openstack
/
openstack-doc-tools
Code Issues Proposed changes

validate modified files and build affected books 

At the moment the validation script simply validates all
files and builds all books afterwards. This results in a
very long runtime (appr. 30 minutes at the moment).

With this patch only modified files are validated and only
affected books are built. A book is affected if one XML
file includes a modified file. If no book is directly
affected all books are built.

In this way, resources can be saved and validation is
performed much faster.

Change-Id: I17dcf119c0bafc4656c9f267f0d3f5ae884ea76e
This commit is contained in:
Christian Berendt 2013-08-09 13:18:15 +02:00
parent d0abb8c290
commit 7c4b4bd4a3
1 changed files with 101 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

116
validate.py
View File

@ -34,6 +34,28 @@ FILE_EXCEPTIONS = ['ha-guide-docinfo.xml','bk001-ch003-associate-general.xml']
BOOK_EXCEPTIONS = []
# NOTE(berendt): check_output as provided in Python 2.7.5 to make script
# usable with Python < 2.7
def check_output(*popenargs, **kwargs):
"""Run command with arguments and return its output as a byte string.
If the exit code was non-zero it raises a CalledProcessError. The
CalledProcessError object will have the return code in the returncode
attribute and output in the output attribute.
"""
if 'stdout' in kwargs:
raise ValueError('stdout argument not allowed, it will be overridden.')
process = subprocess.Popen(stdout=subprocess.PIPE, *popenargs, **kwargs)
output, unused_err = process.communicate()
retcode = process.poll()
if retcode:
cmd = kwargs.get("args")
if cmd is None:
cmd = popenargs[0]
raise CalledProcessError(retcode, cmd, output=output)
return output
def get_schema():
"""Return the DocBook RELAX NG schema"""
url = "http://www.oasis-open.org/docbook/xml/5.0b5/rng/docbookxi.rng"
@ -112,10 +134,27 @@ def error_message(error_log):
return "\n".join(errs)
def get_modified_files():
try:
args = ["git", "diff", "--name-only", "HEAD", "HEAD~1"]
modified_files = check_output(args).strip().split()
except (CalledProcessError, OSError) as e:
print("git failed: %s" % e)
sys.exit(1)
modified_files = map(lambda x: os.path.abspath(x), modified_files)
return modified_files
def validate_individual_files(rootdir, exceptions):
schema = get_schema()
any_failures = False
modified_files = get_modified_files()
print("\nFollowing modified files were found:")
for f in modified_files:
print(">>> %s" % f)
print("")
for root, dirs, files in os.walk(rootdir):
# Don't descend into 'target' subdirectories
@ -127,44 +166,86 @@ def validate_individual_files(rootdir, exceptions):
for f in files:
# Ignore maven files, which are called pom.xml
if f.endswith('.xml') and f != 'pom.xml' \
and f not in exceptions:
if (f.endswith('.xml') and
f != 'pom.xml' and
f not in exceptions):
try:
path = os.path.abspath(os.path.join(root, f))
if path not in modified_files:
print((">>> %s not modified. "
"Skipping validation." % path))
continue
doc = etree.parse(path)
if validation_failed(schema, doc):
any_failures = True
print error_message(schema.error_log)
print(error_message(schema.error_log))
verify_section_tags_have_xmid(doc)
verify_nice_usage_of_whitespaces(os.path.join(root, f))
except etree.XMLSyntaxError as e:
any_failures = True
print "%s: %s" % (path, e)
print("%s: %s" % (path, e))
except ValueError as e:
any_failures = True
print "%s: %s" % (path, e)
print("%s: %s" % (path, e))
if any_failures:
sys.exit(1)
def build_all_books(rootdir, exceptions):
""" Build all of the books.
Looks for all directories with "pom.xml" in them and runs
def build_affected_books(rootdir, book_exceptions, file_exceptions):
"""Build all the books which are affected by modified files.
Looks for all directories with "pom.xml" and checks if a
XML file in the directory includes a modified file. If at least
one XML file includes a modified file the method calls
"mvn clean generate-sources" in that directory.
This will throw an exception if a book fails to build
"""
modified_files = get_modified_files()
print("\nFollowing modified files were found:")
for f in modified_files:
print(">>> %s" % f)
print("")
affected_books = []
books = []
for root, dirs, files in os.walk(rootdir):
book = os.path.basename(root)
if ("pom.xml" in files) and (book not in exceptions):
print "Building %s" % book
if ("pom.xml" in files and
os.path.basename(root) not in book_exceptions):
books.append(root)
os.chdir(root)
subprocess.check_call(["mvn", "clean", "generate-sources"])
for f in files:
if (f.endswith('.xml') and
f != 'pom.xml' and
f not in file_exceptions):
path = os.path.abspath(os.path.join(root, f))
doc = etree.parse(path)
ns = {"xi": "http://www.w3.org/2001/XInclude"}
for node in doc.xpath('//xi:include', namespaces=ns):
href = node.get('href')
if (href.endswith('.xml') and
f not in file_exceptions and
os.path.abspath(href) in modified_files):
affected_books.append(root)
break
if os.path.basename(root) in affected_books:
break
if affected_books:
books = affected_books
else:
print("No books are affected by modified files. Building all books.")
for book in books:
print("Building %s\n" % os.path.basename(book))
os.chdir(book)
subprocess.check_call(["mvn", "clean", "generate-sources"])
def main(rootdir):
validate_individual_files(rootdir, FILE_EXCEPTIONS)
build_all_books(rootdir, BOOK_EXCEPTIONS)
build_affected_books(rootdir, BOOK_EXCEPTIONS, FILE_EXCEPTIONS)
def default_root():
@ -172,8 +253,13 @@ def default_root():
The current working directory must be inside of the openstack-manuals
repository for this method to succeed"""
args = ["git", "rev-parse", "--show-toplevel"]
gitroot = subprocess.check_output(args).rstrip()
try:
args = ["git", "rev-parse", "--show-toplevel"]
gitroot = check_output(args).rstrip()
except (CalledProcessError, OSError) as e:
print("git failed: %s" % e)
sys.exit(1)
return os.path.join(gitroot, "doc/src/docbkx")
if __name__ == '__main__':