Debian Codegraph @

texmf-dist/doc/luatex/base/luatex-fontloader.tex

raw
1 % language=uk
2
3 \environment luatex-style
4
5 \startcomponent luatex-fontloader
6
7 \startchapter[reference=fontloader,title={The fontloader}]
8
9 \topicindex {fonts+loading}
10
11 The fontloader library is sort of independent of the rest in the sense that it
12 can load font into a \LUA\ table that then can be converted into a table suitable
13 for \TEX. The library is an adapted subset of \FONTFORGE\ and as such gives a
14 similar view on a font (which has advantages when you want to debug). We will not
15 discuss \OPENTYPE\ in detail here as the \MICROSOFT\ website offers enough
16 information about it. The tables returned by the loader are not that far from the
17 standard. We have no plans to extend the loader (it may even become an external
18 module at some time).
19
20 \startsection[title={Getting quick information on a font}][library=fontloader]
21
22 \topicindex {fonts+information}
23
24 \libindex{info}
25
26 When you want to locate font by name you need some basic information that is
27 hidden in the font files. For that reason we provide an efficient helper that
28 gets the basic information without loading all of the font. Normally this helper
29 is used to create a font (name) database.
30
31 \startfunctioncall
32 <table> info =
33 fontloader.info(<string> filename)
34 \stopfunctioncall
35
36 This function returns either \type {nil}, or a \type {table}, or an array of
37 small tables (in the case of a \TRUETYPE\ collection). The returned table(s) will
38 contain some fairly interesting information items from the font(s) defined by the
39 file:
40
41 \starttabulate[|l|l|p|]
42 \DB key \BC type \BC explanation \NC \NR
43 \TB
44 \NC \type{fontname} \NC string \NC the \POSTSCRIPT\ name of the font\NC \NR
45 \NC \type{fullname} \NC string \NC the formal name of the font\NC \NR
46 \NC \type{familyname} \NC string \NC the family name this font belongs to\NC \NR
47 \NC \type{weight} \NC string \NC a string indicating the color value of the font\NC \NR
48 \NC \type{version} \NC string \NC the internal font version\NC \NR
49 \NC \type{italicangle} \NC float \NC the slant angle\NC \NR
50 \NC \type{units_per_em} \NC number \NC 1000 for \POSTSCRIPT-based fonts, usually 2048 for \TRUETYPE\NC \NR
51 \NC \type{pfminfo} \NC table \NC (see \in{section}[fontloaderpfminfotable])\NC \NR
52 \LL
53 \stoptabulate
54
55 Getting information through this function is (sometimes much) more efficient than
56 loading the font properly, and is therefore handy when you want to create a
57 dictionary of available fonts based on a directory contents.
58
59 \stopsection
60
61 \startsection[title={Loading an \OPENTYPE\ or \TRUETYPE\ file}][library=fontloader]
62
63 \topicindex {\OPENTYPE}
64 \topicindex {\TRUETYPE}
65
66 \libindex{open}
67 \libindex{close}
68 \libindex{to_table}
69
70 If you want to use an \OPENTYPE\ font, you have to get the metric information
71 from somewhere. Using the \type {fontloader} library, the simplest way to get
72 that information is thus:
73
74 \starttyping
75 function load_font (filename)
76 local metrics = nil
77 local font = fontloader.open(filename)
78 if font then
79 metrics = fontloader.to_table(font)
80 fontloader.close(font)
81 end
82 return metrics
83 end
84
85 myfont = load_font('/opt/tex/texmf/fonts/data/arial.ttf')
86 \stoptyping
87
88 The main function call is
89
90 \startfunctioncall
91 <userdata> f, <table> w = fontloader.open(<string> filename)
92 <userdata> f, <table> w = fontloader.open(<string> filename, <string> fontname)
93 \stopfunctioncall
94
95 The first return value is a userdata representation of the font. The second
96 return value is a table containing any warnings and errors reported by fontloader
97 while opening the font. In normal typesetting, you would probably ignore the
98 second argument, but it can be useful for debugging purposes.
99
100 For \TRUETYPE\ collections (when filename ends in 'ttc') and \DFONT\ collections,
101 you have to use a second string argument to specify which font you want from the
102 collection. Use the \type {fontname} strings that are returned by \type
103 {fontloader.info} for that.
104
105 To turn the font into a table, \type {fontloader.to_table} is used on the font
106 returned by \type {fontloader.open}.
107
108 \startfunctioncall
109 <table> f = fontloader.to_table(<userdata> font)
110 \stopfunctioncall
111
112 This table cannot be used directly by \LUATEX\ and should be turned into another
113 one as described in~\in {chapter} [fonts]. Do not forget to store the \type
114 {fontname} value in the \type {psname} field of the metrics table to be returned
115 to \LUATEX, otherwise the font inclusion backend will not be able to find the
116 correct font in the collection.
117
118 See \in {section} [fontloadertables] for details on the userdata object returned
119 by \type {fontloader.open} and the layout of the \type {metrics} table returned
120 by \type {fontloader.to_table}.
121
122 The font file is parsed and partially interpreted by the font loading routines
123 from \FONTFORGE. The file format can be \OPENTYPE, \TRUETYPE, \TRUETYPE\
124 Collection, \CFF, or \TYPEONE.
125
126 There are a few advantages to this approach compared to reading the actual font
127 file ourselves:
128
129 \startitemize
130
131 \startitem
132 The font is automatically re|-|encoded, so that the \type {metrics} table for
133 \TRUETYPE\ and \OPENTYPE\ fonts is using \UNICODE\ for the character indices.
134 \stopitem
135
136 \startitem
137 Many features are pre|-|processed into a format that is easier to handle than
138 just the bare tables would be.
139 \stopitem
140
141 \startitem
142 \POSTSCRIPT|-|based \OPENTYPE\ fonts do not store the character height and
143 depth in the font file, so the character boundingbox has to be calculated in
144 some way.
145 \stopitem
146
147 \stopitemize
148
149 A loaded font is discarded with:
150
151 \startfunctioncall
152 fontloader.close(<userdata> font)
153 \stopfunctioncall
154
155 \stopsection
156
157 \startsection[title={Applying a \quote{feature file}}][library=fontloader]
158
159 \libindex{apply_featurefile}
160
161 You can apply a \quote{feature file} to a loaded font:
162
163 \startfunctioncall
164 <table> errors = fontloader.apply_featurefile(<userdata> font, <string> filename)
165 \stopfunctioncall
166
167 A \quote {feature file} is a textual representation of the features in an
168 \OPENTYPE\ font. See
169
170 \starttyping
171 http://www.adobe.com/devnet/opentype/afdko/topic_feature_file_syntax.html
172 \stoptyping
173
174 and
175
176 \starttyping
177 http://fontforge.sourceforge.net/featurefile.html
178 \stoptyping
179
180 for a more detailed description of feature files.
181
182 If the function fails, the return value is a table containing any errors reported
183 by fontloader while applying the feature file. On success, \type {nil} is
184 returned.
185
186 \stopsection
187
188 \startsection[title={Applying an \quote{\AFM\ file}}][library=fontloader]
189
190 \topicindex {\TYPEONE}
191
192 \libindex{apply_afmfile}
193
194 You can apply an \quote {\AFM\ file} to a loaded font:
195
196 \startfunctioncall
197 <table> errors = fontloader.apply_afmfile(<userdata> font, <string> filename)
198 \stopfunctioncall
199
200 An \AFM\ file is a textual representation of (some of) the meta information
201 in a \TYPEONE\ font. See
202
203 \starttyping
204 ftp://ftp.math.utah.edu/u/ma/hohn/linux/postscript/5004.AFM_Spec.pdf
205 \stoptyping
206
207 for more information about \AFM\ files.
208
209 Note: If you \type {fontloader.open} a \TYPEONE\ file named \type {font.pfb},
210 the library will automatically search for and apply \type {font.afm} if it exists
211 in the same directory as the file \type {font.pfb}. In that case, there is no
212 need for an explicit call to \type {apply_afmfile()}.
213
214 If the function fails, the return value is a table containing any errors reported
215 by fontloader while applying the AFM file. On success, \type {nil} is returned.
216
217 \stopsection
218
219 \startsection[title={Fontloader font tables},reference=fontloadertables][library=fontloader]
220
221 \topicindex {fontloader+tables}
222
223 \libindex{fields}
224
225 As mentioned earlier, the return value of \type {fontloader.open} is a userdata
226 object. One way to have access to the actual metrics is to call \type
227 {fontloader.to_table} on this object, returning the table structure that is
228 explained in the following sections. In the following sections we will not
229 explain each field in detail. Most fields are self descriptive and for the more
230 technical aspects you need to consult the relevant font references.
231
232 It turns out that the result from \type {fontloader.to_table} sometimes needs
233 very large amounts of memory (depending on the font's complexity and size) so it
234 is possible to access the userdata object directly.
235
236 \startitemize
237 \startitem
238 All top|-|level keys that would be returned by \type {to_table()}
239 can also be accessed directly.
240 \stopitem
241 \startitem
242 The top|-|level key \quote {glyphs} returns a {\it virtual\/} array that
243 allows indices from \type {f.glyphmin} to (\type {f.glyphmax}).
244 \stopitem
245 \startitem
246 The items in that virtual array (the actual glyphs) are themselves also
247 userdata objects, and each has accessors for all of the keys explained in the
248 section \quote {Glyph items} below.
249 \stopitem
250 \startitem
251 The top|-|level key \quote {subfonts} returns an {\it actual} array of userdata
252 objects, one for each of the subfonts (or nil, if there are no subfonts).
253 \stopitem
254 \stopitemize
255
256 A short example may be helpful. This code generates a printout of all
257 the glyph names in the font \type {PunkNova.kern.otf}:
258
259 \starttyping
260 local f = fontloader.open('PunkNova.kern.otf')
261 print (f.fontname)
262 local i = 0
263 if f.glyphcnt > 0 then
264 for i=f.glyphmin,f.glyphmax do
265 local g = f.glyphs[i]
266 if g then
267 print(g.name)
268 end
269 i = i + 1
270 end
271 end
272 fontloader.close(f)
273 \stoptyping
274
275 In this case, the \LUATEX\ memory requirement stays below 100MB on the test
276 computer, while the internal structure generated by \type {to_table()} needs more
277 than 2GB of memory (the font itself is 6.9MB in disk size).
278
279 Only the top|-|level font, the subfont table entries, and the glyphs are virtual
280 objects, everything else still produces normal \LUA\ values and tables.
281
282 If you want to know the valid fields in a font or glyph structure, call the \type
283 {fields} function on an object of a particular type (either glyph or font):
284
285 \startfunctioncall
286 <table> fields = fontloader.fields(<userdata> font)
287 <table> fields = fontloader.fields(<userdata> font_glyph)
288 \stopfunctioncall
289
290 For instance:
291
292 \startfunctioncall
293 local fields = fontloader.fields(f)
294 local fields = fontloader.fields(f.glyphs[0])
295 \stopfunctioncall
296
297 \stopsection
298
299 \startsection[title={Table types}][library=fontloader]
300
301 \startsubsection[title={The main table}]
302
303 The top|-|level keys in the returned table are (the explanations in this part of
304 the documentation are not yet finished):
305
306 \starttabulate[|l|l|p|]
307 \DB key \BC type \NC explanation \NC \NR
308 \TB
309 \NC \type{table_version} \NC string \NC indicates the metrics version (currently~0.3)\NC \NR
310 \NC \type{fontname} \NC string \NC \POSTSCRIPT\ font name\NC \NR
311 \NC \type{fullname} \NC string \NC official (human-oriented) font name\NC \NR
312 \NC \type{familyname} \NC string \NC family name\NC \NR
313 \NC \type{weight} \NC string \NC weight indicator\NC \NR
314 \NC \type{copyright} \NC string \NC copyright information\NC \NR
315 \NC \type{filename} \NC string \NC the file name\NC \NR
316 \NC \type{version} \NC string \NC font version\NC \NR
317 \NC \type{italicangle} \NC float \NC slant angle\NC \NR
318 \NC \type{units_per_em} \NC number \NC 1000 for \POSTSCRIPT-based fonts, usually 2048 for \TRUETYPE\NC \NR
319 \NC \type{ascent} \NC number \NC height of ascender in \type {units_per_em}\NC \NR
320 \NC \type{descent} \NC number \NC depth of descender in \type {units_per_em}\NC \NR
321 \NC \type{upos} \NC float \NC \NC \NR
322 \NC \type{uwidth} \NC float \NC \NC \NR
323 \NC \type{uniqueid} \NC number \NC \NC \NR
324 \NC \type{glyphs} \NC array \NC \NC \NR
325 \NC \type{glyphcnt} \NC number \NC number of included glyphs\NC \NR
326 \NC \type{glyphmax} \NC number \NC maximum used index the glyphs array\NC \NR
327 \NC \type{glyphmin} \NC number \NC minimum used index the glyphs array\NC \NR
328 \NC \type{notdef_loc} \NC number \NC location of the \type {.notdef} glyph
329 or \type {-1} when not present \NC \NR
330 \NC \type{hasvmetrics} \NC number \NC \NC \NR
331 \NC \type{onlybitmaps} \NC number \NC \NC \NR
332 \NC \type{serifcheck} \NC number \NC \NC \NR
333 \NC \type{isserif} \NC number \NC \NC \NR
334 \NC \type{issans} \NC number \NC \NC \NR
335 \NC \type{encodingchanged} \NC number \NC \NC \NR
336 \NC \type{strokedfont} \NC number \NC \NC \NR
337 \NC \type{use_typo_metrics} \NC number \NC \NC \NR
338 \NC \type{weight_width_slope_only} \NC number \NC \NC \NR
339 \NC \type{head_optimized_for_cleartype} \NC number \NC \NC \NR
340 \NC \type{uni_interp} \NC enum \NC \nod {unset}, \type {none}, \type {adobe},
341 \type {greek}, \type {japanese}, \type {trad_chinese},
342 \type {simp_chinese}, \type {korean}, \type {ams}\NC \NR
343 \NC \type{origname} \NC string \NC the file name, as supplied by the user\NC \NR
344 \NC \type{map} \NC table \NC \NC \NR
345 \NC \type{private} \NC table \NC \NC \NR
346 \NC \type{xuid} \NC string \NC \NC \NR
347 \NC \type{pfminfo} \NC table \NC \NC \NR
348 \NC \type{names} \NC table \NC \NC \NR
349 \NC \type{cidinfo} \NC table \NC \NC \NR
350 \NC \type{subfonts} \NC array \NC \NC \NR
351 \NC \type{commments} \NC string \NC \NC \NR
352 \NC \type{fontlog} \NC string \NC \NC \NR
353 \NC \type{cvt_names} \NC string \NC \NC \NR
354 \NC \type{anchor_classes} \NC table \NC \NC \NR
355 \NC \type{ttf_tables} \NC table \NC \NC \NR
356 \NC \type{ttf_tab_saved} \NC table \NC \NC \NR
357 \NC \type{kerns} \NC table \NC \NC \NR
358 \NC \type{vkerns} \NC table \NC \NC \NR
359 \NC \type{texdata} \NC table \NC \NC \NR
360 \NC \type{lookups} \NC table \NC \NC \NR
361 \NC \type{gpos} \NC table \NC \NC \NR
362 \NC \type{gsub} \NC table \NC \NC \NR
363 \NC \type{mm} \NC table \NC \NC \NR
364 \NC \type{chosenname} \NC string \NC \NC \NR
365 \NC \type{macstyle} \NC number \NC \NC \NR
366 \NC \type{fondname} \NC string \NC \NC \NR
367 %NC \type{design_size} \NC number \NC \NC \NR
368 \NC \type{fontstyle_id} \NC number \NC \NC \NR
369 \NC \type{fontstyle_name} \NC table \NC \NC \NR
370 %NC \type{design_range_bottom} \NC number \NC \NC \NR
371 %NC \type{design_range_top} \NC number \NC \NC \NR
372 \NC \type{strokewidth} \NC float \NC \NC \NR
373 \NC \type{mark_classes} \NC table \NC \NC \NR
374 \NC \type{creationtime} \NC number \NC \NC \NR
375 \NC \type{modificationtime} \NC number \NC \NC \NR
376 \NC \type{os2_version} \NC number \NC \NC \NR
377 \NC \type{math} \NC table \NC \NC \NR
378 \NC \type{validation_state} \NC table \NC \NC \NR
379 \NC \type{horiz_base} \NC table \NC \NC \NR
380 \NC \type{vert_base} \NC table \NC \NC \NR
381 \NC \type{extrema_bound} \NC number \NC \NC \NR
382 \NC \type{truetype} \NC boolean \NC signals a \TRUETYPE\ font \NC \NR
383 \LL
384 \stoptabulate
385
386 \stopsubsection
387
388 \startsubsection[title={\type {glyphs}}]
389
390 The \type {glyphs} is an array containing the per|-|character
391 information (quite a few of these are only present if non|-|zero).
392
393 \starttabulate[|l|l|p|]
394 \DB key \BC type \BC explanation \NC \NR
395 \TB
396 \NC \type{name} \NC string \NC the glyph name \NC \NR
397 \NC \type{unicode} \NC number \NC unicode code point, or -1 \NC \NR
398 \NC \type{boundingbox} \NC array \NC array of four numbers, see note below \NC \NR
399 \NC \type{width} \NC number \NC only for horizontal fonts \NC \NR
400 \NC \type{vwidth} \NC number \NC only for vertical fonts \NC \NR
401 \NC \type{tsidebearing} \NC number \NC only for vertical ttf/otf fonts, and only if non|-|zero \NC \NR
402 \NC \type{lsidebearing} \NC number \NC only if non|-|zero and not equal to boundingbox[1] \NC \NR
403 \NC \type{class} \NC string \NC one of "none", "base", "ligature", "mark", "component"
404 (if not present, the glyph class is \quote {automatic}) \NC \NR
405 \NC \type{kerns} \NC array \NC only for horizontal fonts, if set \NC \NR
406 \NC \type{vkerns} \NC array \NC only for vertical fonts, if set \NC \NR
407 \NC \type{dependents} \NC array \NC linear array of glyph name strings, only if nonempty\NC \NR
408 \NC \type{lookups} \NC table \NC only if nonempty \NC \NR
409 \NC \type{ligatures} \NC table \NC only if nonempty \NC \NR
410 \NC \type{anchors} \NC table \NC only if set \NC \NR
411 \NC \type{comment} \NC string \NC only if set \NC \NR
412 \NC \type{tex_height} \NC number \NC only if set \NC \NR
413 \NC \type{tex_depth} \NC number \NC only if set \NC \NR
414 \NC \type{italic_correction} \NC number \NC only if set \NC \NR
415 \NC \type{top_accent} \NC number \NC only if set \NC \NR
416 \NC \type{is_extended_shape} \NC number \NC only if this character is part of a math extension list \NC \NR
417 \NC \type{altuni} \NC table \NC alternate \UNICODE\ items \NC \NR
418 \NC \type{vert_variants} \NC table \NC \NC \NR
419 \NC \type{horiz_variants} \NC table \NC \NC \NR
420 \NC \type{mathkern} \NC table \NC \NC \NR
421 \LL
422 \stoptabulate
423
424 On \type {boundingbox}: The boundingbox information for \TRUETYPE\ fonts and
425 \TRUETYPE-based \OTF\ fonts is read directly from the font file.
426 \POSTSCRIPT-based fonts do not have this information, so the boundingbox of
427 traditional \POSTSCRIPT\ fonts is generated by interpreting the actual bezier
428 curves to find the exact boundingbox. This can be a slow process, so the
429 boundingboxes of \POSTSCRIPT-based \OTF\ fonts (and raw \CFF\ fonts) are
430 calculated using an approximation of the glyph shape based on the actual glyph
431 points only, instead of taking the whole curve into account. This means that
432 glyphs that have missing points at extrema will have a too|-|tight boundingbox,
433 but the processing is so much faster that in our opinion the tradeoff is worth
434 it.
435
436 The \type {kerns} and \type {vkerns} are linear arrays of small hashes:
437
438 \starttabulate[|l|l|p|]
439 \DB key \BC type \BC explanation \NC \NR
440 \TB
441 \NC \type{char} \NC string \NC \NC \NR
442 \NC \type{off} \NC number \NC \NC \NR
443 \NC \type{lookup} \NC string \NC \NC \NR
444 \LL
445 \stoptabulate
446
447 The \type {lookups} is a hash, based on lookup subtable names, with
448 the value of each key inside that a linear array of small hashes:
449
450 % TODO: fix this description
451
452 \starttabulate[|l|l|p|]
453 \DB key \BC type \BC explanation \NC \NR
454 \TB
455 \NC \type{type} \NC enum \NC \type {position}, \type {pair}, \type
456 {substitution}, \type {alternate}, \type
457 {multiple}, \type {ligature}, \type
458 {lcaret}, \cbk {kerning}, \type {vkerning},
459 \type {anchors}, \type {contextpos}, \type
460 {contextsub}, \type {chainpos}, \type
461 {chainsub}, \type {reversesub}, \type
462 {max}, \type {kernback}, \type {vkernback}
463 \NC \NR
464 \NC \type{specification} \NC table \NC extra data \NC \NR
465 \LL
466 \stoptabulate
467
468 For the first seven values of \type {type}, there can be additional
469 sub|-|information, stored in the sub-table \type {specification}:
470
471 \starttabulate[|l|l|p|]
472 \DB value \BC type \BC explanation \NC \NR
473 \TB
474 \NC \type{position} \NC table \NC a table of the \type {offset_specs} type \NC \NR
475 \NC \type{pair} \NC table \NC one string: \type {paired}, and an array of one
476 or two \type {offset_specs} tables: \type {offsets} \NC \NR
477 \NC \type{substitution} \NC table \NC one string: \type {variant} \NC \NR
478 \NC \type{alternate} \NC table \NC one string: \type {components} \NC \NR
479 \NC \type{multiple} \NC table \NC one string: \type {components} \NC \NR
480 \NC \type{ligature} \NC table \NC two strings: \type {components}, \type {char} \NC \NR
481 \NC \type{lcaret} \NC array \NC linear array of numbers \NC \NR
482 \LL
483 \stoptabulate
484
485 Tables for \type {offset_specs} contain up to four number|-|valued fields: \type
486 {x} (a horizontal offset), \type {y} (a vertical offset), \type {h} (an advance
487 width correction) and \type {v} (an advance height correction).
488
489 The \type {ligatures} is a linear array of small hashes:
490
491 \starttabulate[|l|l|p|]
492 \DB key \BC type \BC explanation \NC \NR
493 \TB
494 \NC \type{lig} \NC table \NC uses the same substructure as a single item in
495 the \type {lookups} table explained above \NC \NR
496 \NC \type{char} \NC string \NC \NC \NR
497 \NC \type{components} \NC array \NC linear array of named components \NC \NR
498 \NC \type{ccnt} \NC number \NC \NC \NR
499 \LL
500 \stoptabulate
501
502 The \type {anchor} table is indexed by a string signifying the anchor type, which
503 is one of:
504
505 \starttabulate[|l|l|p|]
506 \DB key \BC type \BC explanation \NC \NR
507 \TB
508 \NC \type{mark} \NC table \NC placement mark \NC \NR
509 \NC \type{basechar} \NC table \NC mark for attaching combining items to a base char \NC \NR
510 \NC \type{baselig} \NC table \NC mark for attaching combining items to a ligature \NC \NR
511 \NC \type{basemark} \NC table \NC generic mark for attaching combining items to connect to \NC \NR
512 \NC \type{centry} \NC table \NC cursive entry point \NC \NR
513 \NC \type{cexit} \NC table \NC cursive exit point \NC \NR
514 \LL
515 \stoptabulate
516
517 The content of these is a short array of defined anchors, with the
518 entry keys being the anchor names. For all except \type {baselig}, the
519 value is a single table with this definition:
520
521 \starttabulate[|l|l|p|]
522 \DB key \BC type \BC explanation \NC \NR
523 \TB
524 \NC \type{x} \NC number \NC x location \NC \NR
525 \NC \type{y} \NC number \NC y location \NC \NR
526 \NC \type{ttf_pt_index} \NC number \NC truetype point index, only if given \NC \NR
527 \LL
528 \stoptabulate
529
530 For \type {baselig}, the value is a small array of such anchor sets sets, one for
531 each constituent item of the ligature.
532
533 For clarification, an anchor table could for example look like this :
534
535 \starttyping
536 ['anchor'] = {
537 ['basemark'] = {
538 ['Anchor-7'] = { ['x']=170, ['y']=1080 }
539 },
540 ['mark'] ={
541 ['Anchor-1'] = { ['x']=160, ['y']=810 },
542 ['Anchor-4'] = { ['x']=160, ['y']=800 }
543 },
544 ['baselig'] = {
545 [1] = { ['Anchor-2'] = { ['x']=160, ['y']=650 } },
546 [2] = { ['Anchor-2'] = { ['x']=460, ['y']=640 } }
547 }
548 }
549 \stoptyping
550
551 Note: The \type {baselig} table can be sparse!
552
553 \stopsubsection
554
555 \startsubsection[title={\type {map}}]
556
557 The top|-|level map is a list of encoding mappings. Each of those is a table
558 itself.
559
560 \starttabulate[|l|l|p|]
561 \DB key \BC type \BC explanation \NC \NR
562 \TB
563 \NC \type{enccount} \NC number \NC \NC \NR
564 \NC \type{encmax} \NC number \NC \NC \NR
565 \NC \type{backmax} \NC number \NC \NC \NR
566 \NC \type{remap} \NC table \NC \NC \NR
567 \NC \type{map} \NC array \NC non|-|linear array of mappings\NC \NR
568 \NC \type{backmap} \NC array \NC non|-|linear array of backward mappings\NC \NR
569 \NC \type{enc} \NC table \NC \NC \NR
570 \LL
571 \stoptabulate
572
573 The \type {remap} table is very small:
574
575 \starttabulate[|l|l|p|]
576 \DB key \BC type \BC explanation \NC \NR
577 \TB
578 \NC \type{firstenc} \NC number \NC \NC \NR
579 \NC \type{lastenc} \NC number \NC \NC \NR
580 \NC \type{infont} \NC number \NC \NC \NR
581 \LL
582 \stoptabulate
583
584 The \type {enc} table is a bit more verbose:
585
586 \starttabulate[|l|l|p|]
587 \DB key \BC type \BC explanation \NC \NR
588 \TB
589 \NC \type{enc_name} \NC string \NC \NC \NR
590 \NC \type{char_cnt} \NC number \NC \NC \NR
591 \NC \type{char_max} \NC number \NC \NC \NR
592 \NC \type{unicode} \NC array \NC of \UNICODE\ position numbers\NC \NR
593 \NC \type{psnames} \NC array \NC of \POSTSCRIPT\ glyph names\NC \NR
594 \NC \type{builtin} \NC number \NC \NC \NR
595 \NC \type{hidden} \NC number \NC \NC \NR
596 \NC \type{only_1byte} \NC number \NC \NC \NR
597 \NC \type{has_1byte} \NC number \NC \NC \NR
598 \NC \type{has_2byte} \NC number \NC \NC \NR
599 \NC \type{is_unicodebmp} \NC number \NC only if non|-|zero\NC \NR
600 \NC \type{is_unicodefull} \NC number \NC only if non|-|zero\NC \NR
601 \NC \type{is_custom} \NC number \NC only if non|-|zero\NC \NR
602 \NC \type{is_original} \NC number \NC only if non|-|zero\NC \NR
603 \NC \type{is_compact} \NC number \NC only if non|-|zero\NC \NR
604 \NC \type{is_japanese} \NC number \NC only if non|-|zero\NC \NR
605 \NC \type{is_korean} \NC number \NC only if non|-|zero\NC \NR
606 \NC \type{is_tradchinese} \NC number \NC only if non|-|zero [name?]\NC \NR
607 \NC \type{is_simplechinese} \NC number \NC only if non|-|zero\NC \NR
608 \NC \type{low_page} \NC number \NC \NC \NR
609 \NC \type{high_page} \NC number \NC \NC \NR
610 \NC \type{iconv_name} \NC string \NC \NC \NR
611 \NC \type{iso_2022_escape} \NC string \NC \NC \NR
612 \LL
613 \stoptabulate
614
615 \stopsubsection
616
617 \startsubsection[title={\type {private}}]
618
619 This is the font's private \POSTSCRIPT\ dictionary, if any. Keys and values are
620 both strings.
621
622 \stopsubsection
623
624 \startsubsection[title={\type {cidinfo}}]
625
626 \starttabulate[|l|l|p|]
627 \DB key \BC type \BC explanation \NC \NR
628 \TB
629 \NC \type{registry} \NC string \NC \NC \NR
630 \NC \type{ordering} \NC string \NC \NC \NR
631 \NC \type{supplement} \NC number \NC \NC \NR
632 \NC \type{version} \NC number \NC \NC \NR
633 \LL
634 \stoptabulate
635
636 \stopsubsection
637
638 \startsubsection[reference=fontloaderpfminfotable,title={\type {pfminfo}}]
639
640 The \type {pfminfo} table contains most of the OS/2 information:
641
642 \starttabulate[|l|l|p|]
643 \DB key \BC type \BC explanation \NC \NR
644 \TB
645 \NC \type{pfmset} \NC number \NC \NC \NR
646 \NC \type{winascent_add} \NC number \NC \NC \NR
647 \NC \type{windescent_add} \NC number \NC \NC \NR
648 \NC \type{hheadascent_add} \NC number \NC \NC \NR
649 \NC \type{hheaddescent_add} \NC number \NC \NC \NR
650 \NC \type{typoascent_add} \NC number \NC \NC \NR
651 \NC \type{typodescent_add} \NC number \NC \NC \NR
652 \NC \type{subsuper_set} \NC number \NC \NC \NR
653 \NC \type{panose_set} \NC number \NC \NC \NR
654 \NC \type{hheadset} \NC number \NC \NC \NR
655 \NC \type{vheadset} \NC number \NC \NC \NR
656 \NC \type{pfmfamily} \NC number \NC \NC \NR
657 \NC \type{weight} \NC number \NC \NC \NR
658 \NC \type{width} \NC number \NC \NC \NR
659 \NC \type{avgwidth} \NC number \NC \NC \NR
660 \NC \type{firstchar} \NC number \NC \NC \NR
661 \NC \type{lastchar} \NC number \NC \NC \NR
662 \NC \type{fstype} \NC number \NC \NC \NR
663 \NC \type{linegap} \NC number \NC \NC \NR
664 \NC \type{vlinegap} \NC number \NC \NC \NR
665 \NC \type{hhead_ascent} \NC number \NC \NC \NR
666 \NC \type{hhead_descent} \NC number \NC \NC \NR
667 \NC \type{os2_typoascent} \NC number \NC \NC \NR
668 \NC \type{os2_typodescent} \NC number \NC \NC \NR
669 \NC \type{os2_typolinegap} \NC number \NC \NC \NR
670 \NC \type{os2_winascent} \NC number \NC \NC \NR
671 \NC \type{os2_windescent} \NC number \NC \NC \NR
672 \NC \type{os2_subxsize} \NC number \NC \NC \NR
673 \NC \type{os2_subysize} \NC number \NC \NC \NR
674 \NC \type{os2_subxoff} \NC number \NC \NC \NR
675 \NC \type{os2_subyoff} \NC number \NC \NC \NR
676 \NC \type{os2_supxsize} \NC number \NC \NC \NR
677 \NC \type{os2_supysize} \NC number \NC \NC \NR
678 \NC \type{os2_supxoff} \NC number \NC \NC \NR
679 \NC \type{os2_supyoff} \NC number \NC \NC \NR
680 \NC \type{os2_strikeysize} \NC number \NC \NC \NR
681 \NC \type{os2_strikeypos} \NC number \NC \NC \NR
682 \NC \type{os2_family_class} \NC number \NC \NC \NR
683 \NC \type{os2_xheight} \NC number \NC \NC \NR
684 \NC \type{os2_capheight} \NC number \NC \NC \NR
685 \NC \type{os2_defaultchar} \NC number \NC \NC \NR
686 \NC \type{os2_breakchar} \NC number \NC \NC \NR
687 \NC \type{os2_vendor} \NC string \NC \NC \NR
688 \NC \type{codepages} \NC table \NC A two-number array of encoded code pages \NC \NR
689 \NC \type{unicoderages} \NC table \NC A four-number array of encoded unicode ranges \NC \NR
690 \NC \type{panose} \NC table \NC \NC \NR
691 \LL
692 \stoptabulate
693
694 The \type {panose} subtable has exactly 10 string keys:
695
696 \starttabulate[|l|l|p|]
697 \DB key \BC type \BC explanation \NC \NR
698 \TB
699 \NC \type{familytype} \NC string \NC Values as in the \OPENTYPE\ font
700 specification: \type {Any}, \type {No Fit},
701 \type {Text and Display}, \type {Script},
702 \type {Decorative}, \type {Pictorial} \NC
703 \NR
704 \NC \type{serifstyle} \NC string \NC See the \OPENTYPE\ font specification for
705 values \NC \NR
706 \NC \type{weight} \NC string \NC idem \NC \NR
707 \NC \type{proportion} \NC string \NC idem \NC \NR
708 \NC \type{contrast} \NC string \NC idem \NC \NR
709 \NC \type{strokevariation} \NC string \NC idem \NC \NR
710 \NC \type{armstyle} \NC string \NC idem \NC \NR
711 \NC \type{letterform} \NC string \NC idem \NC \NR
712 \NC \type{midline} \NC string \NC idem \NC \NR
713 \NC \type{xheight} \NC string \NC idem \NC \NR
714 \LL
715 \stoptabulate
716
717 \stopsubsection
718
719 \startsubsection[reference=fontloadernamestable,title={\type {names}}]
720
721 Each item has two top|-|level keys:
722
723 \starttabulate[|l|l|p|]
724 \DB key \BC type \BC explanation \NC \NR
725 \TB
726 \NC \type{lang} \NC string \NC language for this entry \NC \NR
727 \NC \type{names} \NC table \NC \NC \NR
728 \LL
729 \stoptabulate
730
731 The \type {names} keys are the actual \TRUETYPE\ name strings. The possible keys
732 are: \type {copyright}, \type {family}, \type {subfamily}, \type {uniqueid},
733 \type {fullname}, \type {version}, \type {postscriptname}, \type {trademark},
734 \type {manufacturer}, \type {designer}, \type {descriptor}, \type {venderurl},
735 \type {designerurl}, \type {license}, \type {licenseurl}, \type {idontknow},
736 \type {preffamilyname}, \type {prefmodifiers}, \type {compatfull}, \type
737 {sampletext}, \type {cidfindfontname}, \type {wwsfamily} and \type
738 {wwssubfamily}.
739
740 \stopsubsection
741
742 \startsubsection[title={\type {anchor_classes}}]
743
744 The anchor_classes classes:
745
746 \starttabulate[|l|l|p|]
747 \DB key \BC type \BC explanation \NC \NR
748 \TB
749 \NC \type{name} \NC string \NC a descriptive id of this anchor class\NC \NR
750 \NC \type{lookup} \NC string \NC \NC \NR
751 \NC \type{type} \NC string \NC one of \nod {mark}, \type {mkmk}, \type {curs}, \type {mklg} \NC \NR
752 \LL
753 \stoptabulate
754
755 % type is actually a lookup subtype, not a feature name. Officially, these
756 % strings should be gpos_mark2mark etc.
757
758 \stopsubsection
759
760 \startsubsection[title={\type {gpos}}]
761
762 The \type {gpos} table has one array entry for each lookup. (The \type {gpos_}
763 prefix is somewhat redundant.)
764
765 \starttabulate[|l|l|p|]
766 \DB key \BC type \BC explanation \NC \NR
767 \TB
768 \NC \type{type} \NC string \NC one of \type {gpos_single}, \type {gpos_pair},
769 \type {gpos_cursive}, \type {gpos_mark2base},\crlf
770 \type {gpos_mark2ligature}, \type {gpos_mark2mark}, \type {gpos_context},\crlf \type {gpos_contextchain} \NC \NR
771 \NC \type{flags} \NC table \NC \NC \NR
772 \NC \type{name} \NC string \NC \NC \NR
773 \NC \type{features} \NC array \NC \NC \NR
774 \NC \type{subtables} \NC array \NC \NC \NR
775 \LL
776 \stoptabulate
777
778 The flags table has a true value for each of the lookup flags that is actually
779 set:
780
781 \starttabulate[|l|l|p|]
782 \DB key \BC type \BC explanation \NC \NR
783 \TB
784 \NC \type{r2l} \NC boolean \NC \NC \NR
785 \NC \type{ignorebaseglyphs} \NC boolean \NC \NC \NR
786 \NC \type{ignoreligatures} \NC boolean \NC \NC \NR
787 \NC \type{ignorecombiningmarks} \NC boolean \NC \NC \NR
788 \NC \type{mark_class} \NC string \NC \NC \NR
789 \LL
790 \stoptabulate
791
792 The features subtable items of gpos have:
793
794 \starttabulate[|l|l|p|]
795 \DB key \BC type \BC explanation \NC \NR
796 \TB
797 \NC \type{tag} \NC string \NC \NC \NR
798 \NC \type{scripts} \NC table \NC \NC \NR
799 \LL
800 \stoptabulate
801
802 The scripts table within features has:
803
804 \starttabulate[|l|l|p|]
805 \DB key \BC type \BC explanation \NC \NR
806 \TB
807 \NC \type{script} \NC string \NC \NC \NR
808 \NC \type{langs} \NC array of strings \NC \NC \NR
809 \LL
810 \stoptabulate
811
812 The subtables table has:
813
814 \starttabulate[|l|l|p|]
815 \DB key \BC type \BC explanation \NC \NR
816 \TB
817 \NC \type{name} \NC string \NC \NC \NR
818 \NC \type{suffix} \NC string \NC (only if used)\NC \NR % used by gpos_single to get a default
819 \NC \type{anchor_classes} \NC number \NC (only if used)\NC \NR
820 \NC \type{vertical_kerning} \NC number \NC (only if used)\NC \NR
821 \NC \type{kernclass} \NC table \NC (only if used)\NC \NR
822 \LL
823 \stoptabulate
824
825 The kernclass with subtables table has:
826
827 \starttabulate[|l|l|p|]
828 \DB key \BC type \BC explanation \NC \NR
829 \TB
830 \NC \type{firsts} \NC array of strings \NC \NC \NR
831 \NC \type{seconds} \NC array of strings \NC \NC \NR
832 \NC \type{lookup} \NC string or array \NC associated lookup(s) \NC \NR
833 \NC \type{offsets} \NC array of numbers \NC \NC \NR
834 \LL
835 \stoptabulate
836
837 Note: the kernclass (as far as we can see) always has one entry so it could be one level
838 deep instead. Also the seconds start at \type {[2]} which is close to the fontforge
839 internals so we keep that too.
840
841 \stopsubsection
842
843 \startsubsection[title={\type {gsub}}]
844
845 This has identical layout to the \type {gpos} table, except for the
846 type:
847
848 \starttabulate[|l|l|p|]
849 \DB key \BC type \BC explanation \NC \NR
850 \TB
851 \NC \type{type} \NC string \NC one of \type {gsub_single}, \type {gsub_multiple},
852 \type {gsub_alternate}, \type {gsub_ligature},\crlf
853 \type {gsub_context}, \type {gsub_contextchain},
854 \type {gsub_reversecontextchain} \NC \NR
855 \LL
856 \stoptabulate
857
858 \stopsubsection
859
860 \startsubsection[title={\type {ttf_tables} and \type {ttf_tab_saved}}]
861
862 \starttabulate[|l|l|p|]
863 \DB key \BC type \BC explanation \NC \NR
864 \TB
865 \NC \type{tag} \NC string \NC \NC \NR
866 \NC \type{len} \NC number \NC \NC \NR
867 \NC \type{maxlen} \NC number \NC \NC \NR
868 \NC \type{data} \NC number \NC \NC \NR
869 \LL
870 \stoptabulate
871
872 \stopsubsection
873
874 \startsubsection[title={\type {mm}}]
875
876 \starttabulate[|l|l|p|]
877 \DB key \BC type \BC explanation \NC \NR
878 \TB
879 \NC \type{axes} \NC table \NC array of axis names \NC \NR
880 \NC \type{instance_count} \NC number \NC \NC \NR
881 \NC \type{positions} \NC table \NC array of instance positions
882 (\#axes * instances )\NC \NR
883 \NC \type{defweights} \NC table \NC array of default weights for instances \NC \NR
884 \NC \type{cdv} \NC string \NC \NC \NR
885 \NC \type{ndv} \NC string \NC \NC \NR
886 \NC \type{axismaps} \NC table \NC \NC \NR
887 \LL
888 \stoptabulate
889
890 The \type {axismaps}:
891
892 \starttabulate[|l|l|p|]
893 \DB key \BC type \BC explanation \NC \NR
894 \TB
895 \NC \type{blends} \NC table \NC an array of blend points \NC \NR
896 \NC \type{designs} \NC table \NC an array of design values \NC \NR
897 \NC \type{min} \NC number \NC \NC \NR
898 \NC \type{def} \NC number \NC \NC \NR
899 \NC \type{max} \NC number \NC \NC \NR
900 \LL
901 \stoptabulate
902
903 \stopsubsection
904
905 \startsubsection[title={\type {mark_classes}}]
906
907 The keys in this table are mark class names, and the values are a
908 space|-|separated string of glyph names in this class.
909
910 \stopsubsection
911
912 \startsubsection[title={\type {math}}]
913
914 The math table has the variables that are also discussed in the chapter about
915 math: \type {ScriptPercentScaleDown}, \type {ScriptScriptPercentScaleDown}, \type
916 {DelimitedSubFormulaMinHeight}, \type {DisplayOperatorMinHeight}, \type
917 {MathLeading}, \type {AxisHeight}, \type {AccentBaseHeight}, \type
918 {FlattenedAccentBaseHeight}, \type {SubscriptShiftDown}, \type {SubscriptTopMax},
919 \type {SubscriptBaselineDropMin}, \type {SuperscriptShiftUp}, \type
920 {SuperscriptShiftUpCramped}, \type {SuperscriptBottomMin}, \type
921 {SuperscriptBaselineDropMax}, \type {SubSuperscriptGapMin}, \type
922 {SuperscriptBottomMaxWithSubscript}, \type {SpaceAfterScript}, \type
923 {UpperLimitGapMin}, \type {UpperLimitBaselineRiseMin}, \type {LowerLimitGapMin},
924 \type {LowerLimitBaselineDropMin}, \type {StackTopShiftUp}, \type
925 {StackTopDisplayStyleShiftUp}, \type {StackBottomShiftDown}, \type
926 {StackBottomDisplayStyleShiftDown}, \type {StackGapMin}, \type
927 {StackDisplayStyleGapMin}, \type {StretchStackTopShiftUp}, \type
928 {StretchStackBottomShiftDown}, \type {StretchStackGapAboveMin}, \type
929 {StretchStackGapBelowMin}, \type {FractionNumeratorShiftUp}, \type
930 {FractionNumeratorDisplayStyleShiftUp}, \type {FractionDenominatorShiftDown},
931 \type {FractionDenominatorDisplayStyleShiftDown}, \type
932 {FractionNumeratorGapMin}, \type {FractionNumeratorDisplayStyleGapMin} \type
933 {FractionRuleThickness}, \type {FractionDenominatorGapMin}, \type
934 {FractionDenominatorDisplayStyleGapMin}, \type {SkewedFractionHorizontalGap},
935 \type {SkewedFractionVerticalGap}, \type {OverbarVerticalGap}, \type
936 {OverbarRuleThickness}, \type {OverbarExtraAscender}, \type
937 {UnderbarVerticalGap}, \type {UnderbarRuleThickness}, \type
938 {UnderbarExtraDescender}, \type {RadicalVerticalGap}, \type
939 {RadicalDisplayStyleVerticalGap}, \type {RadicalRuleThickness}, \type
940 {RadicalExtraAscender}, \type {RadicalKernBeforeDegree}, \type
941 {RadicalKernAfterDegree}, \type {RadicalDegreeBottomRaisePercent}, \type
942 {MinConnectorOverlap}, \type {FractionDelimiterSize} and \type
943 {FractionDelimiterDisplayStyleSize}.
944
945 \stopsubsection
946
947 \startsubsection[title={\type {validation_state}}]
948
949 This is just a bonus table with keys: \type {bad_ps_fontname}, \type {bad_glyph_table}, \type {bad_cff_table}, \type {bad_metrics_table}, \type {bad_cmap_table}, \type {bad_bitmaps_table}, \type {bad_gx_table}, \type {bad_ot_table}, \type {bad_os2_version} and \type {bad_sfnt_header}.
950
951 \stopsubsection
952
953 \startsubsection[title={\type {horiz_base} and \type {vert_base}}]
954
955 \starttabulate[|l|l|p|]
956 \DB key \BC type \BC explanation \NC \NR
957 \TB
958 \NC \type{tags} \NC table \NC an array of script list tags\NC \NR
959 \NC \type{scripts} \NC table \NC \NC \NR
960 \LL
961 \stoptabulate
962
963 The \type {scripts} subtable:
964
965 \starttabulate[|l|l|p|]
966 \DB key \BC type \BC explanation \NC \NR
967 \TB
968 \NC \type{baseline} \NC table \NC \NC \NR
969 \NC \type{default_baseline} \NC number \NC \NC \NR
970 \NC \type{lang} \NC table \NC \NC \NR
971 \LL
972 \stoptabulate
973
974 The \type {lang} subtable:
975
976 \starttabulate[|l|l|p|]
977 \DB key \BC type \BC explanation \NC \NR
978 \TB
979 \NC \type{tag} \NC string \NC a script tag \NC \NR
980 \NC \type{ascent} \NC number \NC \NC \NR
981 \NC \type{descent} \NC number \NC \NC \NR
982 \NC \type{features} \NC table \NC \NC \NR
983 \LL
984 \stoptabulate
985
986 The \type {features} points to an array of tables with the same layout except
987 that in those nested tables, the tag represents a language.
988
989 \stopsubsection
990
991 \startsubsection[title={\type {altuni}}]
992
993 An array of alternate \UNICODE\ values. Inside that array are hashes with:
994
995 \starttabulate[|l|l|p|]
996 \DB key \BC type \BC explanation \NC \NR
997 \TB
998 \NC \type{unicode} \NC number \NC this glyph is also used for this unicode \NC \NR
999 \NC \type{variant} \NC number \NC the alternative is driven by this unicode selector \NC \NR
1000 \LL
1001 \stoptabulate
1002
1003 \stopsubsection
1004
1005 \startsubsection[title={\type {vert_variants} and \type {horiz_variants}}]
1006
1007 \starttabulate[|l|l|p|]
1008 \DB key \BC type \BC explanation \NC \NR
1009 \TB
1010 \NC \type{variants} \NC string \NC \NC \NR
1011 \NC \type{italic_correction} \NC number \NC \NC \NR
1012 \NC \type{parts} \NC table \NC \NC \NR
1013 \LL
1014 \stoptabulate
1015
1016 The \type {parts} table is an array of smaller tables:
1017
1018 \starttabulate[|l|l|p|]
1019 \DB key \BC type \BC explanation \NC \NR
1020 \TB
1021 \NC \type{component} \NC string \NC \NC \NR
1022 \NC \type{extender} \NC number \NC \NC \NR
1023 \NC \type{start} \NC number \NC \NC \NR
1024 \NC \type{end} \NC number \NC \NC \NR
1025 \NC \type{advance} \NC number \NC \NC \NR
1026 \LL
1027 \stoptabulate
1028
1029 \stopsubsection
1030
1031 \startsubsection[title={\type {mathkern}}]
1032
1033 \starttabulate[|l|l|p|]
1034 \DB key \BC type \BC explanation \NC \NR
1035 \TB
1036 \NC \type{top_right} \NC table \NC \NC \NR
1037 \NC \type{bottom_right} \NC table \NC \NC \NR
1038 \NC \type{top_left} \NC table \NC \NC \NR
1039 \NC \type{bottom_left} \NC table \NC \NC \NR
1040 \LL
1041 \stoptabulate
1042
1043 Each of the subtables is an array of small hashes with two keys:
1044
1045 \starttabulate[|l|l|p|]
1046 \DB key \BC type \BC explanation \NC \NR
1047 \TB
1048 \NC \type{height} \NC number \NC \NC \NR
1049 \NC \type{kern} \NC number \NC \NC \NR
1050 \LL
1051 \stoptabulate
1052
1053 \stopsubsection
1054
1055 \startsubsection[title={\type {kerns}}]
1056
1057 Substructure is identical to the per|-|glyph subtable.
1058
1059 \stopsubsection
1060
1061 \startsubsection[title={\type {vkerns}}]
1062
1063 Substructure is identical to the per|-|glyph subtable.
1064
1065 \stopsubsection
1066
1067 \startsubsection[title={\type {texdata}}]
1068
1069 \starttabulate[|l|l|p|]
1070 \DB key \BC type \BC explanation \NC \NR
1071 \TB
1072 \NC \type{type} \NC string \NC \nod {unset}, \type {text}, \nod {math}, \type {mathext} \NC \NR
1073 \NC \type{params} \NC array \NC 22 font numeric parameters \NC \NR
1074 \LL
1075 \stoptabulate
1076
1077 \stopsubsection
1078
1079 \startsubsection[title={\type {lookups}}]
1080
1081 Top|-|level \type {lookups} is quite different from the ones at character level.
1082 The keys in this hash are strings, the values the actual lookups, represented as
1083 dictionary tables.
1084
1085 \starttabulate[|l|l|p|]
1086 \DB key \BC type \BC explanation \NC \NR
1087 \TB
1088 \NC \type{type} \NC string \NC \NC \NR
1089 \NC \type{format} \NC enum \NC one of \type {glyphs}, \type {class}, \type {coverage},
1090 \type {reversecoverage} \NC \NR
1091 \NC \type{tag} \NC string \NC \NC \NR
1092 \NC \type{current_class} \NC array \NC \NC \NR
1093 \NC \type{before_class} \NC array \NC \NC \NR
1094 \NC \type{after_class} \NC array \NC \NC \NR
1095 \NC \type{rules} \NC array \NC an array of rule items\NC \NR
1096 \LL
1097 \stoptabulate
1098
1099 Rule items have one common item and one specialized item:
1100
1101 \starttabulate[|l|l|p|]
1102 \DB key \BC type \BC explanation \NC \NR
1103 \TB
1104 \NC \type{lookups} \NC array \NC a linear array of lookup names \NC \NR
1105 \NC \type{glyphs} \NC array \NC only if the parent's format is \type {glyphs} \NC \NR
1106 \NC \type{class} \NC array \NC only if the parent's format is \type {class} \NC \NR
1107 \NC \type{coverage} \NC array \NC only if the parent's format is \type {coverage} \NC \NR
1108 \NC \type{reversecoverage} \NC array \NC only if the parent's format is \type {reversecoverage} \NC \NR
1109 \LL
1110 \stoptabulate
1111
1112 A glyph table is:
1113
1114 \starttabulate[|l|l|p|]
1115 \DB key \BC type \BC explanation \NC \NR
1116 \TB
1117 \NC \type{names} \NC string \NC \NC \NR
1118 \NC \type{back} \NC string \NC \NC \NR
1119 \NC \type{fore} \NC string \NC \NC \NR
1120 \LL
1121 \stoptabulate
1122
1123 A class table is:
1124
1125 \starttabulate[|l|l|p|]
1126 \DB key \BC type \BC explanation \NC \NR
1127 \TB
1128 \NC \type{current} \NC array \NC of numbers \NC \NR
1129 \NC \type{before} \NC array \NC of numbers \NC \NR
1130 \NC \type{after} \NC array \NC of numbers \NC \NR
1131 \LL
1132 \stoptabulate
1133
1134 for coverage:
1135
1136 \starttabulate[|l|l|p|]
1137 \DB key \BC type \BC explanation \NC \NR
1138 \TB
1139 \NC \type{current} \NC array \NC of strings \NC \NR
1140 \NC \type{before} \NC array \NC of strings \NC \NR
1141 \NC \type{after} \NC array \NC of strings \NC \NR
1142 \LL
1143 \stoptabulate
1144
1145 and for reverse coverage:
1146
1147 \starttabulate[|l|l|p|]
1148 \DB key \BC type \BC explanation \NC \NR
1149 \TB
1150 \NC \type{current} \NC array \NC of strings \NC \NR
1151 \NC \type{before} \NC array \NC of strings \NC \NR
1152 \NC \type{after} \NC array \NC of strings \NC \NR
1153 \NC \type{replacements} \NC string \NC \NC \NR
1154 \LL
1155 \stoptabulate
1156
1157 \stopsubsection
1158
1159 \stopsection
1160
1161 \stopchapter
1162
1163 \stopcomponent