Merge pull request #7 from codeclassroom/dev

0.4 release

CHANGELOG.md

@@ -1,6 +1,13 @@
 # Changelog
 
 
+## [0.4] - March 10, 2020
+
+### Changed [⚠️ Breaking Changes]
+- `getShareScores` & `getInsights` have been decoupled from the check class, they now have to imported separately.
+- Minor changes in the `analyze.py` module.
+
+
 ## [0.3] - Jan 1, 2020
 
 ### Added

demo.py

@@ -1,26 +1,30 @@
 """Usage example"""
 import os
 import pprint
-from plagcheck import plagcheck
+from plagcheck.plagcheck import check, insights, share_scores
 
 from dotenv import load_dotenv
 load_dotenv()
 
-language = "python"
+language = "java"
 userid = os.environ["USER_ID"]
 
 
-moss = plagcheck.check(language, userid)
+moss = check(language, userid)
 
-moss.addFilesByWildCard("testfiles/test_python*.py")
+moss.addFilesByWildCard("testfiles/test_java*.java")
 
 # or moss.addFile("testfiles/test_python.py")
 
 moss.submit()
 
 print(moss.getHomePage())
-pprint.pprint(moss.getResults())
-# print frequency of each shared solution
-pprint.pprint(moss.getShareScores())
+
+result = moss.getResults()
+
+pprint.pprint(result)
+
 # print potential distributor-culprit relationships
-pprint.pprint(moss.getInsights())
+pprint.pprint(insights(result))
+# print frequency of each shared solution
+pprint.pprint(share_scores(result))

docs/changelog.md

@@ -1,6 +1,13 @@
 # Changelog
 
 
+## [0.4] - March 10, 2020
+
+### Changed [⚠️ Breaking Changes]
+- `getShareScores` & `getInsights` have been decoupled from the check class, they now have to imported separately.
+- Minor changes in the `analyze.py` module.
+
+
 ## [0.3] - Jan 1, 2020
 
 ### Added

docs/insights.md

@@ -0,0 +1,51 @@
+# Insights
+
+PlagCheck provides algorithmic analysis of Moss results.
+
+### Terminologies
+
+### 1. Node
+Nodes are results returned by Moss i.e every
+individual file.
+
+### 2. Tags
+Tags are roles which a file serves i.e. a tag is
+a potential distributor or potential culprit or
+both.
+
+### 3. M-group
+m-groups (moss-groups) are groups of solution which have similar code.
+For example A student who solves a programming problem may share their
+solution with 3 of his/her friends, that is a single m-group with 4 nodes.
+
+For example if you run [demo.py](https://github.com/codeclassroom/PlagCheck/blob/master/demo.py), `insights()` will return the following data:
+```java
+
+{'DCtoC Paths': [('testfiles/test_java5.java', 'testfiles/test_java2.java'),
+                 ('testfiles/test_java4.java', 'testfiles/test_java2.java')],
+ 'DtoC Paths': [('testfiles/test_java3.java', 'testfiles/test_java2.java'),
+                ('testfiles/test_java3.java', 'testfiles/test_java.java'),
+                ('testfiles/test_java7.java', 'testfiles/test_java6.java')],
+ 'DtoDC Paths': [('testfiles/test_java3.java', 'testfiles/test_java5.java'),
+                 ('testfiles/test_java3.java', 'testfiles/test_java4.java')]}
+
+```
+
+This analysis can be visualized into following _Disconnected Directed Graph_
+
+![moss results](https://drive.google.com/uc?export=view&id=1Lc8obgjihfo7EGimn300mTtqfmHK0Zem)
+
+We assign Tags to every individual Node.
+
+1. D - Distributor
+Student(s) who distributed their
+code in a group.
+2. C - Culprit
+Student(s) who copied the shared
+code.
+3. DC - Both a Distributor & Culprit
+
+In the above depicted graph, there are 2 unique _m-groups_.
+
+1. Group 1 : [1, 2, 3, 4, 5]
+2. Group 2 : [7, 6]

docs/installation.md

@@ -2,16 +2,24 @@
 
 Installing plagcheck is pretty simple, just run
 
-`pip install plagcheck`
+```bash
+pip install plagcheck
+```
 
 Install a specific verison 
 
-`pip install plagcheck==0.2`
+```bash
+pip install plagcheck==0.4
+```
 
 or directly from GitHub if you cannot wait to test new features
 
-`pip install git+https://github.com/codeclassroom/PlagCheck.git`
+```bash
+pip install git+https://github.com/codeclassroom/PlagCheck.git
+```
 
-If you have already installed it and want to update
+If you have a old version, update it using
 
-`pip install --upgrade plagcheck`
+```bash
+pip install --upgrade plagcheck
+```

docs/usage.md

@@ -1,6 +1,6 @@
 # Usage
 
-plagcheck provides the following classes:
+plagcheck provides the following classes & methods:
 
 ### check(files, lang, user_id)
 
@@ -16,29 +16,33 @@ plagcheck provides the following classes:
 """Usage example"""
 import os
 import pprint
-from plagcheck import plagcheck
+from plagcheck.plagcheck import check, insights, share_scores
 
 from dotenv import load_dotenv
 load_dotenv()
 
-language = "python"
+language = "java"
 userid = os.environ["USER_ID"]
 
 
-moss = plagcheck.check(language, userid)
+moss = check(language, userid)
 
-moss.addFilesByWildCard("testfiles/test_python*.py")
+moss.addFilesByWildCard("testfiles/test_java*.java")
 
 # or moss.addFile("testfiles/test_python.py")
 
 moss.submit()
 
 print(moss.getHomePage())
-pprint.pprint(moss.getResults())
-# print frequency of each shared solution
-pprint.pprint(moss.getShareScores())
+
+result = moss.getResults()
+
+pprint.pprint(result)
+
 # print potential distributor-culprit relationships
-pprint.pprint(moss.getInsights())
+pprint.pprint(insights(result))
+# print frequency of each shared solution
+pprint.pprint(share_scores(result))
 
 ```
 
@@ -72,18 +76,6 @@ c.getHomePage()
 ```python
 
 c.getResults()
-"""
-[
-  {
-    "file1":"filename1.py",
-    "file2":"filename2.py",
-    "percentage": 34,
-    "no_of_lines_matched": 3,
-    "lines_matched":[["2-3", "10-11"]]
-  },
-....
-]
-"""
 
 ```
 
@@ -162,14 +154,16 @@ program code that also appears in the base file is not counted in matches.
 code for an assignment. Multiple Base files are allowed.
 - You should use a base file if it is convenient; base files improve results, but are not usually necessary for obtaining useful information.
 
-### 7. getShareScores()
-**Parameters** : `None` <br>
+<hr>
+
+### share_scores()
+**Parameters** : `Moss Results`(returned by `getResults()`) <br>
 **Return Type** : `Dict` <br>
 **Description**: Share Score is a utility which returns frequency of every individual file.<br>
 **Demo**:
 ```python
 
-c.getShareScores()
+print(share_scores(moss_data))
 
 # Will return
 """
@@ -179,4 +173,15 @@ c.getShareScores()
 """
 ```
 Share Score is basically the frequency of each file appearing in Moss Results.
-i.e Higher the frequency, the more is that solution "shared" by different files.
+i.e Higher the frequency, the more is that solution "shared" by different files.
+
+### insights()
+**Parameters** : `Moss Results`(returned by `getResults()`) <br>
+**Return Type** : `Dict` <br>
+**Description**: See [Insights](/insights).<br>
+**Demo**:
+```python
+
+print(insights(moss_data))
+
+```

mkdocs.yml

@@ -8,6 +8,7 @@ nav:
     - Documentation: index.md
     - Installation: installation.md
     - Usage: usage.md
+    - PlagCheck Insights: insights.md
     - Moss: moss.md
     - Changelog: changelog.md
     - About: about.md

plagcheck/__init__.py

@@ -1,2 +1,2 @@
 """The MOSS interface package for CodeClassroom"""
-from plagcheck.plagcheck import check
+from plagcheck.plagcheck import check, insights, share_scores

plagcheck/analyze.py

@@ -28,7 +28,7 @@ def __init__(self):
         self.nodes = []
         self.nodeCount = 0
 
-    def relatesTo(self, P1, P2, node1, node2):
+    def relate(self, P1, P2, node1, node2):
         """Set a path between two file nodes"""
         node_obj_dict = {}
 

plagcheck/plagcheck.py

@@ -43,6 +43,45 @@ def request(url: str):
     return req.decode("utf-8")
 
 
+def share_scores(moss_data: dict) -> dict:
+    """Share Score Insights"""
+    similar_code_files = []
+    for result in moss_data:
+        similar_code_files.append(result["file1"])
+        similar_code_files.append(result["file2"])
+
+    # frequency of files which are similar
+    share_score = collections.Counter(similar_code_files)
+
+    return dict(share_score)
+
+
+def insights(moss_data: dict) -> dict:
+    """Analysis for Moss"""
+    mg = Mgroups()
+    similar_code_files = set()
+    insights = {}
+
+    for r in moss_data:
+        similar_code_files.add(r["file1"])
+        similar_code_files.add(r["file2"])
+
+    mg.createNodes(similar_code_files)
+
+    for r in moss_data:
+        mg.relate(
+            r["percentage_file1"], r["percentage_file2"], r["file1"], r["file2"]
+        )
+
+    mg.set_tags()
+
+    insights["DtoC Paths"] = mg.d2c()
+    insights["DtoDC Paths"] = mg.d2dc()
+    insights["DCtoC Paths"] = mg.dc2c()
+
+    return insights
+
+
 class check:
     """
     Args:
@@ -133,40 +172,3 @@ def getResults(self) -> Tuple[str, Results]:
         """Return the result as a list of dictionary"""
 
         return self.moss_results
-
-    def getShareScores(self):
-        """Share Score Insights"""
-        similar_code_files = []
-        for result in self.moss_results:
-            similar_code_files.append(result["file1"])
-            similar_code_files.append(result["file2"])
-
-        # frequency of files which are similar
-        share_score = collections.Counter(similar_code_files)
-
-        return dict(share_score)
-
-    def getInsights(self):
-        """Analysis for Moss"""
-        mg = Mgroups()
-        similar_code_files = set()
-        insights = {}
-
-        for r in self.moss_results:
-            similar_code_files.add(r["file1"])
-            similar_code_files.add(r["file2"])
-
-        mg.createNodes(similar_code_files)
-
-        for r in self.moss_results:
-            mg.relatesTo(
-                r["percentage_file1"], r["percentage_file2"], r["file1"], r["file2"]
-            )
-
-        mg.set_tags()
-
-        insights["DtoC Paths"] = mg.d2c()
-        insights["DtoDC Paths"] = mg.d2dc()
-        insights["DCtoC Paths"] = mg.dc2c()
-
-        return insights