Skip to content

Latest commit

 

History

History
266 lines (219 loc) · 15.5 KB

CHANGELOG.md

File metadata and controls

266 lines (219 loc) · 15.5 KB

MDAPL Changelog

For the latest release of “Mastering Dyalog APL”, visit https://github.com/Dyalog/MDAPL/releases.

This changelog keeps track of two types of changes. In the first instance, it lists the changes between consecutive releases of this book. However, because this book is a rework of the first edition by Bernard Legrand, this changelog also marks with [n] content that is new in this rework.

0.7.0

  • Add chapter “Mathematical Functions”:
    • refactor “Sorting and Searching Data”:
      • [n] explain how “Grade Up/Down” sort numeric arrays of arbitrary rank;
      • [n] add subsection on how “Grade Up” inverts a permutation;
    • remove subsection “Encode & Decode > Applications > Calculating Positions in a Matrix” (rendered obsolete by high-rank “Index Of”);
    • refactor “Randomised Values”:
      • [n] say “Deal” ?0 generates decimal numbers between 0 and 1;
      • move Specialist's Subsection “Random Link” to the main section (the book uses ⎕RL a lot to generate variables, so it makes sense to describe it in the main text).
    • refactor subsection “Some More Maths”:
      • [n] list the “Circle” functions for complex numbers;
      • [n] add the identity (×≡∨×∧);
      • [n] mention “Unique” next to “Union” and “Intersection”.
    • refactor “Exercises”:
      • [n] add exercise on representing sparse matrices and using “Encode” and “Decode” (replaces old “Encode & Decode > Applications > Calculating Positions in a Matrix”).
    • refactor “The Specialist's Section”:
      • move “Random Link” out;
      • [n] write about total array ordering.

0.6.1

  • Add section “More about DISPLAY” to chapter “Nested Arrays (Continued)”.
  • Fix bad cross-reference.

0.6.0

  • [n] Add chapter “Tacit Programming”:
    • [n] add section “Function Composition”:
      • add section “Operators > Compose”;
      • [n] add “Atop” and “Over” ;
    • [n] add section “Binding”:
      • [n] add example of binding in the right operand of @, e.g. in 0@(>∘5) to replace values larger than 5;
    • add old “Operators > Commute”:
      • [n] distinguish between “Commute” and “Selfie”;
    • [n] add section “Constant” on A⍨;
    • [n] write about trains;
    • [n] add exercises.

0.5.0

  • Add chapter “Operators”:
    • prefer “operand(s)” over “argument(s)” when talking about operators (the book seldom uses “operands”);
    • refactor section on “Outer Product”:
      • remove sub-subsection “Applications” > “Dispatching Items into Categories” (obsolete with the introduction of interval index, dyadic );
      • [n] add exercises on outer product;
    • refactor section on “Inner Product”:
      • include inner product explanatory diagrams as figures (there is no good way of including the explanatory diagrams as markdown tables or MathJax tables);
      • remove obsolete example about ∧.= (a long example showing how to use ∧.= became obsolete with the extension of index of, dyadic , to higher-rank arrays – instead, this example was moved to the exercises);
      • reword sub-subsection “Is a Graph Contiguous?” (make the text more mathematically precise while keeping it light for people with no/little mathematical background);
    • refactor intermission exercises:
      • [n] add exercise about ∧.= where we try to implement dyadic on matrices with inner products, scans, and reductions;
    • remove sections “Compose” and “Commute” (moved to chapter “Tacit Programming”)
    • [n] add section “At” about @;
    • [n] add section “Rank” about ;
      • [n] explain rank in terms of frames;
      • [n] define cells;
      • [n] explain rank in terms of cells;
    • [n] add section “Key” about ;
    • [n] add section “Stencil” about ;
    • refactor section “The Specialist's Section”:
      • [n] write about rank outer products F⍤99 j⍤k 99;
      • [n] write about a length-3 right operand to rank ;
    • [n] add exercises:
      • [n] redo the exercise on cross counting with key ;
      • [n] add exercises to test new operators;
      • [n] add exercises to compute frames when using rank ;
  • [n] Write about “Choose Indexing” and “Reach Indexing” in the chapter “Nested Arrays (Continued)” (these types of indexing come in handy in later sections):
    • [n] mention these techniques at “Data and Variables > Indexing > Even More Indexing”;
  • Remove quotes from figure captions.
  • Refactor “Variables” appendix:
    • explicitly order variables by their first appearance;
    • add cross-references to usage locations;
  • Add custom admonition type for footnotes.
  • Restyle footnote in Appendix with new custom admonition (footnotes, often marked with an asterisk * in the original book, are being marked with asterisks and then written out with an admonition called a footnote).
  • Fix bug with preprocessing script (standalone headers were not getting anchor links).
  • Fix references in “Working with Data Shape”.

0.4.5

  • Moved source files into dedicated folder docs.
  • Simplified build process by removing resource migration.

0.4.4

  • Improve exercise presentation:
    • provide an admonition for exercises with auto-numbering;
    • provide an admonition for solutions;
    • link solutions to corresponding exercises;
  • Improve phrasing of several exercises and provide more concrete instructions.

0.4.3.1

  • Use local APL385 font when possible.

0.4.3

  • Stylistic improvements to online book:
    • now uses the APL385 Unicode font;
    • fixed improperly styled admonitions.
  • Unify “tip”/“advice”/“suggestion”/“recommendation”/etc admonitions as “tip”.
  • Fix NameError in preprocessing script.

0.4.2

  • Fix #24 by improving preprocessing script.
  • Improve formatting and wording in the chapter (fix #23 and #26).
  • Fix typos (fix #23).
  • Fix #21.

0.4.1

  • Fix typos.
  • Fix duplicate headers in “Some Primitive Functions”:
    • Under “Axis Specification”, there were 2 almost-consecutive duplicate level 4 headers under the same level 3 header, and it made sense to join them together;
    • Rename “Special Notations” to “Replicate with Axis” under the section on replicate, to make the header more explicit;

v0.4

  • Add chapter “Nested Arrays (Continued)”:
    • refactor “First Contact” > “More About DISPLAY” (see #13);
    • remove subsection “Depth” > “Match & Natch” (dyadic and are already covered earlier in the book);
    • [n] add subsection “Depth” > “The Depth of an Array, Take 2” (introduce default vector values technique; try fixing the issue with Process by giving default values to the missing positions of the argument vector);
    • [n] add section “Nest” (the function Process was used to introduce the concept of depth; we can use the same example to motivate monadic as a way to do some "argument validation"):
      • [n] introduce that encloses if the argument is simple;
      • [n] add subsection “Argument Homogenisation” (fixes the issue with Process by using );
      • [n] add subsection “Nesting a Scalar” (explains why it may look like "doesn't work" on simple scalars).
    • [n] establish the connection between mix with axis ↑[k] and mix followed by dyadic transpose;
    • make a section out of the intermission exercises (so that they are easier to find in the table of contents of the chapter);
    • remove section “First & Type” (The “First” portion was moved to the section “Pick” and the “Type” portion was split into “Prototype, Fill Item” and the “Specialist's Section”);
    • refactor section “Prototype, Fill Item”:
      • rename it to “Type, Prototype, Fill Item” (before we introduce the notions of prototype and fill item, we define the type of an array); and
      • [n] define the type of an array in a descriptive manner (we can't just say that the type of array is ∊array because is the enlist function).
    • refactor section “Pick”:
      • rename subsection “Beware!” to “Left Argument Length” (more descriptive name);
      • rename subsection “Important” to “Disclosed Result” (ditto);
      • take ]box effects into consideration in subsections “Left Argument Length” and “Disclosed Result”; and
      • add subsection “Pick First”, containing the “First” portion of the old “First & Type” section.
    • refactor section “Partition & Partitioned Enclose”:
      • swap title order to match order of contents;
      • [n] introduce partitioned enclose pattern ⊂ array as a series of sub-subsections covering the several nuances of its behaviour; and
      • use the old subsection “The IBM Definition” as the introduction to partition .
    • refactor section “Enlist” (remove references to ⎕ML, as those were moved to the Specialist's Section);
    • refactor exercises:
      • [n] added exercise 7 to find the result of an empty partitioned enclose, like 0⊂'Partition';
      • [n] added exercise 8 using nest ⊆Y for argument homogenisation;
    • refactor “The Specialist's Section”:
      • [n] add subsection "Computing the Type and Prototype" (after talking about ⎕ML):
        • use ⊃0⍴⊂ / with ⎕ML ← 0 to figure out the type of an array;
        • use ⊃0⍴ / ∊⊃ with ⎕ML ← 0 to figure out the prototype of an array; and
        • [n] include recursive dfn to compute type when ⎕ML ≥ 0.
      • refactor “The IBM Partition on Matrices”:
        • rename it to “High-rank Partition”;
        • [n] add remark about result similarity if ]box is OFF;
        • point out main difference between and ( creates a vector of sub-arrays, while creates a sub-array of vectors);
        • [n] add rules to characterise the result of a partition operation;
  • Include email address to where feedback/errata can be sent.
  • Update TOC format to comply with Jupyter Book's upgrade.
  • Fix styling issues with “Working on Data Shape” (a “rules” admonition and listing some matrices on the solutions for the exercises).

v0.3

  • Add chapter “Special Syntax”:
    • change modified assignment variable names to "make more sense" out of context, and include example values;
    • [n] include a note saying that array notation could supersede the usage of ,← to define long vectors;
    • [n] move the in-depth discussion about the disadvantages of not using parentheses in multiple assignment to this chapter, from the chapter “Data and Variables”;
    • [n] add enlist and table to the list of primitives that work with selective assignment;
  • Add the selective assignment appendix to the appendices.

v0.2.1

  • [n] Add example in “Data and Variables” that explains why multiple assignment without parenthesis can be ambiguous to read.
  • Do not call "mirror" to the primitives and .
  • Make text a bit less Windows-centric.
  • Format APL terms consistently throughout the book (with italics only).
  • Prefer "catenate" over "concatenate" for the primitive ,.
  • Prefer the construct "the function " over "the function" – the "function" qualifier before the function name makes it clearer, earlier, that we are talking about a specific function.
  • Fix reference issues with images that had empty captions.

v0.2.0

  • Add chapter “Working on Data Shape”:
    • [n] replace the "type" primitive ( with ⎕ML ← 0) with a reference to ⎕DR;
    • [n] updated spec of take and drop because they now conform to leading axis theory (the length of the left argument now can be smaller than the rank of the right argument);
    • [n] include exercises to write expressions that determine ⍴ns↑array and ⍴ns↓array from ns, ≢ns, and ⍴array;
    • [n] include exercises to transform the left argument to take into a left argument for drop and vice-versa, given a fixed right argument;
    • [n] include more natural example of when laminate is useful to "show vectors";
    • don't call mirror to reverse and reverse first ;
    • [n] include remark about not being a no-op for vectors, although visually one might expect that;
    • regarding the solutions for the problems:
      • move them to the end of the chapter;
    • [n] write them as dfns;
    • [n] modernise some of the solutions;
    • [n] include explanations of how to arrive at the solutions;

v0.1.1

  • Fix issues #1, #3, and #4.
  • Fix other minor issues and typos.
  • [n] Add a row about shy results on the comparative table between dfns and tradfns.
  • Acknowledge contributions to this rework in the README file.

v0.1.0

Version 0.1.0 includes a significant number of changes because substantial work on the first chapters had already been done when versioning of the MDAPL rework started.

  • Add chapter “Execute and Format Control”:

    • make original Spe 2.2 a subsection on dyadic ;
    • specification 'E' is next to all other specifications (was Spe 2.3);
    • [n] Specialist's Section "Dyadic Execute" now mentions the left argument to can be a reference to a namespace;
    • [n] section on ⎕FMT now contains a quick reference to all specifications;
    • 'Kn' listed as a qualifier for ⎕FMT instead of a specification, which is now in agreement with the documentation (cf. http://help.dyalog.com/latest/#Language/System%20Functions/Format%20Dyadic.htm#GFormat);
    • [n] mention which qualifiers work with the specification 'G';
    • [n] remark that common FORMAT ERROR is the mixture of incompatible qualifiers/specifications;
    • [n] suggest not using for weird programming tricks, instead use functions that receive arguments and return values;
    • [n] add section on controlling session output with user commands and the dfns namespace;
  • Add chapter “First-Aid Kit”:

    • update debugging instructions for Windows;
    • [n] add debugging instructions for RIDE;
    • (possibly other changes);
  • Add chapter “User-Defined Functions”:

    • reorder sections on tradfns and dfns to give more emphasis on dfns;
    • [n] add sections on dfns;
    • [n] add exercises on dfns;
    • [n] comparative table between dfns and tradfns;
    • (possibly other changes);
  • Add chapter “Some Primitive Functions”:

    • introduce match and depth () and not-match and tally ();
    • [n] mention unique that renders the ((⍳≢vector) = vector ⍳ vector) / vector example as obsolete;
    • (possibly other changes);
  • Add chapter “Data and Variables”:

    • (possibly other changes);
  • Add chaper “Getting Started”:

    • [n] add instructions for Linux and MacOS users;
    • [n] add instructions for RIDE;
    • [n] add instructions for Jupyter notebooks;
    • [n] add more information on typing APL glyphs;
    • (possibly other changes);
  • Add chapter “Will You Play APL With Me?”:

    • (possibly other changes);
  • Add chapter “Appendices”:

    • [n] add appendix with all event numbers;
    • [n] add appendix with all the variables that are commonly used in the book and too long to type;