos),
- tf.reduce_sum(non_padding_nor_eos)),
'whole_tgt_accuracy':
(tf.reduce_sum(whole_tgt_correct) /
tf.cast(whole_tgt_correct.shape[0], whole_tgt_correct.dtype), 1.0
|
Remove suggestion to switch to FastAPI branch
Branch does not exist | @@ -28,7 +28,6 @@ Download this repo and install the dependencies:
git clone https://github.com/lnbits/lnbits-legend.git
cd lnbits-legend/
# ensure you have virtualenv installed, on debian/ubuntu 'apt install python3-venv' should work
-# for now you'll need to `git checkout FastAPI`
python3 -m venv venv
./venv/bin/pip install -r requirements.txt
cp .env.example .env
|
Store windows in list instead of set, fixes
Since set is unordered, output of TaskList like widget was unpredictable.
Before:
window 4 | window 1 | window 3| window 2 |
After:
window 1 | window 2 | window 3 | window 4 | | @@ -44,7 +44,7 @@ class _Group(CommandObject):
self.name = name
self.label = name if label is None else label
self.custom_layout = layout # will be set on _configure
- self.windows = set()
+ self.windows = []
self.qtile = None
self.layouts = []
self.floating_layout = None
@@ -61,7 +61,7 @@ class _Group(CommandObject):
self.screen = None
self.current_layout = 0
self.focus_history = []
- self.windows = set()
+ self.windows = []
self.qtile = qtile
self.layouts = [i.clone(self) for i in layouts]
self.floating_layout = floating_layout
@@ -235,7 +235,8 @@ class _Group(CommandObject):
def add(self, win, focus=True, force=False):
hook.fire("group_window_add", self, win)
- self.windows.add(win)
+ if win not in self.windows:
+ self.windows.append(win)
win.group = self
if self.qtile.config.auto_fullscreen and win.wants_to_fullscreen:
win._float_state = FloatStates.FULLSCREEN
|
change: add a function in .cli.parse_args to get parser and parsed result
Add a function .cli.parse_args.parse to get an argument parser and its
parsed result. | @@ -97,4 +97,13 @@ def make_parser(defaults: typing.Optional[typing.Dict] = None
help='Verbose mode; -v or -vv (more verbose)')
return apsr
+
+def parse(argv: typing.List[str]
+ ) -> typing.Tuple[argparse.ArgumentParser, argparse.Namespace]:
+ """
+ Parse given arguments ``argv`` and return it with the parser.
+ """
+ psr = make_parser()
+ return (psr, psr.parse_args(argv))
+
# vim:sw=4:ts=4:et:
|
email tooltip: Adjust background color of email toolip in dark-mode.
This commit adjust the email tooltip of popover in dark-mode. | @@ -399,6 +399,14 @@ body.dark-mode #out-of-view-notification {
border: 1px solid 1px solid hsl(144, 45%, 62%);
}
+body.dark-mode .email_tooltip {
+ background-color: #404c59;
+}
+
+body.dark-mode .email_tooltip:after {
+ border-bottom-color: #404c59 !important;
+}
+
@-moz-document url-prefix() {
body.dark-mode #settings_page select {
background-color: hsla(0, 0%, 0%, 0.2);
|
Skip install prereqs when installing stable
The current stable release (1.8.8) breaks an existing Docker 1.13
install with `--install-prereqs`. See | @@ -333,7 +333,9 @@ class VpcClusterUpgradeTest:
with logger.scope("install dcos"):
# Use the CLI installer to set exhibitor_storage_backend = zookeeper.
- test_util.cluster.install_dcos(cluster, self.stable_installer_url, api=False,
+ # Don't install prereqs since stable breaks Docker 1.13. See
+ # https://jira.mesosphere.com/browse/DCOS_OSS-743.
+ test_util.cluster.install_dcos(cluster, self.stable_installer_url, api=False, install_prereqs=False,
add_config_path=self.config_yaml_override_install)
master_list = [h.private_ip for h in cluster.masters]
|
Update issue templates
A little sizing and reordering. | @@ -9,28 +9,32 @@ assignees: ''
<!--- Please search existing bugs before creating a new one -->
-# Bug Report
-**Install Source:**
+
+### Environment
+- **System OS:** <!--- Windows/OSX/Linux/Heroku/Docker -->
+
+- **Python Version:** <!--- Python Version can be found by running "py -V" -->
+
+- **Install Source:**
<!--- Did you download from pip, the master branch on github, or a different branch?) -->
-**Version:**
+- **Version OR Commit hash:**
<!--- If from pip, what is the version? Run "pip show spotdl" -->
-**OR Commit hash:**
<!--- If not from pip, what is the commit hash? -->
-## Expected Behavior
+------------------------------------------------------------
+
+### Expected Behaviour
<!--- What did you expect to happen? -->
-## Actual Behavior
+### Actual Behaviour
<!--- What actually happened? -->
-## Steps to Reproduce
+### Steps to Reproduce
1.
2.
3.
-## Environment
-**System OS:** <!--- Windows/OSX/Linux/Heroku/Docker -->
-**Python Version:** <!--- Python Version can be found by running "py -V" -->
+--------------------------------------------------------------------
-## Any other information:
+### Any other information:
|
Minor tweak to Tools.xml - correct referrring text [ci skip]
In two places a reference to a section is made using "below"
implying the section is in same doc, but Tools.xml definitions are
distributed to both docs and this is incorrect for User Guide.
Use "in manpage" instead. | @@ -282,8 +282,8 @@ Note that the source files will be scanned
according to the suffix mappings in the
<classname>SourceFileScanner</classname>
object.
-See the section "Scanner Objects,"
-below, for more information.
+See the manpage section "Scanner Objects"
+for more information.
</para>
</summary>
</builder>
@@ -387,8 +387,8 @@ Note that the source files will be scanned
according to the suffix mappings in the
<classname>SourceFileScanner</classname>
object.
-See the section "Scanner Objects,"
-below, for more information.
+See the manpage section "Scanner Objects"
+for more information.
</para>
</summary>
</builder>
|
exclude some caffe2 modules from libtorch mobile build
Summary:
Pull Request resolved:
ghimport-source-id:
Test Plan:
verified libtorch mobile library builds and links successfully;
Imported from OSS | @@ -59,14 +59,18 @@ endif()
# addressed yet.
if (NOT BUILD_ATEN_ONLY)
- add_subdirectory(proto)
- add_subdirectory(contrib)
add_subdirectory(core)
+ add_subdirectory(proto)
+ add_subdirectory(serialize)
add_subdirectory(utils)
+ add_subdirectory(perfkernels)
+
+ # Skip modules that are not used by libtorch mobile yet.
+ if (NOT INTERN_BUILD_MOBILE OR BUILD_CAFFE2_MOBILE)
+ add_subdirectory(contrib)
add_subdirectory(predictor)
add_subdirectory(predictor/emulator)
add_subdirectory(core/nomnigraph)
- add_subdirectory(serialize)
if (USE_NVRTC)
add_subdirectory(cuda_rtc)
endif()
@@ -92,7 +96,6 @@ if(NOT BUILD_ATEN_ONLY)
endif()
endif()
add_subdirectory(opt)
- add_subdirectory(perfkernels)
add_subdirectory(python)
add_subdirectory(queue)
add_subdirectory(sgd)
@@ -100,6 +103,7 @@ if(NOT BUILD_ATEN_ONLY)
# add_subdirectory(test) # todo: use caffe2_gtest_main instead of gtest_main because we will need to call GlobalInit
add_subdirectory(transforms)
endif()
+endif()
# Advanced: if we have white list specified, we will do intersections for all
# main lib srcs.
|
Fix a bug dependent on glibc version
I changed adam.py so that bugs that depend on the version of glibc do not occur. | @@ -76,8 +76,8 @@ class AdamRule(optimizer.UpdateRule):
@property
def lr(self):
- fix1 = 1. - self.hyperparam.beta1 ** self.t
- fix2 = 1. - self.hyperparam.beta2 ** self.t
+ fix1 = 1. - math.pow(self.hyperparam.beta1, self.t)
+ fix2 = 1. - math.pow(self.hyperparam.beta2, self.t)
return self.hyperparam.alpha * math.sqrt(fix2) / fix1
@@ -116,6 +116,6 @@ class Adam(optimizer.GradientMethod):
@property
def lr(self):
- fix1 = 1. - self.hyperparam.beta1 ** self.t
- fix2 = 1. - self.hyperparam.beta2 ** self.t
+ fix1 = 1. - math.pow(self.hyperparam.beta1, self.t)
+ fix2 = 1. - math.pow(self.hyperparam.beta2, self.t)
return self.hyperparam.alpha * math.sqrt(fix2) / fix1
|
update up.sh
Fix up.sh to point to binpash | @@ -7,7 +7,7 @@ set -e
# will install dependencies locally.
PLATFORM=$(uname | tr '[:upper:]' '[:lower:]')
-URL='https://github.com/andromeda/pash/archive/refs/heads/main.zip'
+URL='https://github.com/binpash/pash/archive/refs/heads/main.zip'
VERSION='latest'
DL=$(command -v curl >/dev/null 2>&1 && echo curl || echo 'wget -qO-')
@@ -21,7 +21,7 @@ if [ "$PLATFORM" = "darwin" ]; then
fi
set +e
-git clone [email protected]:andromeda/pash.git
+git clone [email protected]:binpash/pash.git
if [ $? -ne 0 ]; then
echo 'SSH clone failed; attempting HTTPS'
git clone https://github.com/andromeda/pash.git
|
Update simulation.py
To increase readability, make use of pandas index method `get_loc`
instead of relying on `np.where`. | @@ -26,7 +26,7 @@ def get_dynamics(adata, key="fit", extrapolate=False, sorted=False, t=None):
def compute_dynamics(
adata, basis, key="true", extrapolate=None, sort=True, t_=None, t=None
):
- idx = np.where(adata.var_names == basis)[0][0] if isinstance(basis, str) else basis
+ idx = adata.var_names.get_loc(basis) if isinstance(basis, str) else basis
key = "fit" if f"{key}_gamma" not in adata.var_keys() else key
alpha, beta, gamma, scaling, t_ = get_vars(adata[:, basis], key=key)
@@ -95,7 +95,7 @@ def show_full_dynamics(
)
(line,) = ax.plot(st, ut, color=color, linewidth=linewidth, label=label)
- idx = np.where(adata.var_names == basis)[0][0]
+ idx = adata.var_names.get_loc(basis)
beta, gamma = adata.var[f"{key}_beta"][idx], adata.var[f"{key}_gamma"][idx]
xnew = np.linspace(np.min(st), np.max(st))
ynew = gamma / beta * (xnew - np.min(xnew)) + np.min(ut)
@@ -135,7 +135,7 @@ def simulation(
1, ncols, pl.figure(None, (figsize[0] * ncols, figsize[1]), dpi=dpi)
)
):
- idx = np.where(adata.var_names == var_names[i])[0][0]
+ idx = adata.var_names.get_loc(var_names[i])
alpha, ut, st = compute_dynamics(adata, idx)
t = (
adata.obs[xkey]
|
Fixes
Ensure if target_edges is a list, we don't run into a
`UnboundLocalError` because `edges` is not defined | #
# ================================================================================================
+try:
+ import collections.abc as abc
+except ImportError:
+ import collections as abc
+
import dwave_networkx as dnx
import networkx as nx
@@ -90,9 +95,9 @@ def find_clique_embedding(k, m, n=None, t=None, target_edges=None):
elif len(nodes) == 2:
# If k == 2 we simply return two one-qubit chains that are the endpoints of a randomly sampled coupler.
- if not isinstance(target_edges, list):
- edges = list(target_edges)
- edge = edges[random.randrange(len(edges))]
+ if not isinstance(target_edges, abc.Sequence):
+ target_edges = list(target_edges)
+ edge = random.choice(target_edges)
embedding = [[edge[0]], [edge[1]]]
else:
|
Update ppo_cart_pole.gin
with reward scaling and mini_batch_size=128, it will stably convert to 300 in 30 seconds | @@ -4,6 +4,10 @@ include 'ppo.gin'
create_environment.env_name="CartPole-v0"
create_environment.num_parallel_environments=8
+# reward scaling
+ActorCriticAlgorithm.reward_shaping_fn = @reward_scaling
+common.reward_scaling.scale = 0.01
+
# algorithm config
PPOLoss.entropy_regularization=1e-4
PPOLoss.gamma=0.98
@@ -16,7 +20,7 @@ create_ac_algorithm.learning_rate=0.001
# training config
off_policy_trainer.train.mini_batch_length=1
off_policy_trainer.train.num_steps_per_iter=256
-off_policy_trainer.train.mini_batch_size=256
+off_policy_trainer.train.mini_batch_size=128
off_policy_trainer.train.num_iterations=1000
off_policy_trainer.train.summary_interval=5
off_policy_trainer.train.checkpoint_interval=100000
|
Update test_ec2_role_crud to match recent aws auth resp parameters
See current response parameters at:
This behavior was changed in vault 0.9.6.
See:
"[...] to keep consistency with input and output, when reading a role the binds will now be
returned as string arrays rather than strings." | @@ -854,21 +854,21 @@ class IntegrationTest(TestCase):
assert ('qux' in roles['data']['keys'])
foo_role = self.client.get_ec2_role('foo')
- assert (foo_role['data']['bound_ami_id'] == 'ami-notarealami')
+ assert (foo_role['data']['bound_ami_id'] == ['ami-notarealami'])
assert ('ec2rolepolicy' in foo_role['data']['policies'])
bar_role = self.client.get_ec2_role('bar')
- assert (bar_role['data']['bound_account_id'] == '123456789012')
+ assert (bar_role['data']['bound_account_id'] == ['123456789012'])
assert ('ec2rolepolicy' in bar_role['data']['policies'])
baz_role = self.client.get_ec2_role('baz')
- assert (baz_role['data']['bound_iam_role_arn'] == 'arn:aws:iam::123456789012:role/mockec2role')
+ assert (baz_role['data']['bound_iam_role_arn'] == ['arn:aws:iam::123456789012:role/mockec2role'])
assert ('ec2rolepolicy' in baz_role['data']['policies'])
qux_role = self.client.get_ec2_role('qux')
assert (
- qux_role['data']['bound_iam_instance_profile_arn'] == 'arn:aws:iam::123456789012:instance-profile/mockprofile')
+ qux_role['data']['bound_iam_instance_profile_arn'] == ['arn:aws:iam::123456789012:instance-profile/mockprofile'])
assert ('ec2rolepolicy' in qux_role['data']['policies'])
# teardown
|
Add ?next= param for GitHub login
This fixes by adding `?next=/asd` query for GitHub login button.
It will redirect to pointed resource after successful log in. | <div class="col-12 col-md-5">
<h4>For The Carpentries Instructors</h4>
- <p><a class="btn btn-primary w-100" href="/login/github/"><i class="fab fa-github"></i> Log in with your GitHub account</a></p>
+ <p><a class="btn btn-primary w-100" href="{% url 'social:begin' 'github' %}{% if next %}?next={{ next }}{% endif %}"><i class="fab fa-github"></i> Log in with your GitHub account</a></p>
<p>Having trouble logging in? Contact us at <a href="mailto:[email protected]">[email protected]</a>.</p>
</div>
</div>
|
[nixio] Add close() function
Closes underlying nix file and cleans up maps and read_block tracking.
__del__ added that calls the close function. | @@ -16,7 +16,7 @@ This IO supports both writing and reading of NIX files. Reading is supported
only if the NIX file was created using this IO.
"""
-from __future__ import absolute_import, print_function
+from __future__ import absolute_import
import time
from datetime import datetime
@@ -1289,3 +1289,19 @@ class NixIO(BaseIO):
strupdate(type(obj).__name__)
return objhash.hexdigest()
+
+ def close(self):
+ """
+ Closes the open nix file and resets maps.
+ """
+ if (hasattr(self, "nix_file") and
+ self.nix_file and self.nix_file.is_open()):
+ self.nix_file.close()
+ self.nix_file = None
+ self._object_map = None
+ self._lazy_loaded = None
+ self._object_hashes = None
+ self._block_read_counter = None
+
+ def __del__(self):
+ self.close()
|
Add the propose of the inssue
Add a balloon tip with the price break.
Missing yet some information when is more convenient buy the next price break and not your intended quantity. | @@ -1213,6 +1213,22 @@ def add_dist_to_worksheet(wks, wrk_formats, index, start_row, start_col,
purch_qty=xl_rowcol_to_cell(row, purch_qty_col),
qtys=','.join(qtys),
prices=','.join(prices)), wrk_formats['currency'])
+ # Add comment if the price break
+ price_break_info = 'Price break:\n'
+ for count_price_break in range(0,len(qtys)):
+ if int(qtys[count_price_break]) != 0: # 0 qnty information is not userful to show.
+ price_break_info += qtys[count_price_break] + ' - ' + '$' + prices[count_price_break] + '\n'
+ wks.write_comment( row, unit_price_col, price_break_info[:-1])
+ # Conditional format, if the price of the next price break is less than the actual
+ # unity price by the quantity chosen, put the unit price red.
+# wks.conditional_format(row, unit_price_col, row, unit_price_col, { #TODO
+# 'type': 'cell',
+# 'criteria': '<=',
+# 'value': xl_rowcol_to_cell(row, 7),
+# # This is the global data cell holding the minimum unit price for this part.
+# 'format': wrk_formats['best_price']
+# })
+
# Conditionally format the unit price cell that contains the best price.
wks.conditional_format(row, unit_price_col, row, unit_price_col, {
'type': 'cell',
|
[Stress Tester XFails] Remove and apple/swift#36943
The issues have been fixed | ],
"issueUrl" : "https://bugs.swift.org/browse/SR-8898"
},
- {
- "path" : "*\/MovieSwift\/MovieSwift\/MovieSwift\/views\/components\/bottomMenu\/BottomMenu.swift",
- "issueDetail" : {
- "kind" : "codeComplete",
- "offset" : 1723
- },
- "applicableConfigs" : [
- "main"
- ],
- "issueUrl" : "https://bugs.swift.org/browse/SR-14494"
- },
- {
- "path" : "*\/MovieSwift\/MovieSwift\/Packages\/Backend\/Sources\/Backend\/services\/ImageService.swift",
- "issueDetail" : {
- "kind" : "codeComplete",
- "offset" : 1093
- },
- "applicableConfigs" : [
- "main"
- ],
- "issueUrl" : "https://github.com/apple/swift/pull/36943"
- },
{
"path" : "*\/ACHNBrowserUI\/ACHNBrowserUI\/ACHNBrowserUI\/packages\/UI\/Sources\/UI\/ProgressView.swift",
"issueDetail" : {
|
DOC: updated changelog
Updated changelog with a description of the changes in this branch. | @@ -26,8 +26,11 @@ This project adheres to [Semantic Versioning](https://semver.org/).
docstrings from an instantiated Instrument in an interactive session
* Bug Fix
* Fixed default MetaLabel specification in `pysat.utils.load_netcdf4`
+ * Fixed `parse_delimited_filename` output consistency and ability to handle
+ leading and trailing non-parsed text in filenames (e.g., file extensions)
* Maintenance
* Added missing unit tests for `pysat.utils.time`
+ * Added missing unit tests for `pysat.utils.file.parse_delimited_filename`
[3.0.1] - 2021-XX-XX
--------------------
|
Add link to GitHub discussion forum to docs
Update "Getting help" topic to include a link
to the new public discussion forum on GitHub. | @@ -19,6 +19,10 @@ limitations under the License.
Thank you for your interest in Elyra!
+### General questions
+
+Share your questions and ideas with the community in the [GitHub discussion forum](https://github.com/elyra-ai/elyra/discussions).
+
### Create an issue or feature request
If you encounter a problem or have suggestions for improvements please [open an issue on GitHub](https://github.com/elyra-ai/elyra/issues).
|
Subsets and Splits