Create a new class, attempt to edit the class comment ==> Error: RemoteString past end of file
Surely I cannot be the first person to have edited a class comment in the last month or so ... I hope.
Dave
Whenever I type an open-paren $( in a Squeak 4.5 workspace, some mysterious agent of code format rectitude steps in and supplies the matching close-paren. This is hugely annoying and I can't seem to find any preferences or workspace menu settings to make it stop.
Does anyone know how to make this feature go away?
Dave
On 09-04-2014, at 5:11 PM, David T. Lewis lewis@mail.msen.com wrote:
Whenever I type an open-paren $( in a Squeak 4.5 workspace, some mysterious agent of code format rectitude steps in and supplies the matching close-paren. This is hugely annoying and I can't seem to find any preferences or workspace menu settings to make it stop.
Does anyone know how to make this feature go away?
Helpful code can be so unhelpful, right? See if by any chance the syntaxHighlightinasyoutype preference is on - in the ‘browsing’ section of the prefs browser. There’s also shoutstylinginworkspace. Neither of them was on in my 4.5 image but y’never know.
tim -- tim Rowledge; tim@rowledge.org; http://www.rowledge.org/tim Strange OpCodes: XYZZY: Branch and Play Adventure
On Wed, Apr 09, 2014 at 05:18:27PM -0700, tim Rowledge wrote:
On 09-04-2014, at 5:11 PM, David T. Lewis lewis@mail.msen.com wrote:
Whenever I type an open-paren $( in a Squeak 4.5 workspace, some mysterious agent of code format rectitude steps in and supplies the matching close-paren. This is hugely annoying and I can't seem to find any preferences or workspace menu settings to make it stop.
Does anyone know how to make this feature go away?
Helpful code can be so unhelpful, right? See if by any chance the syntaxHighlightinasyoutype preference is on - in the ?browsing? section of the prefs browser. There?s also shoutstylinginworkspace. Neither of them was on in my 4.5 image but y?never know.
I tried both of those, no joy. Actually, I looked through every preference in the Preference Browser, and all the settings in the workspace menu. I must be missing something, but I can't make this feature go away.
Dave
On Wed, Apr 9, 2014 at 7:37 PM, David T. Lewis lewis@mail.msen.com wrote:
I tried both of those, no joy. Actually, I looked through every preference in the Preference Browser, and all the settings in the workspace menu. I must be missing something, but I can't make this feature go away.
You want the "Auto Enclose" preference in category "Morphic".
I can't stand it either. Surprisingly little of my coding consists of typing code out linearly.
On 09-04-2014, at 5:54 PM, Colin Putney colin@wiresong.com wrote:
On Wed, Apr 9, 2014 at 7:37 PM, David T. Lewis lewis@mail.msen.com wrote:
I tried both of those, no joy. Actually, I looked through every preference in the Preference Browser, and all the settings in the workspace menu. I must be missing something, but I can't make this feature go away.
You want the "Auto Enclose" preference in category "Morphic".
I can't stand it either. Surprisingly little of my coding consists of typing code out linearly.
Likewise. I really, really, hate this supposedly ‘helpful’ kind of thing.
tim -- tim Rowledge; tim@rowledge.org; http://www.rowledge.org/tim A bug in the code is worth two in the documentation.
Bonk! Of course, I _love_ Auto Enclose! :-)
Because I'm not "typing a stream of code", I'm "composing nested expressions".
Now, when it included the less-than sign, I resented it, because it assuemed I was typing HTML. But for '"([ and {, they always go in pairs. The machine keeps my nestings correct and I can simply right-arrow to the next outer-level and press Command+Space to expose and "work on" that expression. Love it.
Of the curiously strongly negative reactions, though, I was able to glean just one tidbit of feedback (Colins).
On Wed, Apr 9, 2014 at 7:57 PM, tim Rowledge tim@rowledge.org wrote:
On 09-04-2014, at 5:54 PM, Colin Putney colin@wiresong.com wrote:
On Wed, Apr 9, 2014 at 7:37 PM, David T. Lewis lewis@mail.msen.com wrote:
I tried both of those, no joy. Actually, I looked through every preference in the Preference Browser, and all the settings in the workspace menu. I must be missing something, but I can't make this feature go away.
You want the "Auto Enclose" preference in category "Morphic".
I can't stand it either. Surprisingly little of my coding consists of typing code out linearly.
Likewise. I really, really, hate this supposedly ‘helpful’ kind of thing.
tim
tim Rowledge; tim@rowledge.org; http://www.rowledge.org/tim A bug in the code is worth two in the documentation.
On Wed, Apr 09, 2014 at 08:19:37PM -0500, Chris Muller wrote:
Bonk! Of course, I _love_ Auto Enclose! :-)
Because I'm not "typing a stream of code", I'm "composing nested expressions".
Ha! It makes sense to me now that you explain it that way. I guess that's why it's a preference :-)
For myself, much of what I do in workspaces has little to do with writing code, which is why it comes across to me as an annoyance.
Dave
Now, when it included the less-than sign, I resented it, because it assuemed I was typing HTML. But for '"([ and {, they always go in pairs. The machine keeps my nestings correct and I can simply right-arrow to the next outer-level and press Command+Space to expose and "work on" that expression. Love it.
Of the curiously strongly negative reactions, though, I was able to glean just one tidbit of feedback (Colins).
On Wed, Apr 9, 2014 at 7:57 PM, tim Rowledge tim@rowledge.org wrote:
On 09-04-2014, at 5:54 PM, Colin Putney colin@wiresong.com wrote:
On Wed, Apr 9, 2014 at 7:37 PM, David T. Lewis lewis@mail.msen.com wrote:
I tried both of those, no joy. Actually, I looked through every preference in the Preference Browser, and all the settings in the workspace menu. I must be missing something, but I can't make this feature go away.
You want the "Auto Enclose" preference in category "Morphic".
I can't stand it either. Surprisingly little of my coding consists of typing code out linearly.
Likewise. I really, really, hate this supposedly ???helpful??? kind of thing.
tim
tim Rowledge; tim@rowledge.org; http://www.rowledge.org/tim A bug in the code is worth two in the documentation.
On 10 April 2014 02:19, Chris Muller asqueaker@gmail.com wrote:
Bonk! Of course, I _love_ Auto Enclose! :-)
Because I'm not "typing a stream of code", I'm "composing nested expressions".
Hey, for a change I agree with Chris! For any Lisp users out there, the equivalent is a paredit-mode lite. Paredit (an Emacs mode) is all about composing and manipulating nested expressions. What would make auto-enclose much more paredit like is automating the manipulation of parens. Not just adding a ) every time you type a ( but letting e you "slurp" expressions into the enclosing paren (whether [], {} or () parens) or "burp" them out, deleting empty pairs of parens, and so on. Paredit takes a bit of getting used to - precisely because you're leaning on the IDE more to manipulate expressions rather than adding/deleting characters - but once you get used to it, it's awesome.
frank
>...Not just adding a ) every time you type a ( but letting e you >"slurp" expressions into the enclosing paren (whether [], {} or () >parens) or "burp" them out, deleting empty pairs of parens, and so on.
Throw in a few microwave burritos and coding smalltalk will sound downright symphonic.
Having worked in AutoLisp for quite some times I heavily used this feature. Enclose parts of expressions when I got precedence wrong or to temporarily commenting out code with this feature was really handy.
It's been nasty with < and with ' in comments.
So when it was gone and I didn't find the preference I just stopped using the feature :-))
Cheers
Herbert
Am 10.04.2014 11:47, schrieb Frank Shearar:
On 10 April 2014 02:19, Chris Muller asqueaker@gmail.com wrote:
Bonk! Of course, I _love_ Auto Enclose! :-)
Because I'm not "typing a stream of code", I'm "composing nested expressions".
Hey, for a change I agree with Chris! For any Lisp users out there, the equivalent is a paredit-mode lite. Paredit (an Emacs mode) is all about composing and manipulating nested expressions. What would make auto-enclose much more paredit like is automating the manipulation of parens. Not just adding a ) every time you type a ( but letting e you "slurp" expressions into the enclosing paren (whether [], {} or () parens) or "burp" them out, deleting empty pairs of parens, and so on. Paredit takes a bit of getting used to - precisely because you're leaning on the IDE more to manipulate expressions rather than adding/deleting characters - but once you get used to it, it's awesome.
frank
Likewise. I really, really, hate this supposedly ‘helpful’ kind of thing.
+1 How we can do to promote reading instead of (fast) writing? ...and to delegate/refactor instead of composing expressions/objects.
IMHO, Help while writing is short term help.
Ale.
----- Original Message ----- From: "tim Rowledge" tim@rowledge.org To: "The general-purpose Squeak developers list" squeak-dev@lists.squeakfoundation.org Sent: Wednesday, April 09, 2014 9:57 PM Subject: Re: [squeak-dev] How do I convince a Workspace to stop "helping" me?
On 09-04-2014, at 5:54 PM, Colin Putney colin@wiresong.com wrote:
On Wed, Apr 9, 2014 at 7:37 PM, David T. Lewis lewis@mail.msen.com wrote:
I tried both of those, no joy. Actually, I looked through every preference in the Preference Browser, and all the settings in the workspace menu. I must be missing something, but I can't make this feature go away.
You want the "Auto Enclose" preference in category "Morphic".
I can't stand it either. Surprisingly little of my coding consists of typing code out linearly.
Likewise. I really, really, hate this supposedly ‘helpful’ kind of thing.
tim -- tim Rowledge; tim@rowledge.org; http://www.rowledge.org/tim A bug in the code is worth two in the documentation.
On Wed, Apr 09, 2014 at 07:54:35PM -0500, Colin Putney wrote:
On Wed, Apr 9, 2014 at 7:37 PM, David T. Lewis lewis@mail.msen.com wrote:
I tried both of those, no joy. Actually, I looked through every preference in the Preference Browser, and all the settings in the workspace menu. I must be missing something, but I can't make this feature go away.
You want the "Auto Enclose" preference in category "Morphic".
Thank you, thank you! That was the problem.
Somehow it never occurred to me that this feature would be related to Morphic, possibly because it isn't. But the feature is not implemented in MVC, so I guess that makes it a Morphic preference ;-)
I can't stand it either. Surprisingly little of my coding consists of typing code out linearly.
I also find it quite interesting to see how much attention is paid to "making coding more productive," as if entering more words per minute into code panes would somehow relate in any positive way to productivity.
Dave
I believe way back in about early 4.4, i was able to do the following in order to get that functionality, I sorta got used to the auto add of the closing bracket.
From my migration script:
"==================================" "OCompletion provides source code completion as you type" "Has some issues loading into 10759" "(Installer ss project: 'OCompletion') install: 'Ocompletion'. (Smalltalk at: #ECToolSet) register. (Smalltalk at: #ToolSet) default: (Smalltalk at: #ECToolSet)."
"This is needed in order to reload the saved prefs file" "Installer ss project: 'MetacelloRepository'; install: 'ConfigurationOfOCompletion'. #(project stableVersion load) inject: (Smalltalk at: #ConfigurationOfOCompletion) into: [ :object :selector | object perform: selector ]." "OCompletion provides source code completion as you type" (Installer ss project: 'OCompletion') install: 'Ocompletion'. (Smalltalk at: #ECToolSet) register. (Smalltalk at: #ToolSet) default: (Smalltalk at: #ECToolSet). "==================================“
I’m sorta remember an eCompletion as well, and maybe also something in OmniBrowser?
Hope this helps.
Ken G. Brown
On Apr 9, 2014, at 18:37, David T. Lewis lewis@mail.msen.com wrote:
On Wed, Apr 09, 2014 at 05:18:27PM -0700, tim Rowledge wrote:
On 09-04-2014, at 5:11 PM, David T. Lewis lewis@mail.msen.com wrote:
Whenever I type an open-paren $( in a Squeak 4.5 workspace, some mysterious agent of code format rectitude steps in and supplies the matching close-paren. This is hugely annoying and I can't seem to find any preferences or workspace menu settings to make it stop.
Does anyone know how to make this feature go away?
Helpful code can be so unhelpful, right? See if by any chance the syntaxHighlightinasyoutype preference is on - in the ?browsing? section of the prefs browser. There?s also shoutstylinginworkspace. Neither of them was on in my 4.5 image but y?never know.
I tried both of those, no joy. Actually, I looked through every preference in the Preference Browser, and all the settings in the workspace menu. I must be missing something, but I can't make this feature go away.
Dave
I got that the other day, but I think it was when I was playing around with condenseSources and condenseChanges. After re-unzipping from the zip, changing class comment worked for me.
On Wed, Apr 9, 2014 at 6:52 PM, David T. Lewis lewis@mail.msen.com wrote:
Create a new class, attempt to edit the class comment ==> Error: RemoteString past end of file
Surely I cannot be the first person to have edited a class comment in the last month or so ... I hope.
Dave
Hi,
works for me in two images on two OS.
Cheers,
Herbert
Am 10.04.2014 01:52, schrieb David T. Lewis:
Create a new class, attempt to edit the class comment ==> Error: RemoteString past end of file
Surely I cannot be the first person to have edited a class comment in the last month or so ... I hope.
Dave
WriteStream>>nextChunkPut
add a self flush at the end will fix it.
Discussion several months ago fingered the gcc 4 compiler so something.
cheers.
tty
---- On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis<lewis@mail.msen.com> wrote ----
Create a new class, attempt to edit the class comment ==> Error: RemoteString past end of file
Surely I cannot be the first person to have edited a class comment in the last month or so ... I hope.
Dave
Ah, right, thank you! I do recall the discussion now that you mention it. I use VMs compiled on my own machine, which happens to be a new Ubuntu system at the moment. It probably has the buggy gcc library.
Thanks a lot, Dave
WriteStream>>nextChunkPut
add a self flush at the end will fix it.
Discussion several months ago fingered the gcc 4 compiler so something.
cheers.
tty
---- On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis<lewis@mail.msen.com> wrote ----
Create a new class, attempt to edit the class comment ==> Error: RemoteString past end of file
Surely I cannot be the first person to have edited a class comment in the last month or so ... I hope.
Dave
Okay, I was just getting this again in our release image. Strange -- it happens on one laptop but not the other..! The suggested patch seemed to work for me, don't we need to push this fix to trunk and squeak45?
On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis lewis@mail.msen.com wrote:
Ah, right, thank you! I do recall the discussion now that you mention it. I use VMs compiled on my own machine, which happens to be a new Ubuntu system at the moment. It probably has the buggy gcc library.
Thanks a lot, Dave
WriteStream>>nextChunkPut
add a self flush at the end will fix it.
Discussion several months ago fingered the gcc 4 compiler so something.
cheers.
tty
---- On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis<lewis@mail.msen.com> wrote ----
Create a new class, attempt to edit the class comment ==> Error: RemoteString past end of file
Surely I cannot be the first person to have edited a class comment in the last month or so ... I hope.
Dave
I should double check to be sure, but I think this was a bug in the run time libraries of the operating system, as opposed to something in the VM or the image. I can't say I like the idea of putting a kludge into the image to work around a bug in Ubuntu, although it might be worthwhile in this case just to prevent new users from having a bad experience.
Come to think of it, we should be able to write a unit test that would detect this problem when running Squeak on a buggy OS.
Dave
Okay, I was just getting this again in our release image. Strange -- it happens on one laptop but not the other..! The suggested patch seemed to work for me, don't we need to push this fix to trunk and squeak45?
On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis lewis@mail.msen.com wrote:
Ah, right, thank you! I do recall the discussion now that you mention it. I use VMs compiled on my own machine, which happens to be a new Ubuntu system at the moment. It probably has the buggy gcc library.
Thanks a lot, Dave
WriteStream>>nextChunkPut
add a self flush at the end will fix it.
Discussion several months ago fingered the gcc 4 compiler so something.
cheers.
tty
---- On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis<lewis@mail.msen.com> wrote ----
Create a new class, attempt to edit the class comment ==> Error: RemoteString past end of file
Surely I cannot be the first person to have edited a class comment in the last month or so ... I hope.
Dave
Strange I encountered the bug on Ubuntu 14.04 instead of 12.04 instead of the other way around.. Actually not, Ubuntu seems to be getting worse in some ways..
On Fri, May 2, 2014 at 4:38 PM, David T. Lewis lewis@mail.msen.com wrote:
I should double check to be sure, but I think this was a bug in the run time libraries of the operating system, as opposed to something in the VM or the image. I can't say I like the idea of putting a kludge into the image to work around a bug in Ubuntu, although it might be worthwhile in this case just to prevent new users from having a bad experience.
Come to think of it, we should be able to write a unit test that would detect this problem when running Squeak on a buggy OS.
Dave
Okay, I was just getting this again in our release image. Strange -- it happens on one laptop but not the other..! The suggested patch seemed to work for me, don't we need to push this fix to trunk and squeak45?
On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis lewis@mail.msen.com wrote:
Ah, right, thank you! I do recall the discussion now that you mention it. I use VMs compiled on my own machine, which happens to be a new Ubuntu system at the moment. It probably has the buggy gcc library.
Thanks a lot, Dave
WriteStream>>nextChunkPut
add a self flush at the end will fix it.
Discussion several months ago fingered the gcc 4 compiler so something.
cheers.
tty
---- On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis<lewis@mail.msen.com> wrote ----
Create a new class, attempt to edit the class comment ==> Error: RemoteString past end of file
Surely I cannot be the first person to have edited a class comment in the last month or so ... I hope.
Dave
On Sat, May 03, 2014 at 11:18:11AM -0500, Chris Muller wrote:
On Fri, May 2, 2014 at 4:38 PM, David T. Lewis lewis@mail.msen.com wrote:
On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis lewis@mail.msen.com wrote:
On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis lewis@mail.msen.com wrote:
Create a new class, attempt to edit the class comment ==> Error: RemoteString past end of file
Surely I cannot be the first person to have edited a class comment in the last month or so ... I hope.
WriteStream>>nextChunkPut
add a self flush at the end will fix it.
Discussion several months ago fingered the gcc 4 compiler so something.
cheers.
tty
Ah, right, thank you! I do recall the discussion now that you mention it. I use VMs compiled on my own machine, which happens to be a new Ubuntu system at the moment. It probably has the buggy gcc library.
Thanks a lot, Dave
Okay, I was just getting this again in our release image. Strange -- it happens on one laptop but not the other..! The suggested patch seemed to work for me, don't we need to push this fix to trunk and squeak45?
I should double check to be sure, but I think this was a bug in the run time libraries of the operating system, as opposed to something in the VM or the image. I can't say I like the idea of putting a kludge into the image to work around a bug in Ubuntu, although it might be worthwhile in this case just to prevent new users from having a bad experience.
Come to think of it, we should be able to write a unit test that would detect this problem when running Squeak on a buggy OS.
Dave
Strange I encountered the bug on Ubuntu 14.04 instead of 12.04 instead of the other way around.. Actually not, Ubuntu seems to be getting worse in some ways..
I am having a hard time reducing this problem to something I can put into a unit test. I can't see any obvious problem in the way the writes are being performed on a file stream, and the actual failure seems to be happening when creating the initial RemoteStream for a class comment. When a new class is created without a class comment, the classComment in the ClassOrganizer is nil. When accepting a new comment in a browser for the first time, a RemoteStream is created to reference the comment that will be saved in the changes file. Something gets out of whack at that point, but I have not yet spotted the cause.
Dave
On Sat, May 03, 2014 at 02:01:57PM -0400, David T. Lewis wrote:
On Sat, May 03, 2014 at 11:18:11AM -0500, Chris Muller wrote:
On Fri, May 2, 2014 at 4:38 PM, David T. Lewis lewis@mail.msen.com wrote:
On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis lewis@mail.msen.com wrote:
On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis lewis@mail.msen.com wrote: > > Create a new class, attempt to edit the class comment ==> Error: > RemoteString past end of file > > Surely I cannot be the first person to have edited a class comment in > the > last month or so ... I hope. >
WriteStream>>nextChunkPut
add a self flush at the end will fix it.
Discussion several months ago fingered the gcc 4 compiler so something.
cheers.
tty
Ah, right, thank you! I do recall the discussion now that you mention it. I use VMs compiled on my own machine, which happens to be a new Ubuntu system at the moment. It probably has the buggy gcc library.
Thanks a lot, Dave
Okay, I was just getting this again in our release image. Strange -- it happens on one laptop but not the other..! The suggested patch seemed to work for me, don't we need to push this fix to trunk and squeak45?
I should double check to be sure, but I think this was a bug in the run time libraries of the operating system, as opposed to something in the VM or the image. I can't say I like the idea of putting a kludge into the image to work around a bug in Ubuntu, although it might be worthwhile in this case just to prevent new users from having a bad experience.
Come to think of it, we should be able to write a unit test that would detect this problem when running Squeak on a buggy OS.
Dave
Strange I encountered the bug on Ubuntu 14.04 instead of 12.04 instead of the other way around.. Actually not, Ubuntu seems to be getting worse in some ways..
I am having a hard time reducing this problem to something I can put into a unit test. I can't see any obvious problem in the way the writes are being performed on a file stream, and the actual failure seems to be happening when creating the initial RemoteStream for a class comment. When a new class is created without a class comment, the classComment in the ClassOrganizer is nil. When accepting a new comment in a browser for the first time, a RemoteStream is created to reference the comment that will be saved in the changes file. Something gets out of whack at that point, but I have not yet spotted the cause.
I pushed the fix^h^h^hworkaround for this to the inbox. I bothers me a bit that I can't come up with a reproduceable test case for this, but the change seems harmless.
If no objections, this will go into trunk.
Dave
Magma performance is definitely affected by #flush -- it takes care care to not flush more than once every 5 seconds.
I want to research whether this will affect that..
On Mon, May 5, 2014 at 11:48 AM, David T. Lewis lewis@mail.msen.com wrote:
On Sat, May 03, 2014 at 02:01:57PM -0400, David T. Lewis wrote:
On Sat, May 03, 2014 at 11:18:11AM -0500, Chris Muller wrote:
On Fri, May 2, 2014 at 4:38 PM, David T. Lewis lewis@mail.msen.com wrote:
On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis lewis@mail.msen.com wrote:
> On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis lewis@mail.msen.com wrote: >> >> Create a new class, attempt to edit the class comment ==> Error: >> RemoteString past end of file >> >> Surely I cannot be the first person to have edited a class comment in >> the >> last month or so ... I hope. >> > > WriteStream>>nextChunkPut > > add a self flush at the end will fix it. > > Discussion several months ago fingered the gcc 4 compiler so something. > > cheers. > > tty >
Ah, right, thank you! I do recall the discussion now that you mention it. I use VMs compiled on my own machine, which happens to be a new Ubuntu system at the moment. It probably has the buggy gcc library.
Thanks a lot, Dave
Okay, I was just getting this again in our release image. Strange -- it happens on one laptop but not the other..! The suggested patch seemed to work for me, don't we need to push this fix to trunk and squeak45?
I should double check to be sure, but I think this was a bug in the run time libraries of the operating system, as opposed to something in the VM or the image. I can't say I like the idea of putting a kludge into the image to work around a bug in Ubuntu, although it might be worthwhile in this case just to prevent new users from having a bad experience.
Come to think of it, we should be able to write a unit test that would detect this problem when running Squeak on a buggy OS.
Dave
Strange I encountered the bug on Ubuntu 14.04 instead of 12.04 instead of the other way around.. Actually not, Ubuntu seems to be getting worse in some ways..
I am having a hard time reducing this problem to something I can put into a unit test. I can't see any obvious problem in the way the writes are being performed on a file stream, and the actual failure seems to be happening when creating the initial RemoteStream for a class comment. When a new class is created without a class comment, the classComment in the ClassOrganizer is nil. When accepting a new comment in a browser for the first time, a RemoteStream is created to reference the comment that will be saved in the changes file. Something gets out of whack at that point, but I have not yet spotted the cause.
I pushed the fix^h^h^hworkaround for this to the inbox. I bothers me a bit that I can't come up with a reproduceable test case for this, but the change seems harmless.
If no objections, this will go into trunk.
Dave
Hi Chris,
On Mon, May 5, 2014 at 11:17 AM, Chris Muller ma.chris.m@gmail.com wrote:
Magma performance is definitely affected by #flush -- it takes care care to not flush more than once every 5 seconds.
I want to research whether this will affect that..
I doubt it very much. David's fix s to add a flush to nextChunkPut: and nextChunkPut: is used exclusively to write source code to sources/changes files.
On Mon, May 5, 2014 at 11:48 AM, David T. Lewis lewis@mail.msen.com wrote:
On Sat, May 03, 2014 at 02:01:57PM -0400, David T. Lewis wrote:
On Sat, May 03, 2014 at 11:18:11AM -0500, Chris Muller wrote:
On Fri, May 2, 2014 at 4:38 PM, David T. Lewis lewis@mail.msen.com
wrote:
On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis <
lewis@mail.msen.com> wrote:
> >> On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis <
lewis@mail.msen.com> wrote:
>>> >>> Create a new class, attempt to edit the class comment ==>
Error:
>>> RemoteString past end of file >>> >>> Surely I cannot be the first person to have edited a class
comment in
>>> the >>> last month or so ... I hope. >>> >> >> WriteStream>>nextChunkPut >> >> add a self flush at the end will fix it. >> >> Discussion several months ago fingered the gcc 4 compiler so
something.
>> >> cheers. >> >> tty >> > > Ah, right, thank you! I do recall the discussion now that you
mention it.
> I use VMs compiled on my own machine, which happens to be a new
Ubuntu
> system at the moment. It probably has the buggy gcc library. > > Thanks a lot, > Dave > Okay, I was just getting this again in our release image. Strange
--
it happens on one laptop but not the other..! The suggested patch seemed to work for me, don't we need to push this fix to trunk and squeak45?
I should double check to be sure, but I think this was a bug in the
run
time libraries of the operating system, as opposed to something in
the VM
or the image. I can't say I like the idea of putting a kludge into
the
image to work around a bug in Ubuntu, although it might be
worthwhile in
this case just to prevent new users from having a bad experience.
Come to think of it, we should be able to write a unit test that
would
detect this problem when running Squeak on a buggy OS.
Dave
Strange I encountered the bug on Ubuntu 14.04 instead of 12.04 instead of the other way around.. Actually not, Ubuntu seems to be getting worse in some ways..
I am having a hard time reducing this problem to something I can put
into
a unit test. I can't see any obvious problem in the way the writes are
being
performed on a file stream, and the actual failure seems to be happening when creating the initial RemoteStream for a class comment. When a new
class
is created without a class comment, the classComment in the
ClassOrganizer
is nil. When accepting a new comment in a browser for the first time, a RemoteStream is created to reference the comment that will be saved in
the
changes file. Something gets out of whack at that point, but I have not
yet
spotted the cause.
I pushed the fix^h^h^hworkaround for this to the inbox. I bothers me a
bit that
I can't come up with a reproduceable test case for this, but the change
seems
harmless.
If no objections, this will go into trunk.
Dave
On 5 May 2014 17:48, David T. Lewis lewis@mail.msen.com wrote:
On Sat, May 03, 2014 at 02:01:57PM -0400, David T. Lewis wrote:
On Sat, May 03, 2014 at 11:18:11AM -0500, Chris Muller wrote:
On Fri, May 2, 2014 at 4:38 PM, David T. Lewis lewis@mail.msen.com wrote:
On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis lewis@mail.msen.com wrote:
> On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis lewis@mail.msen.com wrote: >> >> Create a new class, attempt to edit the class comment ==> Error: >> RemoteString past end of file >> >> Surely I cannot be the first person to have edited a class comment in >> the >> last month or so ... I hope. >> > > WriteStream>>nextChunkPut > > add a self flush at the end will fix it. > > Discussion several months ago fingered the gcc 4 compiler so something. > > cheers. > > tty >
Ah, right, thank you! I do recall the discussion now that you mention it. I use VMs compiled on my own machine, which happens to be a new Ubuntu system at the moment. It probably has the buggy gcc library.
Thanks a lot, Dave
Okay, I was just getting this again in our release image. Strange -- it happens on one laptop but not the other..! The suggested patch seemed to work for me, don't we need to push this fix to trunk and squeak45?
I should double check to be sure, but I think this was a bug in the run time libraries of the operating system, as opposed to something in the VM or the image. I can't say I like the idea of putting a kludge into the image to work around a bug in Ubuntu, although it might be worthwhile in this case just to prevent new users from having a bad experience.
Come to think of it, we should be able to write a unit test that would detect this problem when running Squeak on a buggy OS.
Dave
Strange I encountered the bug on Ubuntu 14.04 instead of 12.04 instead of the other way around.. Actually not, Ubuntu seems to be getting worse in some ways..
I am having a hard time reducing this problem to something I can put into a unit test. I can't see any obvious problem in the way the writes are being performed on a file stream, and the actual failure seems to be happening when creating the initial RemoteStream for a class comment. When a new class is created without a class comment, the classComment in the ClassOrganizer is nil. When accepting a new comment in a browser for the first time, a RemoteStream is created to reference the comment that will be saved in the changes file. Something gets out of whack at that point, but I have not yet spotted the cause.
I pushed the fix^h^h^hworkaround for this to the inbox. I bothers me a bit that I can't come up with a reproduceable test case for this, but the change seems harmless.
This is the same issue that cropped up just after 4.4's release, right? IIRC I patched trunk & 4.4, but then soon reverted the workarounds because the community considered the workaround a Bad Thing (tm), possibly making the proper fix more difficult to find.
I'm happy if folks have since changed their mind.
If no objections, this will go into trunk.
Not from me.
frank
Dave
On Mon, May 05, 2014 at 09:48:49PM +0100, Frank Shearar wrote:
On 5 May 2014 17:48, David T. Lewis lewis@mail.msen.com wrote:
On Sat, May 03, 2014 at 02:01:57PM -0400, David T. Lewis wrote:
On Sat, May 03, 2014 at 11:18:11AM -0500, Chris Muller wrote:
On Fri, May 2, 2014 at 4:38 PM, David T. Lewis lewis@mail.msen.com wrote:
On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis lewis@mail.msen.com wrote: > >> On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis lewis@mail.msen.com wrote: >>> >>> Create a new class, attempt to edit the class comment ==> Error: >>> RemoteString past end of file >>> >>> Surely I cannot be the first person to have edited a class comment in >>> the >>> last month or so ... I hope. >>> >> >> WriteStream>>nextChunkPut >> >> add a self flush at the end will fix it. >> >> Discussion several months ago fingered the gcc 4 compiler so something. >> >> cheers. >> >> tty >> > > Ah, right, thank you! I do recall the discussion now that you mention it. > I use VMs compiled on my own machine, which happens to be a new Ubuntu > system at the moment. It probably has the buggy gcc library. > > Thanks a lot, > Dave > Okay, I was just getting this again in our release image. Strange -- it happens on one laptop but not the other..! The suggested patch seemed to work for me, don't we need to push this fix to trunk and squeak45?
I should double check to be sure, but I think this was a bug in the run time libraries of the operating system, as opposed to something in the VM or the image. I can't say I like the idea of putting a kludge into the image to work around a bug in Ubuntu, although it might be worthwhile in this case just to prevent new users from having a bad experience.
Come to think of it, we should be able to write a unit test that would detect this problem when running Squeak on a buggy OS.
Dave
Strange I encountered the bug on Ubuntu 14.04 instead of 12.04 instead of the other way around.. Actually not, Ubuntu seems to be getting worse in some ways..
I am having a hard time reducing this problem to something I can put into a unit test. I can't see any obvious problem in the way the writes are being performed on a file stream, and the actual failure seems to be happening when creating the initial RemoteStream for a class comment. When a new class is created without a class comment, the classComment in the ClassOrganizer is nil. When accepting a new comment in a browser for the first time, a RemoteStream is created to reference the comment that will be saved in the changes file. Something gets out of whack at that point, but I have not yet spotted the cause.
I pushed the fix^h^h^hworkaround for this to the inbox. I bothers me a bit that I can't come up with a reproduceable test case for this, but the change seems harmless.
This is the same issue that cropped up just after 4.4's release, right? IIRC I patched trunk & 4.4, but then soon reverted the workarounds because the community considered the workaround a Bad Thing (tm), possibly making the proper fix more difficult to find.
Let me guess ... it was probably me griping about it when you originally fixed it in 4.4, right?
I'm happy if folks have since changed their mind.
If it was me, then I guess changed my mind. No point in punishing the user just because they are using Ubuntu.
If no objections, this will go into trunk.
Not from me.
frank
Thanks. Let's let Chris check it out to make sure it does not affect Magma, then move it (back) to trunk.
Dave
On 6 May 2014 01:48, David T. Lewis lewis@mail.msen.com wrote:
On Mon, May 05, 2014 at 09:48:49PM +0100, Frank Shearar wrote:
On 5 May 2014 17:48, David T. Lewis lewis@mail.msen.com wrote:
On Sat, May 03, 2014 at 02:01:57PM -0400, David T. Lewis wrote:
On Sat, May 03, 2014 at 11:18:11AM -0500, Chris Muller wrote:
On Fri, May 2, 2014 at 4:38 PM, David T. Lewis lewis@mail.msen.com wrote:
> > On Thu, Apr 10, 2014 at 10:06 AM, David T. Lewis lewis@mail.msen.com wrote: >> >>> On Wed, 09 Apr 2014 16:52:54 -0700 David T. Lewis lewis@mail.msen.com wrote: >>>> >>>> Create a new class, attempt to edit the class comment ==> Error: >>>> RemoteString past end of file >>>> >>>> Surely I cannot be the first person to have edited a class comment in >>>> the >>>> last month or so ... I hope. >>>> >>> >>> WriteStream>>nextChunkPut >>> >>> add a self flush at the end will fix it. >>> >>> Discussion several months ago fingered the gcc 4 compiler so something. >>> >>> cheers. >>> >>> tty >>> >> >> Ah, right, thank you! I do recall the discussion now that you mention it. >> I use VMs compiled on my own machine, which happens to be a new Ubuntu >> system at the moment. It probably has the buggy gcc library. >> >> Thanks a lot, >> Dave >> > Okay, I was just getting this again in our release image. Strange -- > it happens on one laptop but not the other..! The suggested patch > seemed to work for me, don't we need to push this fix to trunk and > squeak45? > I should double check to be sure, but I think this was a bug in the run time libraries of the operating system, as opposed to something in the VM or the image. I can't say I like the idea of putting a kludge into the image to work around a bug in Ubuntu, although it might be worthwhile in this case just to prevent new users from having a bad experience.
Come to think of it, we should be able to write a unit test that would detect this problem when running Squeak on a buggy OS.
Dave
Strange I encountered the bug on Ubuntu 14.04 instead of 12.04 instead of the other way around.. Actually not, Ubuntu seems to be getting worse in some ways..
I am having a hard time reducing this problem to something I can put into a unit test. I can't see any obvious problem in the way the writes are being performed on a file stream, and the actual failure seems to be happening when creating the initial RemoteStream for a class comment. When a new class is created without a class comment, the classComment in the ClassOrganizer is nil. When accepting a new comment in a browser for the first time, a RemoteStream is created to reference the comment that will be saved in the changes file. Something gets out of whack at that point, but I have not yet spotted the cause.
I pushed the fix^h^h^hworkaround for this to the inbox. I bothers me a bit that I can't come up with a reproduceable test case for this, but the change seems harmless.
This is the same issue that cropped up just after 4.4's release, right? IIRC I patched trunk & 4.4, but then soon reverted the workarounds because the community considered the workaround a Bad Thing (tm), possibly making the proper fix more difficult to find.
Let me guess ... it was probably me griping about it when you originally fixed it in 4.4, right?
My father would say "no name, no pack drill" :) I think you did object, and one or two other folks, and for good enough reason that I retracted the fix.
I'm happy if folks have since changed their mind.
If it was me, then I guess changed my mind. No point in punishing the user just because they are using Ubuntu.
Yep, yep.
frank
If no objections, this will go into trunk.
Not from me.
frank
Thanks. Let's let Chris check it out to make sure it does not affect Magma, then move it (back) to trunk.
Dave
squeak-dev@lists.squeakfoundation.org
-
Alejandro F. Reimondo -
Chris Muller -
Chris Muller -
Colin Putney -
David T. Lewis -
Eliot Miranda -
Frank Shearar -
gettimothy -
Herbert König -
Ken G. Brown -
tim Rowledge