March 2015 - Pharo-users - lists.pharo.org

Re: [Pharo-users] Question about NeoJSON

by Mariano Martinez Peck

On Thu, Mar 12, 2015 at 6:09 PM, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: > I don't really understand why you want to keep the original JSON. It is a > bit like you don't trust the parsing and/or mapping. Now, the mapping I > understand (somewhat). > > Yes, there are a few domain-specific reasons, but let me outline a few: 1) Imagine that from this JSON I have a list of objects that I build. Imagine for a minute these are "InvestmentTransactions". The way I build these objects from JSON could be quite complex. And can even have different kind of JSONs that I receive to build same type of objects. So sometimes..it is very handy to have the original JSON string to see why an object was built that way, why it has certain state, etc. 2) The app user could later MODIFY the created instances of InvestmentTransactions. However, it is nice to see which was the original values in case of the transactions that were imported via JSON. 3) These JSON I receive may have some query / API limits (queries per second etc). And I may need to replicate this to multiple other apps of mine. If all my apps were all calling the API I may likely reach limits. So one possibility was to implement our own REST / JSON server for our local apps. In that case, having the original JSON would allow me a "repeater server" very easily. BTW...note that having a cache in each server does not solve the ratio limit problem. 4) Imagine that so far the InvestmentTransction object have 20 instVars but the JSON element of a transaction brings 50 fields. I don't want to right now store everything...just the 20 instVars I need. In addition, the server that sends the JSON may send you only a time-window..say..the transactions of the last month. So..maybe.. in the future, my app has more features and then suddenly I need other properties I was not yet mapping. But the server only sends me the last month. Therefore, if I store the whole response string, I can have access to all data even later. Ok..these were a few reasons. Not all scenarios are like that, and maybe storing the original string is not the best solution either. I am just explaining why I find useful to store it. > So I would parse once without mapping, which would give you pure Arrays > and Dictionaries that you are guaranteed can be used to reproduce the full > original input. Pick the second level in that structure and use it. > > Indeed, that was what I was thinking too. Thanks for the alternative. > > On 12 Mar 2015, at 21:47, Mariano Martinez Peck <marianopeck(a)gmail.com> > wrote: > > > > > > > > On Thu, Mar 12, 2015 at 5:32 PM, Sven Van Caekenberghe <sven(a)stfx.eu> > wrote: > > Mariano, > > > > I assume you what this while reading JSON. That won't be possible. > NeoJSON is designed as an efficient stream parser, working as it goes. So > it never knows the whole JSON expression. > > > > Uhhhh I imagined that :( > > > > > > I would do it manually, but that can only be done at the top level. > > > > Well...my list of TestObject is not at the root, but second level..... > > But in any case... how would you do that? > > > > Thanks! > > > > > > Sven > > > > > On 12 Mar 2015, at 21:22, Mariano Martinez Peck <marianopeck(a)gmail.com> > wrote: > > > > > > > > > > > > On Thu, Mar 12, 2015 at 5:16 PM, Esteban A. Maringolo < > emaringolo(a)gmail.com> wrote: > > > Assuming you store the original JSON string as a String object in an > > > instance variable named "json" you could map it as any other string. > > > > > > > > > But how do I get the JSON string associated to the TestObject > instance? In other words... > > > > > > mapper for: TestObject do: [ :mapping | > > > > > > "HERE how do I get the JSON string of TestObject in order to map it to > 'jsonOriginalString' ? " > > > > > > ] > > > > > > > > > Or are you planning to store it as JSON object itself? > > > > > > no, a string is OK. > > > > > > Thanks in advance, > > > > > > > > > > > > Esteban A. Maringolo > > > > > > > > > 2015-03-12 16:50 GMT-03:00 Mariano Martinez Peck < > marianopeck(a)gmail.com>: > > > > Hi, > > > > > > > > In this link: > > > > https://github.com/svenvc/docs/blob/master/neo/neo-json-paper.md > > > > > > > > I see an example of a custom mapping like this one: > > > > > > > > mapper for: TestObject do: [ :mapping | > > > > mapping mapInstVars: #(id name). > > > > (mapping mapInstVar: #timestamp to: 'created-at') valueSchema: > > > > DateAndTime. > > > > (mapping mapInstVar: #points) valueSchema: #ArrayOfPoints. > > > > (mapping mapInstVar: #bytes) valueSchema: ByteArray ]. > > > > > > > > > > > > I need something similar, but I need to store the original JSON > string that > > > > was used to build a particular TestObject instance, in an instance > variable > > > > of TestObject too. In other words, I need an instVar > "originalResponse" in > > > > TestObject that would be set with the original JSON string. > > > > > > > > Any ideas how can I implement this? > > > > > > > > Thanks in advance, > > > > > > > > -- > > > > Mariano > > > > http://marianopeck.wordpress.com > > > > > > > > > > > > > > > -- > > > Mariano > > > http://marianopeck.wordpress.com > > > > > > > > > > > > -- > > Mariano > > http://marianopeck.wordpress.com > > > -- Mariano http://marianopeck.wordpress.com

March 13, 2015

Re: [Pharo-users] When to use a stream for concatenating text..

by Sven Van Caekenberghe

The breaking point seems to be around 5 concatenations: str:= Character alphabet. [ str, str, str, str, str ] bench. "'1,020,043 per second'" [ String new: str size * 5 streamContents: [ :out | 5 timesRepeat: [ out nextPutAll: str ] ] ] bench. "'949,738 per second'" > On 13 Mar 2015, at 08:52, Marcus Denker <marcus.denker(a)inria.fr> wrote: > >> >> On 13 Mar 2015, at 08:41, stepharo <stepharo(a)free.fr> wrote: >> >> use >> >> String streamContents: [:s | >> s nextPutAll: 'jlklkjkl' ] >> >> >> or >> >> String streamContents: [:s | >> s << 'jlklkjkl' ] >> >> >> it is a great method for manipulating >> >> Le 10/3/15 19:09, sergio_101 a Ã©crit : >>> >>> it seems that in more cases than not, i find that developers use a stream when concatenating some text strings. >>> >>> I am wondering if this is a smalltalk thing, or is there a real speed benefit when using streams in this way. >>> > > When adding to a string, you create a new one with the right size and copy over the old content. > > Using a stream avoids that. But there is a cost for handling the stream, too: > > [String streamContents: [:s | > s nextPutAll: 'jlklkjkl'; > nextPutAll: 'jlklkjkl' ]] bench. "'1,866,774 per second'" > > > ['jlklkjkl' , 'jlklkjkl'] bench "'5,433,931 per secondâ" > > > So it is only faster e.g. when adding to a string in a loop or on larger strings. > But it reads quite nice, too, so I tend to use it without thinking about performance. > > Marcus

March 13, 2015

[OT] DB Open Data-Train Challenge in Germany

by Torsten Bergmann

Similar to the Dataton in Paris (http://lists.pharo.org/pipermail/pharo-users_lists.pharo.org/2015-March/017…) with data from Africa to visualize possible with Roassal there is in germany next friday (20.3.2015) a day in Frankfurt organized by german railway, also with open data (timetables, train data, ...): https://www.eventbrite.com/e/db-open-data-train-challenge-tickets-160262449… So if one might be interested to visit the event. BTW: German railway is using Smalltalk applications a lot to do all their timetable planning and train simulations (see http://www.uic.org/html/it/e-business_2005/docs/woelfle.pdf for some screenshots) Maybe this is a good chance to demo Pharo or use Roassal to visualize data.

March 13, 2015

Re: [Pharo-users] Cleaning changes file?

by Marcus Denker

> On 13 Mar 2015, at 09:18, stepharo <stepharo(a)free.fr> wrote: > > >>>>> >>>>> You can do that using the expression: >>>>> >>>>> PharoChangesCondenser condense >>>>> >>>>> Although that also means you lose the version history of methods inside your image. This is typically not done a lot, maybe for production deploy, maybe for release of Pharo itself. >>>>> >>>>> >>>> Yes, but we need to change that. The artefact on the CI server needs to be the artefact of release. We should not have *anything* of the kind âand a day before release we call Xâ. >>>> Because for sure it will be broken. And we waste resources: the changes are the history of the revision control system. >>> >>> OK so a silly and not-so-innocent question: Why not include that in the build script *right now* ? It has to be done some time, and now the discussion is open why not make use of the opportunity? >>> >>> Also, my use of the changes file is uniquely to recover in case of a crash. With condensed changes for every daily build I download this would make my recovery just that bit easier. One small change â¦ >>> >> >> From a size point of view it would make sense: a âPharo4.sourcesâ is even a bit smaller than the current .changes (which itself would be empty). Compressed it would be 15,7 instead of 13,7 due to the .changes >> being better compressible (it contains much useless things even by that metricâ¦) >> >> The problem is that we would need to see how it plays with the get.pharo.org <http://get.pharo.org/> where the vm download contains all of Pharo1, Pharo2 and Pharo3 sourcesâ¦ > > I do not understand your stress. If we produce a condense change files what is the problem. > You are talking about the problem of having multiple .sources? yes. we download always *all* sources files with the VM. Marcus

March 13, 2015

Re: [Pharo-users] Cleaning changes file?

by stepharo

>>>> >>>> You can do that using the expression: >>>> >>>> PharoChangesCondenser condense >>>> >>>> Although that also means you lose the version history of methods >>>> inside your image. This is typically not done a lot, maybe for >>>> production deploy, maybe for release of Pharo itself. >>>> >>>> >>> Yes, but we need to change that. The artefact on the CI server needs >>> to be the artefact of release. We should not have *anything* of the >>> kind âand a day before release we call Xâ. >>> Because for sure it will be broken. And we waste resources: the >>> changes are the history of the revision control system. >> >> OK so a silly and not-so-innocent question: Why not include that in >> the build script *right now* ? It has to be done some time, and now >> the discussion is open why not make use of the opportunity? >> >> Also, my use of the changes file is uniquely to recover in case of a >> crash. With condensed changes for every daily build I download this >> would make my recovery just that bit easier. One small change â¦ >> > > From a size point of view it would make sense: a âPharo4.sourcesâ is > even a bit smaller than the current .changes (which itself would be > empty). Compressed it would be 15,7 instead of 13,7 due to the .changes > being better compressible (it contains much useless things even by > that metricâ¦) > > The problem is that we would need to see how it plays with the > get.pharo.org <http://get.pharo.org> where the vm download contains > all of Pharo1, Pharo2 and Pharo3 sourcesâ¦ I do not understand your stress. If we produce a condense change files what is the problem. You are talking about the problem of having multiple .sources? > > This is a real messâ¦ I am locking forward to a time when we just have > one .pharo image. > > Marcus > >

March 13, 2015

Re: [Pharo-users] Notebook/slideshow morph

by stepharo

Pay attention that my code was an experiement to build test first a mini powerpoint. So there is not much. > Is there a way to save and reload the slide show. Like in the good old > squeak days (TM) with a *.morph file? No but nothing prevent you to plug fuel for that? Now I was thinking that I want to rebuild it on bloc but I should not start before finishing a long list of todo. BTW I navigated so much in this old ugly code that "it is not the good old days." Sorry but I cannot let you say that. We are in the good new days! > > Would be a nice addition to recreate some of the presentation and > multimedia features known from Squeak now also in Pharo. To recreate. rebuild, redesign, of course! not to copy code! Pharo has never been against multimedia. Our goal is to have a solid base for inventing new things and to build a meta platform . I would have love to have Sophie running on Pharo. This is not by accident that we are working on - Athens - TxText - Bloc - Event touch - SDL and not just web app I remember that one person did a new desktop or something like that to build presentations. Esse https://www.youtube.com/watch?v=OVjx9KMd3i8 > > Thanks > T. > >

March 13, 2015

Re: [Pharo-users] When to use a stream for concatenating text..

by Marcus Denker

> On 13 Mar 2015, at 08:41, stepharo <stepharo(a)free.fr> wrote: > > use > > String streamContents: [:s | > s nextPutAll: 'jlklkjkl' ] > > > or > > String streamContents: [:s | > s << 'jlklkjkl' ] > > > it is a great method for manipulating > > Le 10/3/15 19:09, sergio_101 a Ã©crit : >> >> it seems that in more cases than not, i find that developers use a stream when concatenating some text strings. >> >> I am wondering if this is a smalltalk thing, or is there a real speed benefit when using streams in this way. >> When adding to a string, you create a new one with the right size and copy over the old content. Using a stream avoids that. But there is a cost for handling the stream, too: [String streamContents: [:s | s nextPutAll: 'jlklkjkl'; nextPutAll: 'jlklkjkl' ]] bench. "'1,866,774 per second'" ['jlklkjkl' , 'jlklkjkl'] bench "'5,433,931 per secondâ" So it is only faster e.g. when adding to a string in a loop or on larger strings. But it reads quite nice, too, so I tend to use it without thinking about performance. Marcus

March 13, 2015

Re: [Pharo-users] When to use a stream for concatenating text..

by stepharo

use String streamContents: [:s | s nextPutAll: 'jlklkjkl' ] or String streamContents: [:s | s << 'jlklkjkl' ] it is a great method for manipulating Le 10/3/15 19:09, sergio_101 a Ã©crit : > > it seems that in more cases than not, i find that developers use a > stream when concatenating some text strings. > > I am wondering if this is a smalltalk thing, or is there a real speed > benefit when using streams in this way. > > Thanks! >

March 13, 2015

Re: [Pharo-users] Settings Browser: Moose 5.1 exported settings raises error in Pharo 4.0

by stepharo

The problem is also when you rename a class. Having explicit class name is a problem. Damien C we wwna to extract and revisit the configuration approach of pillar to offer a nice way to support configurations. Now it is based on Ston and it will make the approach based on a little package. So for now ignore error is the best way. Stef Le 12/3/15 20:29, Juraj Kubelka a Ã©crit : > Hi, > > If I export settings from Settings Browser in Moose 5.1 and then open Pharo 4.0 image, it raises an error "GTGenericStackDebugger not found in SystemDictionaryâ. > > The problem is a setting: > (Smalltalk at: #GTGenericStackDebugger) perform: #enableStackColoring: with: (true) > > which apparently does not exists in Pharo 4.0. > > I think if we want to share settings between images, it should ignore any setting that raise an error, right? > > Thanks! > Juraj >

March 13, 2015

configuration browser descriptions

by Sebastian Heidbrink

Hi! Here is another information based on my last beginners session. One should discuss if it might be useful and easy to implement to have a configuration browser that includes descriptions for the offered configurations. Beginners have no idea what Seaside, AltBrowser, Artefact,... stand for. Cheers! Sebastian

March 13, 2015