Make git-freeze bail out if the user has too much untracked data.

git_common.py

Index: git_common.py
diff --git a/git_common.py b/git_common.py
index 6abb5bad0b7f20a50baa4f84012155b965cc3b70..d9ad7ce8608e6e19f93b28e66095169d973334c0 100644
--- a/git_common.py
+++ b/git_common.py
@@ -32,8 +32,10 @@ import threading
 
 import subprocess2
 
-ROOT = os.path.abspath(os.path.dirname(__file__))
+from cStringIO import StringIO
+
 
+ROOT = os.path.abspath(os.path.dirname(__file__))
 IS_WIN = sys.platform == 'win32'
 GIT_EXE = ROOT+'\\git.bat' if IS_WIN else 'git'
 TEST_MODE = False
@@ -389,6 +391,36 @@ def diff(oldrev, newrev, *args):
 
 def freeze():
   took_action = False
+  key = 'depot-tools.freeze-size-limit'
+  MB = 2**20
+  limit_mb = get_config_int(key, 100)
+  untracked_size = 0
+
+  for f, s in status():
+    if is_unmerged(s):
+      die("Cannot freeze unmerged changes!")
+    if limit_mb > 0:
+      if s.lstat == '?':
+        untracked_size += os.stat(f).st_size / MB

[tandrii(chromium), 2016/06/21 14:43:37]

don't divide here, otherwise 3 files 400KB each will amount to 0MB.

[agable, 2016/06/22 11:16:51]

Done

+      if untracked_size > limit_mb:

[tandrii(chromium), 2016/06/21 14:43:37]

divide by MB here, and better yet multiply by limit_mb by MB.

[agable, 2016/06/22 11:16:51]

Done

+        die("""\
+          You appear to have too much untracked+unignored data in your git
+          checkout: %d/%dMB.

[tandrii(chromium), 2016/06/21 14:43:37]

personal preference: s/%d/%.1f and make MB a float and then untracked_size/MB.

[agable, 2016/06/22 11:16:51]

Gonna leave limit in MB, since letting it be a decimal is vaguely silly. But did
change the format to print a float for the amount of untracked data.

+
+          Run `git status` to see what it is.
+
+          In addition to making many git commands slower, this will prevent
+          depot_tools from freezing your in-progress changes.
+
+          You should add untracked data that you want to ignore to your repo's
+            .git/info/excludes
+          file. See `git help ignore` for the format of this file.
+
+          If this data is indended as part of your commit, you may adjust the
+          freeze limit by running:
+            git config %s <new_limit>
+          Where <new_limit> is an integer threshold in megabytes.""",
+          untracked_size, limit_mb, key)
 
   try:
     run('commit', '--no-verify', '-m', FREEZE + '.indexed')
@@ -499,6 +531,13 @@ def is_dormant(branch):
   return branch_config(branch, 'dormant', 'false') != 'false'
 
 
+def is_unmerged(stat_value):
+  return (
+      'U' in (stat_value.lstat, stat_value.rstat) or
+      ((stat_value.lstat == stat_value.rstat) and stat_value.lstat in 'AD')
+  )
+
+
 def manual_merge_base(branch, base, parent):
   set_branch_config(branch, 'base', base)
   set_branch_config(branch, 'base-upstream', parent)
@@ -708,6 +747,50 @@ def is_dirty_git_tree(cmd):
   return False
 
 
+def status():
+  """Returns a parsed version of git-status.
+
+  Returns a generator of (current_name, (lstat, rstat, src)) pairs where:
+    * current_name is the name of the file
+    * lstat is the left status code letter from git-status
+    * rstat is the left status code letter from git-status
+    * src is the current name of the file, or the original name of the file
+      if lstat == 'R'
+  """
+  stat_entry = collections.namedtuple('stat_entry', 'lstat rstat src')
+
+  def tokenizer(stream):
+    acc = StringIO()
+    c = None
+    while c != '':
+      c = stream.read(1)
+      if c in (None, '', '\0'):
+        s = acc.getvalue()

[tandrii(chromium), 2016/06/21 14:43:37]

nit: if you'd had used plain StringIO, which for this use case seems totally
fine, you could then rewrite this block as:

if acc.len:
  yield acc.getvalue()
  acc = StringIO()

[agable, 2016/06/22 11:16:51]

Good point, done.

+        acc = StringIO()
+        if s:
+          yield s
+      else:
+        acc.write(c)
+
+  def parser(tokens):
+    END = object()
+    tok = lambda: next(tokens, END)
+    while True:
+      status_dest = tok()
+      if status_dest is END:
+        return
+      stat, dest = status_dest[:2], status_dest[3:]

[tandrii(chromium), 2016/06/21 14:43:37]

suggestion: why not plain old except to detect END?

while True:
  try:
    status_dest = next(tokens)
  except StopIteration:
    return

and below
  src = next(tokens)
  no need for assert, as above line will raise if not end.

[agable, 2016/06/22 11:16:51]

Nice, done.

+      lstat, rstat = stat
+      if lstat == 'R':
+        src = tok()
+        assert src is not END
+      else:
+        src = dest
+      yield (dest, stat_entry(lstat, rstat, src))
+
+  return parser(tokenizer(run_stream('status', '-z', bufsize=-1)))
+
+
 def squash_current_branch(header=None, merge_base=None):
   header = header or 'git squash commit.'
   merge_base = merge_base or get_or_create_merge_base(current_branch())