blob: a6155ff1bdc265cc7edbf01a6d13211c4c4807d6 [file] [log] [blame]
Daniel Veillard01791d51998-07-24 19:24:09 +00001
Daniel Veillardad8f99d2000-01-15 14:20:03 +00002 TODO for the XML parser and stuff:
3 ==================================
Daniel Veillard01791d51998-07-24 19:24:09 +00004
Daniel Veillard361d8452000-04-03 19:48:13 +00005
Daniel Veillard75c29212000-03-14 18:33:47 +00006TODO:
7=====
8
Daniel Veillarde0854c32000-08-27 21:12:29 +00009- performances: there is still improvements needed when parsing Docbook DTD
10 a single function to optimize/avoid.
Daniel Veillard5d211f42000-04-07 17:00:24 +000011- DOM needs
Daniel Veillard5d211f42000-04-07 17:00:24 +000012 int xmlPruneProp(xmlNodePtr node, xmlAtttrPtr attr);
Daniel Veillarde0854c32000-08-27 21:12:29 +000013- listing all attributes in a node.
Daniel Veillard75c29212000-03-14 18:33:47 +000014- Correct standalone checking/emitting (hard)
15 2.9 Standalone Document Declaration
Daniel Veillard75c29212000-03-14 18:33:47 +000016- Better checking of external parsed entities TAG 1234
17- Find way of representing PERefs in the Dtd so that %entity; can
18 be saved back.
19- Go through erratas and do the cleanup.
Daniel Veillarde0854c32000-08-27 21:12:29 +000020 http://www.w3.org/XML/xml-19980210-errata ... started ...
Daniel Veillard361d8452000-04-03 19:48:13 +000021- Handle undefined namespaces in entity contents better ... at least
22 issue a warning
Daniel Veillard06047432000-04-24 11:33:38 +000023- fix --disable-corba configure switch handling, and use XML_WITHOUT_CORBA
24 not WITHOUT_CORBA flag
Daniel Veillardf0cc7cc2000-08-26 21:40:43 +000025- reported by Michael, update of doc node when pasting on a new document
26 There can be far more than simply the doc pointer which refers to
27 the old document, for example namespace declarations or entities
28 references can also be a nasty problem, far more than updating the
29 doc values.
Daniel Veillard46057e12000-09-24 18:49:59 +000030- jamesh suggestion: SAX like functions to save a document ie. call a
31 function to open a new element with given attributes, write character
32 data, close last element, etc
Daniel Veillardad8f99d2000-01-15 14:20:03 +000033
34TODO:
35=====
36
Daniel Veillard75c29212000-03-14 18:33:47 +000037- Get OASIS testsuite to a more friendly result, check all the results
Daniel Veillard496a1cf2000-05-03 14:20:55 +000038 once stable. Current state at:
Daniel Veillard75c29212000-03-14 18:33:47 +000039 http://xmlsoft.org/conf/result.html
Daniel Veillardad8f99d2000-01-15 14:20:03 +000040
Daniel Veillard3dd82e72000-03-20 11:48:04 +000041- Optimization of tag strings allocation ?
Daniel Veillard75c29212000-03-14 18:33:47 +000042
43- maintain coherency of namespace when doing cut'n paste operations
44 => the functions are coded, but need testing
45
Daniel Veillard3dd82e72000-03-20 11:48:04 +000046- function to rebuild the ID table
47- functions to rebuild the DTD hash tables (after DTD changes).
Daniel Veillard75c29212000-03-14 18:33:47 +000048
Daniel Veillard75c29212000-03-14 18:33:47 +000049
50EXTENSIONS:
51===========
Daniel Veillard3dd82e72000-03-20 11:48:04 +000052
Daniel Veillard75c29212000-03-14 18:33:47 +000053- Tools to produce man pages from the SGML docs.
Daniel Veillardad8f99d2000-01-15 14:20:03 +000054- Finish XPath
55 => attributes addressing troubles
56 => defaulted attributes handling
57 => namespace axis ?
58
59- Add Xpointer recognition/API
60
61- Add Xlink recognition/API
62 => started adding an xlink.[ch] with a unified API for XML and HTML.
Daniel Veillard496a1cf2000-05-03 14:20:55 +000063 it's crap :-(
Daniel Veillardad8f99d2000-01-15 14:20:03 +000064
65- Implement XSLT
66 => seems that someone volunteered ?!?
67
68- Implement XSchemas
Daniel Veillard496a1cf2000-05-03 14:20:55 +000069 => Really need to be done <grin/>
Daniel Veillardad8f99d2000-01-15 14:20:03 +000070
71- O2K parsing;
72 => this is a somewhat ugly mix of HTML and XML, adding a specific
73 routine in the comment parsing code of HTML and plug the XML
74 parsing one in-there should not be too hard. Key point is to get
75 XSL to transform all this to something decent ...
76
Daniel Veillardad8f99d2000-01-15 14:20:03 +000077- extend the shell with:
78 - edit
79 - load/save
80 - mv (yum, yum, but it's harder because directories are ordered in
81 our case, mvup and mvdown would be required)
82
Daniel Veillardad8f99d2000-01-15 14:20:03 +000083- Add HTML validation using the XHTML DTD
84 - problem: do we want to keep and maintain the code for handling
85 DTD/System ID cache directly in libxml ?
86
87- Add a DTD cache prefilled with xhtml DTDs and entities and a program to
88 manage them -> like the /usr/bin/install-catalog from SGML
89 right place seems $datadir/xmldtds
Daniel Veillard361d8452000-04-03 19:48:13 +000090 Maybe this is better left to user apps
Daniel Veillardad8f99d2000-01-15 14:20:03 +000091
92- Add output to XHTML in case of HTML documents.
93
Daniel Veillarde3d88ef2000-01-24 13:55:06 +000094
Daniel Veillardad8f99d2000-01-15 14:20:03 +000095Done:
96=====
97
Daniel Veillard1e851392000-10-15 10:02:56 +000098- HTML: handling of Script and style data elements, need special code in
99 the parser and saving functions (handling of < > " ' ...):
100 http://www.w3.org/TR/html4/types.html#type-script
101 Attributes are no problems since entities are accepted.
Daniel Veillarde0854c32000-08-27 21:12:29 +0000102- DOM needs
103 xmlAttrPtr xmlNewDocProp(xmlDocPtr doc, const xmlChar *name, const xmlChar *value)
Daniel Veillard2f2bf412000-08-20 15:11:02 +0000104- problem when parsing hrefs with & with the HTML parser (IRC ac)
105- If the internal encoding is not UTF8 saving to a given encoding doesn't
106 work => fix to force UTF8 encoding ...
107 done, added documentation too
Daniel Veillard32bc74e2000-07-14 14:49:25 +0000108- Add an ASCII I/O encoder (asciiToUTF8 and UTF8Toascii)
109- Issue warning when using non-absolute namespaces URI.
110- the html parser should add <head> and <body> if they don't exist
111 started, not finished.
112 Done, the automatic closing is added and 3 testcases were inserted
Daniel Veillard3f6f7f62000-06-30 17:58:25 +0000113- Command to force the parser to stop parsing and ignore the rest of the file.
Daniel Veillard32bc74e2000-07-14 14:49:25 +0000114 xmlStopParser() should allow this, mostly untested
Daniel Veillardbe803962000-06-28 23:40:59 +0000115- support for HTML empty attributes like <hr noshade>
116- plugged iconv() in for support of a large set of encodings.
117- xmlSwitchToEncoding() rewrite done
Daniel Veillard496a1cf2000-05-03 14:20:55 +0000118- URI checkings (no fragments) rfc2396.txt
Daniel Veillard5d211f42000-04-07 17:00:24 +0000119- Added a clean mechanism for overload or added input methods:
120 xmlRegisterInputCallbacks()
Daniel Veillard361d8452000-04-03 19:48:13 +0000121- dynamically adapt the alloc entry point to use g_alloc()/g_free()
122 if the programmer wants it:
123 - use xmlMemSetup() to reset the routines used.
124- Check attribute normalization especially xmlGetProp()
125- Validity checking problems for NOTATIONS attributes
126- Validity checking problems for ENTITY ENTITIES attributes
127- Parsing of a well balanced chunk xmlParseBalancedChunkMemory()
128- URI module: validation, base, etc ... see uri.[ch]
129- turn tester into a generic program xmllint installed with libxml
Daniel Veillardec303412000-03-24 13:41:54 +0000130- extend validity checks to go through entities content instead of
131 just labelling them PCDATA
Daniel Veillard75c29212000-03-14 18:33:47 +0000132- Save Dtds using the children list instead of dumping the tables,
133 order is preserved as well as comments and PIs
134- Wrote a notice of changes requires to go from 1.x to 2.x
135- make sure that all SAX callbacks are disabled if a WF error is detected
136- checking/handling of newline normalization
137 http://localhost/www.xml.com/axml/target.html#sec-line-ends
138- correct checking of '&' '%' on entities content.
139- checking of PE/Nesting on entities declaration
140- checking/handling of xml:space
141 - checking done.
142 - handling done, not well tested
143- Language identification code, productions [33] to [38]
144 => done, the check has been added and report WFness errors
145- Conditional sections in DTDs [61] to [65]
146 => should this crap be really implemented ???
147 => Yep OASIS testsuite uses them
148- Allow parsed entities defined in the internal subset to override
149 the ones defined in the external subset (DtD customization).
150 => This mean that the entity content should be computed only at
151 use time, i.e. keep the orig string only at parse time and expand
152 only when referenced from the external subset :-(
153 Needed for complete use of most DTD from Eve Maler
154- Add regression tests for all WFC errors
155 => did some in test/WFC
156 => added OASIS testsuite routines
157 http://xmlsoft.org/conf/result.html
158
159- I18N: http://wap.trondheim.com/vaer/index.phtml is not XML and accepted
160 by the XML parser, UTF-8 should be checked when there is no "encoding"
161 declared !
162- Support for UTF-8 and UTF-16 encoding
163 => added some convertion routines provided by Martin Durst
164 patched them, got fixes from @@@
165 I plan to keep everything internally as UTF-8 (or ISO-Latin-X)
166 this is slightly more costly but more compact, and recent processors
167 efficiency is cache related. The key for good performances is keeping
168 the data set small, so will I.
169 => the new progressive reading routines call the detection code
170 is enabled, tested the ISO->UTF-8 stuff
Daniel Veillardad8f99d2000-01-15 14:20:03 +0000171- External entities loading:
172 - allow override by client code
173 - make sure it is alled for all external entities referenced
174 Done, client code should use xmlSetExternalEntityLoader() to set
175 the default loading routine. It will be called each time an external
176 entity entity resolution is triggered.
177- maintain ID coherency when removing/changing attributes
178 The function used to deallocate attributes now check for it being an
179 ID and removes it from the table.
180- push mode parsing i.e. non-blocking state based parser
181 done, both for XML and HTML parsers. Use xmlCreatePushParserCtxt()
182 and xmlParseChunk() and html counterparts.
183 The tester program now has a --push option to select that parser
184 front-end. Douplicated tests to use both and check results are similar.
185
186- Most of XPath, still see some troubles and occasionnal memleaks.
187- an XML shell, allowing to traverse/manipulate an XML document with
188 a shell like interface, and using XPath for the anming syntax
189 - use of readline and history added when available
190 - the shell interface has been cleanly separated and moved to debugXML.c
191- HTML parser, should be fairly stable now
192- API to search the lang of an attribute
193- Collect IDs at parsing and maintain a table.
194 PBM: maintain the table coherency
195 PBM: how to detect ID types in absence of DtD !
196- Use it for XPath ID support
197- Add validity checking
198 Should be finished now !
199- Add regression tests with entity substitutions
200
201- External Parsed entities, either XML or external Subset [78] and [79]
202 parsing the xmllang DtD now works, so it should be sufficient for
203 most cases !
204
205- progressive reading. The entity support is a first step toward
Daniel Veillard260a68f1998-08-13 03:39:55 +0000206 asbtraction of an input stream. A large part of the context is still
207 located on the stack, moving to a state machine and putting everyting
208 in the parsing context should provide an adequate solution.
Daniel Veillardad8f99d2000-01-15 14:20:03 +0000209 => Rather than progressive parsing, give more power to the SAX-like
210 interface. Currently the DOM-like representation is built but
211 => it should be possible to define that only as a set of SAX callbacks
212 and remove the tree creation from the parser code.
213 DONE
Daniel Veillard01791d51998-07-24 19:24:09 +0000214
Daniel Veillardad8f99d2000-01-15 14:20:03 +0000215- DOM support, instead of using a proprietary in memory
216 format for the document representation, the parser should
217 call a DOM API to actually build the resulting document.
218 Then the parser becomes independent of the in-memory
219 representation of the document. Even better using RPC's
220 the parser can actually build the document in another
221 program.
222 => Work started, now the internal representation is by default
223 very near a direct DOM implementation. The DOM glue is implemented
224 as a separate module. See the GNOME gdome module.
225
Daniel Veillard01791d51998-07-24 19:24:09 +0000226- C++ support : John Ehresman <jehresma@dsg.harvard.edu>
227- Updated code to follow more recent specs, added compatibility flag
Daniel Veillard260a68f1998-08-13 03:39:55 +0000228- Better error handling, use a dedicated, overridable error
229 handling function.
230- Support for CDATA.
231- Keep track of line numbers for better error reporting.
232- Support for PI (SAX one).
Daniel Veillardad8f99d2000-01-15 14:20:03 +0000233- Support for Comments (bad, should be in ASAP, they are parsed
234 but not stored), should be configurable.
235- Improve the support of entities on save (+SAX).
Daniel Veillard01791d51998-07-24 19:24:09 +0000236