Change the user action file format from .txt to .xml.

In this way, more information can be added (currently added 'description' and 'owner' for each action) 

A few functions are moved from tools/metrics/histograms to tools/metrics/common to be shared by tools/metrics and tools/histograms.

BUG=340735
NOTRY=true

Review URL: https://codereview.chromium.org/149503005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@255357 0039d316-1c4b-4281-b951-d872f2087c98

PRESUBMIT.py

@@ -1014,28 +1014,32 @@ def _CheckCygwinShell(input_api, output_api):
 
 def _CheckUserActionUpdate(input_api, output_api):
   """Checks if any new user action has been added."""
-  if any('chromeactions.txt' == input_api.os_path.basename(f) for f in
+  if any('actions.xml' == input_api.os_path.basename(f) for f in
          input_api.LocalPaths()):
-    # If chromeactions.txt is already included in the changelist, the PRESUBMIT
-    # for chromeactions.txt will do a more complete presubmit check.
+    # If actions.xml is already included in the changelist, the PRESUBMIT
+    # for actions.xml will do a more complete presubmit check.
     return []
 
-  with open('tools/metrics/actions/chromeactions.txt') as f:
-    current_actions = f.read()
-
   file_filter = lambda f: f.LocalPath().endswith(('.cc', '.mm'))
   action_re = r'[^a-zA-Z]UserMetricsAction\("([^"]*)'
+  current_actions = None
   for f in input_api.AffectedFiles(file_filter=file_filter):
     for line_num, line in f.ChangedContents():
       match = input_api.re.search(action_re, line)
       if match:
+        # Loads contents in tools/metrics/actions/actions.xml to memory. It's
+        # loaded only once.
+        if not current_actions:
+          with open('tools/metrics/actions/actions.xml') as actions_f:
+            current_actions = actions_f.read()
+        # Search for the matched user action name in |current_actions|.
         for action_name in match.groups():
-          name_pattern = r'\t%s\n' % action_name
-          if name_pattern not in current_actions:
+          action = 'name="{0}"'.format(action_name)
+          if action not in current_actions:
             return [output_api.PresubmitPromptWarning(
               'File %s line %d: %s is missing in '
-              'tools/metrics/actions/chromeactions.txt. Please run '
-              'tools/metrics/actions/extract_actions.py --hash to update.'
+              'tools/metrics/actions/actions.xml. Please run '
+              'tools/metrics/actions/extract_actions.py to update.'
               % (f.LocalPath(), line_num, action_name))]
   return []
 

tools/metrics/actions/OWNERS

-# per-file rules:
-per-file chromeactions.txt=*

tools/metrics/actions/PRESUBMIT.py

+# Copyright 2014 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+"""Presubmit script for actions.xml.
+
+See http://dev.chromium.org/developers/how-tos/depottools/presubmit-scripts
+for more details on the presubmit API built into gcl.
+"""
+
+
+def CheckChange(input_api, output_api):
+  """Checks that actions.xml is up to date and pretty-printed."""
+  for f in input_api.AffectedTextFiles():
+    p = f.AbsoluteLocalPath()
+    if (input_api.basename(p) == 'actions.xml'
+        and input_api.os_path.dirname(p) == input_api.PresubmitLocalPath()):
+      cwd = input_api.os_path.dirname(p)
+      exit_code = input_api.subprocess.call(
+          ['python', 'extract_actions.py', '--presubmit'], cwd=cwd)
+      if exit_code != 0:
+        return [output_api.PresubmitError(
+            'actions.xml is not up to date or is not formatted correctly; '
+            'run extract_actions.py to fix')]
+  return []
+
+
+def CheckChangeOnUpload(input_api, output_api):
+  return CheckChange(input_api, output_api)
+
+
+def CheckChangeOnCommit(input_api, output_api):
+  return CheckChange(input_api, output_api)

tools/metrics/actions/extract_actions_test.py

+#!/usr/bin/env python
+# Copyright 2014 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+import unittest
+
+import extract_actions
+
+# Empty value to be inserted to |ACTIONS_MOCK|.
+NO_VALUE = ''
+
+ONE_OWNER = '<owner>name1@google.com</owner>\n'
+TWO_OWNERS = """
+<owner>name1@google.com</owner>\n
+<owner>name2@google.com</owner>\n
+"""
+
+DESCRIPTION = '<description>Description.</description>\n'
+TWO_DESCRIPTIONS = """
+<description>Description.</description>\n
+<description>Description2.</description>\n
+"""
+
+OBSOLETE = '<obsolete>Not used anymore. Replaced by action2.</obsolete>\n'
+TWO_OBSOLETE = '<obsolete>Obsolete1.</obsolete><obsolete>Obsolete2.</obsolete>'
+
+COMMENT = '<!--comment-->'
+
+# A format string to mock the input action.xml file.
+ACTIONS_XML = """
+{comment}
+<actions>
+
+<action name="action1">
+{owners}{obsolete}{description}
+</action>
+
+</actions>"""
+
+NO_OWNER_EXPECTED_XML = (
+    '<actions>\n\n'
+    '<action name="action1">\n'
+    '  <owner>Please list the metric\'s owners. '
+    'Add more owner tags as needed.</owner>\n'
+    '  <description>Description.</description>\n'
+    '</action>\n\n'
+    '</actions>\n'
+)
+
+ONE_OWNER_EXPECTED_XML = (
+    '<actions>\n\n'
+    '<action name="action1">\n'
+    '  <owner>name1@google.com</owner>\n'
+    '  <description>Description.</description>\n'
+    '</action>\n\n'
+    '</actions>\n'
+)
+
+TWO_OWNERS_EXPECTED_XML = (
+    '<actions>\n\n'
+    '<action name="action1">\n'
+    '  <owner>name1@google.com</owner>\n'
+    '  <owner>name2@google.com</owner>\n'
+    '  <description>Description.</description>\n'
+    '</action>\n\n'
+    '</actions>\n'
+)
+
+NO_DESCRIPTION_EXPECTED_XML = (
+    '<actions>\n\n'
+    '<action name="action1">\n'
+    '  <owner>name1@google.com</owner>\n'
+    '  <owner>name2@google.com</owner>\n'
+    '  <description>Please enter the description of the metric.</description>\n'
+    '</action>\n\n'
+    '</actions>\n'
+)
+
+OBSOLETE_EXPECTED_XML = (
+    '<actions>\n\n'
+    '<action name="action1">\n'
+    '  <owner>name1@google.com</owner>\n'
+    '  <owner>name2@google.com</owner>\n'
+    '  <description>Description.</description>\n'
+    '  <obsolete>Not used anymore. Replaced by action2.</obsolete>\n'
+    '</action>\n\n'
+    '</actions>\n'
+)
+
+ADD_ACTION_EXPECTED_XML = (
+    '<actions>\n\n'
+    '<action name="action1">\n'
+    '  <owner>name1@google.com</owner>\n'
+    '  <owner>name2@google.com</owner>\n'
+    '  <description>Description.</description>\n'
+    '</action>\n\n'
+    '<action name="action2">\n'
+    '  <owner>Please list the metric\'s owners.'
+    ' Add more owner tags as needed.</owner>\n'
+    '  <description>Please enter the description of the metric.</description>\n'
+    '</action>\n\n'
+    '</actions>\n'
+)
+
+COMMENT_EXPECTED_XML = (
+    '<!--comment-->\n\n'
+    '<actions>\n\n'
+    '<action name="action1">\n'
+    '  <owner>name1@google.com</owner>\n'
+    '  <owner>name2@google.com</owner>\n'
+    '  <description>Description.</description>\n'
+    '</action>\n\n'
+    '</actions>\n'
+)
+
+
+class ActionXmlTest(unittest.TestCase):
+
+  def _GetProcessedAction(self, owner, description, obsolete, new_actions=[],
+                          comment=NO_VALUE):
+    """Forms an actions XML string and returns it after processing.
+
+    It parses the original XML string, adds new user actions (if there is any),
+    and pretty prints it.
+
+    Args:
+      owner: the owner tag to be inserted in the original XML string.
+      description: the description tag to be inserted in the original XML
+        string.
+      obsolete: the obsolete tag to be inserted in the original XML string.
+      new_actions: optional. List of new user actions' names to be inserted.
+      comment: the comment tag to be inserted in the original XML string.
+
+    Returns:
+      An updated and pretty-printed action XML string.
+    """
+    # Form the actions.xml mock content based on the input parameters.
+    current_xml = ACTIONS_XML.format(owners=owner, description=description,
+                                     obsolete=obsolete, comment=comment)
+    actions, actions_dict, comments = extract_actions.ParseActionFile(
+        current_xml)
+    for new_action in new_actions:
+      actions.add(new_action)
+    return extract_actions.PrettyPrint(actions, actions_dict, comments)
+
+  def testNoOwner(self):
+    xml_result = self._GetProcessedAction(NO_VALUE, DESCRIPTION, NO_VALUE)
+    self.assertEqual(NO_OWNER_EXPECTED_XML, xml_result)
+
+  def testOneOwnerOneDescription(self):
+    xml_result = self._GetProcessedAction(ONE_OWNER, DESCRIPTION, NO_VALUE)
+    self.assertEqual(ONE_OWNER_EXPECTED_XML, xml_result)
+
+  def testTwoOwners(self):
+    xml_result = self._GetProcessedAction(TWO_OWNERS, DESCRIPTION, NO_VALUE)
+    self.assertEqual(TWO_OWNERS_EXPECTED_XML, xml_result)
+
+  def testNoDescription(self):
+    xml_result = self._GetProcessedAction(TWO_OWNERS, NO_VALUE, NO_VALUE)
+    self.assertEqual(NO_DESCRIPTION_EXPECTED_XML, xml_result)
+
+  def testTwoDescriptions(self):
+    current_xml = ACTIONS_XML.format(owners=TWO_OWNERS, obsolete=NO_VALUE,
+                                     description=TWO_DESCRIPTIONS,
+                                     comment=NO_VALUE)
+    # Since there are two description tags, the function ParseActionFile will
+    # raise SystemExit with exit code 1.
+    with self.assertRaises(SystemExit) as cm:
+      _, _ = extract_actions.ParseActionFile(current_xml)
+    self.assertEqual(cm.exception.code, 1)
+
+  def testObsolete(self):
+    xml_result = self._GetProcessedAction(TWO_OWNERS, DESCRIPTION, OBSOLETE)
+    self.assertEqual(OBSOLETE_EXPECTED_XML, xml_result)
+
+
+  def testTwoObsoletes(self):
+    current_xml = ACTIONS_XML.format(owners=TWO_OWNERS, obsolete=TWO_OBSOLETE,
+                                     description=DESCRIPTION, comment=NO_VALUE)
+    # Since there are two obsolete tags, the function ParseActionFile will
+    # raise SystemExit with exit code 1.
+    with self.assertRaises(SystemExit) as cm:
+      _, _ = extract_actions.ParseActionFile(current_xml)
+    self.assertEqual(cm.exception.code, 1)
+
+  def testAddNewActions(self):
+    xml_result = self._GetProcessedAction(TWO_OWNERS, DESCRIPTION, NO_VALUE,
+                                          new_actions=['action2'])
+    self.assertEqual(ADD_ACTION_EXPECTED_XML, xml_result)
+
+  def testComment(self):
+    xml_result = self._GetProcessedAction(TWO_OWNERS, DESCRIPTION, NO_VALUE,
+                                          comment=COMMENT)
+    self.assertEqual(COMMENT_EXPECTED_XML, xml_result)
+
+
+if __name__ == '__main__':
+  unittest.main()

tools/metrics/actions/print_style.py

+# Copyright 2014 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+"""Holds the constants for pretty printing actions.xml."""
+
+import os
+import sys
+
+# Import the metrics/common module for pretty print xml.
+sys.path.append(os.path.join(os.path.dirname(__file__), '..', 'common'))
+import pretty_print_xml
+
+# Desired order for tag and tag attributes.
+# { tag_name: [attribute_name, ...] }
+ATTRIBUTE_ORDER = {
+    'action': ['name'],
+    'owner': [],
+    'description': [],
+    'obsolete': [],
+}
+
+# Tag names for top-level nodes whose children we don't want to indent.
+TAGS_THAT_DONT_INDENT = ['actions']
+
+# Extra vertical spacing rules for special tag names.
+# {tag_name: (newlines_after_open, newlines_before_close, newlines_after_close)}
+TAGS_THAT_HAVE_EXTRA_NEWLINE = {
+    'actions': (2, 1, 1),
+    'action': (1, 1, 1),
+}
+
+# Tags that we allow to be squished into a single line for brevity.
+TAGS_THAT_ALLOW_SINGLE_LINE = ['owner', 'description', 'obsolete']
+
+def GetPrintStyle():
+  """Returns an XmlStyle object for pretty printing actions."""
+  return pretty_print_xml.XmlStyle(ATTRIBUTE_ORDER,
+                                   TAGS_THAT_HAVE_EXTRA_NEWLINE,
+                                   TAGS_THAT_DONT_INDENT,
+                                   TAGS_THAT_ALLOW_SINGLE_LINE)

tools/metrics/histograms/diffutil.py → tools/metrics/common/diff_util.py

-# Copyright 2013 The Chromium Authors. All rights reserved.
+# Copyright 2014 The Chromium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style license that can be
 # found in the LICENSE file.
 

tools/metrics/common/pretty_print_xml.py

+# Copyright 2014 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+"""Utility file for pretty print xml file.
+
+The function PrettyPrintNode will be used for formatting both histograms.xml
+and actions.xml.
+"""
+
+import logging
+import textwrap
+import xml.dom.minidom
+
+WRAP_COLUMN = 80
+
+
+class Error(Exception):
+  pass
+
+
+def LastLineLength(s):
+  """Returns the length of the last line in s.
+
+  Args:
+    s: A multi-line string, including newlines.
+
+  Returns:
+    The length of the last line in s, in characters.
+  """
+  if s.rfind('\n') == -1: return len(s)
+  return len(s) - s.rfind('\n') - len('\n')
+
+
+def XmlEscape(s):
+  """XML-escapes the given string, replacing magic characters (&<>") with their
+  escaped equivalents."""
+  s = s.replace("&", "&amp;").replace("<", "&lt;")
+  s = s.replace("\"", "&quot;").replace(">", "&gt;")
+  return s
+
+
+class XmlStyle(object):
+  """A class that stores all style specification for an output xml file."""
+
+  def __init__(self, attribute_order, tags_that_have_extra_newline,
+               tags_that_dont_indent, tags_that_allow_single_line):
+    # List of tag names for top-level nodes whose children are not indented.
+    self.attribute_order = attribute_order
+    self.tags_that_have_extra_newline = tags_that_have_extra_newline
+    self.tags_that_dont_indent = tags_that_dont_indent
+    self.tags_that_allow_single_line = tags_that_allow_single_line
+
+  def PrettyPrintNode(self, node, indent=0):
+    """Pretty-prints the given XML node at the given indent level.
+
+    Args:
+      node: The minidom node to pretty-print.
+      indent: The current indent level.
+
+    Returns:
+      The pretty-printed string (including embedded newlines).
+
+    Raises:
+      Error if the XML has unknown tags or attributes.
+    """
+    # Handle the top-level document node.
+    if node.nodeType == xml.dom.minidom.Node.DOCUMENT_NODE:
+      return '\n'.join([self.PrettyPrintNode(n) for n in node.childNodes])
+
+    # Handle text nodes.
+    if node.nodeType == xml.dom.minidom.Node.TEXT_NODE:
+      # Wrap each paragraph in the text to fit in the 80 column limit.
+      wrapper = textwrap.TextWrapper()
+      wrapper.initial_indent = ' ' * indent
+      wrapper.subsequent_indent = ' ' * indent
+      wrapper.break_on_hyphens = False
+      wrapper.break_long_words = False
+      wrapper.width = WRAP_COLUMN
+      text = XmlEscape(node.data)
+      # Remove any common indent.
+      text = textwrap.dedent(text.strip('\n'))
+      lines = text.split('\n')
+      # Split the text into paragraphs at blank line boundaries.
+      paragraphs = [[]]
+      for l in lines:
+        if len(l.strip()) == 0 and len(paragraphs[-1]) > 0:
+          paragraphs.append([])
+        else:
+          paragraphs[-1].append(l)
+      # Remove trailing empty paragraph if present.
+      if len(paragraphs) > 0 and len(paragraphs[-1]) == 0:
+        paragraphs = paragraphs[:-1]
+      # Wrap each paragraph and separate with two newlines.
+      return '\n\n'.join([wrapper.fill('\n'.join(p)) for p in paragraphs])
+
+    # Handle element nodes.
+    if node.nodeType == xml.dom.minidom.Node.ELEMENT_NODE:
+      newlines_after_open, newlines_before_close, newlines_after_close = (
+          self.tags_that_have_extra_newline.get(node.tagName, (1, 1, 0)))
+      # Open the tag.
+      s = ' ' * indent + '<' + node.tagName
+
+      # Calculate how much space to allow for the '>' or '/>'.
+      closing_chars = 1
+      if not node.childNodes:
+        closing_chars = 2
+
+      # Pretty-print the attributes.
+      attributes = node.attributes.keys()
+      if attributes:
+        # Reorder the attributes.
+        if node.tagName not in self.attribute_order:
+          unrecognized_attributes = attributes
+        else:
+          unrecognized_attributes = (
+              [a for a in attributes
+               if a not in self.attribute_order[node.tagName]])
+          attributes = [a for a in self.attribute_order[node.tagName]
+                        if a in attributes]
+
+        for a in unrecognized_attributes:
+          logging.error(
+              'Unrecognized attribute "%s" in tag "%s"' % (a, node.tagName))
+        if unrecognized_attributes:
+          raise Error()
+
+        for a in attributes:
+          value = XmlEscape(node.attributes[a].value)
+          # Replace sequences of whitespace with single spaces.
+          words = value.split()
+          a_str = ' %s="%s"' % (a, ' '.join(words))
+          # Start a new line if the attribute will make this line too long.
+          if LastLineLength(s) + len(a_str) + closing_chars > WRAP_COLUMN:
+            s += '\n' + ' ' * (indent + 3)
+          # Output everything up to the first quote.
+          s += ' %s="' % (a)
+          value_indent_level = LastLineLength(s)
+          # Output one word at a time, splitting to the next line where
+          # necessary.
+          column = value_indent_level
+          for i, word in enumerate(words):
+            # This is slightly too conservative since not every word will be
+            # followed by the closing characters...
+            if i > 0 and (column + len(word) + 1 + closing_chars > WRAP_COLUMN):
+              s = s.rstrip()  # remove any trailing whitespace
+              s += '\n' + ' ' * value_indent_level
+              column = value_indent_level
+            s += word + ' '
+            column += len(word) + 1
+          s = s.rstrip()  # remove any trailing whitespace
+          s += '"'
+        s = s.rstrip()  # remove any trailing whitespace
+
+      # Pretty-print the child nodes.
+      if node.childNodes:
+        s += '>'
+        # Calculate the new indent level for child nodes.
+        new_indent = indent
+        if node.tagName not in self.tags_that_dont_indent:
+          new_indent += 2
+        child_nodes = node.childNodes
+
+        # Recursively pretty-print the child nodes.
+        child_nodes = [self.PrettyPrintNode(n, indent=new_indent)
+                       for n in child_nodes]
+        child_nodes = [c for c in child_nodes if len(c.strip()) > 0]
+
+        # Determine whether we can fit the entire node on a single line.
+        close_tag = '</%s>' % node.tagName
+        space_left = WRAP_COLUMN - LastLineLength(s) - len(close_tag)
+        if (node.tagName in self.tags_that_allow_single_line and
+            len(child_nodes) == 1 and
+            len(child_nodes[0].strip()) <= space_left):
+          s += child_nodes[0].strip()
+        else:
+          s += '\n' * newlines_after_open + '\n'.join(child_nodes)
+          s += '\n' * newlines_before_close + ' ' * indent
+        s += close_tag
+      else:
+        s += '/>'
+      s += '\n' * newlines_after_close
+      return s
+
+    # Handle comment nodes.
+    if node.nodeType == xml.dom.minidom.Node.COMMENT_NODE:
+      return '<!--%s-->\n' % node.data
+
+    # Ignore other node types. This could be a processing instruction
+    # (<? ... ?>) or cdata section (<![CDATA[...]]!>), neither of which are
+    # legal in the histograms XML at present.
+    logging.error('Ignoring unrecognized node data: %s' % node.toxml())
+    raise Error()

tools/metrics/histograms/pretty_print.py

@@ -14,58 +14,20 @@ and wrapping text nodes, so we implement our own full custom XML pretty-printer.
 
 from __future__ import with_statement
 
-import diffutil
-import json
 import logging
 import os
 import shutil
 import sys
-import textwrap
 import xml.dom.minidom
 
+import print_style
+
 sys.path.insert(1, os.path.join(sys.path[0], '..', '..', 'python'))
 from google import path_utils
 
-WRAP_COLUMN = 80
-
-# Desired order for tag attributes; attributes listed here will appear first,
-# and in the same order as in these lists.
-# { tag_name: [attribute_name, ...] }
-ATTRIBUTE_ORDER = {
-  'enum': ['name', 'type'],
-  'histogram': ['name', 'enum', 'units'],
-  'int': ['value', 'label'],
-  'fieldtrial': ['name', 'separator', 'ordering'],
-  'group': ['name', 'label'],
-  'affected-histogram': ['name'],
-  'with-group': ['name'],
-}
-
-# Tag names for top-level nodes whose children we don't want to indent.
-TAGS_THAT_DONT_INDENT = [
-  'histogram-configuration',
-  'histograms',
-  'fieldtrials',
-  'enums'
-]
-
-# Extra vertical spacing rules for special tag names.
-# {tag_name: (newlines_after_open, newlines_before_close, newlines_after_close)}
-TAGS_THAT_HAVE_EXTRA_NEWLINE = {
-  'histogram-configuration': (2, 1, 1),
-  'histograms': (2, 1, 1),
-  'fieldtrials': (2, 1, 1),
-  'enums': (2, 1, 1),
-  'histogram': (1, 1, 1),
-  'enum': (1, 1, 1),
-  'fieldtrial': (1, 1, 1),
-}
-
-# Tags that we allow to be squished into a single line for brevity.
-TAGS_THAT_ALLOW_SINGLE_LINE = [
-  'summary',
-  'int',
-]
+# Import the metrics/common module.
+sys.path.append(os.path.join(os.path.dirname(__file__), '..', 'common'))
+import diff_util
 
 # Tags whose children we want to alphabetize. The key is the parent tag name,
 # and the value is a pair of the tag name of the children we want to sort,
@@ -84,165 +46,6 @@ class Error(Exception):
   pass
 
 
-def LastLineLength(s):
-  """Returns the length of the last line in s.
-
-  Args:
-    s: A multi-line string, including newlines.
-
-  Returns:
-    The length of the last line in s, in characters.
-  """
-  if s.rfind('\n') == -1: return len(s)
-  return len(s) - s.rfind('\n') - len('\n')
-
-
-def XmlEscape(s):
-  """XML-escapes the given string, replacing magic characters (&<>") with their
-  escaped equivalents."""
-  s = s.replace("&", "&amp;").replace("<", "&lt;")
-  s = s.replace("\"", "&quot;").replace(">", "&gt;")
-  return s
-
-
-def PrettyPrintNode(node, indent=0):
-  """Pretty-prints the given XML node at the given indent level.
-
-  Args:
-    node: The minidom node to pretty-print.
-    indent: The current indent level.
-
-  Returns:
-    The pretty-printed string (including embedded newlines).
-
-  Raises:
-    Error if the XML has unknown tags or attributes.
-  """
-  # Handle the top-level document node.
-  if node.nodeType == xml.dom.minidom.Node.DOCUMENT_NODE:
-    return '\n'.join([PrettyPrintNode(n) for n in node.childNodes])
-
-  # Handle text nodes.
-  if node.nodeType == xml.dom.minidom.Node.TEXT_NODE:
-    # Wrap each paragraph in the text to fit in the 80 column limit.
-    wrapper = textwrap.TextWrapper()
-    wrapper.initial_indent = ' ' * indent
-    wrapper.subsequent_indent = ' ' * indent
-    wrapper.break_on_hyphens = False
-    wrapper.break_long_words = False
-    wrapper.width = WRAP_COLUMN
-    text = XmlEscape(node.data)
-    # Remove any common indent.
-    text = textwrap.dedent(text.strip('\n'))
-    lines = text.split('\n')
-    # Split the text into paragraphs at blank line boundaries.
-    paragraphs = [[]]
-    for l in lines:
-      if len(l.strip()) == 0 and len(paragraphs[-1]) > 0:
-        paragraphs.append([])
-      else:
-        paragraphs[-1].append(l)
-    # Remove trailing empty paragraph if present.
-    if len(paragraphs) > 0 and len(paragraphs[-1]) == 0:
-      paragraphs = paragraphs[:-1]
-    # Wrap each paragraph and separate with two newlines.
-    return '\n\n'.join([wrapper.fill('\n'.join(p)) for p in paragraphs])
-
-  # Handle element nodes.
-  if node.nodeType == xml.dom.minidom.Node.ELEMENT_NODE:
-    newlines_after_open, newlines_before_close, newlines_after_close = (
-      TAGS_THAT_HAVE_EXTRA_NEWLINE.get(node.tagName, (1, 1, 0)))
-    # Open the tag.
-    s = ' ' * indent + '<' + node.tagName
-
-    # Calculate how much space to allow for the '>' or '/>'.
-    closing_chars = 1
-    if not node.childNodes:
-      closing_chars = 2
-
-    # Pretty-print the attributes.
-    attributes = node.attributes.keys()
-    if attributes:
-      # Reorder the attributes.
-      if not node.tagName in ATTRIBUTE_ORDER:
-        unrecognized_attributes = attributes;
-      else:
-        unrecognized_attributes = (
-          [a for a in attributes if not a in ATTRIBUTE_ORDER[node.tagName]])
-        attributes = (
-          [a for a in ATTRIBUTE_ORDER[node.tagName] if a in attributes])
-
-      for a in unrecognized_attributes:
-        logging.error(
-            'Unrecognized attribute "%s" in tag "%s"' % (a, node.tagName))
-      if unrecognized_attributes:
-        raise Error()
-
-      for a in attributes:
-        value = XmlEscape(node.attributes[a].value)
-        # Replace sequences of whitespace with single spaces.
-        words = value.split()
-        a_str = ' %s="%s"' % (a, ' '.join(words))
-        # Start a new line if the attribute will make this line too long.
-        if LastLineLength(s) + len(a_str) + closing_chars > WRAP_COLUMN:
-          s += '\n' + ' ' * (indent + 3)
-        # Output everything up to the first quote.
-        s += ' %s="' % (a)
-        value_indent_level = LastLineLength(s)
-        # Output one word at a time, splitting to the next line where necessary.
-        column = value_indent_level
-        for i, word in enumerate(words):
-          # This is slightly too conservative since not every word will be
-          # followed by the closing characters...
-          if i > 0 and (column + len(word) + 1 + closing_chars > WRAP_COLUMN):
-            s = s.rstrip()  # remove any trailing whitespace
-            s += '\n' + ' ' * value_indent_level
-            column = value_indent_level
-          s += word + ' '
-          column += len(word) + 1
-        s = s.rstrip()  # remove any trailing whitespace
-        s += '"'
-      s = s.rstrip()  # remove any trailing whitespace
-
-    # Pretty-print the child nodes.
-    if node.childNodes:
-      s += '>'
-      # Calculate the new indent level for child nodes.
-      new_indent = indent
-      if node.tagName not in TAGS_THAT_DONT_INDENT:
-        new_indent += 2
-      child_nodes = node.childNodes
-
-      # Recursively pretty-print the child nodes.
-      child_nodes = [PrettyPrintNode(n, indent=new_indent) for n in child_nodes]
-      child_nodes = [c for c in child_nodes if len(c.strip()) > 0]
-
-      # Determine whether we can fit the entire node on a single line.
-      close_tag = '</%s>' % node.tagName
-      space_left = WRAP_COLUMN - LastLineLength(s) - len(close_tag)
-      if (node.tagName in TAGS_THAT_ALLOW_SINGLE_LINE and
-          len(child_nodes) == 1 and len(child_nodes[0].strip()) <= space_left):
-        s += child_nodes[0].strip()
-      else:
-        s += '\n' * newlines_after_open + '\n'.join(child_nodes)
-        s += '\n' * newlines_before_close + ' ' * indent
-      s += close_tag
-    else:
-      s += '/>'
-    s += '\n' * newlines_after_close
-    return s
-
-  # Handle comment nodes.
-  if node.nodeType == xml.dom.minidom.Node.COMMENT_NODE:
-    return '<!--%s-->\n' % node.data
-
-  # Ignore other node types. This could be a processing instruction (<? ... ?>)
-  # or cdata section (<![CDATA[...]]!>), neither of which are legal in the
-  # histograms XML at present.
-  logging.error('Ignoring unrecognized node data: %s' % node.toxml())
-  raise Error()
-
-
 def unsafeAppendChild(parent, child):
   """Append child to parent's list of children, ignoring the possibility that it
   is already in another node's childNodes list.  Requires that the previous
@@ -304,14 +107,14 @@ def PrettyPrint(raw_xml):
   """Pretty-print the given XML.
 
   Args:
-    xml: The contents of the histograms XML file, as a string.
+    raw_xml: The contents of the histograms XML file, as a string.
 
   Returns:
     The pretty-printed version.
   """
   tree = xml.dom.minidom.parseString(raw_xml)
   tree = TransformByAlphabetizing(tree)
-  return PrettyPrintNode(tree)
+  return print_style.GetPrintStyle().PrettyPrintNode(tree)
 
 
 def main():
@@ -356,7 +159,7 @@ def main():
     logging.info('%s is not formatted correctly; run pretty_print.py to fix.' %
                  histograms_filename)
     sys.exit(1)
-  if not diffutil.PromptUserToAcceptDiff(
+  if not diff_util.PromptUserToAcceptDiff(
       xml, pretty,
       'Is the prettified version acceptable?'):
     logging.error('Aborting')

tools/metrics/histograms/print_style.py

+# Copyright 2014 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style license that can be
+# found in the LICENSE file.
+
+"""Holds the constants for pretty printing histograms.xml."""
+
+import os
+import sys
+
+# Import the metrics/common module for pretty print xml.
+sys.path.append(os.path.join(os.path.dirname(__file__), '..', 'common'))
+import pretty_print_xml
+
+# Desired order for tag attributes; attributes listed here will appear first,
+# and in the same order as in these lists.
+# { tag_name: [attribute_name, ...] }
+ATTRIBUTE_ORDER = {
+  'enum': ['name', 'type'],
+  'histogram': ['name', 'enum', 'units'],
+  'int': ['value', 'label'],
+  'fieldtrial': ['name', 'separator', 'ordering'],
+  'group': ['name', 'label'],
+  'affected-histogram': ['name'],
+  'with-group': ['name'],
+}
+
+# Tag names for top-level nodes whose children we don't want to indent.
+TAGS_THAT_DONT_INDENT = [
+  'histogram-configuration',
+  'histograms',
+  'fieldtrials',
+  'enums'
+]
+
+# Extra vertical spacing rules for special tag names.
+# {tag_name: (newlines_after_open, newlines_before_close, newlines_after_close)}
+TAGS_THAT_HAVE_EXTRA_NEWLINE = {
+  'histogram-configuration': (2, 1, 1),
+  'histograms': (2, 1, 1),
+  'fieldtrials': (2, 1, 1),
+  'enums': (2, 1, 1),
+  'histogram': (1, 1, 1),
+  'enum': (1, 1, 1),
+  'fieldtrial': (1, 1, 1),
+}
+
+# Tags that we allow to be squished into a single line for brevity.
+TAGS_THAT_ALLOW_SINGLE_LINE = [
+  'summary',
+  'int',
+]
+
+

tools/metrics/histograms/update_extension_functions.py

@@ -13,9 +13,11 @@ import re
 import sys
 
 from xml.dom import minidom
+import print_style
 
-from diffutil import PromptUserToAcceptDiff
-from pretty_print import PrettyPrintNode
+# Import the metrics/common module.
+sys.path.append(os.path.join(os.path.dirname(__file__), '..', 'common'))
+from diff_util import PromptUserToAcceptDiff
 
 HISTOGRAMS_PATH = 'histograms.xml'
 ENUM_NAME = 'ExtensionFunctions'
@@ -136,7 +138,8 @@ def main():
   UpdateHistogramDefinitions(histogram_values, histograms_doc)
 
   Log('Writing out new histograms file.')
-  new_xml = PrettyPrintNode(histograms_doc)
+  new_xml = print_style.GetPrintStyle().PrettyPrintNode(histograms_doc)
+
   if PromptUserToAcceptDiff(xml, new_xml, 'Is the updated version acceptable?'):
     with open(HISTOGRAMS_PATH, 'wb') as f:
       f.write(new_xml)

tools/metrics/histograms/update_policies.py

@@ -8,6 +8,7 @@ definitions read from policy_templates.json.
 If the file was pretty-printed, the updated version is pretty-printed too.
 """
 
+import os
 import re
 import sys
 
@@ -15,12 +16,15 @@ from ast import literal_eval
 from optparse import OptionParser
 from xml.dom import minidom
 
-from diffutil import PromptUserToAcceptDiff
-from pretty_print import PrettyPrintNode
+import print_style
+
+# Import the metrics/common module.
+sys.path.append(os.path.join(os.path.dirname(__file__), '..', 'common'))
+from diff_util import PromptUserToAcceptDiff
 
 HISTOGRAMS_PATH = 'histograms.xml'
-POLICY_TEMPLATES_PATH =
-    '../../../components/policy/resources/policy_templates.json'
+POLICY_TEMPLATES_PATH = (
+    '../../../components/policy/resources/policy_templates.json')
 ENUM_NAME = 'EnterprisePolicies'
 
 class UserError(Exception):
@@ -118,8 +122,7 @@ def main():
     xml = f.read()
 
   UpdateHistogramDefinitions(policy_templates, histograms_doc)
-
-  new_xml = PrettyPrintNode(histograms_doc)
+  new_xml = print_style.GetPrintStyle().PrettyPrintNode(histograms_doc)
   if PromptUserToAcceptDiff(xml, new_xml, 'Is the updated version acceptable?'):
     with open(HISTOGRAMS_PATH, 'wb') as f:
       f.write(new_xml)