From 7fde0426dfc9d966cfaabb00192019ee3c81b84d Mon Sep 17 00:00:00 2001
From: s-kawamura-w664 <s-kawamura_p@nec.com>
Date: Tue, 30 Mar 2021 07:12:44 +0000
Subject: [PATCH] Add script for detecting bad characters.

Co-authored-by: Shu Muto <shu.mutow@nec.com>
---
 scripts/README.md         | 26 +++++++++++++++++++
 scripts/check-ctrlcode.py | 54 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 80 insertions(+)
 create mode 100755 scripts/check-ctrlcode.py

diff --git a/scripts/README.md b/scripts/README.md
index 4f7ee7eec3..02e3e66ff0 100644
--- a/scripts/README.md
+++ b/scripts/README.md
@@ -11,6 +11,7 @@
 | `linkchecker.py` | This a link checker for Kubernetes documentation website. |
 | `lsync.sh` | This script checks if the English version of a page has changed since a localized page has been committed. |
 | `replace-capture.sh` | This script sets K8S_WEBSITE in your env to your docs website root or rely on this script to determine it automatically |
+| `check-ctrlcode.py` | This script finds control-code(0x00-0x1f) in text files. |
 
 
 
@@ -152,3 +153,28 @@ The following command checks a subdirectory:
 
     ./scripts/lsync.sh content/zh/docs/concepts/
 
+## check-ctrlcode.py
+
+This script finds control-code(0x00-0x1f) in text files.
+It will display illegal character in browser.
+
+```
+Usage: ./check-ctrlcode.py <dir> <ext>
+
+  <dir>    Specify the directory to check.
+  <ext>    Specify the extension to check.
+
+For example, we can execute as following.
+
+  ./check-ctrlcode.py ../content/en/ .md
+
+The output is following format.
+
+  "{0} <L{1}:{2}:{3}>: {4}"
+
+  {0} : The path of file that a control-code exists.
+  {1} : The line number that a control-code exists.
+  {2} : The column number that a control-code exists.
+  {3} : The found control-code.
+  {4} : The one-line strings in the file.
+```
\ No newline at end of file
diff --git a/scripts/check-ctrlcode.py b/scripts/check-ctrlcode.py
new file mode 100755
index 0000000000..b2705a409a
--- /dev/null
+++ b/scripts/check-ctrlcode.py
@@ -0,0 +1,54 @@
+#!/usr/bin/env python3
+
+import os
+import sys
+import re
+
+def main():
+    args = sys.argv
+    if (len(args) != 3):
+        print("Usage: ./check-ctrlcode.py <dir> <ext>")
+        sys.exit(1)
+
+    dirpath = args[1]
+    ext = args[2]
+
+    fullpath = os.path.abspath(dirpath)
+    if (os.path.isdir(fullpath) is not True):
+        print("Directory not found.")
+        sys.exit(1)
+
+    check_dir(fullpath, ext)
+
+def check_dir(path, ext):
+    for f in os.listdir(path):
+        if(f[0] == "."):
+            continue
+        fullpath = os.path.join(path, f)
+        if(os.path.isdir(fullpath)):
+            check_dir(fullpath, ext)
+            continue
+        exts = os.path.splitext(f)
+        if(exts[1] != ext):
+            continue
+        check_ctrlcode(fullpath)
+
+def check_ctrlcode(filepath):
+    line = 0
+    with open(filepath, encoding='utf-8') as f:
+        while True:
+            str = f.readline()
+            if(str == ""):
+                break
+            line = line + 1
+            # check 0x00-0x1f except 0x09(HT), 0x0a(LF), 0x0d(CR)
+            pattern = re.compile('[\u0000-\u0008\u000b\u000c\u000e-\u001f]')
+            m = pattern.search(str)
+            if(m == None):
+                continue
+            pos = m.end()
+            ctrl = m.group().encode("utf-8")
+            print("{0} <L{1}:{2}:{3}>: {4}\n".format(filepath, line, pos, ctrl, str.replace('\n','')))
+
+
+main()