Forum

    Discussion regarding the Eiffel programming language and various supporting tools.
  • Feb 24
    Antw: [EXT] [eiffel-users] Some thing make pass 3 of the compiler very slow for class depending of vision 2
    Hi! I have no answer for this, but questions: Does the Eiffel runtime have provisions to measure where time was spent (a tool like gprof)? If so, is there a way to let the compiler explain where it spent its time? My only idea for the problem is (as Eiffel may be resource-hungry): Do you have
  • Feb 23
    RE: [eiffel-users] Some thing make pass 3 of the compiler very slow for class depending of vision 2
    Sounds like global dataflow analysis cost when changing a class inherited by and/or used by a large number of other classes. Each dependency needs to be verified. R -------- Original Message -------- Subject: [eiffel-users] Some thing make pass 3 of the compiler very slow for class depending of
  • Feb 23
    Some thing make pass 3 of the compiler very slow for class depending of vision 2
    I just change the color value in a class which is a descendent of EV_WIDGET. Only one correction on one class ! mardi 23 février 2021, 19:15:39 (UTC+0100) Eiffel Compilation Manager Version 20.11.10.5048 - linux-x86-64 Degree 6: Examining System Degree 5: Parsing Classes Degree 4:
  • Feb 23
    Antw: [EXT] [eiffel-users] Is this look-ahead adder SCOOPable?
    numbers, > a and b: > > loop > sum [i] := a [i] + b [i] + carry > if sum [i] < a [i] then -- NATURAL wraps around past zero if > too big > carry := 1 > else > carry := 0 > end > i := i + 1 > end > > The above
  • Feb 22
    EShell like JShell or Python Shell, but for Eiffel
    Does anyone know of an Eiffel-based shell program similar to JShell? See: tryjshell.org
  • Feb 21
    Re: [eiffel-users] Is this look-ahead adder SCOOPable?
    Hey, thanks. Will look at that. But I'm still curious if about SCOOP jjj On Saturday, February 20, 2021 at 11:14:22 AM UTC-7 Finnian Reilly wrote: > > I have a class BIG_NUMBER implemented as an ARRAY [NATURAL_8] where each > array `item' represents a "digit" in base-256. To naively add two
  • Feb 21
    Re: [eiffel-users] Is this look-ahead adder SCOOPable?
    Yes, NATURAL_8 was for illustration only. "a separate region for each digit is too costly." Sure, but I just wanted to know if it were possible and how to do it, so I could better understand SCOOP. I thought perhaps a look-ahead for a segment of digits, where a normal add within the segment.
  • Feb 20
    Re: [eiffel-users] Is this look-ahead adder SCOOPable?
    I have a class BIG_NUMBER implemented as an ARRAY [NATURAL_8] where each array `item' represents a "digit" in base-256. To naively add two numbers, a and b: Maybe class INTEGER_X found in "$ISE_EIFFEL/contrib/library/math/eapml/eapml.ecf" already has what you need?
  • Feb 20
    Re: [eiffel-users] Is this look-ahead adder SCOOPable?
    Unless the numbers have thousands/millions of digits making the addition using SCOOP would not make the program faster. It would be much more efficient to use NATURAL_64 instead of NATURAL_8 instead. Also, an optimized implementation for a specific number length could make the difference, e.g.
  • Feb 20
    Re: [eiffel-users] New Video
    Hi Anders! Thank you for your kind response and feedback! Yes—it is interesting how far AI-based text-to-speech has come. I can still detect that it is a generated voice because I know what to watch out for while listening. BUT—that line is getting more and more blurry. To the unsuspecting, I
  • Feb 20
    Is this look-ahead adder SCOOPable?
    Question to the SCOOP experts out there: I have a class BIG_NUMBER implemented as an ARRAY [NATURAL_8] where each array `item' represents a "digit" in base-256. To naively add two numbers, a and b: loop sum [i] := a [i] + b [i] + carry if sum [i] < a [i] then -- NATURAL
  • Feb 19
    Re: [eiffel-users] New Video
    Hi Larry Nice video. I did not understand that it was a computer generated voice when I listened to it. Vänligen Anders Persson BSharp AB Linked in profile +46 763 17 23 25 On Fri, 19 Feb 2021 at 11:48, Larry Rix wrote: > I am using an AI-generated voice for
  • Feb 19
    Re: [eiffel-users] New Video
    I am using an AI-generated voice for the narration. I chose this particular voice because I love Australian accents and wanted a female voice rather than a male. In honesty—I chose a female voice for the same reason the US Air Force chose a female voice for their cockpit voice system because men
  • Feb 19
    Re: [eiffel-users] New Video
    Nice! Completely understandable, and a great sales job IMHO. My narrator was a British-accented speaker, I'm guessing because my Region is New Zealand? Or is that just your preferred narrator? On Fri, Feb 19, 2021 at 4:24 AM Larry Rix wrote: > https://youtu.be/b5AS
  • Feb 18
    Re: [eiffel-users] pick and drop
    LOL ... Oh my ... I totally love the "COVID-19 Negative" tag on your contact info!!! That is truly hilarious!!! On Monday, February 8, 2021 at 1:37:52 AM UTC-5 johns...@wolterworks.com wrote: > > Mr. Rix, 6:48 PM? What part of planet Earth R U located? > > > > Cheers, John S. Wolter > ---------
  • Feb 18
    Re: [eiffel-users] pick and drop
    Savannah, GA USA On Monday, February 8, 2021 at 1:37:52 AM UTC-5 johns...@wolterworks.com wrote: > > Mr. Rix, 6:48 PM? What part of planet Earth R U located? > > > > Cheers, John S. Wolter > ------------------------------------------------------------ > Wolter Works, LLC > EMail: johns...@wolter
  • Feb 18
    New Video
    https://youtu.be/b5ASMYWrhSE
  • Feb 17
    Multiplier operator missing
    Hi all. Although REAL_32 etc. have a feature like product alias "*" alias "×" (other: REAL_32): REAL_32, the operator "×" (Unicode 00D7) is not present in the dialog "Insert symbols". I've been wondering... why? Was it overlooked? Some technical reason? Does any one know? Best
  • Feb 14
    Re: [eiffel-users] Re: Base Ext2
    Dear Finnian! All of what you have stated is largely true. 15-plus years of coding effort is excellent and worthwhile. It really needs to be usable to 20.11 and beyond. Given all of this, there is a possible dove-tail. There is a consideration out there to make a pitch to possible clients where
  • Feb 14
    Re: [eiffel-users] Re: Base Ext2
    Dear Larry I have spent another hour-plus this morning trying to isolate even a small part of Eiffel-loop to where I can get it to compile in a version of Eiffel Studio beyond 16.05. I have once again failed. Sorry you experienced this frustration. It might be easier with a tool such as I have
  • Feb 14
    Re: [eiffel-users] Easing the path of migrating Eiffel-Loop to 20.11
    More thoughts Or the tool can even be made more granular and only copy one class at a time, between prompts. Or the tool can be easily made to create a tabular menu of all the classes in the current dependency layer, and lets you choose which one you want to work on next, perhaps by first
  • Feb 14
    Re: [eiffel-users] Easing the path of migrating Eiffel-Loop to 20.11
    Dear Larry my apologies again for confusing with someone called "Irix". Silly me! I have spent another hour-plus this morning trying to isolate even a small part of Eiffel-loop to where I can get it to compile in a version of Eiffel Studio beyond 16.05. I have once again failed. I have an idea!
  • Feb 14
    Re: [eiffel-users] Re: Base Ext2
    Hi Larry, All of this is really great. However, have you had any success at all in making Eiffel-loop Void-safe and SCOOP-friendly? Unfortunately not, but if enough people are interested to sponsor this gargantuan task here is my offer: I am willing to undertake the task for € 24.00 EUR/hour
  • Feb 14
    Re: Base Ext2
    I have spent another hour-plus this morning trying to isolate even a small part of Eiffel-loop to where I can get it to compile in a version of Eiffel Studio beyond 16.05. I have once again failed. I would love to use some of this lovely and well-crafted code you have created, but asking me (any
  • Feb 14
    Re: Base Ext2
    HI Finnian! All of this is really great. However, have you had any success at all in making Eiffel-loop Void-safe and SCOOP-friendly? Kindest regards, Larry
  • Feb 14
    Re: Base Ext2
    Regarding ARRAY_COMPARABLE [G -> HASHABLE] I am not sure I understand why you would want to sort things by hash code. Can you give a use case for this please. Incidentally you can achieve this with routine `order_by' from the EL_ARRAYED_LIST class in Eiffel-Loop feature -- Reorder items
  • Feb 14
    Re: Base Ext2
    Incidentally lately I have further optimized ZSTRING and the benchmarks now compare very favorably to STRING_32 with many operations performing faster or only marginally slower, depending on the characters being used. In all cases there is a hugh memory saving which is the whole raison d'etre of
  • Feb 14
    Re: Base Ext2
    Class ZSTRING in Eiffel-Loop also has some features inspired by Python. Most notably is the tuple substitution operator #$ which functions like the Python % operator for Python tuples: Eg. print "%s = \"%s\"" % ('a', 1) [image: Screenshot.png] This sample code is from the class EL_SOFTWARE_VERSION
  • Feb 13
    Re: Base Ext2
    BTW/FYI -- The {TUPLE}.arrayed and {TUPLE}.item being "detachable separate ANY" is a royal pain in the backside! It makes what ought to be simple exchanges between ARRAY and TUPLE turn into a hot-mess of conversion code (see the library).
  • Feb 13
    Base Ext2
    A library where I am capturing new array and string features based on inspiration from Python. https://github.com/ljr1981/base_ext2
  • See more ...