- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Adding a new catalog entry

by Pharo4Stef

> well we will have to clean and simplify all that in the future. > This is why versionner should be good so that people use it and we can simplify the declarations. > > Stef > > > But I had the feeling that Versionner UI is currently very limited vs Metacello options (especially w.r.t. cross dialect support) The problem is that metacello is too complex. :) And the API could be better. > Or do you have a different POV, like only Pharo3.0 stuff should go to MetaRepoForPharo3.0, and the ConfigurationOf... should be expunged from any other dialect/version ? For now this is ok to have multiple dialects in the metarepo because people may want to have gemstone and squeak version. Now for the core Pharo tool configruations should be as simple as possible (RB, â¦) Stef > >> >> >> Bye >> T. >> >> Gesendet: Sonntag, 16. MÃ¤rz 2014 um 17:31 Uhr >> Von: "Nicolas Cellier" <nicolas.cellier.aka.nice(a)gmail.com> >> An: "Pharo Development List" <pharo-dev(a)lists.pharo.org> >> Betreff: Re: [Pharo-dev] Adding a new catalog entry >> >> 2014-03-16 16:06 GMT+01:00 Sven Van Caekenberghe <sven(a)stfx.eu>: >> Nicolas, >> >> On 16 Mar 2014, at 10:08, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: >> >> > To get access your StHub account, if you have any, should be part of the Pharo team, but I don't think I can do that. >> >> I now see that 'nice' is part of 'Pharo', so I think you should be able to save to MetaRepoForPharo30. >> >> >> Thanks, I think I messed up when entering my credentials... >> I have now published with a symbolic stable version. >> >> Sven

March 17, 2014

Re: [Pharo-dev] EyeTreeInspector not ready for prime-time

by Sven Van Caekenberghe

On 13 Mar 2014, at 12:51, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > I'm glad we're re-inventing the tools! And, I'm struggling with this tool in > its current form... > > As others have mentioned, the "class" pseudo-variable is confusing, but here > are some that I find worse: > - The receiver does not change with the selection. For me, this was a key > feature of the explorer. > - It does not properly update e.g. if you are inspecting a class containing > a dictionary and add an item to the dictionary, collapsing and expanding the > dictionary node does not show the added item. It seems you must open an > entirely new inspector. > - If you toggle from EyeTreeInspector to EyeInspector, your code in the > bottom pane is lost With the following two cases integrated https://pharo.fogbugz.com/f/cases/13076/Optimize-the-information-displayed-… https://pharo.fogbugz.com/f/cases/13086/EyeTreeInspector-text-pane-doItRece… we're up to the next two open cases https://pharo.fogbugz.com/f/cases/13090/EyeTreeInspector-should-open-with-o… https://pharo.fogbugz.com/f/cases/13091/EyeInspector-switching-should-prese… which would leave the updating issue, which has to be described. Any help appreciated ! Sven

March 17, 2014

Re: [Pharo-dev] WhatsUp from: 2014-03-17 until: 2014-03-31

by Stephan Eggermont

Added Pier3 to MetaRepoForPharo30 Stephan

March 17, 2014

Re: [Pharo-dev] WhatsUp from: 2014-03-17 until: 2014-03-31

by Stefan Marr

Hi: On 17 Mar 2014, at 07:00, seaside(a)rmod.lille.inria.fr wrote: > ### Hereâs what I've been up to since the last WhatsUp: - Fixed the CogVM simulator and the related build setup for OSX. This means, you can go to: https://github.com/pharo-project/pharo-vm/pull/20 grab the source and build the VM on OSX with Bochs plugin and start debugging the VM in the simulator. The build still needs to be adapted to the other platforms as well. Best regards Stefan -- Stefan Marr INRIA Lille - Nord Europe http://stefan-marr.de/research/

March 17, 2014

Re: [Pharo-dev] Image's size in pillar

by Damien Cassou

Hi Natalia, On Sun, Mar 16, 2014 at 2:23 PM, Natalia Tymchuk <natalia.tymchuk(a)unikernel.net> wrote: > but with changing the value width I didn't get changes of my images sizes. How can I make my images take half of the page's width? I fixed this a month ago in the development version of Pillar but didn't release a stable version yet. -- Damien Cassou http://damiencassou.seasidehosting.st "Success is the ability to go from one failure to another without losing enthusiasm." Winston Churchill

March 17, 2014

WhatsUp from: 2014-03-17 until: 2014-03-31

by seaside＠rmod.lille.inria.fr

Hi! We're sending this automatic email twice a month, to give the community an opportunity to easily know what's happening and to coordinate efforts. Just answer informally, and feel free to spawn discussions thereafter! ### Here's what I've been up to since the last WhatsUp: - $HEROIC_ACHIEVEMENTS_OR_DISMAL_FAILURES_OR_SIMPLE_BORING_NECESSARY_TASKS ### What's next, until 2014-03-31 (*): - $NEXT_STEPS_TOWARDS_WORLD_DOMINATION (*) we'll be expecting results by then ;)

March 17, 2014

Re: [Pharo-dev] [ANN] ZTimezone support added to ZTimestamp

by Tudor Girba

Thanks, Sven! This might like a look small thing, but if we think at the amount of energy invested around fixing the current date/timestamp, we will get a different picture :). Doru On Sun, Mar 16, 2014 at 10:12 PM, Sven Van Caekenberghe <sven(a)stfx.eu>wrote: > Hi, > > I just added a first version of timezone support to the ZTimestamp project > - a small, lightweight, non-intrusive, self-dependent alternative to > DateAndTime and friends. With this addition, it is now possible to work > with UTC/GMT timestamps internally yet present correctly localised > timestamps in any timezone, based on the standard Olsen database. > ZTimestampFormat was extended with optional timezone support. > > About ZTimestamp > > > https://ci.inria.fr/pharo-contribution/job/PharoProjectCatalog/HTML_Report/… > > https://github.com/svenvc/docs/blob/master/neo/ztimestamp.md > > ZTimestamp can be loaded most easily from the Pharo 2 or 3 Configuration > Browser. > > StHub > > > http://www.smalltalkhub.com/#!/~SvenVanCaekenberghe/Neo/packages/ZTimestamp > > CI job > > https://ci.inria.fr/pharo-contribution/job/ZTimestamp/ > > From the class comment of ZTimezone > > > https://github.com/svenvc/ztimestamp/tree/master/repository/ZTimestamp.pack… > > === > > I am ZTimezone, representing the timezone information in the standard > Olsen database. > > http://en.wikipedia.org/wiki/Tz_database > > Usage > > You reference timezones by their ID. The list of supported identifiers is > accessible using > > ZTimezone timezoneIdentifiers. > > To access a timezone do > > ZTimezone id: 'Europe/Brussels'. > > The necessary information will be loaded, parsed and cached from a binary > file of the zoneinfo database (see also man tzfile). This should work > automagically on Mac OS X and Unix, on Windows you have to download the > necessary files and specify their location > > ZTimezone zoneInfoLocation: FileLocator C / 'foo' / 'bar' / 'zoneinfo'. > > Once you get a handle on a timezone, the main operation is to query the > sub timezone that is applicable at a certain point in time > > (ZTimezone id: 'Europe/Brussels') subzoneForTimestamp: ZTimestamp now. > > The ZSubTimezone instance returned contains information like the UTC > offset. Since you'll probably only be interested in that aspect there is a > convenience method > > (ZTimezone id: 'Europe/Brussels') offsetForTimestamp: DateAndTime now. > > The flow is that for every GMT timestamp, you get the concrete offset to > use for a specific timezone. Note that this is not a constant, it depends > on the time periode the timestamp falls in. > > The are 2 more convenience methods to quickly convert between GTM and > local wall time > > (ZTimezone id: #'Europe/Brussels') gmtToLocal: ZTimestamp now. > (ZTimezone id: #'Europe/Brussels') localToGmt: DateAndTime now. > > Also note the zoneTab and the timezones are cached in the image. When the > TZ database changes, it might be necessary to either call #cleanUp or > #reloadAll. When moving images between machines, either all info should be > loaded and cached, or it might be necessary to use #zoneInfoLocation: again. > > Implementation > > A chronological array of transition points in unix time specifies which > sub zone is active from that point on to the next. > > === > > Sven > > > -- www.tudorgirba.com "Every thing has its own flow"

March 17, 2014

Interesting link

by S Krish

1. Since then, Smalltalk (SqueakNOS <http://squeaknos.blogspot.com/>), Forth (colorForth <http://www.colorforth.com/cf.htm>), and Lisp (Genera<http://en.wikipedia.org/wiki/Genera_(operating_system)>) have all flirted with becoming operating systems, and Oberon<http://en.wikipedia.org/wiki/Oberon_(programming_language)> was designed<http://www.inf.ethz.ch/personal/wirth/ProjectOberon/PO.System.pdf> to be one <http://en.wikipedia.org/wiki/Oberon_(operating_system)> from the start. But none achieved economic success, for the simple reason that none of the projects involved attempted to provide value to people. They solved technical problems to validate that their concepts can work in the real world, but did not pursue the delivery of better solutions to real-world problems than would otherwise be possible.â©<http://davidad.github.io/blog/2014/03/12/the-operating-system-is-out-of-dat…> Pharo hopefully corrects the course.. >From : http://davidad.github.io/blog/2014/03/12/the-operating-system-is-out-of-dat… Most of all, letâs rethink the received wisdom that you should teach your computer to do things in a programming language and run the resulting program on an operating system. A righteous operating system should be a programming language. And for goodness' sake, letâs not use the entire network stack just to talk to another process on the same machine which is responsible for managing a database using the filesystem stack. At least letâs use shared memory (thatâs what itâs *for*!). But if we believe in the future â if we believe in ourselves â letâs dare to ask why, anyway, does the operating system give you this âfilesystemâ thing thatâs no good as a database and expect you to just accept that âstuff on computers goes in folders, lahâ? Any decent software environment ought to have a fully featured database, built in, and no need for a âfilesystemâ.

March 17, 2014

Re: [Pharo-dev] [ANN] ZTimezone support added to ZTimestamp

by Sebastian Sastre

fantaaaastic Sven! thanks for sharing sebastian o/ On Mar 16, 2014, at 6:12 PM, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: > Hi, > > I just added a first version of timezone support to the ZTimestamp project - a small, lightweight, non-intrusive, self-dependent alternative to DateAndTime and friends. With this addition, it is now possible to work with UTC/GMT timestamps internally yet present correctly localised timestamps in any timezone, based on the standard Olsen database. ZTimestampFormat was extended with optional timezone support. > > About ZTimestamp > > https://ci.inria.fr/pharo-contribution/job/PharoProjectCatalog/HTML_Report/… > > https://github.com/svenvc/docs/blob/master/neo/ztimestamp.md > > ZTimestamp can be loaded most easily from the Pharo 2 or 3 Configuration Browser. > > StHub > > http://www.smalltalkhub.com/#!/~SvenVanCaekenberghe/Neo/packages/ZTimestamp > > CI job > > https://ci.inria.fr/pharo-contribution/job/ZTimestamp/ > > From the class comment of ZTimezone > > https://github.com/svenvc/ztimestamp/tree/master/repository/ZTimestamp.pack… > > === > > I am ZTimezone, representing the timezone information in the standard Olsen database. > > http://en.wikipedia.org/wiki/Tz_database > > Usage > > You reference timezones by their ID. The list of supported identifiers is accessible using > > ZTimezone timezoneIdentifiers. > > To access a timezone do > > ZTimezone id: 'Europe/Brussels'. > > The necessary information will be loaded, parsed and cached from a binary file of the zoneinfo database (see also man tzfile). This should work automagically on Mac OS X and Unix, on Windows you have to download the necessary files and specify their location > > ZTimezone zoneInfoLocation: FileLocator C / 'foo' / 'bar' / 'zoneinfo'. > > Once you get a handle on a timezone, the main operation is to query the sub timezone that is applicable at a certain point in time > > (ZTimezone id: 'Europe/Brussels') subzoneForTimestamp: ZTimestamp now. > > The ZSubTimezone instance returned contains information like the UTC offset. Since you'll probably only be interested in that aspect there is a convenience method > > (ZTimezone id: 'Europe/Brussels') offsetForTimestamp: DateAndTime now. > > The flow is that for every GMT timestamp, you get the concrete offset to use for a specific timezone. Note that this is not a constant, it depends on the time periode the timestamp falls in. > > The are 2 more convenience methods to quickly convert between GTM and local wall time > > (ZTimezone id: #'Europe/Brussels') gmtToLocal: ZTimestamp now. > (ZTimezone id: #'Europe/Brussels') localToGmt: DateAndTime now. > > Also note the zoneTab and the timezones are cached in the image. When the TZ database changes, it might be necessary to either call #cleanUp or #reloadAll. When moving images between machines, either all info should be loaded and cached, or it might be necessary to use #zoneInfoLocation: again. > > Implementation > > A chronological array of transition points in unix time specifies which sub zone is active from that point on to the next. > > === > > Sven > >

March 17, 2014

Re: [Pharo-dev] Adding a new catalog entry

by Nicolas Cellier

2014-03-16 23:56 GMT+01:00 Sven Van Caekenberghe <sven(a)stfx.eu>: > > On 16 Mar 2014, at 22:44, Nicolas Cellier < > nicolas.cellier.aka.nice(a)gmail.com> wrote: > > > Or do you have a different POV, like only Pharo3.0 stuff should go to > MetaRepoForPharo3.0, and the ConfigurationOf... should be expunged from any > other dialect/version ? > > MetaRepoForPharo3 should only contain configurations that load/work > without problems in Pharo 3 and that are/will be supported. > > Of course, if a configuration contains supports for other platforms as > well, that is OK. > Yes, I see that is the current practice. But I mean one could imagine another workflow: - an automated tool extracts pharoX.Y configuration from a multi-dialect configuration This is what is published in MetaRepoForPharoX.Y (maybe a pharoX.Y user don't care that much about squeakZ.T or gemstone nor of a very old version dedicated to Pharo0.1) - the pharoX.Y only configuration is managed via a simple Versionner UI and committed into MetaRepoForPharoX.Y in short cycle until it works well enough - an automated tool merge pharoX.Y configuration back into multi-dialect configuration stored somewhere else I don't think the 1st item is difficult (doesn't Metacello does exactly that when loading) 2nd item is almost what you have currently with Versionner. The 3rd item is the difficult point, it would have to track differences between dialectA and dialectB or dialect version X and dialect version Z, but also consistency of package baselines and versions. However I'm sure automation would be a great improvment in this area (compared to manual dissection of configuration).

March 16, 2014