Commit Graph

44 Commits

Author SHA1 Message Date
Jan Kundrát
0d5f1c7d80 Split JSON export from service XLS reading
We have a better module for this.

Change-Id: Id4b68d3ddb119f27df3dfac52277981558fc5e50
2020-06-08 18:47:15 +02:00
Jan Kundrát
5bc42332cd tests: Do not create JSON files in the source tree
Change-Id: I5ef71cc82466367fa9e9eea4d3f02453a4d2f469
2020-06-08 18:31:24 +02:00
Jan Kundrát
754be7ca08 Do not create *_auto_design.json by default
It's a bad habit to write files into the source code repository. It will
also become impossible if gnpy is installed into a systemwide, possible
read-only location.

The old behavior can be reactivated by using an extra option to tell
GNPy where to put the generated file.

Change-Id: I9ad43890ca5886f516885de5938a4778870a06c5
2020-06-08 18:28:59 +02:00
Jan Kundrát
1b2b048b47 reorganization: move request processing and request disjunction handling into topology.requests
Change-Id: I14902a2e15cc5fd27530cb294f4f549f6974fd49
2020-05-23 22:03:23 +02:00
Jan Kundrát
7da4ec08d8 reorganization: move JSON-specific bits from path_requests_run to tools.json_io
Change-Id: I03cf4e298300387ee5d56251e2e552e111b59f65
2020-05-23 21:28:06 +02:00
Jan Kundrát
648039521e reorganization: move all JSON processing into an extra module
We agreed that `gnpy.core` should only contain stuff for propagating
wavelengths. Conceptually, JSON parsing and even instantiating these
network elements from data obtained through JSON is *not* something that
is on the same level -- and this will become more important when we move
into YANG format in future.

Also, instead of former `gnpy.core.equipment.common`, use
`gnpy.tools.json_io._JsonThing`. It is not really an awesome name :),
but I think it sucks less than a thing called "common" which would be no
really longer any "common" in that new file.

Change-Id: Ifd85ea4423d418c14c8fae3d5054c5cb5638d283
2020-05-23 20:44:47 +02:00
Jan Kundrát
9faf6430a5 reorganization: gnpy/{core => tools}/service_sheet.py
Change-Id: I88559cc718536f222b8ea9829bcc72a425c062ca
2020-05-23 15:23:27 +02:00
Jan Kundrát
01c566a325 reorganization: gnpy/{core => topology}/spectrum_assignment.py
Change-Id: Ic6194ce639dcb2f9419372febe0f2b58473edb38
2020-05-23 15:05:42 +02:00
Jan Kundrát
0823f8de46 Move automatic_nch, automatic_fmax into utils
I think that gnpy/core/equipment.py should contain only stuff which
prepares the equipment_config, not anything "lower level" that is reused
from other places.

Change-Id: I0cd593fd3e5558178ddd0ad8fff5c596e022894a
2020-05-23 13:08:24 +02:00
Jan Kundrát
46f89aa770 coding style: autopep8 in an aggressive mode (-aaaaaaaaaa)
I decided to skip the following chunk of the diff because I think that
it would actually made the code a bit harder to read:

diff --git gnpy/core/service_sheet.py gnpy/core/service_sheet.py
index 9965840..9834111 100644
--- gnpy/core/service_sheet.py
+++ gnpy/core/service_sheet.py
@@ -41,8 +41,22 @@ logger = getLogger(__name__)

 class Request(namedtuple('Request', 'request_id source destination trx_type mode \
     spacing power nb_channel disjoint_from nodes_list is_loose path_bandwidth')):
-    def __new__(cls, request_id, source, destination, trx_type,  mode=None, spacing=None, power=None, nb_channel=None, disjoint_from='',  nodes_list=None, is_loose='', path_bandwidth=None):
-        return super().__new__(cls, request_id, source, destination, trx_type, mode, spacing, power, nb_channel, disjoint_from,  nodes_list, is_loose, path_bandwidth)
+    def __new__(
+            cls,
+            request_id,
+            source,
+            destination,
+            trx_type,
+            mode=None,
+            spacing=None,
+            power=None,
+            nb_channel=None,
+            disjoint_from='',
+            nodes_list=None,
+            is_loose='',
+            path_bandwidth=None):
+        return super().__new__(cls, request_id, source, destination, trx_type, mode, spacing,
+                               power, nb_channel, disjoint_from, nodes_list, is_loose, path_bandwidth)

 # Type for output data:  // from dutc

diff --git tests/test_automaticmodefeature.py tests/test_automaticmodefeature.py
index 0e5f633..5ba5881 100644
--- tests/test_automaticmodefeature.py
+++ tests/test_automaticmodefeature.py
@@ -32,7 +32,26 @@ eqpt_library_name = Path(__file__).parent.parent / 'tests/data/eqpt_config.json'
 @pytest.mark.parametrize("net", [network_file_name])
 @pytest.mark.parametrize("eqpt", [eqpt_library_name])
 @pytest.mark.parametrize("serv", [service_file_name])
-@pytest.mark.parametrize("expected_mode", [['16QAM', 'PS_SP64_1', 'PS_SP64_1', 'PS_SP64_1', 'mode 2 - fake', 'mode 2', 'PS_SP64_1', 'mode 3', 'PS_SP64_1', 'PS_SP64_1', '16QAM', 'mode 1', 'PS_SP64_1', 'PS_SP64_1', 'mode 1', 'mode 2', 'mode 1', 'mode 2', 'nok']])
+@pytest.mark.parametrize("expected_mode",
+                         [['16QAM',
+                           'PS_SP64_1',
+                           'PS_SP64_1',
+                           'PS_SP64_1',
+                           'mode 2 - fake',
+                           'mode 2',
+                           'PS_SP64_1',
+                           'mode 3',
+                           'PS_SP64_1',
+                           'PS_SP64_1',
+                           '16QAM',
+                           'mode 1',
+                           'PS_SP64_1',
+                           'PS_SP64_1',
+                           'mode 1',
+                           'mode 2',
+                           'mode 1',
+                           'mode 2',
+                           'nok']])
 def test_automaticmodefeature(net, eqpt, serv, expected_mode):
     equipment = load_equipment(eqpt)
     network = load_network(net, equipment)

Change-Id: I522c45c079b3a9540568657e2ae0a4bfc5fb1272
2020-05-19 12:53:11 +02:00
Jan Kundrát
3548ed74e2 coding style: autopep --in-place --recursive --jobs 4 --max-line-length 120 gnpy/ tests/
Change-Id: I2f0fca5aa1314f9bb546a3e6dc712a42580cd562
2020-05-19 12:40:00 +02:00
Jan Kundrát
145653df6e reorganization: gnpy/{core => topology}/request.py
Change-Id: Ib399549479b56634c681930aa444b657e5f58ca7
2020-05-19 11:56:02 +02:00
Jan Kundrát
63a6256b5e pep8: rename classes to use CamelCase
Change-Id: Ia696e05b72f1bc5feb570996f492042dafab262d
2020-05-19 11:54:27 +02:00
Jan Kundrát
8b1d8b3479 reorganization: XLS conversion goes to gnpy.tools
Change-Id: Ibbaddacd24a2d0f6f6d98fdc30d57da3be188338
2020-05-19 11:54:23 +02:00
EstherLerouzic
de58d7d7c2 Enable ILA and FUSED type in routing constraints with xls input
- builds correspondance dicts between input name from excel
  and names created with convert.py and autodesign in network.py
- correct the corresp_name dicts according to the effective
  network autodesign. This supports the case of fiber splitting
  and of fused elements
- include the case of parrallel links with only one hop
- interpret the node list constraint given by the user with the dict
- filter the constraints that are not applicable
- add tests for constraints
- correct equipment sheet of mesh_example_topologyv2.xls: morlaix and
  loudeac should not appear in node A column since they are fused

ILA and FUSED constraints must be filled with the next node
information in order to avoid confusion on the direction.
for example
     eg    a----b-----c
           |    |     |
           i    j     k
           |    |     |
           e----f-----g

a constraint 'j' given for service i to k leads to 2 possible direction:
i-a-b-j-f-g-k
i-e-f-j-b-c-k
the user must indicate the chossen direction. This ambiguity does not
exist with network input in json format (names are unique).

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
Change-Id: Ia7cf2026e569c8b566459791fc89546b91fb117c
2020-04-23 11:15:37 +02:00
Jan Kundrát
82ce3384ba tests: Use XLSX for the CORONET topology
Now that XLSX is officially supported, it should be tested as well.

Change-Id: I9286d3cb56950d554582d2eaf8153da5ffdd771a
2020-03-21 13:19:24 +01:00
Jan Kundrát
1f1877f7a9 tests: do not hardcode file suffix lengths
This will become useful for XLS -> XLSX conversion.

Change-Id: I025e4c24d00526d3bb48c23dcbdc82a65be9a477
2020-03-21 13:19:24 +01:00
Jan Kundrát
fe811f725c tests: use native pytest features for exception handling
Using `with pytest.raises` is better than open coding the equivalent
feature. Similarly, when a block is not expected to raise an exception,
let's just let it run outside of a `try` block and rely on the test
framework to report a possible failure when hitting an unhandled
exception.

Change-Id: Icb1bb83e649733b56fcdc9168cabf88c9cf8d478
2020-01-10 18:16:53 +00:00
EstherLerouzic
c577a75725 second set of modification due to codacy report
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-10 09:02:19 +01:00
EstherLerouzic
31e634615b Complete test on response comparison
If request is bidir and 'z-a-path-metric' is missing, raises an error
If present, should not raise an error

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-10 09:02:19 +01:00
EstherLerouzic
5b939bc57a Improve compare_response function
remove some prints, indicate which from actual or expected key
is printed.
preceise that test is also performed on values

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-10 09:02:19 +01:00
EstherLerouzic
42ba3eb98d Add a test to go through the new added code in spectrum_assignment
Verify that there is no raised error if M=0 and the blocking attribute is there

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-10 09:02:19 +01:00
EstherLerouzic
37bd5d0404 Code coverage improvement test service creation w/o sync vector
add one service  only excel file + changes on compare.py to support
no synchronization vector case

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 19:17:35 +01:00
EstherLerouzic
f788b81d21 Code coverage improvement test json generation with Z to A direction
add aa bidir request to test service error handling
TODO: write more detailed test on the bidir case

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 19:17:35 +01:00
EstherLerouzic
2ff1ce6b34 Code coverage improvement test ServiceError handling
For this purpose we create a wrong request with M=0 and verify
that serviceError is correctly raised when calling Result_element class

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 19:17:35 +01:00
EstherLerouzic
898eada097 Update test_parser
function compute_path_with_disjunction needs an additional reversed path
and results must contain wavelength assignment

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 19:17:35 +01:00
EstherLerouzic
87cc3dac00 Corrections with respect to second review
Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 19:09:39 +01:00
EstherLerouzic
89e28cc7be Correct bug in parser: csv header comparison
previous comparison was done on the result from .sort(), ie None.
list.sort() method modifies the list in-place and returns None.

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 15:38:42 +01:00
EstherLerouzic
6a7a04ebb1 syntax correction
reordering imports call according to pylint3 + extra line removed

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 15:38:42 +01:00
EstherLerouzic
0366fc2956 Adding a no-path case for test coverage
add a no path case (request 6) in requests and expected responses.
response is also generated if path is not feasible: checks
that it is correctly handled in csv and json responses

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 15:38:42 +01:00
EstherLerouzic
48b7d71f02 add a test on json response generation
create a json response based on test file and compare it to expected
response.
comparison first checks that there is no missing or extra key
and then compares keys

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 15:38:42 +01:00
EstherLerouzic
715baf2a1c correct compare response header test to support any order
previous test assumed same order for header fields.
order the headers in the same way in order to support different
types of order

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 15:38:42 +01:00
EstherLerouzic
e55cea776e improve code quality
remove unused import, use correct indents, remove extra spaces
add function description

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 15:38:42 +01:00
EstherLerouzic
b388d143fd change assert to raise AssertionError
use
  if not condition:
  raise AssertionError()
instead of
  assert condition
according to codacy recommendation

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 15:38:42 +01:00
EstherLerouzic
c592c572d8 adding a test for the csv creation
using pandas package to have an easy ordering of response column:
- test that the generated header is as expected
- read the response. In order to support different orders wrt
  response answer and field answer, test function frst orders lines
  according to response index and then to columns (fields of the answer).
check that the answers are as expected

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-10-09 15:38:42 +01:00
EstherLerouzic
79f198d6fe reduce the number of test files
- use the same file to test different configuration: with and without Eqpt sheet defined,
   add the abcdfgh topo for disjunction tests in the same file
 - change the name of files to avoid mixing with examples/meshTopologyExampleV2.xxx

TODO: test if examples/ files pass for both path_requests_run and trasmission_main_example

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-04-09 17:13:40 +01:00
EstherLerouzic
315a12b9df Update parser test files in order wrt novel road_equalization branch
- update eqpt_config.json to power-mode : false to take into account gain target
    of the eqpt sheet
  - correct eqpt-config.json with new parameters
  - add gain target in meshTopolgyExampleV2Eqpt.xls
  - correct test_parser.py test function names
  - suppress Exceltestfile
  - change power_mode in test_parser to True in eqpt_config.json for all tests
    except for parser: changed in the function to False to enable
    Eqpt files gain target reading

Next: reduce the number of test files and include examples/ files in the tests

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-04-09 17:13:40 +01:00
EstherLerouzic
e14d145f2c Adding a test on autodesign self consistency
tests it the autodesign applied on an autodesigned file gives exactly the same result

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-04-09 17:13:29 +01:00
EstherLerouzic
f839da39f0 adding a test on autodesign verfication
- tests if networ autodesign does not change some reference files
- changed the meshTopologyExampleV2 files to account for newest features:
  empty columns in mode, power, nb channels, disjuntion examples ...
- added a tsp type with more modes for diversification of tests

Signed-off-by: EstherLerouzic <esther.lerouzic@orange.com>
2019-04-09 17:13:29 +01:00
Jean-Luc Auge
4c2d61bb9b update test files
Signed-off-by: Jean-Luc Auge <jeanluc.auge@orange.com>
2019-01-14 17:14:17 +01:00
James Powell
bcf93e1d9f enforce utf-8 encoding for reading/writing JSON 2018-10-29 11:14:31 -04:00
James Powell
f8e146b9b9 duplicate service sheet to simplify structure 2018-10-14 20:55:25 -04:00
James Powell
bed4e9f1e1 clean up JSON comparator & remove redundant files 2018-07-02 17:48:57 -04:00
James Powell
8fccbb0ac2 rename test files 2018-07-02 16:28:06 -04:00