This is an archive of the discontinued Mercurial Phabricator instance.

Differential D12089

encoding: fix trim() to be O(n) instead of O(n^2)
ClosedPublic

Authored by martinvonz on Jan 26 2022, 1:29 PM.

Details

Reviewers
None
Group Reviewers
hg-reviewers
Commits
rHGf1ed5c304f45: encoding: fix trim() to be O(n) instead of O(n^2)
Summary

encoding.trim() iterated over the possible lengths smaller than the
input and created a slice for each. It then calculated the column
width of the result, which is of course O(n), so the overall algorithm
was O(n). This patch rewrites it to iterate over the unicode
characters, keeping track of the length so far. Also, the old
algorithm started from the end of the string, which made it much worse
when the input is large and the limit is small (such as the typical 72
we pass to it).

You can time it by running something like this:

time python3 -c 'from mercurial.utils import stringutil; print(stringutil.ellipsis(b"0123456789" * 1000, 5))'

That drops from 4.05 s to 83 ms with this patch (and most of that is
of course startup time).

Diff Detail

Repository
rHG Mercurial
Lint
Automatic diff as part of commit; lint not applicable.
Unit
Automatic diff as part of commit; unit tests not applicable.

Event Timeline

martinvonz created this revision.Jan 26 2022, 1:29 PM
martinvonz added a commit: rHGf1ed5c304f45: encoding: fix trim() to be O(n) instead of O(n^2).Jan 26 2022, 2:14 PM
This revision was not accepted when it landed; it landed in state Needs Review.
Closed by commit rHGf1ed5c304f45: encoding: fix trim() to be O(n) instead of O(n^2) (authored by martinvonz). · Explain Why
This revision was automatically updated to reflect the committed changes.

Revision Contents
Changeset List

PathPackages
Mmercurial/encoding.py (24 lines)

Diff 31982

mercurial/encoding.py

if ucolwidth(u) <= width: # trimming is not needed if ucolwidth(u) <= width: # trimming is not needed
return s return s
width -= len(ellipsis) width -= len(ellipsis)
if width <= 0: # no enough room even for ellipsis if width <= 0: # no enough room even for ellipsis
return ellipsis[: width + len(ellipsis)] return ellipsis[: width + len(ellipsis)]
chars = list(u)
if leftside: if leftside:
uslice = lambda i: u[i:] chars.reverse()
concat = lambda s: ellipsis + s width_so_far = 0
else: for i, c in enumerate(chars):
uslice = lambda i: u[:-i] width_so_far += ucolwidth(c)
concat = lambda s: s + ellipsis if width_so_far > width:
for i in pycompat.xrange(1, len(u)): break
usub = uslice(i) chars = chars[:i]
if ucolwidth(usub) <= width: if leftside:
return concat(usub.encode(_sysstr(encoding))) chars.reverse()
return ellipsis # no enough room for multi-column characters u = u''.join(chars).encode(_sysstr(encoding))
if leftside:
return ellipsis + u
return u + ellipsis
class normcasespecs(object): class normcasespecs(object):
"""what a platform's normcase does to ASCII strings """what a platform's normcase does to ASCII strings
This is specified per platform, and should be consistent with what normcase This is specified per platform, and should be consistent with what normcase
on that platform actually does. on that platform actually does.