[r2002-11-01 21:19:43 by cvscast] Empty log message

Original author: cvscast
Date: 2002-11-01 21:19:43+00:00

runtime/load_xml.ml

@@ -65,8 +65,6 @@ let run s =
       | E_start_tag (name,att,_) ->  get (); parse_elt name att
       | _ -> get (); parse_doc ()
   in
-  prerr_endline "load_xml: start";
   get ();
-  let r = parse_doc () in
-  prerr_endline "load_xml: done";
-  r
+  parse_doc ()
+

tests/xhtml-special.ent

+<!-- Special characters for XHTML -->
+
+<!-- Character entity set. Typical invocation:
+     <!ENTITY % HTMLspecial PUBLIC
+        "-//W3C//ENTITIES Special for XHTML//EN"
+        "http://www.w3.org/TR/xhtml1/DTD/xhtml-special.ent">
+     %HTMLspecial;
+-->
+
+<!-- Portions (C) International Organization for Standardization 1986:
+     Permission to copy in any form is granted for use with
+     conforming SGML systems and applications as defined in
+     ISO 8879, provided this notice is included in all copies.
+-->
+
+<!-- Relevant ISO entity set is given unless names are newly introduced.
+     New names (i.e., not in ISO 8879 list) do not clash with any
+     existing ISO 8879 entity names. ISO 10646 character numbers
+     are given for each character, in hex. values are decimal
+     conversions of the ISO 10646 values and refer to the document
+     character set. Names are Unicode names. 
+-->
+
+<!-- C0 Controls and Basic Latin -->
+<!ENTITY quot    "&#34;"> <!--  quotation mark, U+0022 ISOnum -->
+<!ENTITY amp     "&#38;#38;"> <!--  ampersand, U+0026 ISOnum -->
+<!ENTITY lt      "&#38;#60;"> <!--  less-than sign, U+003C ISOnum -->
+<!ENTITY gt      "&#62;"> <!--  greater-than sign, U+003E ISOnum -->
+<!ENTITY apos	 "&#39;"> <!--  apostrophe = APL quote, U+0027 ISOnum -->
+
+<!-- Latin Extended-A -->
+<!ENTITY OElig   "&#338;"> <!--  latin capital ligature OE,
+                                    U+0152 ISOlat2 -->
+<!ENTITY oelig   "&#339;"> <!--  latin small ligature oe, U+0153 ISOlat2 -->
+<!-- ligature is a misnomer, this is a separate character in some languages -->
+<!ENTITY Scaron  "&#352;"> <!--  latin capital letter S with caron,
+                                    U+0160 ISOlat2 -->
+<!ENTITY scaron  "&#353;"> <!--  latin small letter s with caron,
+                                    U+0161 ISOlat2 -->
+<!ENTITY Yuml    "&#376;"> <!--  latin capital letter Y with diaeresis,
+                                    U+0178 ISOlat2 -->
+
+<!-- Spacing Modifier Letters -->
+<!ENTITY circ    "&#710;"> <!--  modifier letter circumflex accent,
+                                    U+02C6 ISOpub -->
+<!ENTITY tilde   "&#732;"> <!--  small tilde, U+02DC ISOdia -->
+
+<!-- General Punctuation -->
+<!ENTITY ensp    "&#8194;"> <!-- en space, U+2002 ISOpub -->
+<!ENTITY emsp    "&#8195;"> <!-- em space, U+2003 ISOpub -->
+<!ENTITY thinsp  "&#8201;"> <!-- thin space, U+2009 ISOpub -->
+<!ENTITY zwnj    "&#8204;"> <!-- zero width non-joiner,
+                                    U+200C NEW RFC 2070 -->
+<!ENTITY zwj     "&#8205;"> <!-- zero width joiner, U+200D NEW RFC 2070 -->
+<!ENTITY lrm     "&#8206;"> <!-- left-to-right mark, U+200E NEW RFC 2070 -->
+<!ENTITY rlm     "&#8207;"> <!-- right-to-left mark, U+200F NEW RFC 2070 -->
+<!ENTITY ndash   "&#8211;"> <!-- en dash, U+2013 ISOpub -->
+<!ENTITY mdash   "&#8212;"> <!-- em dash, U+2014 ISOpub -->
+<!ENTITY lsquo   "&#8216;"> <!-- left single quotation mark,
+                                    U+2018 ISOnum -->
+<!ENTITY rsquo   "&#8217;"> <!-- right single quotation mark,
+                                    U+2019 ISOnum -->
+<!ENTITY sbquo   "&#8218;"> <!-- single low-9 quotation mark, U+201A NEW -->
+<!ENTITY ldquo   "&#8220;"> <!-- left double quotation mark,
+                                    U+201C ISOnum -->
+<!ENTITY rdquo   "&#8221;"> <!-- right double quotation mark,
+                                    U+201D ISOnum -->
+<!ENTITY bdquo   "&#8222;"> <!-- double low-9 quotation mark, U+201E NEW -->
+<!ENTITY dagger  "&#8224;"> <!-- dagger, U+2020 ISOpub -->
+<!ENTITY Dagger  "&#8225;"> <!-- double dagger, U+2021 ISOpub -->
+<!ENTITY permil  "&#8240;"> <!-- per mille sign, U+2030 ISOtech -->
+<!ENTITY lsaquo  "&#8249;"> <!-- single left-pointing angle quotation mark,
+                                    U+2039 ISO proposed -->
+<!-- lsaquo is proposed but not yet ISO standardized -->
+<!ENTITY rsaquo  "&#8250;"> <!-- single right-pointing angle quotation mark,
+                                    U+203A ISO proposed -->
+<!-- rsaquo is proposed but not yet ISO standardized -->
+
+<!-- Currency Symbols -->
+<!ENTITY euro   "&#8364;"> <!--  euro sign, U+20AC NEW -->

tests/xhtml.xml

+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE html SYSTEM "xhtml1-strict.dtd">
+<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
+  <head>
+    <title>Virtual Library</title>
+  </head>
+  <body>
+    <p>Moved to <a href="http://example.org/">example.org</a>.</p>
+  </body>
+</html>
+

tools/pull.ml

+(* To test PXP pull parsers *)
+
+open Pxp_yacc
+open Pxp_lexer_types
+open Pxp_types
+open Printf
+
+
+(* dump_event: dumps a single parsing event *)
+
+let dump_event =
+  function
+      E_start_doc(v,sa,dtd) ->
+	printf "E_start_doc version=%s standalone=%b\n" v sa
+    | E_end_doc ->
+	printf "E_end_doc\n"
+    | E_start_tag(name,attlist,_) ->
+	printf "E_start_tag %s %s\n" name 
+	  (String.concat " " (List.map (fun (n,v) -> n ^ "=" ^ v) attlist))
+    | E_end_tag(name,_) ->
+	printf "E_end_tag %s\n" name
+    | E_char_data data ->
+	printf "E_char_data %s\n" data
+    | E_pinstr(target,data) ->
+	printf "E_pinstr %s %s\n" target data
+    | E_comment data ->
+	printf "E_comment %s\n" data
+    | E_position(ent,line,col) ->
+	printf "E_position %s line=%d col=%d\n" ent line col
+    | E_error e ->
+	printf "E_error %s\n" (Pxp_types.string_of_exn e)
+    | E_end_of_stream ->
+	printf "E_end_of_stream\n"
+;;
+
+
+(* parse: prints the events while parsing the passed string *)
+
+let parse s =
+  let config = default_config in
+  let mgr = create_entity_manager config (from_file s) in
+  let next_event = 
+    create_pull_parser config (`Entry_document [`Extend_dtd_fully]) mgr in
+  let event = ref (Some E_end_of_stream) in
+  while !event <> None do
+    event := next_event();
+    match !event with
+	Some e -> dump_event e
+      | None -> ()
+  done
+;;
+
+
+let () = parse Sys.argv.(1);;