Thank you very much for your response, and sorry for the delay... Juliusz Chroboczek wrote:
> EM> For example, Netscape 2/3/4 agreed to download only fonts which it > EM> "recognized" their encodings. And "8859-8" was not among them > EM> (even by using "User-Defined"). > > What are you speaking about? Netscape 2/3/4 has no support whatsoever > for downloading Type 1 fonts. It worked great for many people. Since the release of a font server for X (even before it was called XFS, but just FS), Type1 has worked transparently with all of the X applications. The only problem was that some applications agreed, for example, to use only fonts that their name ends with "8859-1", etc. But this problem was not specific to Type1 only, but also to PCF. > EM> In any case, we didn't discuss this issue, but another thing: > EM> The DICT encoding attribute of the Type1 files. > > What is the ``DICT encoding attribute''? Are you speaking of the font > dictionary's ``/Encoding'' entry? Exactly. > XFree86 4.0 will respect its value > if the font is declared as ``-adobe-fontspecific''; the way this works > has not changed since X11R5. This is exactly the problem. Real world fonts can't end with ``-adobe-fontspecific'', because each font must have its own name, encoding, etc., that will be also respected by applications (a similar problem to what I wrote above about apps that refuse to use fonts with specific encodings, such as "8859-8"). So font coders must use names like "agrave" for ASCII code 224 (decimal), for example. Sometime, probably between XF86-3 and 4.0, these names changed. I don't have any problem to change names. Even 40-50 names of chars. My only problem is that now I'll have to maintain two sets of fonts, because all the older or other implementations of X, don't respect these new names. > EM> I just know that hundreds, or even thousands, UNIX and Linux > EM> users, have used these fonts for years. Until they stopped to > EM> work suddenly > > X11 has been broken w.r.t. international Type 1 fonts since X11R5. > XFree86 4.0 fixes support for such fonts. > > Unfortunately, a number of fonts have been designed to work around the > limitations of X11R5. Workarounds for using such fonts in XFree86 4.0 > are indicated in the README.fonts document. Are you talking about a document which is included in the XFree86 package? If not, where can I find it? > EM> a long-years-standard that STOPPED to work. > > No. You're talking about a well-established bug workaround that is no > longer needed. > > I am sorry that this fix broke compatibility; we did discuss ways of > being correct while preserving compatibility, but we didn't find a > satisfactory solution, which is why we decided to just do the right > thing. > > The glyph names used by the Type 1 backend are defined by an Adobe > standard, with a few changes for compatibility with X11R5 (e.g. U+0027 > maps to a closing quotation mark rather than a vertical apostrophe, > U+00B5 maps to "mu1" rather than "mu"). All the local changes are in > the ASCII (``Basic Latin'') and Latin-1 ranges, as this is the subset > of Unicode supported by the Type 1 backend in X11R5. Thanks for the explanation. > EM> And this is not the issue of this thread, but just the issue of why > EM> "agrave" and its friends are not supported anymore. > > Could you clarify this? /agrave and friends are perfectly supported > in all encodings that include the associated character -- but as far > as I know, there's no LATIN SMALL LETTER A WITH GRAVE in ISO 8859-8. I'll try to summarize the problem: For years, Hebrew fonts used the name "agrave" for the ASCII code 224, aacute for 225, etc. Of course, there is no A with GRAVE in Hebrew; but there was no other way to relate to Alef or to Beit, so the names which were recognized by the renderers, were used. You could use 8859-8 for the font (the normal encoding), or 8859-1 (in case you wanted cheat a special app, like Netscape, to agree to use this font). In both cases, the "agrave" etc. worked. Recently, the fonts stopped to work. In both cases (-8 and -1). Using "afii57664" and its friends, resolved the problem, but broke the fonts for older and other implementations. Thank you very much for your response! -- Eli Marmor [EMAIL PROTECTED] CTO, Founder Netmask (El-Mar) Internet Technologies Ltd. __________________________________________________________ Tel.: +972-9-766-1020 8 Yad-Harutzim St. Fax.: +972-9-766-1314 P.O.B. 7004 Mobile: +972-50-23-7338 Kfar-Saba 44641, Israel _______________________________________________ Fonts mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/fonts
