October 2019 - Pharo-users - lists.pharo.org

Re: [Pharo-users] Aconcagua Canonical Repo(s)

by Gabriel Cotelli

I don't know if call it canonical. But certainly it is the version the community is maintaining. It's a fork of the mtaborda repo and converted to tonel format. I don't know how it was migrated from Sthub, but it was before any tool for migrating the history existed. I think it must contain all the functionality available in Sthub. Regarding Units package we never made a thorough comparison so I cannot tell the differences. But this package is intended to be usable for things beyond standard units. We're using it for financial measurements besides the physical ones. Hope it helps in clarifying a bit the situation. On Mon, Oct 14, 2019, 14:13 Sean P. DeNigris <sean(a)clipperadams.com> wrote: > It seems that this is now canonical: https://github.com/ba-st/aconcagua > > I assume its predecessor [1] was a port from this StHub repo [2]. There are > two other repos on StHub [3] - the first of which has changes after the > last > in [1] and the other has changes which may have been/need to be merged. Can > anyone shed further light? > > 1. https://github.com/mtaborda/aconcagua > 2. http://smalltalkhub.com/#!/~maxi/Aconcagua > 3. http://smalltalkhub.com/#!/~HernanWilkinson/Aconcagua > http://smalltalkhub.com/#!/~BenoitAstruc/Aconcagua > > p.s. IMHO when porting from MCZ to GH, a good practice is to state in the > GH > README that it's a port, including the URL of the source and a way to > clarify the last commit (maybe the date could often be sufficient) > > > > ----- > Cheers, > Sean > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html > >

Oct. 14, 2019

Re: [Pharo-users] NVARCHAR and Glorp

by eftomi

Thanks, Sven. I found the "ultimate" place where strings are being prepared for INSERT INTO clause - this happens in String>>#glorpPrintSQLOn:. Not so easy to parametrize at a first glance. Maybe with WideString>>#glorpPrintSQLOn: but I'm not sure how other platforms would be happy. I decided to work on other issues right now so that I can publish a relatively decent package ASAP. An interesting observation about UTF-8 and SQL Server 2019 is here: https://sqlquantumleap.com/2018/09/28/native-utf-8-support-in-sql-server-20…. Best wishes, Tomaz -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Oct. 14, 2019

Re: [Pharo-users] How to install stripe library on Pharo 7 64bit ?

by Paul DeBruicker

Are you sure you're using a released version of Pharo 7? Running the load script in a stable released version of Pharo 7 loads with no errors for me. I'm not in a position to be able to support un-released or non-stable or outdated versions of Pharo (e.g. pharo 8). To get a stable version of Pharo 7 run this from a command line on mac or linux: curl https://get.pharo.org/64/ | bash The missing baseUrl is my fault. I override it in my own code to make nginx do the ssl termination for the https connections. I've uploaded a corrected version. IT should be StripeSystem>>#baseUrl ^'https://api.stripe.com/v1' Also the API reference docs are here: https://stripe.com/docs/api For any Stripe object I haven't created a class for the code should create an instance of a StripeObject class Hope this helps Paul Pharo Smalltalk Users mailing list wrote > OK, did some more testing and here are the results. > > > Pharo 7 > - Image hangs when loading Zinc HTTP Components. I tried loading Zinc > HTTP > Components separately through Metacello with the same results so I don't > think the problem is with your code. > > > Pharo 8 > - Stripe Loads fine, > - 22/29 Stripe tests fail (On debug it says that StripeSystem class does > not > understand #baseUrl message) > - Seaside not launched & can't right click Seaside Control Panel to start > server (Fixed by reloading Seaside3 from Metacello.) > - Stripe example apps (stripe & stripe-form) crash (Seaside Walkback > Error: > Instances of UndefinedObject are not indexable) > > > > > > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Oct. 14, 2019

Re: [Pharo-users] Aconcagua Canonical Repo(s)

by Cyril Ferlicot

What is the relation with Units maintained by Zweindenker? On Mon 14 Oct 2019 at 19:13, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > It seems that this is now canonical: https://github.com/ba-st/aconcagua > > I assume its predecessor [1] was a port from this StHub repo [2]. There are > two other repos on StHub [3] - the first of which has changes after the > last > in [1] and the other has changes which may have been/need to be merged. Can > anyone shed further light? > > 1. https://github.com/mtaborda/aconcagua > 2. http://smalltalkhub.com/#!/~maxi/Aconcagua > 3. http://smalltalkhub.com/#!/~HernanWilkinson/Aconcagua > http://smalltalkhub.com/#!/~BenoitAstruc/Aconcagua > > p.s. IMHO when porting from MCZ to GH, a good practice is to state in the > GH > README that it's a port, including the URL of the source and a way to > clarify the last commit (maybe the date could often be sufficient) > > > > ----- > Cheers, > Sean > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html > > -- Cyril Ferlicot https://ferlicot.fr

Oct. 14, 2019

Re: [Pharo-users] Aconcagua Canonical Repo(s)

by Sean P. DeNigris

To slightly further complicate matters, canonical does not quite have all commits from its predecessor. ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Oct. 14, 2019

Aconcagua Canonical Repo(s)

by Sean P. DeNigris

It seems that this is now canonical: https://github.com/ba-st/aconcagua I assume its predecessor [1] was a port from this StHub repo [2]. There are two other repos on StHub [3] - the first of which has changes after the last in [1] and the other has changes which may have been/need to be merged. Can anyone shed further light? 1. https://github.com/mtaborda/aconcagua 2. http://smalltalkhub.com/#!/~maxi/Aconcagua 3. http://smalltalkhub.com/#!/~HernanWilkinson/Aconcagua http://smalltalkhub.com/#!/~BenoitAstruc/Aconcagua p.s. IMHO when porting from MCZ to GH, a good practice is to state in the GH README that it's a port, including the URL of the source and a way to clarify the last commit (maybe the date could often be sufficient) ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Oct. 14, 2019

Re: [Pharo-users] Transcript: printString or asString

by Richard Sargent

Samuel, One thing I don't recall seeing in this thread is a discussion of the semantics of the methods names. I find it helpful to consider #asWhatever to be a conversion method, used to convert one object to another essentially compatible class. e.g. #asFloat sent to an integer would be expected to yield a floating point number with the same value, but #asInteger sent to a floating point number would lose precision but might still be sensible for some purposes. [more on this below] But, #printString and its ilk (#displayString, #storeString, others) are not intended to be conversion methods. They are to provide an external representation of a format suitable for the intended audience. Typically, (and I say that guardedly), #printString's intended audience is the programmers working on the system and #displayString is intended for the system's users. In simple cases, #printString is reversible and can be parsed to recover the value of the original object. In general, it doesn't include enough information to do so. The same is true of #displayString. If we consider a slightly more complex example than a number, consider a User Profile. Its #printString might be 'UserProfile(John Doe)'. Its #displayString might be 'John Doe'. Neither of those provide all the details of the object, although they could be parsed and so used to look up an existing object. ---------- Continuing from [more on this below] above... I find #asString is one of the most problematic methods that I have ever seen. It is deceptive, an "attractive nuisance", as one colleague liked to say. Clarity comes from being explicit in one's programming. Say what you mean and mean what you say. If you are trying to convert from one object form to another, use a conversion method whose name is precise. If you are trying to provide information for a programmer about an object, #printString is the choice.(e.g. in a stack dump). As a general rule, #asWhatever methods that are generally applicable to the set of possible values. e.g. number conversion methods are a good example, even though there are exceptions (loss of precision, NaN, and Inf are some). I have seen a lot of #asWhatever methods added to the String hierarchy, but almost all of them are dubious, in my opinion of course. Most strings are not numbers, for example. String>>#asNumber is one of the worst things I have ever seen, both from a consistency perspective and an implementation perspective. 'abc123def456' asNumber has more variant results than I can enumerate. On Mon, Oct 14, 2019 at 9:28 AM Samuel Teixeira Santos <arcanosam(a)gmail.com> wrote: > Very clear to me now. > > Thank you and to others too by your previous answers >

Oct. 14, 2019

Re: [Pharo-users] Transcript: printString or asString

by Samuel Teixeira Santos

Very clear to me now. Thank you and to others too by your previous answers

Oct. 14, 2019

Re: [Pharo-users] OSProcess and CommandShell available on GitHub for Pharo users

by Sean P. DeNigris

David T. Lewis wrote > they can now be loading in Pharo... In Pharo 7.0.4, the following chugged for about an hour cycling back and forth between OSP and Command Shell without completing: baseline: 'OSProcess' with: [ spec repository: 'github://dtlewis290/OSProcess-Tonel'; loads: #(#'Core with Output') ] I wonder if it's related to the old circular dependency problem we had which IIRC was resolved with "Pipeability"... ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Users-f1310670.html

Oct. 14, 2019

Re: [Pharo-users] Transcript: printString or asString

by Christopher Fuhrman

On Mon, Oct 14, 2019, 08:49 Samuel Teixeira Santos <arcanosam(a)gmail.com> wrote: > when you say when: "what this object converted *to a string means in my > domain"* - what really means, specially about when you say *'domain'?* > > Could you give some example for this? > > Thanks > A good example might be PLU codes [1] on produce in North America. They're 4 or 5 digits, and employees of a grocery store usually know them by heart (they're in the "employee of a grocery store" domain). Customers, however, are interested in knowing what the vegetable or fruit is, not the code. So, depending on the application domain, #asString might print the code (for grocery store employees) or the name of the item (for customers). Delving deeper, we find these codes even contain info about whether produce is GMO, organic, etc. which could also be used by #asString depending on requirements. [1]: http://innvista.com/health/foods/plu-codes-alphabetical-order/ >

Oct. 14, 2019