[libvirt PATCH v2 1/3] scripts: Check spelling

[Tim Wiederhake]

This is a wrapper for codespell [1], a spell checker for source code.
Codespell does not compare words to a dictionary, but rather works by
checking words against a list of common typos, making it produce fewer
false positives than other solutions.

The script in this patch works around the lack of per-directory ignore
lists and some oddities regarding capitalization in ignore lists.

[1] (https://github.com/codespell-project/codespell/)

Signed-off-by: Tim Wiederhake <twiederh at redhat.com>
---
 scripts/check-spelling.py | 119 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 119 insertions(+)
 create mode 100755 scripts/check-spelling.py

diff --git a/scripts/check-spelling.py b/scripts/check-spelling.py
new file mode 100755
index 0000000000..0480a506e8
--- /dev/null
+++ b/scripts/check-spelling.py
@@ -0,0 +1,119 @@
+#!/usr/bin/env python3
+
+import argparse
+import re
+import subprocess
+import os
+
+
+IGNORE_LIST = [
+    # ignore all translation files
+    ("/po/", []),
+
+    # ignore all git files
+    ("/.git/", []),
+
+    # ignore this script
+    ("/scripts/check-spelling.py", []),
+
+    # 3rd-party: keycodemapdb
+    ("/src/keycodemapdb/", []),
+
+    # 3rd-party: VirtualBox SDK
+    ("/src/vbox/vbox_CAPI", []),
+
+    # 3rd-party: qemu
+    ("/tests/qemucapabilitiesdata/caps_", []),
+
+    # other
+    ("/", ["msdos", "MSDOS", "wan", "WAN", "hda", "HDA", "inout"]),
+    ("/NEWS.rst", "crashers"),
+    ("/docs/gitdm/companies/others", "Archiv"),
+    ("/docs/glib-adoption.rst", "preferrable"),
+    ("/docs/js/main.js", "whats"),
+    ("/examples/polkit/libvirt-acl.rules", ["userA", "userB", "userC"]),
+    ("/src/libvirt-domain.c", "PTD"),
+    ("/src/libxl/libxl_logger.c", "purposedly"),
+    ("/src/nwfilter/nwfilter_dhcpsnoop.c", "ether"),
+    ("/src/nwfilter/nwfilter_ebiptables_driver.c", "parm"),
+    ("/src/nwfilter/nwfilter_learnipaddr.c", "ether"),
+    ("/src/qemu/qemu_agent.c", "crypted"),
+    ("/src/qemu/qemu_agent.h", "crypted"),
+    ("/src/qemu/qemu_process.c", "wee"),
+    ("/src/security/apparmor/libvirt-lxc", "devic"),
+    ("/src/security/apparmor/libvirt-qemu", "readby"),
+    ("/src/storage_file/storage_file_probe.c", "conectix"),
+    ("/src/util/virnetdevmacvlan.c", "calld"),
+    ("/src/util/virtpm.c", "parm"),
+    ("/tests/qemuagenttest.c", "IST"),
+    ("/tests/storagepoolxml2xml", "cant"),
+    ("/tests/sysinfodata/", "sie"),
+    ("/tests/testutils.c", "nIn"),
+    ("/tests/vircgroupdata/ovirt-node-6.6.mounts", "hald"),
+    ("/tests/virhostcpudata/", "sie"),
+    ("/tools/virt-host-validate-common.c", "sie"),
+]
+
+
+def ignore(filename, linenumber, word, suggestion):
+    if len(word) <= 2:
+        return True
+
+    for f, w in IGNORE_LIST:
+        if not filename.startswith(f):
+            continue
+        if word in w or not w:
+            return True
+    return False
+
+
+def main():
+    line_pattern = re.compile("^(.*):(.*): (.*) ==> (.*)$")
+    output_template = "(\"{0}\", \"{2}\"),\t# line {1}, \"{3}\"?"
+
+    parser = argparse.ArgumentParser(description="Check spelling")
+    parser.add_argument(
+        "dir",
+        help="Path to source directory. "
+        "Defaults to parent directory of this script",
+        type=os.path.realpath,
+        nargs='?')
+    args = parser.parse_args()
+
+    if not args.dir:
+        args.dir = os.path.dirname(os.path.dirname(os.path.realpath(__file__)))
+
+    process = subprocess.run(
+        ["codespell", args.dir],
+        stdout=subprocess.PIPE,
+        stderr=subprocess.PIPE,
+        universal_newlines=True)
+
+    if process.returncode not in (0, 65):
+        exit("error: unexpected returncode %s" % process.returncode)
+
+    if process.stderr:
+        exit("error: unexpected output to stderr: \"%s\"" % process.stderr)
+
+    findings = 0
+    for line in process.stdout.split("\n"):
+        line = line.strip().replace(args.dir, "")
+        if not line:
+            continue
+
+        match = line_pattern.match(line)
+        if not match:
+            exit("error: unexpected line: \"%s\"" % line)
+
+        if ignore(*match.groups()):
+            continue
+
+        print(output_template.format(*match.groups()))
+        findings += 1
+
+    if findings:
+        exit("error: %s spelling errors" % findings)
+
+
+if __name__ == "__main__":
+    main()
-- 
2.31.1