SWI-Prolog -- library(readutil): Read utilities

See also

- library(pure_input) allows for processing files with DCGs.
- library(lazy_lists) for creating lazy lists from input.

This library provides some commonly used reading predicates. As these predicates have proven to be time-critical in some applications we moved them to C. For compatibility as well as to reduce system dependency, we link the foreign code at runtime and fallback to the Prolog implementation if the shared object cannot be found.

[det]read_line_to_codes(+Stream, -Line:codes)

Read the next line of input from Stream. Unify content of the lines as a list of character codes with Line after the line has been read. A line is ended by a newline character or end-of-file. Unlike read_line_to_codes/3, this predicate removes a trailing newline character.

[det]read_line_to_codes(+Stream, -Line, ?Tail)

Difference-list version to read an input line to a list of character codes. Reading stops at the newline or end-of-file character, but unlike read_line_to_codes/2, the newline is retained in the output. This predicate is especially useful for reading a block of lines up to some delimiter. The following example reads an HTTP header ended by a blank line:

read_header_data(Stream, Header) :-
    read_line_to_codes(Stream, Header, Tail),
    read_header_data(Header, Stream, Tail).

read_header_data("\r\n", _, _) :- !.
read_header_data("\n", _, _) :- !.
read_header_data("", _, _) :- !.
read_header_data(_, Stream, Tail) :-
    read_line_to_codes(Stream, Tail, NewTail),
    read_header_data(Tail, Stream, NewTail).

[det]read_line_to_string(+Stream, -String)

Read the next line from Stream into String. String does not contain the line terminator. String is unified with the atom end_of_file if the end of the file is reached.

See also: read_string/5 can be used to read lines with separated records without creating intermediate strings.

[det]read_stream_to_codes(+Stream, -Codes)

[det]read_stream_to_codes(+Stream, -Codes, ?Tail)

Read input from Stream to a list of character codes. The version read_stream_to_codes/3 creates a difference-list.

[det]read_file_to_codes(+Spec, -Codes, +Options)

Read the file Spec into a list of Codes. Options is split into options for absolute_file_name/3 and open/4. In addition, the following option is provided:

tail(?Tail): Read the data into a difference list Codes\Tail.

See also: phrase_from_file/3 and read_file_to_string/3.

[det]read_file_to_string(+Spec, -String, +Options)

Read the file Spec into a the string String. Options is split into options for absolute_file_name/3 and open/4.

See also: phrase_from_file/3 and read_file_to_codes/3.

[det]read_file_to_terms(+Spec, -Terms, +Options)

Read the file Spec into a list of terms. Options is split over absolute_file_name/3, open/4 and read_term/3. In addition, the following option is processed:

tail(?Tail): If present, Terms\Tail forms a difference list.

Note that the output options of read_term/3, such as variable_names or subterm_positions will cause read_file_to_terms/3 to fail if Spec contains multiple terms because the values for the different terms will not unify.

LogicalCaptain said (2021-01-10T21:52:49):

Here is code to query the user with several questions and put the result into a dict, keyed with a "slot name".

input_into_dict.pl

The above collects patient data for an expert system and so a run looks like this:

?-
questions(Dict),prettyprint_dict(Dict).

what is subject name?
|: J.F. Sebastian
what is patient risk? (one of: [high,medium,low]) (default: medium)
|: don't know
Expecting one of: [high,medium,low], but got: don't know
what is patient risk? (one of: [high,medium,low]) (default: medium)
|: LOW
how long have they had the condition for? (one of: [months,weeks,days]) (default: weeks)
|: months
is the trial justifiable? (one of: [yes,no]) (default: yes)
|:
what is their antibody count? (one of: [high,medium,low]) (default: medium)
|: low
how do they live? (one of: [sedentary,active]) (default: sedentary)
|: sedentary
do they have anaemia? (one of: [yes,no]) (default: no)
|:
is their blood pressure raised? (one of: [yes,no]) (default: no)
|:
  anaemia_state   no
  antibody_count  low
  blood_pressure  no
  clinical_trial  yes
  condition_time  months
  life_style      sedentary
  name            J.F. Sebastian
  patient_risk    low

Dict = patient{anaemia_state:no,antibody_count:low,blood_pressure:no,
               clinical_trial:yes,condition_time:months,
               life_style:sedentary,name:"J.F. Sebastian",patient_risk:low}.

A.48 library(readutil): Read utilities