CVS log for Amaya/amaya/f/fetchXMLname_f.h

Up to [Public] / Amaya / amaya / f

Request diff between arbitrary revisions

Keyword substitution: kv
Default branch: MAIN

char* => const char*
(Regis patches + many many others)

When a document has a doctype that accepts Math entities, Amaya generates
these entities even if they are not included within a Math element.

Irene

Too many entities are now generated.

Irene

Amaya never generated HTML entities.
Irene

Add thot schema for templates.

A. Tollenaere

Avoid to export alphanumeric entities when there is no DocType.
Irene

Remove "#ifndef __STDC__" specific prototypes generation (make proto).
Now fuctions's prototypes are by default __STDC__ compliant (even if __STDC__ flag is not defined).
[ This ugly change is necessary for wxWindows port on WINDOWS platform. ]

S. GULLY

This is the Amaya C++ version : ie amaya uses now g++ (c++ version of gcc)

WARNING:
Before compilation do :
    - On debian just install g++ : "apt-get install g++"
    - A cvs update into libwww directory because there is a expat header to update.
    - "rm -rf *"
    - "../configure [options...]"

S. GULLY

Bug fix in xml : a new xml nature was added each time a synchronization
 was made thru the source view.
Laurent

The CSS parser didn't handle correctly enumerated attribute values.
Irene

- Allow annotation bodies to be of type text, svg, and mathml, as well
  as XHTML.
- Send the correct content type when publishing an annotation (we were
  always using text/html)

-jose

Avoid to open a new window when the user click with the right button a
link "mailto:" or a link to a file which won't be displayed by Amaya
Irene

Improvements on generic-XML version :
- Synchronization formatted-view / source-view
- Dynamic allocation of the attributes lists for the extension schemas
- Modification of some presentation rules to consider an XML element as Inline (in P language sense) as soon as it contains a text leaf
- Include specific XML events to the general mechanism
Laurent

Renaming GraphML by SVG
Don't forget to relaunch "../configure", "make clean", "make depend"

Irene

- Initial work for the futur XML-generic version.
- Support of SVG images in (X)HTML documents (not yet completely finished).
Laurent

Irene

Removing CHAR_T, ustr, ufopen, usprintf from these files.
Irene

ISO-latin-1 characters are now exported as entities when the document
encoding is ASCII, as ISO-latin-1 characters when the document encoding is
ISO_8859_1, as UTF8 characters when the document encoding is UTF_8.
Irene

Add a function that accedes to an entity by its code and returns its name.
Laurent

Unify the treatment of XML entities (the exhaustive table for MathML entities is not yet connected).
LC

Block elements are not allowed within an anchor (XML parser).
LC

Test if an unknown attribute in a XML document is caused by the current profile
LC

Reformulation of the XML parsing errors.
Four error types are defined
  1. Error about encoding (unknown / undefined)
  2. Document not well-formed
  3. Unknown tags/attributes
  4. Unknown tags/attributes for the current profile

The treatment of the attributes is not yet finished

LC

Addition of an information in the element mapping tables to indicate if
an element is a character-level one or not.
Addition of 'Input' elements in the HTML mapping table.
LC

Improving the control on the XHTML conformance
Irene

Improving the error awareness in the XML parser.
Preparing the control on attributes.
Irene

Mixing the HTML parser and the XML parser
Irene

Filtering XMHTL-basic elements and attributes
Irene

Transfert of the "Mapping XHTML Element" treatment from XmlNameSpace.c to fetchXMLname.c
LC

Creating new modules to implement XLink
VQ

Amaya crashed when it read an invalid XML document
Irene

Unicode.
New Features:
- html2thot.c
  * GetNextChar allows to get the next Wide Char from a file or a buffer.
  * ParseHTMLError writes a multibyte string into the output file. This file
    is supposed to be UTF 8 for the moment.

- config.c
  New function: GetNextWideCharFromMultibyteString allows to read the right number
  bytes into a string coded in char* (bytes) and gives a Wide Char.
  This function is used in getFirstWord, getSecondWord, SingleWord, getStringAfterColon.
  It also will be used for the HTML parser.


- fileaccess.c
  New function:
  * TtaGetWideChar reads the right number of bytes to give a wide char
  * TtaWriteWideChar transforms a wide char into the right number of bytes and writes
    the bytes into a file.
  Modified functoins:
  * TtaWriteDocIdent
  * TtaCheckDirectory
  * TtaReadOpen

- uio.c
  New function: ufgets allows to read the right number of bytes from a given
  file to give a Wide character string which length is N (parameter of ufgets).
  If N > NB (number of wide characters remaining) then ufgets returns a
  Wide Character String whitch length is NB.
  if EOF then ufgets returns NULL.

- css.c
  Style sheets are handled as Wide Characters now.
  LoadStyleSheets is adapted to handle Wide Character Strings.

- mesage.c
  TtaGetMessageTable allowd to fill a table of Wide Character
  Strings from a file.

- registry.c
  New function WGetEnv.
  Windows provides _wgetenv to get system env vars coded in Wide Characters
  * WGetEnv does the samething on Unix.

  Modified functions:
  * TtaGetEnvString
  * TtaGetDefEnvString

- platforms.c
  Modified function: TtaFileExist.

Work still to do:
- interface with libwww
- interface with motif dialogs.

Ramzi.

Unicode.

Ramzi

Make Amaya able to print with CSS and fix some bugs in the CSS parser:
- Amaya didn't detect complex selectors and managed them as simple selectors
- Amaya tried to interpret CSS comment within CSS rules.
Irene

Creation of 2 new modules to collect parsing functions shared by
the HTML, XML and CSS parsers
Irene