rishi/scryer-prolog
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #1699 from triska/sgml_doc

Browse Source 
DOC: convert library(sgml) documentation to DocLog format
This commit is contained in:
Mark Thom
2023-01-24 22:43:22 +01:00
committed by GitHub
parent 26d39c3617 af9f0f81d8
commit b6204eb6f1
1 changed files with 62 additions and 49 deletions
Download Patch File Download Diff File Expand all files Collapse all files

39
src/lib/sgml.pl
View File

@@ -2,56 +2,69 @@
Predicates for parsing HTML and XML documents. Predicates for parsing HTML and XML documents.
Written 2020-2022 by Markus Triska (triska@metalevel.at) Written 2020-2022 by Markus Triska (triska@metalevel.at)
Part of Scryer Prolog. Part of Scryer Prolog.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - */
/** Predicates for parsing HTML and XML documents.
Currently, two predicates are provided: Currently, two predicates are provided:
- load_html(+Source, -Es, +Options) - `load_html(+Source, -Es, +Options)`
- load_xml(+Source, -Es, +Options) - `load_xml(+Source, -Es, +Options)`
These predicates parse HTML and XML documents, respectively. These predicates parse HTML and XML documents, respectively.
Source must be one of: Source must be one of:
- a list of characters with the document contents - a list of characters with the document contents
- stream(S), specifying a stream S from which to read the content - `stream(S)`, specifying a stream S from which to read the content
- file(Name), where Name is a list of characters specifying a file name. - `file(Name)`, where Name is a list of characters specifying a file name.
Es is unified with the abstract syntax tree of the parsed document, Es is unified with the abstract syntax tree of the parsed document,
represented as a list of elements where each is of the form: represented as a list of elements where each is of the form:
* a list of characters, representing text * a list of characters, representing text
* element(Name, Attrs, Children)
- Name, an atom, is the name of the tag * `element(Name, Attrs, Children)`
- Attrs is a list of Key=Value pairs:
Key is an atom, and Value is a list of characters - `Name`, an atom, is the name of the tag
- Children is a list of elements as specified here.
- `Attrs` is a list of `Key=Value` pairs:
`Key` is an atom, and `Value` is a list of characters
- `Children` is a list of elements as specified here.
Currently, Options are ignored. In the future, more options may be Currently, Options are ignored. In the future, more options may be
provided to control parsing. provided to control parsing.
Example: Example:
```
?- load_html("<html><head><title>Hello!</title></head></html>", Es, []). ?- load_html("<html><head><title>Hello!</title></head></html>", Es, []).
```
Yielding: Yielding:
```
Es = [element(html,[], Es = [element(html,[],
[element(head,[], [element(head,[],
[element(title,[], [element(title,[],
["Hello!"])]), ["Hello!"])]),
element(body,[],[])])]. element(body,[],[])])].
```
library(xpath) provides convenient reasoning about parsed documents. `library(xpath)` provides convenient reasoning about parsed documents.
For example, to fetch the title of the document above, we can use: For example, to fetch the title of the document above, we can use:
```
?- load_html("<html><head><title>Hello!</title></head></html>", Es, []), ?- load_html("<html><head><title>Hello!</title></head></html>", Es, []),
xpath(Es, //title(text), T). xpath(Es, //title(text), T).
```
Yielding T = "Hello!". Yielding `T = "Hello!"`.
Use http_open/3 from library(http/http_open) to read answers from Use `http_open/3` from `library(http/http_open)` to read answers from
web servers via streams. web servers via streams.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - */ */
:- module(sgml, [load_html/3, :- module(sgml, [load_html/3,
load_xml/3]). load_xml/3]).