------------------------------------------------------------------- Charater font encoding in xcircuit ------------------------------------------------------------------- Xcircuit supports ISO-Latin2 encoding (Eastern Europe) since version 2.0a9. This support was extended to ISO-Latin5 (Turkish) in version 2.0a10 (6/14/1999), and Cyrillic (ISO-8859-5) in version 3.9.64 (3/29/2017). Between 1999 and 2017, projects like fontforge made it much easier to obtain and use Type1 PostScript fonts. While the fonts are not generally available on printers, generally files are converted to PDF for printing, and ps2pdf takes care of inserting the proper font glyphs, and most Linux systems have URW fonts installed, or at least they are easy to obtain. In the latest version the above encodings can be reached simply by selecting Text->Encoding->(name of encoding). As of this writing, Cyrillic fonts are not available in xcircuit drawn vector form, but are available from the font rendering engine if xcircuit is compiled with Cairo graphics. Font encodings using standard glyph names (e.g., from URW fonts and others) are kept in text files in the runtime directories and are inserted into the output so that no further manipulation is needed to view the proper text characters. The instructions below are generally deprecated; use of "ogonkify" works only on ISO-Latin-X encodings (not, for example, Cyrillic). It is needed to generate the correct glyphs for sending PostScript output directly to a printer. However, as noted above, modern usage is generally to first convert to PDF format, for which ogonkify is not needed. Version 3.6.36 corrects a long-existing mistake in which only X11 keysyms 0-255 are accepted as keyboard-entry. By converting the X11 keysyms to keycodes, non-ASCII-encoded characters can be typed directly into XCircuit from the keyboard. It would be nice, though, if XCircuit would auto-detect the expected encoding from the keysym and insert the proper font encoding switch, loading fonts if necessary. When creating text in an encoding other than PostScript "standard", be sure to select "Text->Encoding->(encoding name)" from the menu. This can be enabled by default simply by having a .xcircuitrc file in your home directory containing a line such as: label encoding ISOLatin2 ------------------------------------------------------------------- 1) The best practice is to convert all xcircuit output to PDF using, typically, ps2pdf, and then ignore all the instruction below, since the PDF file contains all font information needed for printing. 2) To get proper PostScript output on a printer (when sending PostScript directly---this does not apply to PDF files!), you will need to have the (separate) program "ogonkify", found at: http://www.dcs.ed.ac.uk/home/jec/ogonkify/ ftp://ftp.dcs.ed.ac.uk/pub/jec/programs "ogonkify" is necessary because PostScript does not contain ISO-Latin2 encoding by default. The accents such as the caron and ogonek exist in the font definition, but the correct alignments for putting them together need to be provided. Ogonkify does this best. 3) Every file containing ISO-Latin2 text must be postprocessed through "ogonkify" using the following command, before it will print correctly on a PostScript printer: ogonkify -e L2 -ATH file.ps | perl -ne \ 's/StandardEncoding/ISOLatin2Encoding/g;\ s/ exch defi/ISO2 exch defi/g;print' > file.out.ps Switches: -A does Courier, -T does Times-Roman, and -H does Helvetica; use only what's necessary to keep the output file smaller. "-e L2" signifies Latin2 encoding; this is not strictly necessary as ogonkify defaults to L2. All other Latin encodings require the -e switch. Note that the output is piped through a couple of perl commands. This may seem overly complicated; the first substitution corrects a bug in ogonkify output (which may disappear from releases of ogonkify in the future). The second substitution allows xcircuit to specify the ISO-Latin2 encodings separately from ISO-Latin1 and Adobe Standard encodings, so that any and all encodings can be "mixed and matched" in the same file and even the same label. 4) In a nutshell: xcircuit defines each ISO-Latin2 font name as the original name followed by the string "ISO2". Xcircuit can then load the file and interpret the font encodings unambiguously. However, the new ISO-Latin2 font names are not defined by PostScript. Running "ogonkify" in the manner shown above will add a true PostScript definition for each of the ISO-Latin2 fonts. Without "ogonkify", a PostScript printer or interpreter will either produce an error or else substitute a default font for the unrecognized one. 5) The "ogonkification" of the file will not affect xcircuit's ability to read it back into the program. xcircuit will continue to display the ISO-Latin2 text correctly regardless of whether or not the file has been run through "ogonkify". However, the header material inserted by "ogonkify" will be lost every time xcircuit writes the file back out. 6) Other ISO-LatinX encodings: Currently ISO-Latin5 is supported, and additional encodings will be created as I get requests for them. Follow the instructions above for ISO-Latin2, but replace all occurences of "2" with "5". e.g., ogonkify -e L5 -ATH file.ps | perl -ne \ 's/StandardEncoding/ISOLatin5Encoding/g;\ s/ exch defi/ISO5 exch defi/g;print' > file.out.ps and add lines % times_romaniso5.xfe loadfontencoding % courieriso5.xfe loadfontencoding % helveticaiso5.xfe loadfontencoding to psfiles/builtins.lps (or /usr/local/lib/xcircuit-2.0/builtins.lps, if already installed).