[Geoserver-devel] Jalopy formatting

GeoServer GeoServer Developer
1

Hey all,

If you're on linux, using a UTF-8 based locale, you'll wind up with strange 'mis-formatted' foreign characters. Gabriel and Martin, you guys sometimes get 'weirded' names when jalopy runs, and this causes many many spurrious changes to the source, which in turn causes 'svn status' to be completely useless.

The fix for this (for those few folks who it affects, mostly just Justin and I) is to set your locale to "en_US.ISO8859-1". Usually this just means doing "export LANG=en_US.iso8859-1", but perhaps you might need to do something further.

Now that I've done this, I'm getting lots of OTHER spurrious diffs...but these ones are a little more suspect.

For example, here's a snippit of svn diff on KMLWriter.java after a FRESH checkout and then a build (which jalopies the code):

@@ -1049,7 +1049,7 @@
                 if (symbolizers[m] instanceof TextSymbolizer) {
                     TextSymbolizer ts = (TextSymbolizer) symbolizers[m];
                     Expression ex = ts.getLabel();
- String value = (String) ex.evaluate(feature,String.class);
+ String value = (String) ex.evaluate(feature, String.class);
                     title.append(value);

svn blame says that jgarnett is responsible for that one, and I know he's on windows...is it possible that jalopy produces different output (space between method call arguments?) on different platforms? Are some folks not running jalopy before commits?

It's maybe a 7 out of 10 on the "anoying scale" to me to have some folks running jalopy and some folks not (solely because it makes svn status totally useless, and because if someone else isn't using jalopy, why should my builds have to take an extra minute?) but if I'm the only one bugged by that particular issue, then that's cool and I'll just learn to deal.

Thoughts? Anyone KNOW that they're not using jalopy and have a particular reason why not?

Thanks!
--saul

2

I'm pretty positive that Jalopy is no different on different platforms for whitespace - the whole point of jalopy is for agreed upon whitespace.

I think what Andrea does is if he gets any differences like that he'll just commit right then. See for example: http://fisheye.codehaus.org/changelog/~author=aaime/geoserver/?cs=6666

Personally I don't find it all that annoying if individual people forget to reformat every once in awhile, since someone else will soon come along and do the cleanup automatically. But I'm also not generally the one doing that, so maybe it is annoying (though I actually used to run Jalopy by hand and then commit just those, to keep things clean).

But I'd be fine if we try to get more vigilant about people being sure to run Jalopy before committing. As it's really not that hard...

Chris

Farber, Saul (ENV) wrote:

Hey all,
If you're on linux, using a UTF-8 based locale, you'll wind up with strange 'mis-formatted' foreign characters. Gabriel and Martin, you guys sometimes get 'weirded' names when jalopy runs, and this causes many many spurrious changes to the source, which in turn causes 'svn status' to be completely useless.
The fix for this (for those few folks who it affects, mostly just Justin and I) is to set your locale to "en_US.ISO8859-1". Usually this just means doing "export LANG=en_US.iso8859-1", but perhaps you might need to do something further.
Now that I've done this, I'm getting lots of OTHER spurrious diffs...but these ones are a little more suspect.
For example, here's a snippit of svn diff on KMLWriter.java after a FRESH checkout and then a build (which jalopies the code):
@@ -1049,7 +1049,7 @@
                 if (symbolizers[m] instanceof TextSymbolizer) {
                     TextSymbolizer ts = (TextSymbolizer) symbolizers[m];
                     Expression ex = ts.getLabel();
- String value = (String) ex.evaluate(feature,String.class);
+ String value = (String) ex.evaluate(feature, String.class);
                     title.append(value);

It's maybe a 7 out of 10 on the "anoying scale" to me to have some folks running jalopy and some folks not (solely because it makes svn status totally useless, and because if someone else isn't using jalopy, why should my builds have to take an extra minute?) but if I'm the only one bugged by that particular issue, then that's cool and I'll just learn to deal.
Thoughts? Anyone KNOW that they're not using jalopy and have a particular reason why not?
Thanks!
--saul

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

--
Chris Holmes
The Open Planning Project
http://topp.openplans.org

3

Aha! Good point.

I'll just start commiting those kinds of things right away...am I polluting the svn commit revision number space or anything?

Didn't mean to be a curmudgeon, just didn't think of commiting those fixes.

--saul

________________________________

From: Chris Holmes [mailto:cholmes@anonymised.com]
Sent: Tue 5/8/2007 6:11 PM
To: Farber, Saul (ENV)
Cc: geoserver-devel@lists.sourceforge.net
Subject: Re: [Geoserver-devel] Jalopy formatting

I'm pretty positive that Jalopy is no different on different platforms
for whitespace - the whole point of jalopy is for agreed upon whitespace.

I think what Andrea does is if he gets any differences like that he'll
just commit right then. See for example:
http://fisheye.codehaus.org/changelog/~author=aaime/geoserver/?cs=6666

Personally I don't find it all that annoying if individual people forget
to reformat every once in awhile, since someone else will soon come
along and do the cleanup automatically. But I'm also not generally the
one doing that, so maybe it is annoying (though I actually used to run
Jalopy by hand and then commit just those, to keep things clean).

But I'd be fine if we try to get more vigilant about people being sure
to run Jalopy before committing. As it's really not that hard...

Chris

Farber, Saul (ENV) wrote:

Hey all,

If you're on linux, using a UTF-8 based locale, you'll wind up with strange 'mis-formatted' foreign characters. Gabriel and Martin, you guys sometimes get 'weirded' names when jalopy runs, and this causes many many spurrious changes to the source, which in turn causes 'svn status' to be completely useless.

The fix for this (for those few folks who it affects, mostly just Justin and I) is to set your locale to "en_US.ISO8859-1". Usually this just means doing "export LANG=en_US.iso8859-1", but perhaps you might need to do something further.

Now that I've done this, I'm getting lots of OTHER spurrious diffs...but these ones are a little more suspect.

For example, here's a snippit of svn diff on KMLWriter.java after a FRESH checkout and then a build (which jalopies the code):

@@ -1049,7 +1049,7 @@
                 if (symbolizers[m] instanceof TextSymbolizer) {
                     TextSymbolizer ts = (TextSymbolizer) symbolizers[m];
                     Expression ex = ts.getLabel();
- String value = (String) ex.evaluate(feature,String.class);
+ String value = (String) ex.evaluate(feature, String.class);
                     title.append(value);

svn blame says that jgarnett is responsible for that one, and I know he's on windows...is it possible that jalopy produces different output (space between method call arguments?) on different platforms? Are some folks not running jalopy before commits?

It's maybe a 7 out of 10 on the "anoying scale" to me to have some folks running jalopy and some folks not (solely because it makes svn status totally useless, and because if someone else isn't using jalopy, why should my builds have to take an extra minute?) but if I'm the only one bugged by that particular issue, then that's cool and I'll just learn to deal.

Thoughts? Anyone KNOW that they're not using jalopy and have a particular reason why not?

Thanks!
--saul

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

--
Chris Holmes
The Open Planning Project
http://topp.openplans.org/>

4

Farber, Saul (ENV) wrote:

Aha! Good point.
I'll just start commiting those kinds of things right away...am I polluting the svn commit revision number space or anything?

I see no problem with it, I'd rather have a high revision number than messy code.

Didn't mean to be a curmudgeon, just didn't think of commiting those fixes.

No worries at all.

Chris

--saul

________________________________

From: Chris Holmes [mailto:cholmes@anonymised.com]
Sent: Tue 5/8/2007 6:11 PM
To: Farber, Saul (ENV)
Cc: geoserver-devel@lists.sourceforge.net
Subject: Re: [Geoserver-devel] Jalopy formatting

I'm pretty positive that Jalopy is no different on different platforms
for whitespace - the whole point of jalopy is for agreed upon whitespace.

I think what Andrea does is if he gets any differences like that he'll
just commit right then. See for example:
http://fisheye.codehaus.org/changelog/~author=aaime/geoserver/?cs=6666

Personally I don't find it all that annoying if individual people forget
to reformat every once in awhile, since someone else will soon come
along and do the cleanup automatically. But I'm also not generally the
one doing that, so maybe it is annoying (though I actually used to run
Jalopy by hand and then commit just those, to keep things clean).

But I'd be fine if we try to get more vigilant about people being sure
to run Jalopy before committing. As it's really not that hard...

Chris

Farber, Saul (ENV) wrote:

Hey all,

If you're on linux, using a UTF-8 based locale, you'll wind up with strange 'mis-formatted' foreign characters. Gabriel and Martin, you guys sometimes get 'weirded' names when jalopy runs, and this causes many many spurrious changes to the source, which in turn causes 'svn status' to be completely useless.

The fix for this (for those few folks who it affects, mostly just Justin and I) is to set your locale to "en_US.ISO8859-1". Usually this just means doing "export LANG=en_US.iso8859-1", but perhaps you might need to do something further.

Now that I've done this, I'm getting lots of OTHER spurrious diffs...but these ones are a little more suspect.

For example, here's a snippit of svn diff on KMLWriter.java after a FRESH checkout and then a build (which jalopies the code):

@@ -1049,7 +1049,7 @@
                 if (symbolizers[m] instanceof TextSymbolizer) {
                     TextSymbolizer ts = (TextSymbolizer) symbolizers[m];
                     Expression ex = ts.getLabel();
- String value = (String) ex.evaluate(feature,String.class);
+ String value = (String) ex.evaluate(feature, String.class);
                     title.append(value);

svn blame says that jgarnett is responsible for that one, and I know he's on windows...is it possible that jalopy produces different output (space between method call arguments?) on different platforms? Are some folks not running jalopy before commits?

It's maybe a 7 out of 10 on the "anoying scale" to me to have some folks running jalopy and some folks not (solely because it makes svn status totally useless, and because if someone else isn't using jalopy, why should my builds have to take an extra minute?) but if I'm the only one bugged by that particular issue, then that's cool and I'll just learn to deal.

Thoughts? Anyone KNOW that they're not using jalopy and have a particular reason why not?

Thanks!
--saul

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Geoserver-devel mailing list
Geoserver-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

--
Chris Holmes
The Open Planning Project
http://topp.openplans.org/>

--
Chris Holmes
The Open Planning Project
http://topp.openplans.org

5

Farber, Saul (ENV) ha scritto:

Hey all,
If you're on linux, using a UTF-8 based locale, you'll wind up with strange 'mis-formatted' foreign characters. Gabriel and Martin, you guys sometimes get 'weirded' names when jalopy runs, and this causes many many spurrious changes to the source, which in turn causes 'svn status' to be completely useless.
The fix for this (for those few folks who it affects, mostly just Justin and I) is to set your locale to "en_US.ISO8859-1". Usually this just means doing "export LANG=en_US.iso8859-1", but perhaps you might need to do something further.
Now that I've done this, I'm getting lots of OTHER spurrious diffs...but these ones are a little more suspect.
For example, here's a snippit of svn diff on KMLWriter.java after a FRESH checkout and then a build (which jalopies the code):
@@ -1049,7 +1049,7 @@
                 if (symbolizers[m] instanceof TextSymbolizer) {
                     TextSymbolizer ts = (TextSymbolizer) symbolizers[m];
                     Expression ex = ts.getLabel();
- String value = (String) ex.evaluate(feature,String.class);
+ String value = (String) ex.evaluate(feature, String.class);
                     title.append(value);

It just means he did not run a full build before committing.
That happens sometimes :slight_smile:
Cheers
Andrea