Finished the package changes analysis.

analysis/softenv_analysis.py

@@ -42,7 +42,47 @@ def sources_stats(input_table):
             pkgmgr[row[2]] += 1
     return pkgmgr
 
-# def pkg_changed(pkgname, )
+def pkg_changed(table, artifact_name, pkgname, pkgsource):
+    """
+    Analyzes the given package lists table to determine if the given package
+    changed for the given artifact.
+
+    Parameters
+    ----------
+    table: str
+        Table to analyse.
+
+    artifact_name: str
+        Name of the artifact for which we want to analyze package changes.
+
+    pkgname: str
+        The package we want to track changes.
+
+    pkgsource: str
+        Source of the package, in case there is multiple packages with the
+        same name but different sources.
+
+    Returns
+    -------
+    changed: bool
+        True if the version number of the package changed over time, False
+        otherwise.
+    """
+    changed = False
+    i = 0
+    pkgver = ""
+    while i < len(table) and not changed:
+        row = table[i]
+        # Artifact name is in the 4th column, package name in the first,
+        # and package source in the 3rd:
+        if row[3] == artifact_name and row[0] == pkgname and row[2] == pkgsource:
+            # If the first version number has not been saved yet:
+            if pkgver == "":
+                pkgver = row[1] # Package version is in the 2nd column
+            elif row[1] != pkgver:
+                changed = True
+        i += 1
+    return changed
 
 def pkgs_changes(input_table):
     """
@@ -59,15 +99,27 @@ def pkgs_changes(input_table):
     dict
         Output table of the analysis in the form of a dict with headers as keys.
     """
-    pkgmgr = {}
+    pkgchanges_dict = {}
+    # Key is the artifact name, and value is a list of tuples constituted
+    # of the package that has been checked and its source for this artifact:
+    # FIXME: Memory usage?
+    checked_artifacts = {}
     i = 0
     for row in input_table:
+        artifact_name = row[3] # Artifact name is in the 4th column
+        if artifact_name not in checked_artifacts.keys():
+            checked_artifacts[artifact_name] = []
+        pkgname = row[0] # Package name is in the first column
+        pkgsource = row[2] # Package source is in the 3rd column
+        if (pkgname, pkgsource) not in checked_artifacts[artifact_name]:
+            if pkg_changed(input_table, artifact_name, pkgname, pkgsource):
                 # Third column is the package source:
-        if row[2] not in pkgmgr:
-            pkgmgr[row[2]] = 1
+                if row[2] not in pkgchanges_dict:
+                    pkgchanges_dict[row[2]] = 1
                 else:
-            pkgmgr[row[2]] += 1
-    return pkgmgr
+                    pkgchanges_dict[row[2]] += 1
+            checked_artifacts[artifact_name].append((pkgname, pkgsource))
+    return pkgchanges_dict
 
 def pkgs_per_container(input_table):
     """

ecg.py

@@ -201,7 +201,6 @@ def buildresult_saver(result, buildstatus_path, config_path):
     -------
     None
     """
-    file_exists = os.path.exists(buildstatus_path)
     buildstatus_file = open(buildstatus_path, "a")
     artifact_name = os.path.basename(config_path).split(".")[0]
     # # Writing header in case file didn't exist:
@@ -244,7 +243,6 @@ def build_image(config, src_dir, image_name, docker_cache = False):
     build_command = f"docker build{cache_arg} -t {image_name} ."
     build_process = subprocess.run(build_command.split(" "), cwd=path, capture_output=True)
     build_output = f"stdout:\n{build_process.stdout.decode('utf-8')}\nstderr:\n{build_process.stderr.decode('utf-8')}"
-    # build_output = build_process.stderr.decode("utf-8")
     logging.info(f"Output of '{build_command}':")
     logging.info(build_output)
     return_code = build_process.returncode
@@ -284,7 +282,7 @@ def check_env(config, src_dir, artifact_name, pkglist_path):
     # of the package manager, depending on the package managers.
     # Each package manager is associated with a tuple, the first item being
     # the package manager's command, the second being the arguments for the
-    # query (they must be separate for the "--entrypoint" argument of Docker
+    # query (they must be separated for the "--entrypoint" argument of Docker
     # 'run', see below), and the third one being the command that will format
     # the output of the query command (this one can be an empty string in case
     # the formatting part is already done using the options of the first command).
@@ -453,8 +451,6 @@ def main():
     try:
         config_file = open(config_path, "r")
         config = json.loads(config_file.read())
-        # config = yaml.safe_load(config_file)
-        # print(config)
         config_file.close()
 
         dl_dir = None