Merge pull request #25 from pessimistic-io/stats

Stats

analyzer.py

@@ -29,7 +29,10 @@ def slither_analyzer(output:str) -> dict[list[Finding]]:
                 if not re.match(r"[0-9]{40}", address):
                     address = path
                 findings.append(Finding(address, fname, ",".join([str(l) for l in element['source_mapping']['lines']])))
-            result[detector_result['check']] = findings
+            if detector_result['check'] in result:
+                result[detector_result['check']] += findings
+            else:    
+                result[detector_result['check']] = findings
     elif not 'results' in output:
         raise SlitherOutError('no results')
     return result

runner.py

@@ -14,6 +14,9 @@
 from storage import Storage
 from config import LOGGING_LEVEL
 
+CONTRACT_STAT_TYPE_NAME = 'by_contract'
+FINDING_STAT_TYPE_NAME = 'by_finding'
+
 def process_file(contract: Contract, use_slither: bool = False) -> tuple[Contract, dict[str, list]]:
     """Run subproccess contract processing
     Args:
@@ -77,7 +80,7 @@ def main(output, extra_output, input, skip_duplicates, skip_libs, new_contracts,
         detectors = DETECTORS
     # Use multiprocessing Pool to run slitherin in parallel
     logger.info("starting pool on %d cores contract", pool)
-    detector_statistics = Counter()
+    detector_statistics = { CONTRACT_STAT_TYPE_NAME: Counter(), FINDING_STAT_TYPE_NAME: Counter() }
     start_time = time.time()
     storage = Storage()
     with Pool(pool) as pool:
@@ -95,25 +98,34 @@ def main(output, extra_output, input, skip_duplicates, skip_libs, new_contracts,
                                 f_extra.write(f"{finding.address};{finding.filename};{detector};\"{finding.lines}\"\n")
                         if count_files:
                             increment = len(files_counter)
-                detector_statistics[detector] += increment
-            if not count_files:
-                detector_statistics['total'] += 1
-            else:
-                detector_statistics['total'] += count_sol_files(contract.filename)
-
+                if not count_files:
+                    detector_statistics[CONTRACT_STAT_TYPE_NAME][detector] += increment
+                count_findings = len(findings)
+                if count_findings > 0:
+                    detector_statistics[FINDING_STAT_TYPE_NAME][detector] += count_findings
+            sol_files = count_sol_files(contract.filename)
+            for stat_type in detector_statistics:
+                if count_files and stat_type ==  CONTRACT_STAT_TYPE_NAME:
+                    continue
+                detector_statistics[stat_type]['files'] += sol_files
+                detector_statistics[stat_type]['contracts'] += 1
+
             for detector in contract.detectors:
                 storage.set_contract_checked(contract.address, contract.chain_id, detector)
 
             if timeout is not None and time.time() - start_time > timeout:
                 logger.info("timeout stop, processed %d tasks", detector_statistics['total'])
                 break
+    for stat_type in list(detector_statistics.keys()):
+        if len(detector_statistics[stat_type]) == 0:
+            del detector_statistics[stat_type]
     logger.info("completed pool in %s", str(timedelta(seconds=round(time.time()-start_time))))
     df = pd.DataFrame.from_dict(detector_statistics, orient='index')
     print(df.to_markdown())
     if output is not None:
         logger.info("Save stats to file %s", output)
         with open(output, 'w') as f:
-            f.write(df.to_csv(sep=';'))     
+            f.write(df.to_csv(sep=';', float_format='%.0f'))     
 
 if __name__ == "__main__":
     main()

save_sheet.py

@@ -2,8 +2,10 @@
 import time
 import click
 import subprocess
+import logging
 
 from sheet import Sheet
+from config import LOGGING_LEVEL
 
 DETECTOR_COL_NUM = 2
 
@@ -15,36 +17,55 @@ def get_slitherin_version():
     except Exception as e:
         return ''
 
-@click.command()
-@click.option('-i', '--input', help="file with benchmark results", required=True)
-@click.option('-sa', '--service-account', help="google service account json file", required=True)
-@click.option('-si', '--sheet-id', help="google sheet id", required=True)
-@click.option('-ln', '--list-name', help="google list name", required=True)
-@click.option('-sv', '--slitherin-version', help="slitherin version, default value taken from slitherin --version command", required=False, default=get_slitherin_version())
-def main(input, service_account, sheet_id, list_name, slitherin_version):
+def save_sheet(results, service_account, sheet_id, list_name, slitherin_version):
+    logger = logging.getLogger()
+    logger.info("save results for %d detectors to list %s", len(results), list_name)
+    print(results)
     sheet = Sheet(service_account, sheet_id, list_name)
     values = sheet.get_rows("A1:ZZZ")
     detector_names = values[0][DETECTOR_COL_NUM:]
 
     detector_col_by_name = {detector_names[i]:(i+DETECTOR_COL_NUM) for i in range(0, len(detector_names))}
     new_row = [round(time.time()), slitherin_version] + ['']*len(detector_names)
     new_columns = []
-    with open(input, 'r') as f:
-        for line in f:
-            detector_name, c = line.split(';')
-            if detector_name == '':
-                continue
-            if detector_name in detector_col_by_name:
-                new_row[detector_col_by_name[detector_name]] = c
-            else:
-                new_columns.append([detector_name]+['']*(len(values)-1)+[c])
+    for detector_name in results:
+        if detector_name in detector_col_by_name:
+            new_row[detector_col_by_name[detector_name]] = results[detector_name]
+        else:
+            new_columns.append([detector_name]+['']*(len(values)-1)+[results[detector_name]])
 
     response = sheet.add_row(new_row)
     last_column_num = DETECTOR_COL_NUM + len(detector_names) - 1
     for column in new_columns:
         last_column_num += 1
         sheet.add_column(last_column_num, column)
 
+@click.command()
+@click.option('-i', '--input', help="file with benchmark results", required=True)
+@click.option('-sa', '--service-account', help="google service account json file", required=True)
+@click.option('-si', '--sheet-id', help="google sheet id", required=True)
+@click.option('-ln', '--list-name', help="google list name", required=True)
+@click.option('-sv', '--slitherin-version', help="slitherin version, default value taken from slitherin --version command", required=False, default=get_slitherin_version())
+def main(input, service_account, sheet_id, list_name, slitherin_version):
+    handler = logging.StreamHandler()
+    handler.setFormatter(logging.Formatter("%(levelname)s: %(asctime)s - %(process)s - %(message)s"))
+
+    logger = logging.getLogger()
+    logger.setLevel(LOGGING_LEVEL)
+    logger.addHandler(handler)
+    with open(input, 'r') as f:
+        detector_results = {}
+        for line in f:
+            line_list = line.strip().split(';')
+            stat_type, c = line_list[0], line_list[1:]
+
+            if stat_type == '':
+                detector_names = c
+                continue
+            detector_results[stat_type] = dict(zip(detector_names, c))
+        for stat_type in detector_results:
+            save_sheet(detector_results[stat_type], service_account, sheet_id, f"{list_name}_{stat_type}", slitherin_version)
+
 if __name__ == "__main__":
     main()