Complete terminology rename from TRUE_POSITIVE/FALSE_POSITIVE to ISSUE/NON_ISSUE

Author: GuyZivRH

config/default_config.yaml

@@ -20,7 +20,7 @@ CONFIG_H_PATH: ""
 COMPILE_COMMANDS_JSON_PATH: ""
 SERVICE_ACCOUNT_JSON_PATH: ""
 
-USE_KNOWN_FALSE_POSITIVE_FILE: true
+USE_KNOWN_NON_ISSUES_FILE: true
 CALCULATE_RAGAS_METRICS: false
 WRITE_RESULTS_INCLUDE_NON_FINAL: true
 

deploy/Makefile

@@ -12,7 +12,7 @@ TK := tkn --context $(CONTEXT)
 
 # Pipeline parameters (overrideable on the CLI):
 REPO_REMOTE_URL                       ?= source/code/url
-NON_ISSUES_URL              ?= false/positives/url
+NON_ISSUES_URL              ?= non/issues/url
 
 LLM_URL                          ?= http://<<please-set-llm-url>>
 LLM_MODEL_NAME                   ?= llm-model

deploy/README.md

@@ -121,7 +121,7 @@ make all PROJECT_NAME="systemd" \
  PROJECT_VERSION="257-9" \
  REPO_REMOTE_URL="https://download.devel.redhat.com/brewroot/vol/rhel-10/packages/systemd/257/9.el10/src/systemd-257-9.el10.src.rpm" \
  INPUT_REPORT_FILE_PATH="https://docs.google.com/spreadsheets/d/1NPGmERBsSTdHjQK2vEocQ-PvQlRGGLMds02E_RGF8vY/export?format=csv" \
- NON_ISSUES_URL="https://gitlab.cee.redhat.com/osh/known-false-positives/-/raw/master/systemd/ignore.err"
+ NON_ISSUES_URL="https://gitlab.cee.redhat.com/osh/known-non-issues/-/raw/master/systemd/ignore.err"
 ```
 
 ### 6. Step-by-Step Alternative

deploy/tekton/pipeline.yaml

@@ -51,10 +51,6 @@ spec:
       type: string
       description: "Optional GDrive SA file name"
       default: "service_account.json"
-    - name: USE_KNOWN_FALSE_POSITIVE_FILE
-      type: string
-      description: "Whether to use known false positive file for filtering (true/false)"
-      default: "true"
   workspaces:
     - name: shared-workspace
     - name: gitlab-token-ws
@@ -89,23 +85,23 @@ spec:
         - name: source-workspace
           workspace: shared-workspace
 
-    - name: fetch-false-positives
+    - name: fetch-non-issues
       runAfter:
         - prepare-source
       taskRef:
-        name: fetch-false-positives
+        name: fetch-non-issues
       params:
         - name: NON_ISSUES_URL
           value: "$(params.NON_ISSUES_URL)"
       workspaces:
-        - name: false-positives-workspace
+        - name: non-issues-workspace
           workspace: shared-workspace
         - name: gitlab-token-ws
           workspace: gitlab-token-ws
 
     - name: execute-ai-analysis
       runAfter:
-        - fetch-false-positives
+        - fetch-non-issues
       taskRef:
         name: execute-ai-analysis
       params:
@@ -134,7 +130,7 @@ spec:
       workspaces:
         - name: source-workspace
           workspace: shared-workspace
-        - name: false-positives-workspace
+        - name: non-issues-workspace
           workspace: shared-workspace
         - name: llm-credentials-ws
           workspace: llm-credentials-ws

deploy/tekton/tasks/execute_sast_ai_workflow.yaml

@@ -39,7 +39,7 @@ spec:
       default: "true"
   workspaces:
     - name: source-workspace
-    - name: false-positives-workspace
+    - name: non-issues-workspace
     - name: llm-credentials-ws
     - name: google-sa-json-ws
     - name: cache-workspace
@@ -54,7 +54,7 @@ spec:
         - name: REPO_LOCAL_PATH
           value: "$(params.REPO_LOCAL_PATH)"
         - name: KNOWN_NON_ISSUES_FILE_PATH
-          value: "$(workspaces.false-positives-workspace.path)/ignore.err"
+          value: "$(workspaces.non-issues-workspace.path)/ignore.err"
         - name: USE_KNOWN_NON_ISSUES_FILE
           value: "$(params.USE_KNOWN_NON_ISSUES_FILE)"
         - name: INPUT_REPORT_FILE_PATH

deploy/tekton/tasks/fetch_false_positives.yaml renamed to deploy/tekton/tasks/fetch_non_issues.yaml

@@ -1,27 +1,27 @@
 apiVersion: tekton.dev/v1
 kind: Task
 metadata:
-  name: fetch-false-positives
+  name: fetch-non-issues
 spec:
   params:
     - name: NON_ISSUES_URL
       type: string
       description: "Optional GitLab URL containing known non-issues"
   workspaces:
-    - name: false-positives-workspace
+    - name: non-issues-workspace
       description: "Workspace to store the downloaded ignore.err"
     - name: gitlab-token-ws
       description: "Optional secret mount for GitLab token (if needed)"
       optional: true
   steps:
-    - name: fetch-false-positives
+    - name: fetch-non-issues
       image: curlimages/curl:latest
       script: |
         #!/usr/bin/env sh
         set -euo pipefail
         FP_URL="$(params.NON_ISSUES_URL)"
         if [ -z "$FP_URL" ]; then
-          echo "No falsePositivesUrl provided; skipping fetch..."
+          echo "No nonIssuesUrl provided; skipping fetch..."
           exit 0
         fi
 
@@ -30,14 +30,14 @@ spec:
         if [ -f "$TOKEN_FILE" ]; then
             GITLAB_TOKEN=$(cat "$TOKEN_FILE")
             echo "GitLab token found. Fetching file with authentication..."
-            curl --retry 3 --retry-delay 5 -k -H "PRIVATE-TOKEN: $GITLAB_TOKEN" -fL "$FP_URL" -o "$(workspaces.false-positives-workspace.path)/ignore.err" \
+            curl --retry 3 --retry-delay 5 -k -H "PRIVATE-TOKEN: $GITLAB_TOKEN" -fL "$FP_URL" -o "$(workspaces.non-issues-workspace.path)/ignore.err" \
               || (echo "Error: Could not fetch non-issues file with token." && exit 1)
         else
             echo "No GitLab token file found; attempting unauthenticated fetch..."
-            curl --retry 3 --retry-delay 5 -k -fL "$FP_URL" -o "$(workspaces.false-positives-workspace.path)/ignore.err" \
+            curl --retry 3 --retry-delay 5 -k -fL "$FP_URL" -o "$(workspaces.non-issues-workspace.path)/ignore.err" \
               || (echo "Error: Could not fetch non-issues file unauthenticated." && exit 1)
         fi
-        if [ -f "$(workspaces.false-positives-workspace.path)/ignore.err" ]; then
+        if [ -f "$(workspaces.non-issues-workspace.path)/ignore.err" ]; then
           echo "Success"
           exit 0
         else

deploy/tekton/tasks/validate_urls.yaml

@@ -25,7 +25,7 @@ spec:
         echo "Validating source code URL: $(params.REPO_REMOTE_URL)"
         curl -ksSfL "$(params.REPO_REMOTE_URL)" || (echo "Error: Source code URL is invalid" && exit 1)
 
-    - name: validate-false-positives-url
+    - name: validate-non-issues-url
       image: curlimages/curl:latest
       script: |
         #!/usr/bin/env sh

src/ExcelWriter.py

@@ -263,38 +263,38 @@ def write_results_table(workbook, worksheet, evaluation_summary):
     worksheet.merge_range("A1:B1", "Human Results", cell_formatting(workbook, "#4f8df1"))
     worksheet.write(1, 0, f"Verified {ISSUES_TITLE}", cell_formatting(workbook, "#bfbfbf"))
     worksheet.write(
-        1, 1, len(evaluation_summary.actual_true_positives), cell_formatting(workbook, "#ffffff")
+        1, 1, len(evaluation_summary.actual_issues), cell_formatting(workbook, "#ffffff")
     )
     worksheet.write(2, 0, f"Verified {NON_ISSUES_TITLE}", cell_formatting(workbook, "#bfbfbf"))
     worksheet.write(
-        2, 1, len(evaluation_summary.actual_false_positives), cell_formatting(workbook, "#ffffff")
+        2, 1, len(evaluation_summary.actual_non_issues), cell_formatting(workbook, "#ffffff")
     )
 
     worksheet.merge_range("C1:D1", "AI Results", cell_formatting(workbook, "#4f8df1"))
     worksheet.write(1, 2, f"Predicted {ISSUES_TITLE}", cell_formatting(workbook, "#bfbfbf"))
     worksheet.write(
-        1, 3, len(evaluation_summary.predicted_true_positives), cell_formatting(workbook, "#ffffff")
+        1, 3, len(evaluation_summary.predicted_issues), cell_formatting(workbook, "#ffffff")
     )
     worksheet.write(2, 2, f"Predicted {NON_ISSUES_TITLE}", cell_formatting(workbook, "#bfbfbf"))
     worksheet.write(
         2,
         3,
-        len(evaluation_summary.predicted_false_positives),
+        len(evaluation_summary.predicted_non_issues),
         cell_formatting(workbook, "#ffffff"),
     )
 
 
 def write_confusion_matrix(workbook, worksheet, evaluation_summary):
     worksheet.merge_range("A8:A9", "Human Results", cell_formatting(workbook, "#00b903"))
-    worksheet.write(7, 1, "Actual False Positives", cell_formatting(workbook, "#bfbfbf"))
+    worksheet.write(7, 1, "Actual Non Issues", cell_formatting(workbook, "#bfbfbf"))
     worksheet.write(7, 2, evaluation_summary.tp, cell_formatting(workbook, "#28A745"))
-    worksheet.write(8, 1, "Actual True Positives", cell_formatting(workbook, "#bfbfbf"))
+    worksheet.write(8, 1, "Actual Issues", cell_formatting(workbook, "#bfbfbf"))
     worksheet.write(8, 2, evaluation_summary.fp, cell_formatting(workbook, "#FF0000"))
 
     worksheet.merge_range("C6:D6", "AI Results", cell_formatting(workbook, "#4f8df1"))
-    worksheet.write(6, 2, "Predicted False Positives", cell_formatting(workbook, "#bfbfbf"))
+    worksheet.write(6, 2, "Predicted Non Issues", cell_formatting(workbook, "#bfbfbf"))
     worksheet.write(7, 3, evaluation_summary.fn, cell_formatting(workbook, "#FF0000"))
-    worksheet.write(6, 3, "Predicted True Positives", cell_formatting(workbook, "#bfbfbf"))
+    worksheet.write(6, 3, "Predicted Issues", cell_formatting(workbook, "#bfbfbf"))
     worksheet.write(8, 3, evaluation_summary.tn, cell_formatting(workbook, "#28A745"))
 
 
@@ -320,42 +320,42 @@ def write_model_performance(workbook, worksheet, evaluation_summary, METRICS_STA
 def write_table_key(workbook, worksheet, KEY_START_ROW):
     worksheet.write(KEY_START_ROW, 0, "Table Key:", workbook.add_format({"bold": 1}))
     worksheet.write(
-        KEY_START_ROW + 1, 0, "Verified True Positives =", workbook.add_format({"bold": 1})
+        KEY_START_ROW + 1, 0, "Verified Issues =", workbook.add_format({"bold": 1})
     )
     worksheet.write(
         KEY_START_ROW + 1,
         1,
-        "Human verified as a real issue (true positive)",
+        "Human verified as a real issue (issues)",
         workbook.add_format({"italic": 1}),
     )
 
     worksheet.write(
-        KEY_START_ROW + 2, 0, "Verified False Positives =", workbook.add_format({"bold": 1})
+        KEY_START_ROW + 2, 0, "Verified Non Issues =", workbook.add_format({"bold": 1})
     )
     worksheet.write(
         KEY_START_ROW + 2,
         1,
-        "Human verified as not a real issue (false positive)",
+        "Human verified as not a real issue (non issues)",
         workbook.add_format({"italic": 1}),
     )
 
     worksheet.write(
-        KEY_START_ROW + 3, 0, "Predicted True Positives =", workbook.add_format({"bold": 1})
+        KEY_START_ROW + 3, 0, "Predicted Issues =", workbook.add_format({"bold": 1})
     )
     worksheet.write(
         KEY_START_ROW + 3,
         1,
-        "AI Predicted as a real issue (true positive)",
+        "AI Predicted as a real issue (issues)",
         workbook.add_format({"italic": 1}),
     )
 
     worksheet.write(
-        KEY_START_ROW + 4, 0, "Predicted False Positives =", workbook.add_format({"bold": 1})
+        KEY_START_ROW + 4, 0, "Predicted Non Issues =", workbook.add_format({"bold": 1})
     )
     worksheet.write(
         KEY_START_ROW + 4,
         1,
-        "AI Predicted as not a real issue (false positive)",
+        "AI Predicted as not a real issue (non issues)",
         workbook.add_format({"italic": 1}),
     )
 

src/FilterKnownIssues.py

@@ -41,7 +41,7 @@ def get_relevant_known_issues(self, finding_trace: str, issue_type: str) -> list
             known_issues.append(
                 KnownNonIssue(
                     error_trace=doc.page_content,
-                    reason_of_false_positive=doc.metadata.get("reason_of_false_positive", ""),
+                    reason_of_non_issue=doc.metadata.get("reason_of_non_issue", ""),
                     issue_type=doc.metadata.get("issue_type", ""),
                     issue_cwe=doc.metadata.get("issue_cwe"),
                     similarity_score=None,
@@ -99,9 +99,9 @@ def create_known_issue_retriever(main_process: LLMService, config: Config) -> Kn
     Returns:
         KnownIssueRetriever: A known issue retriever.
     """
-    text_false_positives = read_known_errors_file(config.KNOWN_NON_ISSUES_FILE_PATH)
+    text_non_issues = read_known_errors_file(config.KNOWN_NON_ISSUES_FILE_PATH)
     known_issue_db = main_process.vector_service.create_known_issues_vector_store(
-        text_false_positives, main_process.embedding_llm
+        text_non_issues, main_process.embedding_llm
     )
     known_issue_retriever = KnownIssueRetriever(known_issue_db, config.SIMILARITY_ERROR_THRESHOLD)
     return known_issue_retriever
@@ -116,7 +116,7 @@ def convert_similar_known_issues_to_filter_known_error_context(resp) -> str:
             context_list += KNOWN_NON_ISSUES_TEMPLATES["FILTER_CONTEXT_TEMPLATE"].format(
                 index=index,
                 error_trace=known_issue.error_trace,
-                reason=known_issue.reason_of_false_positive,
+                reason=known_issue.reason_of_non_issue,
             )
         return context_list
 
@@ -139,7 +139,7 @@ def convert_similar_issues_to_examples_context_string(
         formatted_context += KNOWN_NON_ISSUES_TEMPLATES["EXAMPLE_MULTILINE_TEMPLATE"].format(
             number=example_number,
             error_trace=known_issue.error_trace,
-            reason=known_issue.reason_of_false_positive,
+            reason=known_issue.reason_of_non_issue,
             issue_type=known_issue.issue_type,
             issue_cwe=known_issue.issue_cwe,
         )

src/Utils/file_utils.py

@@ -55,28 +55,28 @@ def get_human_verified_results_local_excel(filename):
 
     df.columns = df.columns.str.strip().str.lower()
     expected_issue_id = "issue id"
-    expected_false_positive = "false positive?"
+    expected_non_issue = "non issue?"
 
     if expected_issue_id not in df.columns:
         raise KeyError(
             f"Expected column '{expected_issue_id}' not found in the file. \
                 Found columns: {list(df.columns)}"
         )
-    if expected_false_positive not in df.columns:
+    if expected_non_issue not in df.columns:
         raise KeyError(
-            f"Expected column '{expected_false_positive}' not found in the file. \
+            f"Expected column '{expected_non_issue}' not found in the file. \
                 Found columns: {list(df.columns)}"
         )
 
-    ground_truth = dict(zip(df[expected_issue_id], df[expected_false_positive]))
+    ground_truth = dict(zip(df[expected_issue_id], df[expected_non_issue]))
     logger.info(f"Successfully loaded ground truth from {filename}")
     return ground_truth
 
 
 def get_human_verified_results_google_sheet(service_account_file_path, google_sheet_url):
     """
     Reads a Google Sheet and creates a ground truth dictionary,
-    based on the 'False Positive?' column.
+    based on the 'Non Issue?' column.
 
     NOTE: Assumes the data is in the first sheet (sheet name doesn't matter),
           and that if human-verified data is filled, it is filled for all rows.
@@ -90,17 +90,17 @@ def get_human_verified_results_google_sheet(service_account_file_path, google_sh
     rows = sheet.get_all_records()
 
     # Create ground truth dict in case of human verified data already filled in the google sheet
-    if rows[0].get("False Positive?"):
+    if rows[0].get("Non Issue?"):
         ground_truth = {}
         for idx, row in enumerate(rows, start=1):  # start=1 to get def1, def2, ...
-            is_false_positive = row.get("False Positive?", "").strip().lower()
-            if is_false_positive.lower() not in ALL_VALID_OPTIONS:
+            is_non_issue = row.get("Non Issue?", "").strip().lower()
+            if is_non_issue.lower() not in ALL_VALID_OPTIONS:
                 logger.warning(
-                    f"Warning: def{idx} has invalid value '{is_false_positive}' \
-                        in 'False Positive?' column."
+                    f"Warning: def{idx} has invalid value '{is_non_issue}' \
+                        in 'Non Issue?' column."
                 )
 
-            ground_truth[f"def{idx}"] = is_false_positive
+            ground_truth[f"def{idx}"] = is_non_issue
 
     else:
         ground_truth = None