[ENH] Implement Results Reader #130

DanKotlyar · 2018-05-09T16:43:42Z

The results reader parser, jupyter example notebook and unittest require review.

Attempt to remove the amount of files "changed" due to changing line endings. Automatically convert line endings to \n upon commit.

results.py - parser test_ResultsReader - unittest pwr_res.m, pwr_res_filter.m files used by the test ResultsReader.ipynb - jupyter notebook InnerAssembly_res.m - used by .ipynb file

drewejohnson · 2018-05-09T18:41:39Z

As for the travis build, the third cell (This notebook demonstrates .....) needs to be converted to Markdown so that it gets commented out when converted to python via testNotebooks.sh

drewejohnson · 2018-05-09T18:42:27Z

examples/ResultsReader.ipynb

+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# #  Basic operations"


Extra space here is what is blocking the subtitle formatting

drewejohnson · 2018-05-09T18:42:44Z

examples/ResultsReader.ipynb

+   ]
+  },
+  {
+   "cell_type": "raw",


This cell needs to be markdown, not raw

Most updated develop (#132) is merged into results-reader-3

drewejohnson

This is a fantastic improvement over the previous iteration! Largely, I think we need more test cases to cover all the possible uses of this file. More below

Things blocking approval

Fix travis build
More test cases - problem with no universes and problem with no time dependence
Fewer minor attributes stored on ResultsReader object
More commentary in jupyter notebook, rather than just blocks of code
Convert jupyter notebook to rst and include in documentation
Pre-compile regular expressions
Merge latest changes from develop

More detail

Fewer attributes stored on reader

I think a lot of the methods you use to store positional data, varNames varValues, posFile, etc., are not needed and should be returned from their various methods. My reasoning is largely that we don't care about these values after the file has been read and they used very scarcely, most by a single method. These can then be passed to the various _storeX methods.

More test cases

I noticed that the reader seems intended for files with group constant data over time, and the tests largely reflect that. But this is not the entire scope of use for the results file. I would like to see two additional test files and test cases: one with no universes but with time dependence, and another with no universes nor time dependence. I left some comments on test_ResultsReader,py that should make writing the tests easier for you, since the execution of the tests is largely consistent across all test cases.

Pre-compiled regular expressions

There are a lot of regular expression calls here, and many for good reason. I believe we can speed this up by storing compiled regular expressions, i.e. re.compile(r'^\w+'), either outside the class, or as a class attribute,

class ResultsReader(---):
    _regexScalar = re.compile(----)

These can still call the search and match functions with
self._regexScalar.search(tline). This approach saves a little bit of time spent creating and destroying the regular expressions.

Documentation

The notebook looks good, but I get lost towards the end when vast amounts of code are presented with little to no commentary. These notebooks should be easy to understand for someone who is just picking up the ResultsReader. Furthermore, the notebook needs to be converted and added to documentation.

drewejohnson · 2018-05-10T15:05:30Z

serpentTools/parsers/results.py

    """
    Parser responsible for reading and working with result files.

+     .. note::
+
+    Branching points for unique burnup steps and universes,


Remove this block and add more specific documentation on the results reader, such as the attributes

Attributes ------------ metadata: ... univdata: ...

Rather than describe these variables via comments, describe them in the docstring.

Link to docstrings in developer guide

Attributes ------------ metadata: dict Dictionary with serpent descriptive variables as keys and the corresponding description. Within the _res.m file this data is printed multiple times, but contains the same description and thus is stored only once. e.g., 'version': 'Serpent 2.1.29' resdata: dict Dictionary with serpent time-dependent variables as keys and the corresponding values. The data is unique for each burnup step. Some variables also contain uncertainties. e.g., 'absKeff': [[9.91938E-01, 0.00145],[1.81729E-01, 0.00240]] univdata: dict Dictionary of universe names and their corresponding :py:class:`~serpentTools.objects.containers.HomogUniv` objects. The keys describe a unique state: 'universe', burnup (MWd/kg), burnup index, time (days) ('0', 0.0, 1, 0.0)

Exactly 👍 That block should be included in the ResultsReader docstring

drewejohnson · 2018-05-10T15:06:10Z

serpentTools/parsers/results.py

-        path to the depletion file
+        path to the results file
+
+    Errors


Raises --------- <Type of error that is raised>: <Why this error is raised>

Raises ------ SerpentToolsException: Serpent version is not supported No universes are found in the file No results are collected Corrupted results IOError: file is unexpectedly closes while reading

drewejohnson · 2018-05-10T15:06:38Z

serpentTools/parsers/results.py

+        XSReader.__init__(self, filePath, 'xs')
+        self.__fileObj = None
+        self._counter = {'meta': 0, 'rslt': 0, 'univ': 0}
+        self._univlist = list()


Create an empty list with []

Changed to:
self._univlist = []

drewejohnson · 2018-05-10T15:12:18Z

serpentTools/parsers/results.py

+                    'inxs': 'INF_',
+                    'b1xs': 'B1_'}
+        MapStrVersions = {}
+        for iVer in range(10, 31):


I like the intent here, but this should be stripped down. This creates 10 sub-dictionaries, potentially modifies one of them, and then only uses one of them depending upon the serpent version. Since we know what versions we support, and what version the file is from the user and the _precheck method, we can better create and utilize this dictionary

A (nested) dictionary is now explicitly defined. The mapping is done for each version.

MapStrVersions = {'2.1.29': {'meta': 'VERSION ', 'rslt': 'MIN_MACROXS', 'univ': 'GC_UNIVERSE_NAME', 'days': 'BURN_DAYS', 'burn': 'BURNUP', 'inxs': 'INF_', 'b1xs': 'B1_'}, '2.1.30': {'meta': 'VERSION ', 'rslt': 'MEAN_SRC_WGT', 'univ': 'GC_UNIVERSE_NAME', 'days': 'BURN_DAYS', 'burn': 'BURNUP', 'inxs': 'INF_', 'b1xs': 'B1_'}}

drewejohnson · 2018-05-10T15:13:02Z

serpentTools/parsers/results.py

+            self._keysVersion = MapStrVersions[rc['serpentVersion']]
+        else:
+            raise SerpentToolsException("Version {} is not supported by the "
+                                        "ResultsReader".format(rc['serpentVersion']))


This exception should really be raised in the _precheck method, as well as ensuring that the version presented by the user agrees with that found in the input file

This exception was moved to the _precheck. In addition, SerpentToolsException is raised when the version defined in the setting is not matching the actual serpent version (obtained from the file).

drewejohnson · 2018-05-10T16:04:01Z

serpentTools/parsers/results.py

+        if divmod(self._counter['meta'],self._counter['univ'])[0] != self._counter['rslt']:
+            debug('  found {} universes, {} time-points, and {} overall result points'.
+                  format(self._counter['univ'], self._counter['rslt'], self._counter['meta']))
+            raise SerpentToolsException("The file {} is not complete ".format(self.filePath))


Would this be raised if not universes are present in the file, but the file is in fact complete?

This was not changed, however the postcheck function was slightly modified:
if not self.resdata and not self.metadata and not self.univdata:
raise SerpentToolsException("No results were collected "
"from {}".format(self.filePath))

I think that the debug information should be included in the exception to give information as to why we believe the file to be incomplete.

The debug info is now included in the exception.

raise SerpentToolsException( "The file {} is not complete. The reader found {} universes, " "{} time-points, and {} overall result points ".format(self.filePath, self._counter['univ'], self._counter['rslt'], self._counter['meta']))

drewejohnson · 2018-05-10T16:04:56Z

serpentTools/tests/test_ResultsReader.py

+from serpentTools.messages import SerpentToolsException
+
+class _ResultsTesterHelperFull(unittest.TestCase):
+    """Test case to share setUpClass for testing Results Reader"""


Change the docstrings and names of these helpers to reflect their purpose and scope

""" Class with common tests for the results reader. Expected failures/errors: 1. test_varsMatchSettings: compares the keys defined by the user to those obtained by the reader Raises SerpentToolsException 2. test_metadata: Check that metadata variables and their values are properly stored Raises SerpentToolsException 3. test_resdata: Check that time-dependent results variables and their values are properly stored Raises SerpentToolsException 4. test_univdata: Check that expected states are read i.e., ('univ', bu, buIdx, days) For a single state, check that infExp keys and values are stored. Check that infUnc and metadata are properly stored Raises SerpentToolsException """

drewejohnson · 2018-05-10T16:07:12Z

serpentTools/tests/test_ResultsReader.py

+            badReader.read()
+        os.remove(badFile)
+
+    def test_variables(self):


But what variables are being tested? Change the name to reflect the scope and effect of this test

def test_varsMatchSettings(self):

drewejohnson · 2018-05-10T16:07:26Z

serpentTools/tests/test_ResultsReader.py

+class ResultsTesterFull(_ResultsTesterHelperFull):
+    """Class for testing the results reader."""
+
+    def test_raiseError(self):


But what errors are being tested? Change the name to reflect the scope and effect of this test

drewejohnson · 2018-05-10T16:14:18Z

serpentTools/tests/test_ResultsReader.py

@@ -0,0 +1,233 @@
+"""Test the results reader."""


General comment for these readers. The way the Testers are written, there is no need for the helpers. One specific helper is used to build one specific tester. Since the tests are largely the same, just the data compared is different, the actual tests should be put inside a single helper, with different subclasses that test a specific use case. Minimal example

class _Helper(unittest.TestCase): def test_1(self): <do test 1 using instance attributes> def test_2(self): <do test 2 using instance attributes> ... class Case1Tester(_Helper): @classmethod def setupclass(cls): <get the reader using the specific settings> <set a variety of expected results to be tested against using methods on _Helper> <repeat for all test cases, e.g. full file, filtered file, file with no universes, file with no time-dependence, etc>

Looking great!

DanKotlyar · 2018-05-10T17:27:27Z

Thanks for the comments and suggestions. I'll try to implement these quickly.
Regarding the universe related issues, I think that we can really rely on having this information simply since this is the essence of the results file.

drewejohnson · 2018-05-11T12:20:23Z

Just ran a simple test problem with no explicit set gcu parameter using 2.1.30. Group constants for root universe were in the results file. I now agree that we can safely assume the results file will always have some universe data. A unit test for this case is no longer needed for approval

all the files from examples and the doc/welcome/license

Run `git revert develop <>` on all images in docs/examples/images, and on a few other files modified by commit 679e692

results.py test_ResultsReader pwr_res_noUniv.m for testing

test_ResultsReader.py includes 14 test cases. parser is able to read with and without burnups .ipynb is not ready pwr_res_noBU.m was added for testing

…eader-3

…ues with settings. Jupyter notebook has issues to upload all the settings.

drewejohnson

Thank you for implementing the changes I requested. They look great!

Things blocking approval

Fixing jupyter notebook - also part of Unable to import all settings from jupyter #140
Ability to store uncertainties for items like CMM_DIFFCOEF and CMM_TRANSPXS
Convert notebook to rst and include in documentation
Document results reader in docs/api
Use universes rather than univdata on reader
Explicitly store expected data for testing as sets where appropriate
Better way to extract universes

More Detail

`CMM` and non-metadata related XS

The way in which data is passed to HomogUniv objects is by checking if the variable name essentially begins with INF or B1. If so, we assume that the data has uncertainties and pass that to the universe. Otherwise, we don't assume there are uncertainties.

However, there are some group constants that don't start with those two strings that do contain uncertainties. This information needs to be passsed to the universe object.

Documentation

Once completed, the notebook needs to be converted to rst following the Examples guide in the dev docs. Also need to document the reader itself in docs/api following the guide in API - dev docs.

Also, the dictionary MapStrVersions seems very important and could require changes as we begin supporting new versions of SERPENT. Include some information in docs/develop/serpentVersions.rst under a new section in Version with New File Formats so that we can easily determine what we need to change for supporting previous results files

Better way to extract universes

Some of the universes have very long keys, that must be entered exactly in order to properly extract the universe. From the jupyter notebook, we can see one universe would require a key of ('3102', 0.10000000000000001, 2, 1.20048) to be obtained.

Implementing something like the BranchContainer.getUniv method can make it easier to extract a universe given only the universe ID and the index.

drewejohnson · 2018-05-15T12:35:06Z

serpentTools/parsers/results.py

+        corresponding values. The data is unique for each burnup step.
+        Some variables also contain uncertainties.
+        e.g., 'absKeff': [[9.91938E-01, 0.00145],[1.81729E-01, 0.00240]]
+    univdata: dict


Rename univdata to universes

drewejohnson · 2018-05-15T12:35:56Z

serpentTools/parsers/results.py

+                            No universes are found in the file
+                            No results are collected
+                            Corrupted results
+    IOError: file is unexpectedly closes while reading
    """

    def __init__(self, filePath):


What is the motivation behind removing the declaration of metadata, resdata, and univdata from the __init__ method? This behavior is preferred

It is included.
self.metadata, self.resdata, self.universes = {}, {}, {}

My bad, I was only looking at a partial diff when I made that. Great change

drewejohnson · 2018-05-15T12:38:43Z

serpentTools/parsers/results.py

                return
-            self._getVarValues(tline)  # values to be stored
+            varType, varVals = self._getVarValues(tline)  # values to be stored
            if self._posFile == 'meta':


I don't see self._posFile being used outside of this method. Can you have whereAmI return this value after performing all the checks?

if self._posFile == 'meta': self._storeMetaData(varNamePy, varType, varVals) if self._posFile == 'rslt': self._storeResData(varNamePy, varVals) if self._posFile == 'univ': self._storeUnivData(varNameSer, varVals)

The idea is that the keyword is not found 'self._posFile' remains as it was. The function whereAmI either return the position or does not update the existing 'self._posFile'

drewejohnson · 2018-05-15T12:40:40Z

docs/welcome/license.rst

-=======
-
-.. include:: ../../LICENSE
+.. _license:


Reset this file to develop. This may have been introduced by some of my pushed changes through 😬 I will look in to this

The file was reset

drewejohnson · 2018-05-15T12:41:43Z

serpentTools/parsers/results.py

+
+    def _read(self):
+        """Read through the results file and store requested data."""
+        info('Preparing to read {}'.format(self.filePath))


Remove this info calls. These are done automatically by the BaseReader.read method that, in turn, calls this method.

drewejohnson · 2018-05-15T13:31:11Z

serpentTools/parsers/results.py

+                self._storeUnivData(varNameSer, varVals)
+
+    def _whereAmI(self, tline):
+        """Identify the position in the file."""


I think we can return posFile rather than store it on the object. It is only used in _processResults

posFile is stored on the object simply because the method will not update the position, unless one of the keywords is found. This will lead to returning a non-existing variable.

You could have the method return None if none of the keywords are found.

What about the case where _whereAmI does not update posFile? That would indicate we are not at the beginning of a results, metadata, nor universe block, correct?

The posFile will always be one of the these blocks: results, metadata, universes.
when I scroll through the file and find one of keywords, I update the posFile to its location.
Example:
if 'VERSION' is found I know that this is the metadata block and thus self._posFile will store this.
The next variable is still in the results block, however, the posFile will not be updated. Therefore, the function whereAmI will not change the posFile field.
The posFile will be changed once the results block is found.

Ah, got it. Then we do need to keep this as an attribute. Thanks for explaining!

drewejohnson · 2018-05-15T13:32:21Z

serpentTools/parsers/results.py

+        self.__serpentVersion = rc['serpentVersion']
+        self._counter = {'meta': 0, 'rslt': 0, 'univ': 0}
+        self._univlist = []
+        self.metadata,  self.resdata, self.univdata = {}, {}, {}


Rename univdata to universes

drewejohnson · 2018-05-15T13:32:39Z

serpentTools/parsers/results.py

+
+    def __init__(self, filePath):
+        XSReader.__init__(self, filePath, 'xs')
+        self.__fileObj = None


Delete this

The following was deleted.
self.__fileObj = None

drewejohnson · 2018-05-15T13:33:58Z

serpentTools/tests/test_ResultsReader.py

@@ -0,0 +1,233 @@
+"""Test the results reader."""


Looking great!

drewejohnson · 2018-05-15T13:46:37Z

examples/ResultsReader.ipynb

@@ -0,0 +1,2515 @@
+{


Comments on the jupyter notebook:

Additional commentary looks great.

Use %matplotlib inline instead of notebook. If you navigate to how the image is shown here, nothing is displayed in the plotting window. I'm also not sure how the conversion to rst will work with notebook

Copyright needs to be for all contributors, as they all helped build this project.

Make code render in Markdown with single ticks, not apostrophes i.e.

'keys' # not `keys`'

Why do we need the comparison in block [9], where we compare the version of SERPENT used to read the file and what is stored on the rc object? Is the fact that we were able to read a 2.1.30 file with 2.1.29 a good thing or a bad thing?

Delete the additional call to %matplotlib notebook in block [16]

Rename univdata calls in markdown cells to universes to reflect other changes in this review

Can clean up the demo of the univdata section by storing a HomogUniv object with universe = res.universes['3102, 0.0, 1, 0.0] and then doing all the calls to sub-dictionary keys on this object

We need to get Unable to import all settings from jupyter #140 closed before this notebook is complete

Everything was modified.
In respect to the version used vs. defined: a warning would be raised, but otherwise as long as we support the version, at attempt to read will be made.

All the files were modified to include this change: results.py test_ResultsReader.py ResultReader.ipynb

The parser was update and now include .getUniv method. The unittest include 4 additional tests. Jupyter notebook is done. Missing: container for groupVals and groupUnc and also post check to test if universes field contains data or empty.

results.py includes an additional test that relies on the 'hasData' method inserted to containers.pt. test_ResultsReader was introduced with a test to check the new feature. Additional results file is needed for the new test: pwr_res_emptyAttributes.m. Jupyter notebook includes the new copy rights.

drewejohnson · 2018-05-16T18:41:36Z

serpentTools/parsers/results.py

+        must be given. The method allows to input more than one time parameter,
+        however only one is used to extract the data.
+        e.g., .getUniv('0',0.1, 1, 10), .getUniv('0',burnup=0.1), getUniv('0',timeDays=10)
+


This block is not needed. The sphinx autodoc will automatically create documentation for this based on the method docstring

Major change from univ.metadata to univ.gc, univ.gcUnc, univ.groups, univ.microGroups Unittest tests all these changes. Notebook is updated accordingly.

drewejohnson

Things blocking approval

Changes to jupyter notebook [below]
Converted notebook - rst for documentation
Remove warning block in docs/api/results.rst
Include this PR and a brief summary in docs/changelog.rst

Changes to jupyter notebook

Remove block [2] where the rc keys are printed
Explicitly set the serpent version to be 2.1.30 at the top of the file to remove the warning
Remove the comparison in block [9] and the above markdown cell used to compare serpent version from setting and from file
Remove the disclaimer about plotting not yet available. We don't know how "near term" this could be 😁
Update the link in block [18] to point to the full HomogUniv object - http://serpent-tools.readthedocs.io/en/latest/api/containers.html#serpentTools.objects.containers.HomogUniv
Remove mentions of metadata in comments for blocks 29 and 30

Parser, unittest, notebook and all the documentation is ready.

drewejohnson

Fantastic! Nearly done 🏃‍♂️

Things blocking review

Add code blocks to rst - this one is my fault due to poor explanation of what exactly needs to be changed in the dev docs. It's only the ipython3 string that needs to be removed
Format the MapStrVersions in docs/develop/serpentVersions.rst using a code block

Things that would be absolutely fantastic but not needed

In docs/examples/ResultsReader.rst, when you reference a class, such as the ResultsReader, an attribute, or a method, use the Sphinx python referencing system, example

``ResultsReader`` -> :py:class:`~serpentTools.parsers.results.ResultsReader`
``.resdata`` -> :py:attr:`~serpentTools.parsers.results.ResultsReader.resdata`

This is more conversion on your part, but you can use things like the replace directive to save typing. Check out the depletion reader example for some hints. Not required for review, but eventually we will convert these links.

drewejohnson · 2018-05-16T20:41:07Z

docs/develop/serpentVersions.rst

+- metadata: general description of the simulation
+- resdata: time-dependent values, e.g. k-eff
+- universes: universe dependent values, such as cross-sections
+The mapping through the variable `MapStrVersions` should reflect this. 


For rst, you have to use double ticks to get the code formatted properly - http://docutils.sourceforge.net/docs/ref/rst/roles.html#literal

drewejohnson · 2018-05-16T20:42:04Z

docs/develop/serpentVersions.rst

+- universes: universe dependent values, such as cross-sections
+The mapping through the variable `MapStrVersions` should reflect this. 
+
+The basic mapping definition relies on the following structure:


Add a second colon here and tab the whole block in, like

.....structure:: MapStrVersions = { ....

Further reading - http://www.sphinx-doc.org/en/stable/rest.html?highlight=url%20#source-code

This way the code will be rendered properly in the documentation. The empty line that separates the first line of code from the first line of text is crucial as well

drewejohnson · 2018-05-16T20:47:13Z

docs/examples/ResultsReader.rst

+should ease the analyses.
+
+
+


I should clarify the developer guide more. Only the ipython3 phrase needs to be removed. The .. code:: has to remain or else this the blocks will not be rendered as code at all. This one's on me, and the dev guide will be updated accordingly

drewejohnson · 2018-05-16T20:49:37Z

docs/examples/ResultsReader.rst

+settings option.
+
+A detailed description on how to use the settings can be found on:
+http://serpent-tools.readthedocs.io/en/latest/api/settings.html


[Correct this in the notebook as well] Better link would be http://serpent-tools.readthedocs.io/en/latest/settingsTop.html, as this explains what all the settings are and how they influence the code

serpentVersion.rst was modified to include proper code in the documentation. ResultsReader.rst was modified to have proper documentation. Notebook was updated with a new reference to settings.

ResultsReader.rst in docs/examples was modified.

drewejohnson

The results reader example looks phenomenal now. One final request to change the main heading for that file.
Since many of your sections have the same distinction (line of =) as your main header, sphinx takes them all to be chapter headings. I copied the rendered layout from docs/_build/html/examples/index.rst below

Results Reader

    Basic operations

Metadata (metadata)
Results Data (resdata)

    Plotting Results Data (resdata)

Universe Data (universes)

    Get Universe Data (.getUniv)
    Plotting universes
    User Defined Settings
    Conclusion
    References

The most minor of request, and then we're done here!

drewejohnson · 2018-05-17T11:22:44Z

docs/examples/ResultsReader.rst

+
+.. |metadata| replace:: :py:attr:`~serpentTools.parsers.results.ResultsReader.metadata`
+
+.. |universes| replace:: :py:attr:`~serpentTools.parsers.results.ResultsReader.universes`

 Results Reader
 ==============


Add an empty line and an additional line of '=' over the 'Results Readerheading. This causes all other sections to be directly under this chapter. Some of the sections, likeMetadataare at the same level as this with only a single line of=. After building html with make clean html, if you navigate to docs/_build/html/examples/index.html`, then you can see how this affects the layout of this section

The final change aligned the headings in the .rst file

drewejohnson

🎉 🏆 🎉

DanKotlyar added 2 commits May 9, 2018 12:20

Modifed .gitattributes- auto line endings

49973a2

Attempt to remove the amount of files "changed" due to changing line endings. Automatically convert line endings to \n upon commit.

Results Reader parser, unittest and a jupyter added.

6b7f240

results.py - parser test_ResultsReader - unittest pwr_res.m, pwr_res_filter.m files used by the test ResultsReader.ipynb - jupyter notebook InnerAssembly_res.m - used by .ipynb file

DanKotlyar requested a review from drewejohnson May 9, 2018 16:43

DanKotlyar added the Work in progress label May 9, 2018

drewejohnson reviewed May 9, 2018

View reviewed changes

examples/ResultsReader.ipynb Outdated

]

},

{

"cell_type": "raw",

Copy link

Collaborator

drewejohnson May 9, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This cell needs to be markdown, not raw

drewejohnson mentioned this pull request May 9, 2018

Results Reader #127

Closed

drewejohnson self-assigned this May 9, 2018

Merge branch 'develop' into results-reader-3

48fc2ef

Most updated develop (#132) is merged into results-reader-3

drewejohnson requested changes May 10, 2018

View reviewed changes

drewejohnson added this to the 0.5.0 milestone May 10, 2018

DanKotlyar and others added 11 commits May 11, 2018 08:26

Modified files due to line ending change

72b7d05

all the files from examples and the doc/welcome/license

Merge branch 'develop' into results-reader-3

55cb579

Merge branch 'develop' into results-reader-3

633eb83

Merge branch 'develop' into results-reader-3

f32d489

Auto stash before merge of "develop" and "origin/develop"

679e692

Reverting images and some documentation to develop

4489324

Run `git revert develop <>` on all images in docs/examples/images, and on a few other files modified by commit 679e692

dos2unix docs/welcome/license.rst

f31a0cd

Results Reader parser and the unittest are ready for review.

5f942a2

results.py test_ResultsReader pwr_res_noUniv.m for testing

Results parser and unittest were modified

b95685a

test_ResultsReader.py includes 14 test cases. parser is able to read with and without burnups .ipynb is not ready pwr_res_noBU.m was added for testing

Merge remote-tracking branch 'origin/results-reader-3' into results-r…

643554a

…eader-3

Parser stores version from rc . Notebook was updated bu thas some iss…

1d5480a

…ues with settings. Jupyter notebook has issues to upload all the settings.

drewejohnson requested changes May 15, 2018

View reviewed changes

DanKotlyar added 3 commits May 15, 2018 10:11

Small modification to replace the field .univdata with .universes

154d98a

All the files were modified to include this change: results.py test_ResultsReader.py ResultReader.ipynb

Merge branch 'develop' into results-reader-3

c5a7cf8

Revised parser test and jupyter notebook.

cee9ca1

The parser was update and now include .getUniv method. The unittest include 4 additional tests. Jupyter notebook is done. Missing: container for groupVals and groupUnc and also post check to test if universes field contains data or empty.

DanKotlyar added 2 commits May 16, 2018 11:26

Merge branch 'develop' into results-reader-3

13d02ef

Merge branch 'develop' into results-reader-3

d7a96c2

drewejohnson mentioned this pull request May 16, 2018

[API] Group constant dictionaries for HomogUniv; Criticalities as floats #146

Merged

1 task

drewejohnson reviewed May 16, 2018

View reviewed changes

DanKotlyar added 2 commits May 16, 2018 14:47

Merge branch 'develop' into results-reader-3

c84f31f

ResultsReader review before documentation

28ab3ea

Major change from univ.metadata to univ.gc, univ.gcUnc, univ.groups, univ.microGroups Unittest tests all these changes. Notebook is updated accordingly.

drewejohnson requested changes May 16, 2018

View reviewed changes

ResultsReader package with documentation is ready

1e1c07f

Parser, unittest, notebook and all the documentation is ready.

drewejohnson requested changes May 16, 2018

View reviewed changes

drewejohnson changed the title ~~Completed Results Reader~~ [API] Implement Results Reader May 16, 2018

drewejohnson mentioned this pull request May 17, 2018

[DOC] Update README to render propely; Better explain notebook - rst conversion #147

Closed

DanKotlyar added 2 commits May 16, 2018 23:02

Documentation update to Results Reader

9803559

serpentVersion.rst was modified to include proper code in the documentation. ResultsReader.rst was modified to have proper documentation. Notebook was updated with a new reference to settings.

Copy right forced statement was removed from .rst

fe288dd

ResultsReader.rst in docs/examples was modified.

drewejohnson requested changes May 17, 2018

View reviewed changes

drewejohnson added enhancement reader and removed Work in progress labels May 17, 2018

drewejohnson changed the title ~~[API] Implement Results Reader~~ [ENH] Implement Results Reader May 17, 2018

Headings in ResultsReader.rst was modified.

d32a1b6

The final change aligned the headings in the .rst file

drewejohnson approved these changes May 17, 2018

View reviewed changes

DanKotlyar merged commit e4d30fc into develop May 17, 2018

drewejohnson mentioned this pull request May 17, 2018

[ENH] Results file reader #7

Closed

1 task

drewejohnson mentioned this pull request May 18, 2018

[ENH] Plotting methods for results reader #151

Closed

[ENH] Implement Results Reader #130

[ENH] Implement Results Reader #130

Conversation

DanKotlyar commented May 9, 2018

drewejohnson commented May 9, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

drewejohnson left a comment • edited Loading

Choose a reason for hiding this comment

Things blocking approval

More detail

Fewer attributes stored on reader

More test cases

Pre-compiled regular expressions

Documentation

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

DanKotlyar May 11, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

DanKotlyar May 11, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

DanKotlyar commented May 10, 2018

drewejohnson commented May 11, 2018

drewejohnson left a comment • edited Loading

Choose a reason for hiding this comment

Things blocking approval

More Detail

CMM and non-metadata related XS

Documentation

Better way to extract universes

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

drewejohnson left a comment

Choose a reason for hiding this comment

Things blocking approval

Changes to jupyter notebook

drewejohnson left a comment

Choose a reason for hiding this comment

drewejohnson left a comment •

edited

Loading

DanKotlyar May 11, 2018 •

edited

Loading

DanKotlyar May 11, 2018 •

edited

Loading

drewejohnson left a comment •

edited

Loading

`CMM` and non-metadata related XS