| Did you know ... | Search Documentation: |
 | Profile for user LogicalCaptain |
- HOME
- DOWNLOAD
- DOCUMENTATION
- TUTORIALS
- Beginner▶
- Advanced▶
- Web applications▶
- Semantic web▶
- Graphics▶
- Machine learning▶
- External collections▶
- For packagers▶
- COMMUNITY
- COMMERCIAL
- WIKI
Tags by LogicalCaptain
Comments by LogicalCaptain
| Sec. A.9.7 | "Sudoku in Prolog" by Markus Triska: https://www.youtube.com/watch?v=5KUdEZTu06o ... |
| maplist/5 | Check out maplist/2 for more notes |
| Sec. A.9 | A load of examples at Hakank's (Håkan Kjellerstrand) page on SWI-Prolog: http:// ... |
| Sec. A.1 | ## 0 vs failure As expected, aggregate_all/3, which uses findall/3, yields 0 if ... |
| pack/1 | Pack list is here: https://eu.swi-prolog.org/pack/list |
| same_length/3 | See also same_length/2. A comment there contains test code for this predicate. |
| use_module/1 | Interestingly, a `:- use_module(ModuleB)` is not needed in `ModuleA` if `ModuleB ... |
| string_concat/3 | ## Your choices Your choice of predicate is as follows: **Two concatenable terms as input (can split)** - atom_concat/3 (ISO) - string_concat/3 **Two more general concatenable terms as input (cannot split because arguments 1 and 2 are too general)** - atomic_concat/3 **A list of concatenable terms as input (never split)** - atomic_list_concat/2 - generates atom at argument 2. |
| string_codes/2 | ## See also - atom_codes/2 - atom_chars/2 - string_chars/2 and also - atom_strin ... |
| is_stream/1 | These all succeed: ``` is_stream(user_input). is_stream(user_output). is_stream( ... |
| close/1 | "If Stream is not open, an existence error is raised." So closing a stream is no ... |
| Sec. 9.1 | An directly runnable example from [Stack Overflow](https://stackoverflow.com/que ... |
| Sec. 9.2.1 | The syntax from TZom Frühwirth's book (p. |
| Sec. 2.2.4 | Note the two ways to do this via a declaration in the =|test/2|= options list: * ... |
| round/1 | Funnily, you can round infinity! ``` ?- X is round(-1.0Inf). X = -1.0Inf. ?- X i ... |
| Sec. 2.15.1.7 | ## Doc needs help The regular expression =|[+-]?\sd+[.]\sd+Inf|= just can't be c ... |
| atomic_concat/3 | This nearly the same as atom_concat/3 but it doesn't generate substrings from th ... |
| char_type/2 | Bug: What is both in the class "white" and "space"? ``` ?- bagof(Char-Hex, Code^ ... |
| hex_bytes/2 | If you want to print a nmuber in hexadecimal though, just ise format/2: 0-padded 4 nibbles: ``` ?- Code=6464, format(atom(T),"0x~|~`0t~16r~4+",[Code]). Code = 6464, T = '0x1940'. ``` |
| string_lower/2 | See also downcase_atom/2 for the corresponding predicate yielding atoms on argum ... |
| string_upper/2 | See also upcase_atom/2 for the corresponding predicate yielding atoms on argumen ... |
| upcase_atom/2 | ``` ?- upcase_atom('aAbB12',X). X = 'AABB12'. ?- upcase_atom("aAbB12",X). X = 'A ... |
| downcase_atom/2 | ``` ?- downcase_atom('aAbB12',X). X = aabb12. ?- downcase_atom("aAbB12",X). X = ... |
| term_variables/2 | Of course, we are talking about term *as it is at call time*, not *as it syntact ... |
| subsumes_term/2 | Some examples: ``` ?- subsumes_term(f(_,_),f(1,2)). |
| maybe/0 | When you fail, try, then try again: ``` ?- repeat,format("again!~n"),maybe,forma ... |
| atomics_to_string/2 | ## Your choices Your choice of predicate is as follows: **Two concatenable terms as input (can split)** - atom_concat/3 (ISO) - string_concat/3 **Two more general concatenable terms as input (cannot split because arguments 1 and 2 are too general)** - atomic_concat/3 **A list of concatenable terms as input (never split)** - atomic_list_concat/2 - generates atom at argument 2. |
| format_time/3 | ## Doc needs help We read: > The table below gives some format strings for popul ... |
| nb_setval/2 | ## For example Setting, then getting a list of fresh variables. |
| goal_expansion/2 | Of course, goal_expansion/2 is itself Prolog code. Is it compiled first, without a prior application goal_expansion/2? |
| transpose/2 | This should probably be found in =|library(lists)|= Program above, with comments ... |
| Sec. 3.6 | See also: https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Access-Contr ... |
| thread_self/1 | Just my thread: ``` ?- thread_self(Me). Me = main. ``` Another thread without al ... |
| http_handler/3 | I suppose the handlers are visible to all web-server threads that are running in the same SWIPL instance. |
| comma_list/2 | ## Example ``` ?- comma_list((a,b,c,d),X). X = [a, b, c, d]. ?- comma_list((a,b, ... |
| thread_statistics/2 | For example, using a dict prettyprinter: ``` ?- thread_self(Me),thread_statistic ... |
| if/1 | This is like a flexible version of C's =|#ifdef|= ... An example from file =|sta ... |
| map_list_to_pairs/3 | So, if one considers pairs to as =|Key-Value|=, then _Function_ is a predicate o ... |
| Sec. 7 | Inline code can also be created by using the markdown-style "triple backticks" ( ... |
| time/1 | See comments at time/1 |
| statistics/0 | See comments at statistics/0 |
| statistics/2 | ## See also The note under [Obtaining Runtime Statistics](https://eu.swi-prolog. ... |
| Sec. A.52 | There is some functionality for statistics which is hard to find in the manual. ... |
| Sec. 4.42 | The gprof paper can be found here: [Gprof: A call graph execution profiler](http ... |
| ## Doc needs help The TIPC links are outdated. | |
| Sec. 3.1.4 | What is the link to set_prolog_flag/2 ? Example: ``` ?- rdf_default_graph(G). G ... |
| Sec. 3.14.4 | This is old. I last heard from inetd in the early 00's. After that, we got https ... |
| ## A modifiable version of the "PLTP state diagram" Can be found here - [as SVG] ... | |
| term_to_json/3 | See also the description at: https://eu.swi-prolog.org/pldoc/man?section=prolog- ... |
| pengine_create/1 | This creates a "local" pengine if =|option(Server)|= is missing, otherwise a "re ... |
| encoding/1 | Compare with XML's entry line, the "XML Declaration", encoded itself in UTF-8 or UTF-16 (only used once, which should really be enough; does one really want to switch to another encoding in the middle of a file, which demands appropriate editor support?). |
| Sec. 6 | ## Problems with the client example Evidently, =|close(S)|= should be =|close(St ... |
| Sec. 12.2 | See notes under https://eu.swi-prolog.org/pldoc/man?section=shlib |
| PL_term_type() | The "type of the term" is embedded in the term_t structure. In file =|pl-fli.c|= ... |
| PL_succeed() | These are not C functions but macros. In =|SWI-Prolog.h|=, we find: ``` #ifndef ... |
| PL_register_foreign() | Declared in SWI-Prolog.h as follows: ``` typedef foreign_t (*pl_function_t)(); P ... |
| Sec. 12.4.4.7 | See the note at https://eu.swi-prolog.org/pldoc/man?section=foreign-term-analysi ... |
| Sec. 12.4.4 | We read: > CVT_EXCEPTION: If conversion fails due to a type error, raise a Prolo ... |
| Sec. 12.4.5 | ## Mode indicators? It's a bit unusual to see the C parameter names tagged with ... |
| Sec. 12.4.10 | An example from the JPL Java-Prolog bridge: extracts from file ``` packages/jpl/ ... |
| PL_functor_arity() | As documented, this one returns `size_t` but the declaration in SWI-Prolog.h is ... |
| PL_new_functor() | There is also ``` PL_EXPORT(functor_t) PL_new_functor_sz(atom_t f, size_t a); `` ... |
| PL_atom_wchars() | What does it mean for an atom to be marked as "text" (are there others?) |
| PL_new_atom() | Something went wrong in formatting: ``` atom_t ATOM_nil(A) tomic constant that r ... |
| Sec. 12.4.2.1 | References: [PL_unify_atom_chars](https://eu.swi-prolog.org/pldoc/doc_for?object ... |
| min_member/2 | See also - max_member/2 (with comments) - min_list/2 - max_list/2 (with comments ... |
| ## Doc needs help ### Link to "Prolog Development Tool - PDT" Is actually https: ... | |
| use_foreign_library/1 | ## Putting libraries on swipl's equivalent of LD_LIBRARY_PATH The "foreign/1" ta ... |
| read_line_to_string/2 | ## See also - read_line_to_codes/2 - read_line_to_codes/3 ## Example Can be used to ask the user as a proper replacement for read/1 ``` ask_user :- format("You must answer 'yes' to accept this offer one cannot refuse!~n"), read_line_to_string(user_input,S1), string_lower(S1,S2), ( member(S2,["yes","1","ok","y","ja","oui"]) -> format("You may proceed, good citizen.~n") ; format("I am sorry to hear that.~n"), fail ). |
| peek_code/2 | These are blocking. |
| fill_buffer/1 | On the command line, fill_buffer/1 waits for a terminating newline: ``` ?- sleep ... |
| get_single_char/1 | ## Blocking read This predicate blocks waiting for input and returns -1 on EOF ( ... |
| print_message/2 | ## Uses stderr > The predicate print_message/2 is used by the system and librari ... |
| uri_file_name/2 | ## Examples Doesn't like =|http://|= URLs: ``` ?- uri_file_name('http://www.exam ... |
| uri_components/2 | - The original RFC 3986 "Uniform Resource Identifier (URI): Generic Syntax" htt ... |
| Sec. 15.3 | See also text & comments at https://eu.swi-prolog.org/howto/Pack.html |
| pack_list_installed/0 | This is an "interactive command" in that: - It always succeeds even if there are no installed packages - If there are no installed packages, an _informational_ message is printed to STDOUT using print_message/2 - If there are installed packages, information about them is printed to STDOUT using pack_info. |
| wildcard_match/2 | See also the interface to the PCRE library: [library(pcre)](https://eu.swi-prolo ... |
| when/2 | For an example see [\+](https://eu.swi-prolog.org/pldoc/doc_for?object=%28%5C%2B ... |
| See also the main section in the manual: [Database](https://eu.swi-prolog.org/pl ... | |
| file_directory_name/2 | ## See also file_base_name/2 ## Doc needs help The first arg should be named the same as for file_base_name/2: - file_base_name(+Path, ...) Hence - file_directory_name(+Path, -Directory) |
| file_base_name/2 | ## Doc needs help The second arg should be named BaseName analogous to - file_di ... |
| help/1 | This should probably ALSO accept a predicate call as in ``` help(file_base_name( ... |
| file_name_extension/3 | ## Arg naming "Base" should be called "Core" As in: "Name = CoreName . |
| process_create/3 | ## See also: Take a look at this pack: https://www.swi-prolog.org/pack/list?p=by ... |
| absolute_file_name/3 | This is a bit of a mixture between fish and fowl Nothing on the path needs to ex ... |
| pack_install/2 | ## Bug The input buffer should be emptied after the first answer ``` Verify pack ... |
| process_create/3 | ## See also - shell/2 - [process_create∕3](https://eu.swi-prolog.org/pldoc/doc_f ... |
| -- ## See also: Take a look at this pack: https://www.swi-prolog.org/pack/list?p=by ... | |
| system/2 | See shell/2 |
| system/1 | See shell/2 |
| system/0 | See shell/2 |
| system/1 | See also: shell/2 |
| integer/1 | Do not confuse with the type test predicate integer/1 |
| limit/2 | See also once/1 =|once(Goal)|= should be the same as =|limit(1,Goal)|= |
| last/2 | ## Unsurprising behaviour for proper lists The behaviour is "semidet": The predi ... |
| integer//1 | There is confusing ULR aliasing going on. I'm putting remarks into the comment s ... |
| digits//1 | There is confusing ULR aliasing going on. I'm putting remarks into the comment s ... |
| digit//1 | There is confusing ULR aliasing going on. I'm putting remarks into the comment s ... |
| integer//1 | There is confusing ULR aliasing going on. I'm putting remarks into the comment s ... |
| digits//1 | There is confusing ULR aliasing going on. I'm putting remarks into the comment s ... |
| digit//1 | ## "digits" is greedy and does not backtrack Consider this grammar to preferabl ... |
| code_type/2 | Example: Taking the code for =|0|= : ``` ?- code_type(0'0,X). X = alnum ; X = cs ... |
| directory_file_path/3 behaves badly Examples for mode `(+Directory,+File,-Path)` ... | |
| setup_call_catcher_cleanup/4 | See also: - catch/3 - catch_with_backtrace/3 |
| Sec. 3 | See also these for related libraries in other languages: - https://perldoc.perl. ... |
| exists_file/1 | An example: We have a directory, an actual file, and a symlink to a file and to ... |
| directory_file_path/3 | This is of course a purely _syntactic operation_ on atoms (not strings, although strings are accepted). |
| Sec. 2.4 | ## Doc needs help **What is "home directory"?** It is the directory on which =|s ... |
| write_length/3 | Nonground arguments will give arbitrary length: ``` ?- X=foo(bar(baz(_))),write_ ... |
| write_term/3 | Sadly, the Stream must be a stream id: ``` ?- write_term(string(S),[a,b,c],[]). ... |
| Sec. 15 | ## Doc needs help It is not clear what "attaching packs" means and who/what perf ... |
| ## Naming of file Quite important indeed. | |
| Sec. 7 | Note that a tabled predicate may well **change the order of solutions** if used ... |
| Sec. 6.14 | ## Doc needs help 1 The maplist example is typoed ("G" vs "Goal" and calls modul ... |
| multifile/1 | Also see this Logtalk blogpost: [Multifile predicates: dos and don'ts](https://l ... |
| Sec. 6.3 | ## Wrong name, really. use_module/1 is the wrong name for this functionality - i ... |
| current_module/1 | ## Example ``` ?- setof(M,current_module(M),Modules). Modules = [ansi_term,apply ... |
| predicate_name/2 | Example: ``` ?- predicate_name(is_ordset(_),S). S = "is_ordset/1". ``` When is t ... |
| read/1 | ## Note This corresponds to "serialization/deserialization" in Java and other la ... |
| Sec. 5.3.1 | The first footnote is important and extensive should be moved into the text. |
| Sec. 4.25 | ## See also [Operators and modules](https://eu.swi-prolog.org/pldoc/man?section= ... |
| make/0 | This needs an "automake" which performs whatever make/0 does automatically whene ... |
| **Pack or Package?** "Pack" and "Package" seem to be used alternatingly. | |
| list_undefined/0 | This predicate (correctly) does not complain for cases like these: What is insid ... |
| And also: ``` Technical prose is almost immortal. — Frederick P. | |
| sort/4 | Does this predicate come from ECLIPSE? In the ECLIPSE manual: [sort(+Key, +Order ... |
| atomics_to_string/3 | ## Your choices Your choice of predicate is as follows: **Two concatenable terms as input (can split)** - atom_concat/3 (ISO) - string_concat/3 **Two more general concatenable terms as input (cannot split because arguments 1 and 2 are too general)** - atomic_concat/3 **A list of concatenable terms as input (never split)** - atomic_list_concat/2 - generates atom at argument 2. |
| atom_split/3 | Use atomics_to_string/3 to join again. |
| random_member/2 | ## Doc needs help Typo: > Fails of List is the empty list. |
| pairs_keys_values/3 | ## Doc needs help Bad wording: > free or finite lists should be > open or proper lists ("open" including being an unbound variable) |
| Sec. 5.4.2 | See comments at https://eu.swi-prolog.org/pldoc/man?section=bidicts |
| load_test_files/1 | See the comments under the entry for `load_test_files` under "packages" instead ... |
| dict_keys/2 | ## Note that This predicate behaves like findall/3 (it "relates to" the empty di ... |
| clumped/2 | ## Missed opportunity This doesn't run in reverse, doing run-length decoding!?! ... |
| foldl/4 | ## Doc needs help The example says: > the last two arguments [of foldl/4-7] form a typical difference pair That's definitely not right. |
| must_be/2 | ## See also "Type tests in Prolog" by Markus Triska https://youtu.be/ZIv0G4b1xBQ ... |
| Sec. A.48 | ## Note that This implementation seems to be **replaceable** by the built-in [SW ... |
| Sec. A.4 | assoc_to_list/2 doesn't take kindly to a mix-up in arguments: ``` ?- empty_assoc ... |
| call_dcg/3 | ## Doc needs help The second paragraph is about this (I think): call_dcg/3 exist ... |
| nonground/2 | Grabbing an unbound variable from the term that appears as argument 1: **Standar ... |
| Sec. A.47 | Here is code to query the user with several questions and put the result into a ... |
| max_member/2 | ## Unify, maybe? Confusingly, the argument order is ``` max_member(Max,List) ``` whereas for max_list/2 it is reversed: ``` max_list(List,Max) ``` This hurts my head. |
| not/1 | Note that not/1 is not declared as an operator: == ?- write_canonical(not true). ERROR: Syntax error: Operator expected == whereas \+ is, allowing nicer syntax: == ?- write_canonical(\+ true). \+(true) true. == |
| \/1 | The ISO Standard says on page 124: > '\\'(B1) evaluates the expression B1 with v ... |
| **/2 | ## Naming This operator is called "power" in the ISO standard of 1995, page 119. ## More on operators: As Wouter Beek says: > Notice that the precedence of (**)/2 is xfx, > which is different from that of (^)/2, which is xfy. These operators also have the same precedence, 200 (pretty strong), same as the precedence of unary `-`. |
| ;/2 | ## Coding guidelines From Covington et al., [Coding Guidelines for Prolog](https ... |
| ./2 | Does this still exist? |
| -/1 | Note that there is a parse difference between `-1` and `-(1)` aka `- 1`: ``` ?- ... |
| Sec. 7.2.2 | What is the difference between the "markdown backtick" and the "doxygen backtick ... |
| -- ## Docs neecs some help maybe? The text > It acts as if defined by the following fact `=(Term, Term).` does not make sense to me. | |
| min_list/2 | See comments at max_list/2 |
| op/3 | ## Some examples and an explainer [**About Operators**](https://github.com/dtonh ... |
| permutation/2 | To (deterministically) select any permutation "at random" use random_permutation ... |
| random_permutation/2 | The predicate is indeed symmetric: ``` ?- random_permutation([1,2,3,4],L). L = [ ... |
| merge_options/3 | ## This one is buggy ``` ?- merge_options( [max_depth(100),quoted(true),portray( ... |
| random/1 | See also comments under https://eu.swi-prolog.org/pldoc/doc_for?object=f(random/ ... |
| random/3 | See also comments under https://eu.swi-prolog.org/pldoc/doc_for?object=f(random/ ... |
| random_float/0 | See also comments under https://eu.swi-prolog.org/pldoc/doc_for?object=f(random/ ... |
| crypto_n_random_bytes/2 | The appropriate OpenSSL man page for that would be (I think), "man RAND_bytes": ... |
| option/2 | You can use dicts as =|OptionList|=! |
| call_with_depth_limit/3 | ## Examples true/0 needs depth 1: ``` ?- call_with_depth_limit(true,0,R). R = de ... |
| \+/1 | This operator is a hard nut. I have some notes on the Prolog `\+` here: - [**Not ... |
| sum_list/2 | As per a remark made on stack overflow, this predicate's name - expresses what i ... |
| arg/3 | Also raises type_error if Arg is not compound: ``` ?- arg(1,foo,X). ERROR: Type ... |
| duplicate_term/2 | New empty cells (fresh variables) are created for any unbound variable in the or ... |
| order_by/2 | The description is mangled... > "Spec is a list of terms, where each element is ... |
| compare/3 | See https://eu.swi-prolog.org/pldoc/man?section=standardorder for the "Standard ... |
| Documentation | The "paperback version" if for "SWI Prolog Reference Manual 6.2.2" That's too ol ... |
| read_string/5 | == ?- open_string("Hello@",Input), read_string(Input, ",@", "\t ", End, String) ... |
| text_to_string/2 | This behaves the same as atom_string/2, but only in "string direction" and doesn ... |
| nb_setarg/3 | A problem from Stack Overflow: [Unexpected result for predicate nb_setarg/3](htt ... |
| Sec. D | ## Some corrections "A Prolog variable is a value that‘is not yet bound'." That ... |
| Sec. 2.10.1 | ## Needs correction > "which models each predicate in a Prolog program as a stat ... |
| Sec. 11 | ## For Pengines, see The section on [Pengines](https://eu.swi-prolog.org/pldoc/d ... |
| This is the https://opensource.org/licenses/BSD-2-Clause with one slight differe ... | |
| numbervars/3 | ## documentation needs help The output has changed: ``` ?- Term = f(X,Y,X), numb ... |
| numbervars/3 | ## Recap to fight confusion! **numbervars** Rewrites the term. |
| write_canonical/1 | ## documentation needs help We read: "Equivalent to write_term/2 using the optio ... |
| Sec. 2.9 | ``` ?- $X. ``` seems to ... evaluate to the value of the previous `X`? |
| tmp_file_stream/3 | ## Watch out Getting an == Arguments are not sufficiently instantiated == error? Note that there are two tmp_file_stream/3 predicates which take parameters with different semantics! |
| tmp_file/2 | This predicate **does not really create a file**, it just creates a unique _name ... |
| Sec. 4.1 | ## Mode indicator cheat sheet / overview [Get it from here](https://github.com/d ... |
| sort/2 | ## So many sort predicates! I have gone through the available "sort predicates" ... |
| Sec. 4.14.1.1 | This is pretty hot but maybe more adapted to CHR, which is based on a mutable da ... |
| >>/2 | Here is some functionality to transform integers into lists of 0,1 and the conve ... |
| >>/2 | The underlying bit pattern seems to be 2-complement with an infinite number of b ... |
| Sec. 4.9 | The thought arises that Prolog already _has_ continuations but they are not firs ... |
| reset/3 | ## No matching `Ball` If the argument of shift/1 _cannot_ be unified with any `B ... |
| Sec. 4.8 | In this context, the following, though rather old, is highly readable: **Higher- ... |
| !/0 | ## Doc-needs-help The expression ``` t4 :- \+(a, !, fail). ``` is invalid. |
| Sec. 4.33 | Of course, **all** variables of Prolog are **essentially global** (sometimes thr ... |
| call_with_inference_limit/3 | Try exceeding the limit: == ?- call_with_inference_limit( bagof(X,between(1,inf, ... |
| *->/2 | ## In other words... Compared to `->/2` this construct backtracks over the [prem ... |
| option/3 | ## You can use dicts! |
| Sec. A.30 | Note in particular that you cannot backtrack over the list of options via option ... |
| Sec. 6.1 | ## Modules are objects As noted by Paulo Moura "Prolog Modules" can be seen as " ... |
| partition/4 | A partition_freely/4 predicate which partitions into arbitrarily many partitions by accepting a "partitioning closure" as first argument, returning the result as an SWI-Prolog dict where the dict's keys are the partition keys: - [`partition_freely.pl`](https://github.com/dtonhofer/prolog_notes/blob/master/code/heavycarbon/utils/partition_freely.pl) - [`partition_freely.plt`](https://github.com/dtonhofer/prolog_notes/blob/master/code/heavycarbon/utils/partition_freely.plt) Example: Define a partitioning predicate taking 3 arguments == partition_by_length_modulo_m(Modulo,Atom,Key) :- atom_length(Atom,Length), Key is Length mod Modulo. == A "partitioning closure" might then be for exmaple `partition_by_length_modulo_m(5)`: a partially filled in predicate call. Then: == ?- partition_freely( partition_by_length_modulo_m(5), [silent,puffy,left,damaged,fascinated,deafening, wistful,whip,nest,inquisitive,imperfect,jog,unwieldy,provide,locket,reign], result_tag, DictOut). DictOut = result_tag{ 0:[puffy, fascinated, reign], 1:[silent, inquisitive, locket], 2:[damaged, wistful, provide], 3:[jog, unwieldy], 4:[left, deafening, whip, nest, imperfect]}. == |
| This page does not list all the available howtos tht can presumably be found in ... | |
| Sec. 4.14.1 | See also the howto page: [How to deal with the Prolog dynamic database?](https:/ ... |
| More on GMP at [gmplib.org](https://gmplib.org/) and the [Wikipedia GMP entry](h ... | |
| Sec. 4.17.1 | ## Doc needs help "Initially, these are bound to the same stream as user_input a ... |
| atom_length/2 | ## See also **string_length/2** for the corresponding predicate for terms that a ... |
| normalize_space/2 | As Jan says in the hierarchically-above page (which took me some time to click o ... |
| atom_number/2 | See also number_string/2 |
| number_string/2 | The corresponding "atomikky" predicate, however, is called atom_number/2 Note th ... |
| crypto_n_random_bytes/2 | The appropriate man page would be "man RAND_bytes" https://www.openssl.org/docs/ ... |
| div/2 | See also divmod/4 (_not_ visible in the function in the header list because it i ... |
| Sec. A.44 | ## File under "docs need help" _"SyntaxName is the principal functor of the quas ... |
| with_output_to_chars/2 | Note that, in spite of the name, the output is a list of (unicode) code points, ... |
| Sec. A.53 | ## Doc needs help. The given example has a syntax error - the final parenthesis ... |
| writeq/1 | Escape sequences are inserted into the ouput as appropriate: ``` ?- writeq('hel\ ... |
| Sec. 5.3 | _There does not seem to be a good place to put this remark:_ Multi-line strings ... |
| var/1 | ## That naming is confusing When you ask == var(X). == you are actually asking w ... |
| Sec. 4.17.5 | IRI is an URI, but internationalized: - https://en.wikipedia.org/wiki/Internatio ... |
| abolish/1 | We read: "all clauses of a predicate with functor Functor and arity Arity" Howev ... |
| PL_set_engine() | In jpl.c there is a call == PL_set_engine(NULL, NULL); == What does that do? |
| Sec. 12.8.2 | "Portable applications" should probably be "consistent/reliable applications", b ... |
| Sec. 12.4.19 | The correct format from `SWI-Prolog.h` is ``` int PL_warning(const char *fmt, .. ... |
| Sec. 12.8.1 | For `PL_check_data` and `PL_check_stacks` the formulation is incomplete: I suppo ... |
| Sec. 12.3.1 | "Creates a new term reference to the same term as the argument. |
| Sec. 12 | ## For a C++ interface Take a look at this package: [A C++ interface to SWI-Prol ... |
| Sec. 12.4.18 | Why can one call `PL_set_prolog_flag` for `PL_BOOL`, `PL_ATOM`, `PL_INTEGER`, bu ... |
| Sec. 12.4.17 | The relevant man page for POSIX OS-level signals is - man 7 signal - man 2 sigac ... |
| Sec. 12.4.12 | In predicate_t PL_predicate(const char *name, int arity, const char* module) how is the encoding of "name" and "module" specified (is it always UTF-8?) (nit pick: note the inconsistent sidedness of *) In int PL_call(term_t t, module_t m) the link to "figure 8" is dead. |
| Sec. 12.4.13 | We read: "See figure 8 for an example." But the link for "figure 8" (i.e. |
| getenv/2 | Background reading: https://en.wikipedia.org/wiki/Environment_variable On Linux, the environment variables, as attributes of the process, can be manually inspected by reading file `/proc/$PID/environ` like a normal text file. |
| read_stream_to_codes/3 | "creates a difference-list" implies: - creates an open list `Codes` ending in `T ... |
| expand_file_name/2 | ## References Note that this is based on "csh" filename expansion. |
| open/4 | ## Resolve the home directory Note that the "~" notation (used on Unix to indica ... |
| Sec. 11 | There is also a pack for UUIDs: https://eu.swi-prolog.org/pack/list?p=pluuid Exp ... |
| ord_intersection/2 | The first argument is not the/a [PowerSet](https://en.wikipedia.org/wiki/Power_s ... |
| Sec. 8.1 | ## Explainer & Rewritten code For an explainer and rewritten code, go [**here**] ... |
| list_to_ord_set/2 | This predicate should probably be named list_to_ordset/2 instead of list_to_ord_ ... |
| Sec. 8.1.1 | Note that you can "put" to an unbound variable: == ?- put_attr(X,module,Y),get_a ... |
| attr_unify_hook/2 | "Triska 2016" is right here: https://www.metalevel.at/swiclpb.pdf "The Boolean C ... |
| put_attr/3 | On the toplevel, the predicate prints itself. |
| -- ## See also: The distribution comes with examples Under =|swiplexe_8.3.22/lib/sw ... | |
| Sec. 4.10 | ## My notes Personal notes on exceptions handling: - [About Exceptions](https:// ... |
| split_string/4 | ## Trim/Strip As said in the examples, this predicate can be used to _trim_ stri ... |
| Reference manual | # Documentation orientation ## Sometimes the documentation website is confusing. Main elements: - [*The SWI-Prolog reference manual*](https://eu.swi-prolog.org/pldoc/doc_for?object=manual), with a subsection: - [*The SWI-Prolog library*](https://eu.swi-prolog.org/pldoc/man?section=libpl) (can also be found via the menu system) - There are more at but they don't appear in the content list (only the libraries that appear in the main doc do so). |
| For some reason this page shows up in no "list of contents", one finds it throug ... | |
| statistics/0 | ## See also The note under [Obtaining Runtime Statistics](https://eu.swi-prolog. ... |
| time/1 | ## See also The note under [Obtaining Runtime Statistics](https://eu.swi-prolog. ... |
| integer/1 | This doesn't _really_ belong here but: A special way to generate integers is by ... |
| string_chars/2 | ## See also - atom_codes/2 - atom_chars/2 - string_codes/2 and also - atom_strin ... |
| atom_concat/3 | ## Your choices Your choice of predicate is as follows: **Two concatenable terms as input (can split)** - atom_concat/3 (ISO) - string_concat/3 **Two more general concatenable terms as input (cannot split because arguments 1 and 2 are too general)** - atomic_concat/3 **A list of concatenable terms as input (never split)** - atomic_list_concat/2 - generates atom at argument 2. |
| shell/2 | ## Doc needs help > Please note that the CreateProcess() API does not imply the ... |
| jpl_class_to_classname/2 | == :- begin_tests(jpl_class_to_classname). instance_ref_to_entity_name(InstanceR ... |
| Sec. 10 | See also: - https://en.wikipedia.org/wiki/POSIX_Threads - https://computing.llnl ... |
| threads/0 | Example: == ?- threads. % Thread Status Time Stack use allocated % ------------- ... |
| Sec. 2.2.5 | The latest output for the test given is: == ?- run_tests. % PL-Unit: test ERROR: /home/user/tests.pl:3: test a: assertion failed Assertion: float(8) ERROR: /home/user/tests.pl:3: test a: assertion failed Assertion: 8==9 A done % 2 assertions failed % 1 test failed % 0 tests passed false. == |
| acyclic_term/1 | See the notes under cyclic_term/1 Note that acyclic_term/1 is ISO, but cyclic_te ... |
| get_dict/5 | This predicate is misnamed. |
| Sec. 4.10.4 | ## ISO standard exceptions The ISO standard ISO/IEC 13211-1 is VERY paywalled (I won't say anything more about the reasonableness/ethics/sustainability of this), so: - The following page substantially _is_ the ISO Standard text: https://www.deransart.fr/prolog/exceptions.html - Ulrich Neumerkel lists the various error classes here: http://www.complang.tuwien.ac.at/ulrich/iso-prolog/error_k - The context is a discussion leading up to the second corrigendum of the ISO standard. ## More info I'm trying to keep [**this opinionated page**](https://github.com/dtonhofer/prolog_notes/blob/master/swipl_notes/about_exceptions/README.md) updated with info on Prolog exceptions. ## About the backtrace How do we get a backtrace and how does the exception term have to look to get one? The backtrace is filled in according to SWI-Prolog conventions because the ISO Standard has nothing to say about this. SWI-Prolog wants the second argument of the `error/2` term (given in the ISO standard as `Imp_def`) to look like `context(Location,Message)`. |
| type_error/2 | > Suppose an argument must be a non-negative integer. |
| -- ## Something for the toolbox: What is a character? == character(A) :- atom(A),at ... | |
| is_list/1 | ## The mode indicator seems wrong The [mode indicator](https://eu.swi-prolog.org ... |
| reverse/2 | ## The implementation (Click on the :- button in the doc) == reverse(Xs, Ys) :- ... |
| Sec. 4.21 | [**This page**](https://github.com/dtonhofer/prolog_notes/tree/master/swipl_note ... |
| setarg/3 | The specification should probably be == setarg(+Arg, !Term, +Value) == Position ... |
| Sec. 2.2 | The example `customize/init.pl` file is in `lib/swipl/customize/init.pl` in the ... |
| load_files/2 | ## Load from where? "The specification for a source file is handed to absolute_f ... |
| Sec. 6.5 | Noet that this works for complex goals too. For example, in one module: == :- mo ... |
| load_test_files/1 | ## Doc needs help What are the options? |
| string_length/2 | For "anytext", see [this page](https://eu.swi-prolog.org/pldoc/man?section=strin ... |
| with_output_to/2 | ## If the goal emitting output fails... Evidently, if the goal fails, and you ou ... |
| run_tests/0 | This predicate succeeds if all the tests succeed and fails otherwise (i.e. |
| Sec. 4.39 | ## "The Byrd Box Model" aka. |
| atomic_list_concat/3 | ## Your choices Your choice of predicate is as follows: **Two concatenable terms as input (can split)** - atom_concat/3 (ISO) - string_concat/3 **Two more general concatenable terms as input (cannot split because arguments 1 and 2 are too general)** - atomic_concat/3 **A list of concatenable terms as input (never split)** - atomic_list_concat/2 - generates atom at argument 2. |
| ground/1 | Notice that the anonymous variable stays unground as every new occurrence is rea ... |
| Sec. 7.6 | Here is an excellent explanatory paper for Well-Founded Models by Teodor Przymus ... |
| Sec. 8 | "In the current system, test units are compiled into sub-modules of the module i ... |
| get_dict_ex/3 | This predicate no longer exists (and this page will be removed in a future updat ... |
| debug/1 | Please read important note on **debug topics** in the comment section of **[debu ... |
| Sec. A.14 | ## An important note on topics (I will add this to the manual text later): Jan w ... |
| Sec. 5.4.1.2 | ## Documentation improvements - The parameter for .put/1 should be called "NewKe ... |
| Sec. 5.4.1 | See comments at https://eu.swi-prolog.org/pldoc/man?section=bidicts ## Doc needs help The `New` in `put(+New)`should really be called `NewKeyValuePairs`. I am also recommeding a proper notation for these **method calls** or **dot calls** (which are called **function calls** here, but they really are not, they are method calls): precede the method name with a dot: - `.get/1` (throws if key does not exist) - `.put/1` (evaluates to a new dict with additions/replacements) - `.put/2` (evaluates to a new dict with single addition/replacement) Or maybe - `get./1` (throws if key does not exist) - `put./1` (evaluates to a new dict with additions/replacements) - `put./2` (evaluates to a new dict with single addition/replacement) And pldoc generation would need an update then. ## Some examples using `plunit` [`dict_put_dotcall_testing.pl`](https://github.com/dtonhofer/prolog_notes/blob/master/swipl_notes/about_dicts/code/dict_put_dotcall_testing.pl) |
| dict_create/3 | The difference between dict_create/3 and dict_pairs/3 is: - dict_create/3 takes ... |
| -- Special case: When you create a module to host dict method calls, you don't list those in the "public list" section: Here, just the predicate is listed: == :- module(point, [point_length/2]). M.multiply(F) % Method ".multiply/1" on dict M := point{x:X, y:Y} % Returns a "point" dict :- X is M.x*F, Y is M.y*F. | |
| list_debug_topics/0 | This is a user-oriented predicate: == ?- list_debug_topics. -------------------- ... |
| :</2 | ## Naming IMHO, "select" is not the appropriate name here, the more so because " ... |
| sub_string/5 | Time for test cases! == :- begin_tests(sub_string). test(find_single_bravo,[true ... |
| min/2 | ## See also The predicates min_list/2 and max_list/2 from library [library(lists ... |
| forall/2 | In particular: Àny binding is rolled back on success because it happens behind t ... |
| is_dict/2 | A random atom is not a dict == ?- is_dict(atom,X). false. == A dict tagged "foo" is tagged "foo" indeed == ?- is_dict(foo{},foo). true. == Extracting the tag of a dict == ?- is_dict(foo{},X). X = foo. == An "anonymous dict" is a dict tagged as "foo" (this actually sets the tag of the anonymous dict to "foo" but we will never be able to ascertain this because there is no intruction to the right of the is_dict/2 call). == ?- is_dict(_{},foo). true. == Making another variable to denote the "hole" of an "anonymous dict" == ?- is_dict(Y{},X), X == Y. Y = X. == Setting/Binding/Refining the still fresh tag of the dict? |
| put_dict/3 | ## Improvement The argument `New` should really be called `Updates` or `NewKeyVa ... |
| dict_pairs/3 | ## Standard pairs are generated ``` ?- dict_pairs(_{a:1,b:2},T,P). P = [a-1,b-2] ... |
| random_between/3 | Note that there is no requirement that either of the args be >= 0. ``` ?- random ... |
| Sec. 4.32 | ## For those looking for the correct predicate to use: - [Relevant predicates fo ... |
| Sec. 4.20 | For those looking for the predicate to use: - [Relevant predicates for generatin ... |
| debug_message_context/1 | The flag =message_context= is explained on the page for current_prolog_flag/2 |
| is/2 | is/2 is **arithmetic function evaluation** from the original Prolog But in SWI-P ... |
| atomic/1 | The complement of atomic/1 is compound/1. |
| compound/1 | [**This page**](https://github.com/dtonhofer/prolog_notes/tree/master/swipl_note ... |
| findall/4 | Useful when you want to append to a list using several calls to findall/4. In th ... |
| number_codes/2 | ## Getting the number from the list of character codes == ?- number_codes(N,[49, 50, 51]). |
| ## Some references Note that this is based on: - Wikipedia: [PCRE: Perl Compatib ... | |
| number_chars/2 | This tests whether the "serialized representation" doesn't lose digits: == ?- X ... |
| atom_chars/2 | ## See also - atom_codes/2 - string_chars/2 - string_codes/2 and also - atom_str ... |
| debug/3 | ## Topics Please read important note on **debug topics** in the comment section ... |
| Sec. 4.13 | ## Introduction and tutorials - [DCG Primer](https://www.metalevel.at/prolog/dcg ... |
| char_code/2 | Remember that a Prolog "character" is an atom of length 1. However, this predica ... |
| phrase/3 | ## Example == :- use_module(library(dcg/basics)). :- begin_tests(dcg). % Parsing the (unicode) codepoints in "Codes" with the callable % integer(X) which actually resolves to predicate integer/3, taking % itself "Codes" and "Rest" in addition to "X". |
| -- ## Doc needs fix "Prolog is described in Schrijvers et al., 2013 (preprint PDF). ... | |
| initialization/1 | Example for when you want to call =main= of your source file **immediately after the source file has been loaded** (from swipl-devel/man/select.pl): == :- initialization ( catch(main, E, ( print_message(error, E), fail )) -> halt ; halt(1) ). == |
| Note the following outcomes of calling a CHR constraint from Prolog: - The call ... | |
| "Committed Choice" very concretely means there is no backtracking over the CHR r ... | |
| The links to "further information" should probably not be as in an academic pape ... | |
| term_expansion/2 | A simple example to print out whatever is read during consultation. In this case ... |
| unify_with_occurs_check/2 | "That is, a variable only unifies to a term if this term does not contain the va ... |
| dif/2 | ## Needs text improvement We read: _Finally, if A and B can unify, goals are del ... |
| throw/1 | I put together some comments on exceptions (especially on ISO standard exception ... |
| Sec. 2.10 | ## Companion info [**My notes on the Byrd Box model here**](https://github.com/d ... |
| Sec. 4.14 | ## Also Not a "database" but allows to define "record-like terms": [`library(rec ... |
| instantiation_error/1 | The ISO exception term carries no information (it is just an atom) and thus is r ... |
| module/1 | This predicate should probably also be able to retrieve the name of the current ... |
| freeze/2 | ## Some notes I made on `freeze` [**Right here**](https://github.com/dtonhofer/p ... |
| succ/2 | In fact, succ/2 is made to order for **natural numbers**. - What is the predeces ... |
| setof/3 | See bagof/3 for some commentary. See also: distinct/1 and distinct/2 |
| false/0 | ## Zero-effect clauses Note that writing clauses like == p(1,2) :- false. == has absolutely no effect on the meaning of Prolog program - you might as well leave them out (because everything that is left out is considered false - Closed World Assumption) For example: == p(2,4) :- false. p(A,B) :- between(1,4,A),B is 2*A. == Result: == ?- p(2,4). true. ?- bagof([X,Y],p(X,Y),Bag). Bag = [[1, 2], [2, 4], [3, 6], [4, 8]]. == and it is exactly the same for == p(A,B) :- between(1,4,A),B is 2*A. p(2,4) :- false. == or == p(A,B) :- between(1,4,A),B is 2*A. == ## And now for some inconsistency Using `false` and the cut means you can write Bad and Inconsistent Programs: == f(X,X) :- !, false. f(0,1). == Now you are in trouble: == ?- f(0,1). |
| Sec. 5.3.2 | ## Compound terms of arity 0 vs. |
| ## Works "out of the box"! As an example, try a few test cases. | |
| retractall/1 | This predicate never fails. "If Head refers to a predicate that is not defined, ... |
| clause/2 | Jan Burse writes: "According to ISO core standard clause/2 is also only allowed ... |
| retract/1 | "logical update view" would be more correctly called "transactional update view" ... |
| dynamic/1 | See also: [Database Section](https://www.swi-prolog.org/pldoc/man?section=db) wh ... |
| flag/3 | From the [Database](https://eu.swi-prolog.org/pldoc/man?section=db) page: "flag/ ... |
| ## Note The [Windows Release Notes](https://eu.swi-prolog.org/windows.html) are ... | |
| include/3 | Example: We want the _Keys_ of _Key-Value_ pairs which match a certain _Value_. ... |
| ## Test cases are your track bed through the swampland You do not want to haphaz ... | |
| domain_error/2 | An example for a "domain error": == ?- catch(length(L,-1),C,true). C = error(dom ... |
| between/3 | ## Doc needs help > a feature that is particularly interesting for generating in ... |
| compound_name_arguments/3 | [**This page**](https://github.com/dtonhofer/prolog_notes/tree/master/swipl_note ... |
| =../2 | ## A bit surprising perhaps == % works for atoms and strings ?- atom =.. X. X = ... |
| Sec. A.63 | ## IMPORTANT IMPORTANT IMPORTANT If you have yall expression in your source, mak ... |
| -- This operator is a hard nut. I have some notes on the Prolog `\+` here: - [**Not ... | |
| The SWI-Prolog documentation has more on SWI-Prolog datatypes test predicates he ... | |
| Sec. 4.5 | ## See also - is_of_type/2 - checks that a variable is bound to a term of a give ... |
| repeat/0 | ## See also forall/2 as a better alternative to the failure-driven loop ## Alter ... |
| once/1 | ## "once should be deprecated" https://swi-prolog.discourse.group/t/once-1-shoul ... |
| -- I put together some comments on exceptions here: [On Exceptions](https://github. ... | |
| length/2 | ## Doc needs fix Maybe change the parameter name from `Int` to `Length`. ## Note ... |
| atomic_list_concat/2 | ## Your choices Your choice of predicate is as follows: **Two concatenable terms as input (can split)** - atom_concat/3 (ISO) - string_concat/3 **Two more general concatenable terms as input (cannot split because arguments 1 and 2 are too general)** - atomic_concat/3 **A list of concatenable terms as input (never split)** - atomic_list_concat/2 - generates atom at argument 2. |
| list_to_assoc/2 | Missed opportunity: The predicate **fails** if =Assoc= is set but =Pairs= is fre ... |
| max_list/2 | ## See also - min_list/2 - max_member/2 - min_member/2 ## Unify, maybe? (This re ... |
| append/3 | ## Bad naming! |
| keysort/2 | There should be a link to library(pairs): https://www.swi-prolog.org/pldoc/man?s ... |
| Sec. A.3 | I had some fun writing =foldl/4= and =foldr/4=, including an implementation of = ... |
| mod/2 | See also divmod/4 (_not_ visible in the function in the header list because it i ... |
| maplist/4 | Check out maplist/2 for more notes --- Explainer on maplist/4 and examples for t ... |
| current_prolog_flag/2 | ## Doc needs help From the DCG page: The flags - [`back_quotes`](https://eu.swi- ... |
| same_length/2 | ## Example In the same way that length/2 can generate lists of fresh unbound var ... |
| nth0/3 | ## Doc needs help > type_error(integer, Index) if Index is not an integer or unb ... |
| -- Powerful append/2 is powerful. | |
| list_to_set/2 | See also library(ordset) https://www.swi-prolog.org/pldoc/man?section=ordsets an ... |
| write_canonical/1 | Quite an interesting predicate to see what syntax tree the parser has created fr ... |
| functor/3 | [**This page**](https://github.com/dtonhofer/prolog_notes/tree/master/swipl_note ... |
| compound_name_arity/3 | [**This page**](https://github.com/dtonhofer/prolog_notes/tree/master/swipl_note ... |
| number/1 | Float Infinity is a number: ``` ?- MinusInf is -1.0Inf, number(MinusInf). MinusI ... |
| append/2 | ## Bletch This must be one of the worst-named predicates in the entire galaxy. c ... |
| member/2 | ## Some text vagueness "provides determinism on the last element" should be "pro ... |
| Sec. 5.1 | Do not be confused by '[|]'. |
| Sec. 5.4 | ## Some additional documentation Notes I took. |
| \==/2 | Application: membership check in a difference list, to check whether we have rea ... |
| atom_codes/2 | ## See also - atom_chars/2 - string_chars/2 - string_codes/2 and also - atom_str ... |
| sleep/1 | ## Example sleeptimes Run: == bagof([Sleep,Delta] ,T_begin^T_end^(member(Sleep,[ ... |
| random/1 | Let's make the distinction between the offer of built-in functions and library p ... |
| get_dict/3 | ## Some demo code [`dict_getting_testing.pl`](https://github.com/dtonhofer/prolo ... |
| -- I have been looking for a predicate to split a list L into three pieces: - Front (a list) - Element at Index N (0-based) - Back (a list) such that: == append([Front,[Element],Back],L). == Here is one. | |
| nth0/4 | Replacing an element in a list (with 0-based indexing) from https://stackoverflo ... |
| foreach/2 | The "foreach" example is confusing because "dif/2" freezes until a decision can ... |
| Sec. 2 | ## Errors on reloading source code Note that all the predicates (including helpe ... |
| Sec. 4.27.2.1 | Links to the GMP library: https://gmplib.org/ https://en.wikipedia.org/wiki/GNU_ ... |
| predsort/3 | This absolutely needs to be extended to `predsort(+Pred, +List, -Sorted, +Flags) ... |
| Sec. 12.2.3 | ## Help Try ``` swipl-ld -help ``` for help information. This is not a script, b ... |
| PL_get_atom_chars() | Returns 1 ( =|TRUE|= ) on success, 0 ( =|FALSE|= ) on failure (e.g. |
| PL_register_foreign_in_module() | "After this call returns successfully" means it returns with =|TRUE|= (i.e. |
| Sec. 12.3 | `atom_t` may point to static memory or garbage-collectable memory if I understan ... |
| meta_predicate/1 | Confused by "The argument is a term that is used to reference a predicate with N more arguments than the given argument term. |
| Sec. 6.5 | Confused by "The argument is a term that is used to reference a predicate with N more arguments than the given argument term. |
| nan/0 | ## NaN is a float Note that NaN is a float: ``` ?- X is nan, float(X). X = 1.5Na ... |
| max/2 | ## See also The predicates min_list/2 and max_list/2 from library [library(lists ... |
| =:=/2 | Some examples: Everything must be known at the time of evaluation: == ?- 2*2 =:= X. ERROR: Arguments are not sufficiently instantiated ?- X=4.0, 2*2 =:= X. X = 4.0. ?- cos(X) =\= sin(X). ERROR: Arguments are not sufficiently instantiated % Ok, then! ?- X=pi, cos(X) =\= sin(X). X = pi. == Sadly =:= is a bit fussy. |
| \==/2 | For clarity, this should be written with complimentary parentheses: == \+(Term1 ... |
| copy_term/2 | It is intuitively clear what "renamed (fresh) variables" means but it's not real ... |
| \=/2 | ``` X = Y ``` means: "Proceeding down this path of computation, =|X|= and =|Y|= ... |
| \=/2 | It would be clearer to write == \+(Term1 = Term2) == rather than than == \+Term1 = Term2 == Although of course: == ?- write_canonical(\+Term1 = Term2). \+(=(_,_)) == |
| ->/2 | ## Notes For - the if-then-else `->/2` + `;/2` and - the [soft-cut](https://eu.s ... |
| initialization/2 | Regarding "main", note that an exit status of 0 indicating "success" is a conven ... |
| Sec. 2.11.1.1 | Contra Carlo Capelli's note, the "shebang" is not at all overengineered. As a co ... |
| Sec. 4.36 | Also consider library "filesex" ("extended operations on files", damned PhDs doi ... |
| Sec. A.26 | For the shebang trick where swipl is started indirectly through =env= == #!/usr/ ... |
| Sec. A.26 | For the "initialization" directive, see: https://www.swi-prolog.org/pldoc/doc_fo ... |
| open/3 | Question: What does an empty option list assume about the encoding of the file? |
| setup_call_cleanup/3 | ## If you come from other programming languages If you come from other programmi ... |
| deterministic/1 | Example: == ory :- format("1st clause:\n"), deterministic(D1), format(" Det at ( ... |
| atom/1 | See also: https://www.swi-prolog.org/pldoc/man?section=text-representation Atoms ... |
| Note that contrary to the statement in the text: "Installing from source is ofte ... | |
| Sec. 6.2 | ## For those looking for the "module qualification mechanism" It's here: [defini ... |
| Sec. G | This above just the list of references that appear in the Documentation. I have ... |
| get_time/1 | Returns a float! Compare with Unix `date` command's "seconds since 1970-01-01 00 ... |
| begin_tests/1 | Note that if you add tests from the command line using [user]., you can do so on ... |
| end_tests/1 | What is the rationale behind having to repeat the test name in the block termina ... |
| consult/1 | ## Doc needs help You can actually also pass a series of files (is this inoffici ... |
| Sec. A.24 | I would also point the user to https://www.swi-prolog.org/pldoc/man?section=ords ... |
| predicate_property/2 | This may not be directly clear from the description: ## Calling predicate_proper ... |
| This should really be linked from the "Documentation" intro page. | |
| Sec. 3 | ISO Prolog only defines the "% ..." comments and the "/* ... */" comment block i ... |
| maplist/2 | ## Explainer Explainer including examples for the use of maplist/2 (and maplist/ ... |
| float/1 | "When used around the 2nd argument of is/2, the result will be returned as a flo ... |
| Sec. A.34 | This needs more explanation on the attached "database file". Is it a text file w ... |
| pengine_event_loop/2 | ## See also ignore/1: Calls Goal as once/1, but succeeds, regardless of whether ... |
| Sec. 1.1 | The link = http://www.swi-prolog.org/pldoc/doc/home/vnc/prolog/lib/swipl/library ... |
| catch/3 | ## Problematic Naming While the `Goal` argument is correctly named, `Catcher` is a confusing name for what is the exception term against which a thrown exception shall be against. |
| float/1 | ## See also - [`float/1`](https://eu.swi-prolog.org/pldoc/man?function=float/1) ... |
| Sec. 2.15.1.3 | This section applies to Strings, too == ?- format(">\c | hello<"). >hello< true. == Jan Wielemaker writes: "Although incomplete in the implementation, the overall idea is that predicates that require text input accept all text representations and produce the documented type as output." ---- The \uXXXX encoding (really, NAMING of a glyph in the Unicode BMP) is called "UCS-2" https://en.wikipedia.org/wiki/Universal_Coded_Character_Set More on Unicode here: https://en.wikipedia.org/wiki/Unicode#Architecture_and_terminology ---- Note that the unicode escapes are meant to be used in the context of strings, atoms or chars (atoms of length 1). |
| writef/2 | As noted on the page of format/2, the escape sequences one uses in practice are ... |
| format/2 | ## Print hex codes 0-padded 4-nibbles: ``` ?- Code=6464, format(atom(T),"0x~|~`0 ... |
| atom_string/2 | ## This also is the "comparison operator for stringy things" Suppose you have re ... |
| -- This corresponds to "serialization/deserialization" in Java and other languages. You CAN but shouldn't use this for reading input typed in by the user. Example: == ?- with_output_to( string(Buf), write_term(a(b,c,[1,2,3],var(ZZ)),[])), % serialize a term as string to "Buf" read_term_from_atom(Buf, T, []). | |
| maplist/3 | Check out maplist/2 for more notes --- Explainer on maplist/3 and examples for t ... |
| Sec. 4.6 | Note that "==" and "\==" are really tests about the current state of the computa ... |
| Sec. 2.1 | "Getting started quickly" should point the user to a description/tutorial/howto ... |
| Sec. 2.15.1.9 | See also: https://eu.swi-prolog.org/pldoc/man?section=widechars |
| Sec. 2.15.1 | The title of this section is "ISO Syntax Support" but should probably be "Extens ... |
| Sec. 2.15.1.5 | This is not ISO Prolog, but quite useful. Compare with Java where this trick is ... |
| findall/3 | ## Notes about findall [**Notes about findall/3**](https://github.com/dtonhofer/ ... |
| Sec. 2.12 | I'm always looking for the way make the ### Prolog Toplevel printer print lists ... |
| Sec. 5.2.2 | ## See also - library(strings): https://eu.swi-prolog.org/pldoc/man?section=stri ... |
| call/1 | ## Backtracking works as expected == ?- call(member(X,[1,2,3])). X = 1 ; X = 2 ; X = 3. == ## Doc changes The text should probably be changed from > Note that clauses may have variables as subclauses, which is identical to call/1. to > Note that clauses may have variables as goals. |
| Currently trying "ProDT"... Trying with Eclipse Oxygen (Eclipse 4.7, 2017-12) on Linux Fedora with JDK 1.8.0 The original site is http://prodevtools.sourceforge.net/ - the instructions on that site are a bit stale. | |
| It might be helpful to indicate when this entry was written. | |
| Dead link in the line "Redo the build. | |
| Compiling on Fedora: At first sight, the error log seems to indicate that there ... | |
| Sadly, the link "The design of the XPCE/Prolog interface (Publication in Worksho ... | |
| bagof/3 | ## See also [library(aggregate)](https://eu.swi-prolog.org/pldoc/man?section=agg ... |
| ^/2 | ## Something weird If SWI-Prolog has not been compiled with GMP, then ^/2 perfor ... |
| ==/2 | Confusingly, the explanations uses two adjectives: "equivalent" and "identical". |
| ?=/2 | I think the text "Succeeds if the syntactic equality of Term1 and Term2..." shou ... |
| The link to "Initialising and Managing.." has to be changed to http://www.swi-pr ... | |
| cyclic_term/1 | Note that acyclic_term/1 is ISO, but cyclic_term/1 is not. Consider the evolutio ... |
| memberchk/2 | This predicate **does not backtrack** (which is why it exists). Use this predica ... |
| Sec. 5 | See also [SWI-Prolog's page on predicate descriptors](https://eu.swi-prolog.org/ ... |
| Sec. A.4.3 | Actually missing a way to get the number of elements in the association list... |
| Reference manual | #### SWI-Prolog Wiki entry There is a [wiki entry at the SWI-Prolog Discourse Gr ... |
| assertion/1 | ## Not an ISO Standard exception assertion/1 throws a non-ISO Standard error ter ... |
| file_search_path/2 | Thanks @Jan. |
| And here is the [reference for floating-point notation](http://www.dcc.fc.up.pt/ ... | |
| Here is a another link of interest that predigests the "What Every Computer Scie ... | |
| file_search_path/2 | The definition should be "file_search_path(?Alias, ?Path)" as the following work ... |
Packages by LogicalCaptain
| Pack
tot: 2
| Version
(#older)
| Downloads
tot: 149
(#latest) | Rating
(#votes/
#comments) | Title |
|---|---|---|---|---|
| scasp | 1.1.430 | 14612 | Goal directed ASP solver | |
| onepointfour_basics | 0.9 | 3 | Some basic predicates that may be useful. Includes a replacement of must_be/2 and a dict prettyprinter |
This list contains packages whose author name, e-mail or homepage url matches the profile information.