8299677: Formatter.format might take a long time to format an integer or floating-point #11939

rgiulietti · 2023-01-11T10:47:03Z

This change transforms a O(n^2) path to O(n) when prepending zero padding to decimal outputs, where n is the length of the padding.

Progress

Change must be properly reviewed (1 review required, with at least 1 Reviewer)
Change must not contain extraneous whitespace
Commit message must refer to an issue

Issue

JDK-8299677: Formatter.format might take a long time to format an integer or floating-point

Reviewers

Alan Bateman (@AlanBateman - Reviewer) ⚠️ Review applies to 275582a9
Sergey Tsypanov (@stsypanov - Author)
Joe Darcy (@jddarcy - Reviewer)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk pull/11939/head:pull/11939
$ git checkout pull/11939

Update a local copy of the PR:
$ git checkout pull/11939
$ git pull https://git.openjdk.org/jdk pull/11939/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 11939

View PR using the GUI difftool:
$ git pr show -t 11939

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk/pull/11939.diff

… or floating-point

bridgekeeper · 2023-01-11T10:48:27Z

👋 Welcome back rgiulietti! A progress list of the required criteria for merging this PR into master will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

openjdk · 2023-01-11T10:51:00Z

@rgiulietti The following labels will be automatically applied to this pull request:

core-libs
i18n

When this pull request is ready to be reviewed, an "RFR" email will be sent to the corresponding mailing lists. If you would like to change these labels, use the /label pull request command.

mlbridge · 2023-01-11T10:54:59Z

Webrevs

AlanBateman · 2023-01-11T11:53:58Z

I assume this issue has been there for a long time, just not noticed because it would be unusual to specify a width of this magnitude. The switch to using repeat + one insert looks good. Are we okay for tests in this area?

rgiulietti · 2023-01-11T12:45:08Z

@AlanBateman This is not a functional bug, only an enhancement to overcome the long times for large, unrealistic widths.
Consequently, I have no specific tests that measure the timing against some hard-coded limits, if I understand you correctly.

JimLaskey · 2023-01-11T12:51:26Z

Just as a note, I've been thinking about adding a repeat(String src, int count) method to AbstractStringBuffer for quite some time. I've held off because I would add a fillWith(T[] dst, T[] src, int start, int end) methods to java.util.Arrays first. I think both sets of methods would be generally useful based on use patterns in the JDK but for time and motivation.

AlanBateman · 2023-01-11T13:43:34Z

@AlanBateman This is not a functional bug, only an enhancement to overcome the long times for large, unrealistic widths.
Consequently, I have no specific tests that measure the timing against some hard-coded limits, if I understand you correctly.

My question wasn't phrased very well, I was really just asking if we have good test coverage for width in the format specifier so that it would catch any regressions from changes to this code.

rgiulietti · 2023-01-11T15:15:51Z

Since specific tests for padding seem lacking, I'll add some as part of this PR.
Please expect another commit before marking as reviewed.

… or floating-point

rgiulietti · 2023-01-11T19:06:53Z

Added straightforward tests for padding.

… or floating-point

stsypanov · 2023-01-12T11:40:32Z

src/java.base/share/classes/java/util/Formatter.java

-                    sb.insert(begin, zero);
-                }
+            if (width > sb.length() && Flags.contains(f, Flags.ZERO_PAD)) {
+                String zeros = Character.toString(zero).repeat(width - sb.length());


Character.toString(zero) could be replaced directly with String.valueOf(zero) saving one more method call

is the bug going to be backported to JDK8? Due to the use of 'repeat' this seems to not be possible.

AlanBateman

Thanks for expanding the tests for this.

AlanBateman · 2023-01-12T11:58:57Z

test/jdk/java/util/Formatter/Padding.java

+
+/*
+ * @test
+ * @run junit Padding


Probably should add a @summary tag to this, maybe a @bug too (I think JDK-4906370 added Formatter in Java 5).

openjdk · 2023-01-12T12:01:33Z

@rgiulietti This change now passes all automated pre-integration checks.

ℹ️ This project also has non-automated pre-integration requirements. Please see the file CONTRIBUTING.md for details.

After integration, the commit message for the final commit will be:

8299677: Formatter.format might take a long time to format an integer or floating-point

Reviewed-by: alanb, stsypanov, darcy

You can use pull request commands such as /summary, /contributor and /issue to adjust it as needed.

At the time when this comment was updated there had been 42 new commits pushed to the master branch:

7a85d95: 8298448: UndefinedBehaviorSanitizer
26890d1: 8296478: Rework 8282948 and 8282700 to use the new autoconf UTIL_ARG_WITH
cf00d09: 8299330: Minor improvements in MSYS2 Workflow handling
48c8fb3: 8299978: Remove MethodHandleNatives.getMembers
89d3f3d: 8299125: UnifiedOopRef in JFR leakprofiler should treat thread local oops correctly
4b57334: 8278326: Socket close is not thread safe and other cleanup
036c808: 8298482: Implement ParallelGC NUMAStats for Linux
0ee8cac: 8299672: Enhance HeapDump JFR event
d716ec5: 8299179: ArrayFill with store on backedge needs to reduce length by 1
af8d3fb: 8264684: os::get_summary_cpu_info padded with spaces
... and 32 more: https://git.openjdk.org/jdk/compare/8b0133f2760f67cd968528c041a600408cc26978...master

As there are no conflicts, your changes will automatically be rebased on top of these commits when integrating. If you prefer to avoid this automatic rebasing, please check the documentation for the /integrate command for further details.

➡️ To integrate this PR with the above commit message to the master branch, type /integrate in a new comment.

LanceAndersen

Overall this looks OK.

Please add either a general comment describing what we are trying to test or a comment for each test method. As we add new tests, adding comments for additional clarity will benefit future maintainers

LanceAndersen · 2023-01-12T12:34:41Z

test/jdk/java/util/Formatter/Padding.java

+    @Nested
+    class BlankPaddingInt {
+
+        @Test


Any thought given to using @MethodSource: in order to specify a DataProvider? The benefit would be you could catch each failure vs having to address or comment out the assertEquals that is failing

… or floating-point

rgiulietti · 2023-01-12T15:29:52Z

All concerns were addressed, I hope.

rgiulietti · 2023-01-12T19:18:23Z

/integrate

openjdk · 2023-01-12T19:21:24Z

Going to push as commit 33412c1.
Since your change was applied there have been 42 commits pushed to the master branch:

7a85d95: 8298448: UndefinedBehaviorSanitizer
26890d1: 8296478: Rework 8282948 and 8282700 to use the new autoconf UTIL_ARG_WITH
cf00d09: 8299330: Minor improvements in MSYS2 Workflow handling
48c8fb3: 8299978: Remove MethodHandleNatives.getMembers
89d3f3d: 8299125: UnifiedOopRef in JFR leakprofiler should treat thread local oops correctly
4b57334: 8278326: Socket close is not thread safe and other cleanup
036c808: 8298482: Implement ParallelGC NUMAStats for Linux
0ee8cac: 8299672: Enhance HeapDump JFR event
d716ec5: 8299179: ArrayFill with store on backedge needs to reduce length by 1
af8d3fb: 8264684: os::get_summary_cpu_info padded with spaces
... and 32 more: https://git.openjdk.org/jdk/compare/8b0133f2760f67cd968528c041a600408cc26978...master

Your commit was automatically rebased without conflicts.

openjdk · 2023-01-12T19:21:46Z

@rgiulietti Pushed as commit 33412c1.

💡 You may see a message that your pull request was closed with unmerged commits. This can be safely ignored.

balcanuc

in which JDK versions will the bug-fix back ported? As it is, it is impossible to be back ported to JDK8

balcanuc · 2023-01-15T17:58:34Z

src/java.base/share/classes/java/util/Formatter.java

-                    sb.insert(begin, zero);
-                }
+            if (width > sb.length() && Flags.contains(f, Flags.ZERO_PAD)) {
+                String zeros = Character.toString(zero).repeat(width - sb.length());


is the bug going to be backported to JDK8? Due to the use of 'repeat' this seems to not be possible.

rgiulietti · 2023-01-16T17:50:35Z

@balcanuc Backporting is done in other, separate projects. Whether changes in mainline are backported, and how, is publicly discussed in the respective mailing lists.

In particular, for JDK 8 Updates you might want to have a look at
this mailing list and the this git repo.

If you feel this change is relevant to JDK 8, you are welcome to discuss a solution that does not rely on repeat() on that mailing list, and then contribute your code to the JDK 8 Update project.

8299677: Formatter.format might take a long time to format an integer…

26e05ca

… or floating-point

openjdk bot added the rfr Pull request is ready for review label Jan 11, 2023

openjdk bot added core-libs core-libs-dev@openjdk.org i18n i18n-dev@openjdk.org labels Jan 11, 2023

8299677: Formatter.format might take a long time to format an integer…

e541882

… or floating-point

rgiulietti added 2 commits January 11, 2023 20:24

8299677: Formatter.format might take a long time to format an integer…

8c3e385

… or floating-point

8299677: Formatter.format might take a long time to format an integer…

275582a

… or floating-point

stsypanov reviewed Jan 12, 2023

View reviewed changes

AlanBateman approved these changes Jan 12, 2023

View reviewed changes

openjdk bot added the ready Pull request is ready to be integrated label Jan 12, 2023

LanceAndersen reviewed Jan 12, 2023

View reviewed changes

8299677: Formatter.format might take a long time to format an integer…

56c8f55

… or floating-point

stsypanov approved these changes Jan 12, 2023

View reviewed changes

jddarcy approved these changes Jan 12, 2023

View reviewed changes

openjdk bot added the integrated Pull request has been integrated label Jan 12, 2023

openjdk bot closed this Jan 12, 2023

openjdk bot removed ready Pull request is ready to be integrated rfr Pull request is ready for review labels Jan 12, 2023

rgiulietti deleted the JDK-8299677 branch January 12, 2023 20:38

balcanuc reviewed Jan 16, 2023

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

8299677: Formatter.format might take a long time to format an integer or floating-point #11939

8299677: Formatter.format might take a long time to format an integer or floating-point #11939

rgiulietti commented Jan 11, 2023 •

edited by openjdk bot

Loading

bridgekeeper bot commented Jan 11, 2023

openjdk bot commented Jan 11, 2023

mlbridge bot commented Jan 11, 2023 •

edited

Loading

AlanBateman commented Jan 11, 2023 •

edited

Loading

rgiulietti commented Jan 11, 2023

JimLaskey commented Jan 11, 2023 •

edited

Loading

AlanBateman commented Jan 11, 2023

rgiulietti commented Jan 11, 2023

rgiulietti commented Jan 11, 2023

stsypanov Jan 12, 2023 •

edited

Loading

balcanuc Jan 15, 2023 •

edited by bridgekeeper bot

Loading

AlanBateman left a comment

AlanBateman Jan 12, 2023

openjdk bot commented Jan 12, 2023 •

edited

Loading

LanceAndersen left a comment

LanceAndersen Jan 12, 2023

rgiulietti commented Jan 12, 2023

rgiulietti commented Jan 12, 2023

openjdk bot commented Jan 12, 2023

openjdk bot commented Jan 12, 2023

balcanuc left a comment •

edited by bridgekeeper bot

Loading

balcanuc Jan 15, 2023 •

edited by bridgekeeper bot

Loading

rgiulietti commented Jan 16, 2023

8299677: Formatter.format might take a long time to format an integer or floating-point #11939

8299677: Formatter.format might take a long time to format an integer or floating-point #11939

Conversation

rgiulietti commented Jan 11, 2023 • edited by openjdk bot Loading

Progress

Issue

Reviewers

Reviewing

bridgekeeper bot commented Jan 11, 2023

openjdk bot commented Jan 11, 2023

mlbridge bot commented Jan 11, 2023 • edited Loading

Webrevs

AlanBateman commented Jan 11, 2023 • edited Loading

rgiulietti commented Jan 11, 2023

JimLaskey commented Jan 11, 2023 • edited Loading

AlanBateman commented Jan 11, 2023

rgiulietti commented Jan 11, 2023

rgiulietti commented Jan 11, 2023

stsypanov Jan 12, 2023 • edited Loading

Choose a reason for hiding this comment

balcanuc Jan 15, 2023 • edited by bridgekeeper bot Loading

Choose a reason for hiding this comment

AlanBateman left a comment

Choose a reason for hiding this comment

AlanBateman Jan 12, 2023

Choose a reason for hiding this comment

openjdk bot commented Jan 12, 2023 • edited Loading

LanceAndersen left a comment

Choose a reason for hiding this comment

LanceAndersen Jan 12, 2023

Choose a reason for hiding this comment

rgiulietti commented Jan 12, 2023

rgiulietti commented Jan 12, 2023

openjdk bot commented Jan 12, 2023

openjdk bot commented Jan 12, 2023

balcanuc left a comment • edited by bridgekeeper bot Loading

Choose a reason for hiding this comment

balcanuc Jan 15, 2023 • edited by bridgekeeper bot Loading

Choose a reason for hiding this comment

rgiulietti commented Jan 16, 2023

rgiulietti commented Jan 11, 2023 •

edited by openjdk bot

Loading

mlbridge bot commented Jan 11, 2023 •

edited

Loading

AlanBateman commented Jan 11, 2023 •

edited

Loading

JimLaskey commented Jan 11, 2023 •

edited

Loading

stsypanov Jan 12, 2023 •

edited

Loading

balcanuc Jan 15, 2023 •

edited by bridgekeeper bot

Loading

openjdk bot commented Jan 12, 2023 •

edited

Loading

balcanuc left a comment •

edited by bridgekeeper bot

Loading

balcanuc Jan 15, 2023 •

edited by bridgekeeper bot

Loading