Daniel Veillard | 01791d5 | 1998-07-24 19:24:09 +0000 | [diff] [blame] | 1 | |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 2 | TODO for the XML parser and stuff: |
| 3 | ================================== |
Daniel Veillard | 01791d5 | 1998-07-24 19:24:09 +0000 | [diff] [blame] | 4 | |
Daniel Veillard | 75c2921 | 2000-03-14 18:33:47 +0000 | [diff] [blame] | 5 | TODO: |
| 6 | ===== |
| 7 | |
| 8 | - extend validity checks to go through entities content instead of |
| 9 | just labelling them PCDATA |
| 10 | - add support for the trick from Henry conf/sun/valid/empty.xml |
| 11 | - Correct standalone checking/emitting (hard) |
| 12 | 2.9 Standalone Document Declaration |
| 13 | - URI checkings (no fragments) rfc2396.txt |
| 14 | - Better checking of external parsed entities TAG 1234 |
| 15 | - Find way of representing PERefs in the Dtd so that %entity; can |
| 16 | be saved back. |
| 17 | - Go through erratas and do the cleanup. |
| 18 | http://www.w3.org/XML/xml-19980210-errata ... bummmer |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 19 | |
| 20 | TODO: |
| 21 | ===== |
| 22 | |
Daniel Veillard | 75c2921 | 2000-03-14 18:33:47 +0000 | [diff] [blame] | 23 | - Get OASIS testsuite to a more friendly result, check all the results |
| 24 | once stable. |
| 25 | http://xmlsoft.org/conf/result.html |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 26 | |
Daniel Veillard | 3dd82e7 | 2000-03-20 11:48:04 +0000 | [diff] [blame^] | 27 | - Optimization of tag strings allocation ? |
Daniel Veillard | 75c2921 | 2000-03-14 18:33:47 +0000 | [diff] [blame] | 28 | |
| 29 | - maintain coherency of namespace when doing cut'n paste operations |
| 30 | => the functions are coded, but need testing |
| 31 | |
Daniel Veillard | 3dd82e7 | 2000-03-20 11:48:04 +0000 | [diff] [blame^] | 32 | - function to rebuild the ID table |
| 33 | - functions to rebuild the DTD hash tables (after DTD changes). |
Daniel Veillard | 75c2921 | 2000-03-14 18:33:47 +0000 | [diff] [blame] | 34 | |
Daniel Veillard | 75c2921 | 2000-03-14 18:33:47 +0000 | [diff] [blame] | 35 | |
Daniel Veillard | 75c2921 | 2000-03-14 18:33:47 +0000 | [diff] [blame] | 36 | |
Daniel Veillard | 75c2921 | 2000-03-14 18:33:47 +0000 | [diff] [blame] | 37 | |
| 38 | EXTENSIONS: |
| 39 | =========== |
Daniel Veillard | 3dd82e7 | 2000-03-20 11:48:04 +0000 | [diff] [blame^] | 40 | |
| 41 | - Check attribute normalization especially xmlGetProp() |
| 42 | - Validity checking problems for NOTATIONS attributes |
| 43 | - Validity checking problems for ENTITY ENTITIES attributes |
| 44 | - dynamically adapt the alloc entry point to use g_alloc()/g_free() |
| 45 | if the programmer wants it: |
| 46 | - use xmlMemSetup() to reset the routines used. |
| 47 | - Parsing of a well balanced chunk |
Daniel Veillard | 75c2921 | 2000-03-14 18:33:47 +0000 | [diff] [blame] | 48 | - URI module: validation, base, etc ... |
| 49 | - Tools to produce man pages from the SGML docs. |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 50 | - Finish XPath |
| 51 | => attributes addressing troubles |
| 52 | => defaulted attributes handling |
| 53 | => namespace axis ? |
| 54 | |
| 55 | - Add Xpointer recognition/API |
| 56 | |
| 57 | - Add Xlink recognition/API |
| 58 | => started adding an xlink.[ch] with a unified API for XML and HTML. |
| 59 | |
| 60 | - Implement XSLT |
| 61 | => seems that someone volunteered ?!? |
| 62 | |
| 63 | - Implement XSchemas |
| 64 | |
| 65 | - O2K parsing; |
| 66 | => this is a somewhat ugly mix of HTML and XML, adding a specific |
| 67 | routine in the comment parsing code of HTML and plug the XML |
| 68 | parsing one in-there should not be too hard. Key point is to get |
| 69 | XSL to transform all this to something decent ... |
| 70 | |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 71 | - extend the shell with: |
| 72 | - edit |
| 73 | - load/save |
| 74 | - mv (yum, yum, but it's harder because directories are ordered in |
| 75 | our case, mvup and mvdown would be required) |
| 76 | |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 77 | - Add HTML validation using the XHTML DTD |
| 78 | - problem: do we want to keep and maintain the code for handling |
| 79 | DTD/System ID cache directly in libxml ? |
| 80 | |
| 81 | - Add a DTD cache prefilled with xhtml DTDs and entities and a program to |
| 82 | manage them -> like the /usr/bin/install-catalog from SGML |
| 83 | right place seems $datadir/xmldtds |
| 84 | |
| 85 | - turn tester into a generic program xml-test installed with xml-devel |
| 86 | |
| 87 | - Add output to XHTML in case of HTML documents. |
| 88 | |
Daniel Veillard | e3d88ef | 2000-01-24 13:55:06 +0000 | [diff] [blame] | 89 | |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 90 | Done: |
| 91 | ===== |
| 92 | |
Daniel Veillard | 75c2921 | 2000-03-14 18:33:47 +0000 | [diff] [blame] | 93 | - Save Dtds using the children list instead of dumping the tables, |
| 94 | order is preserved as well as comments and PIs |
| 95 | - Wrote a notice of changes requires to go from 1.x to 2.x |
| 96 | - make sure that all SAX callbacks are disabled if a WF error is detected |
| 97 | - checking/handling of newline normalization |
| 98 | http://localhost/www.xml.com/axml/target.html#sec-line-ends |
| 99 | - correct checking of '&' '%' on entities content. |
| 100 | - checking of PE/Nesting on entities declaration |
| 101 | - checking/handling of xml:space |
| 102 | - checking done. |
| 103 | - handling done, not well tested |
| 104 | - Language identification code, productions [33] to [38] |
| 105 | => done, the check has been added and report WFness errors |
| 106 | - Conditional sections in DTDs [61] to [65] |
| 107 | => should this crap be really implemented ??? |
| 108 | => Yep OASIS testsuite uses them |
| 109 | - Allow parsed entities defined in the internal subset to override |
| 110 | the ones defined in the external subset (DtD customization). |
| 111 | => This mean that the entity content should be computed only at |
| 112 | use time, i.e. keep the orig string only at parse time and expand |
| 113 | only when referenced from the external subset :-( |
| 114 | Needed for complete use of most DTD from Eve Maler |
| 115 | - Add regression tests for all WFC errors |
| 116 | => did some in test/WFC |
| 117 | => added OASIS testsuite routines |
| 118 | http://xmlsoft.org/conf/result.html |
| 119 | |
| 120 | - I18N: http://wap.trondheim.com/vaer/index.phtml is not XML and accepted |
| 121 | by the XML parser, UTF-8 should be checked when there is no "encoding" |
| 122 | declared ! |
| 123 | - Support for UTF-8 and UTF-16 encoding |
| 124 | => added some convertion routines provided by Martin Durst |
| 125 | patched them, got fixes from @@@ |
| 126 | I plan to keep everything internally as UTF-8 (or ISO-Latin-X) |
| 127 | this is slightly more costly but more compact, and recent processors |
| 128 | efficiency is cache related. The key for good performances is keeping |
| 129 | the data set small, so will I. |
| 130 | => the new progressive reading routines call the detection code |
| 131 | is enabled, tested the ISO->UTF-8 stuff |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 132 | - External entities loading: |
| 133 | - allow override by client code |
| 134 | - make sure it is alled for all external entities referenced |
| 135 | Done, client code should use xmlSetExternalEntityLoader() to set |
| 136 | the default loading routine. It will be called each time an external |
| 137 | entity entity resolution is triggered. |
| 138 | - maintain ID coherency when removing/changing attributes |
| 139 | The function used to deallocate attributes now check for it being an |
| 140 | ID and removes it from the table. |
| 141 | - push mode parsing i.e. non-blocking state based parser |
| 142 | done, both for XML and HTML parsers. Use xmlCreatePushParserCtxt() |
| 143 | and xmlParseChunk() and html counterparts. |
| 144 | The tester program now has a --push option to select that parser |
| 145 | front-end. Douplicated tests to use both and check results are similar. |
| 146 | |
| 147 | - Most of XPath, still see some troubles and occasionnal memleaks. |
| 148 | - an XML shell, allowing to traverse/manipulate an XML document with |
| 149 | a shell like interface, and using XPath for the anming syntax |
| 150 | - use of readline and history added when available |
| 151 | - the shell interface has been cleanly separated and moved to debugXML.c |
| 152 | - HTML parser, should be fairly stable now |
| 153 | - API to search the lang of an attribute |
| 154 | - Collect IDs at parsing and maintain a table. |
| 155 | PBM: maintain the table coherency |
| 156 | PBM: how to detect ID types in absence of DtD ! |
| 157 | - Use it for XPath ID support |
| 158 | - Add validity checking |
| 159 | Should be finished now ! |
| 160 | - Add regression tests with entity substitutions |
| 161 | |
| 162 | - External Parsed entities, either XML or external Subset [78] and [79] |
| 163 | parsing the xmllang DtD now works, so it should be sufficient for |
| 164 | most cases ! |
| 165 | |
| 166 | - progressive reading. The entity support is a first step toward |
Daniel Veillard | 260a68f | 1998-08-13 03:39:55 +0000 | [diff] [blame] | 167 | asbtraction of an input stream. A large part of the context is still |
| 168 | located on the stack, moving to a state machine and putting everyting |
| 169 | in the parsing context should provide an adequate solution. |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 170 | => Rather than progressive parsing, give more power to the SAX-like |
| 171 | interface. Currently the DOM-like representation is built but |
| 172 | => it should be possible to define that only as a set of SAX callbacks |
| 173 | and remove the tree creation from the parser code. |
| 174 | DONE |
Daniel Veillard | 01791d5 | 1998-07-24 19:24:09 +0000 | [diff] [blame] | 175 | |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 176 | - DOM support, instead of using a proprietary in memory |
| 177 | format for the document representation, the parser should |
| 178 | call a DOM API to actually build the resulting document. |
| 179 | Then the parser becomes independent of the in-memory |
| 180 | representation of the document. Even better using RPC's |
| 181 | the parser can actually build the document in another |
| 182 | program. |
| 183 | => Work started, now the internal representation is by default |
| 184 | very near a direct DOM implementation. The DOM glue is implemented |
| 185 | as a separate module. See the GNOME gdome module. |
| 186 | |
Daniel Veillard | 01791d5 | 1998-07-24 19:24:09 +0000 | [diff] [blame] | 187 | - C++ support : John Ehresman <jehresma@dsg.harvard.edu> |
| 188 | - Updated code to follow more recent specs, added compatibility flag |
Daniel Veillard | 260a68f | 1998-08-13 03:39:55 +0000 | [diff] [blame] | 189 | - Better error handling, use a dedicated, overridable error |
| 190 | handling function. |
| 191 | - Support for CDATA. |
| 192 | - Keep track of line numbers for better error reporting. |
| 193 | - Support for PI (SAX one). |
Daniel Veillard | ad8f99d | 2000-01-15 14:20:03 +0000 | [diff] [blame] | 194 | - Support for Comments (bad, should be in ASAP, they are parsed |
| 195 | but not stored), should be configurable. |
| 196 | - Improve the support of entities on save (+SAX). |
Daniel Veillard | 01791d5 | 1998-07-24 19:24:09 +0000 | [diff] [blame] | 197 | |