10386 pbchk should catch capitalised "illumos"

10387 pbchk should check commit message spelling
Reviewed by: Toomas Soome <tsoome@me.com>
Reviewed by: Gergő Mihály Doma <domag02@gmail.com>
Approved by: Dan McDonald <danmcd@joyent.com>

2 files changed, 47 insertions, 16 deletions

diff --git a/usr/src/tools/onbld/Checks/Comments.py b/usr/src/tools/onbld/Checks/Comments.py
index 4b4706618b..8285be2daa 100644
--- a/usr/src/tools/onbld/Checks/Comments.py
+++ b/usr/src/tools/onbld/Checks/Comments.py
@@ -26,17 +26,19 @@
 #
 
 # Copyright 2007, 2010 Richard Lowe
-# Copyright 2018 OmniOS Community Edition (OmniOSce) Association.
+# Copyright 2019 OmniOS Community Edition (OmniOSce) Association.
 
 #
 # Check delta comments:
 #	- Have the correct form.
 #	- Have a synopsis matching that of the bug
 #	- Appear only once.
+#	- Do not contain common spelling errors.
 #
 
 import re, sys
 from onbld.Checks.DbLookups import BugDB
+from onbld.Checks.SpellCheck import spellcheck_line
 
 
 bugre = re.compile(r'^(\d{2,7}) (.*)$')
@@ -71,12 +73,16 @@ def comchk(comments, check_db=True, output=sys.stderr):
 		   'mutant': [],
 		   'dup': [],
 		   'nomatch': [],
-		   'nonexistent': [] }
+		   'nonexistent': [],
+		   'spelling': [] }
 	bugs = {}
 	ret = 0
 	blanks = False
 
+	lineno = 0
 	for com in comments:
+		lineno += 1
+
 		# Our input must be newline-free, comments are line-wise.
 		if com.find('\n') != -1:
 			raise ValueError("newline in comment '%s'" % com)
@@ -89,6 +95,10 @@ def comchk(comments, check_db=True, output=sys.stderr):
 			blanks = True
 			continue
 
+		for err in spellcheck_line(com):
+			errors['spelling'].append(
+			    'comment line {} - {}'.format(lineno, err))
+
 		match = bugre.search(com)
 		if match:
 			if match.group(1) not in bugs:
@@ -179,4 +189,10 @@ def comchk(comments, check_db=True, output=sys.stderr):
 			output.write("  should be: '%s'\n" % err[1])
 			output.write("         is: '%s'\n" % err[2])
 
+	if errors['spelling']:
+		ret = 1
+		output.write("Spellcheck:\n")
+		for err in errors['spelling']:
+			output.write('{}\n'.format(err))
+
 	return ret
diff --git a/usr/src/tools/onbld/Checks/SpellCheck.py b/usr/src/tools/onbld/Checks/SpellCheck.py
index 0d07017846..c858947fe6 100644
--- a/usr/src/tools/onbld/Checks/SpellCheck.py
+++ b/usr/src/tools/onbld/Checks/SpellCheck.py
@@ -26,8 +26,9 @@
 
 import re, sys
 
-spellMsg = '%s: Line %d contains "%s", a common misspelling of "%s"\n'
-altMsg = '%s: Line %d contains "%s"; please use "%s" instead for consistency with other documentation\n'
+spellMsg = 'contains "{}", a common misspelling of "{}"'
+altMsg = 'contains "{}"; please use "{}" instead for consistency with other documentation'
+caseMsg = 'contains "{}"; please use "{}" instead'
 
 misspellings = {
 	'absense': 'absence',
@@ -253,8 +254,13 @@ alternates = {
 	'writeable': 'writable'
 }
 
+case = {
+	'Illumos': 'illumos'
+}
+
 misspellingREs = []
 alternateREs = []
+caseREs = []
 
 for misspelling, correct in misspellings.items():
 	regex = re.compile(r'\b%s\b' % (misspelling), re.IGNORECASE)
@@ -266,12 +272,23 @@ for alternate, correct in alternates.items():
 	entry = (regex, alternate, correct)
 	alternateREs.append(entry)
 
-def check(errmsg, output, filename, line, lineno, entry):
-	if entry[0].search(line):
-		output.write(errmsg % (filename, lineno, entry[1], entry[2]))
-		return 1
-	else:
-		return 0
+for alternate, correct in case.items():
+	regex = re.compile(r'\b%s\b' % (alternate))
+	entry = (regex, alternate, correct)
+	caseREs.append(entry)
+
+def spellcheck_line(line):
+	errs = []
+	for entry in misspellingREs:
+		if entry[0].search(line):
+			errs.append(spellMsg.format(entry[1], entry[2]))
+	for entry in alternateREs:
+		if entry[0].search(line):
+			errs.append(altMsg.format(entry[1], entry[2]))
+	for entry in caseREs:
+		if entry[0].search(line):
+			errs.append(caseMsg.format(entry[1], entry[2]))
+	return errs
 
 def spellcheck(fh, filename=None, output=sys.stderr, **opts):
 	lineno = 1
@@ -283,12 +300,10 @@ def spellcheck(fh, filename=None, output=sys.stderr, **opts):
 	fh.seek(0)
 	for line in fh:
 		line = line.decode(errors='replace')
-		for entry in misspellingREs:
-			ret |= check(spellMsg, output, filename, line,
-			    lineno, entry)
-		for entry in alternateREs:
-			ret |= check(altMsg, output, filename, line,
-			    lineno, entry)
+		for err in spellcheck_line(line):
+			output.write('{}: Line {} {}\n'.format(
+			    filename, lineno, err))
+			ret = 1
 		lineno += 1
 
 	return ret