Commit | Line | Data |
---|---|---|
fac14bbe MW |
1 | /* Copyright (c) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, |
2 | * 2005, 2006 by Arkkra Enterprises */ | |
3 | /* All rights reserved */ | |
4 | /* | |
5 | * Name: plutils.c | |
6 | * | |
7 | * Description: This file contains utility functions belonging to the placement | |
8 | * phase. Some of them are also used by other phases. | |
9 | */ | |
10 | ||
11 | #include "defines.h" | |
12 | #include "structs.h" | |
13 | #include "globals.h" | |
14 | ||
15 | static int phrase_tieslur_note P((struct GRPSYL *gs_p, int nidx, int side, | |
16 | int interfere)); | |
17 | static int tied_to_nidx P((struct GRPSYL *gs_p, int nidx)); | |
18 | static int slurred_to_nidx P((struct GRPSYL *gs_p, int nidx, int sidx)); | |
19 | static RATIONAL lefttime P((double count, struct GRPSYL *firstgs_p, | |
20 | int timeden)); | |
21 | static RATIONAL righttime P((double count, struct GRPSYL *firstgs_p, | |
22 | int timeden)); | |
23 | ||
24 | /* | |
25 | * Name: nextnongrace() | |
26 | * | |
27 | * Abstract: Return next nongrace group in a GRPSYL list. | |
28 | * | |
29 | * Returns: pointer to GRPSYL of next nongrace group, 0 if none | |
30 | * | |
31 | * Description: This function loops down the GRPSYL linked list from the given | |
32 | * starting point. It returns the next nongrace GRPSYL, or 0 | |
33 | * if none. | |
34 | */ | |
35 | ||
36 | struct GRPSYL * | |
37 | nextnongrace(gs_p) | |
38 | ||
39 | struct GRPSYL *gs_p; /* current group */ | |
40 | ||
41 | { | |
42 | gs_p = gs_p->next; | |
43 | while (gs_p != 0 && gs_p->grpvalue == GV_ZERO) | |
44 | gs_p = gs_p->next; | |
45 | return (gs_p); | |
46 | } | |
47 | \f | |
48 | /* | |
49 | * Name: prevnongrace() | |
50 | * | |
51 | * Abstract: Return previous nongrace group in a GRPSYL list. | |
52 | * | |
53 | * Returns: pointer to GRPSYL of previous nongrace group, 0 if none | |
54 | * | |
55 | * Description: This function loop up the GRPSYL linked list from the given | |
56 | * starting point. It returns the previous nongrace GRPSYL, or 0 | |
57 | * if none. | |
58 | */ | |
59 | ||
60 | struct GRPSYL * | |
61 | prevnongrace(gs_p) | |
62 | ||
63 | struct GRPSYL *gs_p; /* current group */ | |
64 | ||
65 | { | |
66 | gs_p = gs_p->prev; | |
67 | while (gs_p != 0 && gs_p->grpvalue == GV_ZERO) | |
68 | gs_p = gs_p->prev; | |
69 | return (gs_p); | |
70 | } | |
71 | \f | |
72 | /* | |
73 | * Name: nextglobnongrace() | |
74 | * | |
75 | * Abstract: Return next nongrace group in this voice. | |
76 | * | |
77 | * Returns: pointer to GRPSYL of next nongrace group, 0 if none | |
78 | * | |
79 | * Description: This function, given a nongrace and the MLL structure it hangs | |
80 | * off of, returns the next nongrace in this voice, even if it's in | |
81 | * the next measure. If it is in the next measure, *mll_p_p gets | |
82 | * updated. But if that next measure is a second or later ending, | |
83 | * it's not considered to be a "next" measure, so return 0. | |
84 | */ | |
85 | ||
86 | struct GRPSYL * | |
87 | nextglobnongrace(gs_p, mll_p_p) | |
88 | ||
89 | struct GRPSYL *gs_p; /* current group */ | |
90 | struct MAINLL **mll_p_p; /* MLL structure it is hanging off of */ | |
91 | ||
92 | { | |
93 | do { | |
94 | gs_p = nextgrpsyl(gs_p, mll_p_p); | |
95 | } while (gs_p != 0 && gs_p->grpvalue == GV_ZERO); | |
96 | return (gs_p); | |
97 | } | |
98 | \f | |
99 | /* | |
100 | * Name: prevglobnongrace() | |
101 | * | |
102 | * Abstract: Return previous nongrace group in this voice. | |
103 | * | |
104 | * Returns: pointer to GRPSYL of previous nongrace group, 0 if none | |
105 | * | |
106 | * Description: This function, given a nongrace and the MLL structure it hangs | |
107 | * off of, returns the prev nongrace in this voice, even if it's | |
108 | * in an earlier measure. If we are at the start of an ending, | |
109 | * it skips over any previous ending and goes to the measure | |
110 | * preceding the first ending. If the resulting nongrace is in a | |
111 | * previous measure, *mll_p_p gets updated. | |
112 | */ | |
113 | ||
114 | struct GRPSYL * | |
115 | prevglobnongrace(gs_p, mll_p_p) | |
116 | ||
117 | struct GRPSYL *gs_p; /* current group */ | |
118 | struct MAINLL **mll_p_p; /* MLL structure it is hanging off of */ | |
119 | ||
120 | { | |
121 | do { | |
122 | gs_p = prevgrpsyl(gs_p, mll_p_p); | |
123 | } while (gs_p != 0 && gs_p->grpvalue == GV_ZERO); | |
124 | return (gs_p); | |
125 | } | |
126 | \f | |
127 | /* | |
128 | * Name: drmo() | |
129 | * | |
130 | * Abstract: Detect rightmost one. | |
131 | * | |
132 | * Returns: void | |
133 | * | |
134 | * Description: This function returns the bit position of the rightmost bit | |
135 | * that is a 1 in the given number, the low order bit being | |
136 | * bit 0. The given number must not be 0. | |
137 | */ | |
138 | ||
139 | int | |
140 | drmo(num) | |
141 | ||
142 | register int num; | |
143 | ||
144 | { | |
145 | register int n; | |
146 | ||
147 | for (n = 0; n < 8 * sizeof(int); n++) { | |
148 | if ( (num & (1 << n)) != 0 ) | |
149 | return (n); | |
150 | } | |
151 | pfatal("0 was passed to drmo"); | |
152 | return (0); /* dead code, but keeps lint happy */ | |
153 | } | |
154 | \f | |
155 | /* | |
156 | * Name: tieslurpad() | |
157 | * | |
158 | * Abstract: How much tie/slur padding is needed after this group? | |
159 | * | |
160 | * Returns: Padding in inches. | |
161 | * | |
162 | * Description: This function returns the amount of padding needed after a | |
163 | * group due to ties or slurs, if the given group is tied to the | |
164 | * next group, or any note in it is tied or slurred to a note | |
165 | * in the following group. Otherwise it returns zero. | |
166 | * NOTE: This function ignores staffscale. | |
167 | */ | |
168 | ||
169 | double | |
170 | tieslurpad(staff_p, gs_p) | |
171 | ||
172 | struct STAFF *staff_p; /* the staff the group is connected to */ | |
173 | struct GRPSYL *gs_p; /* the group after which padding may occur */ | |
174 | ||
175 | { | |
176 | struct NOTE *note_p; /* point at a note structure */ | |
177 | struct GRPSYL *gtemp_p; /* temp GRPSYL pointer */ | |
178 | struct GRPSYL *this_p; /* first GRPSYL in this voice */ | |
179 | struct GRPSYL *that_p; /* first GRPSYL in other voice */ | |
180 | RATIONAL starttime; /* time into measure where *gs_p starts */ | |
181 | float pad; /* how much padding is needed */ | |
182 | int interfere; /* does other voice have notes/rests here? */ | |
183 | int stepdiff; /* vertical reach of a curve */ | |
184 | int n; /* index into notes in group */ | |
185 | int s; /* index into notes slurred to */ | |
186 | ||
187 | ||
188 | /* syllables can't have ties or slurs */ | |
189 | if (gs_p->grpsyl != GS_GROUP) | |
190 | return (0); | |
191 | ||
192 | /* rests and spaces can't have ties or slurs */ | |
193 | if (gs_p->grpcont != GC_NOTES) | |
194 | return (0); | |
195 | ||
196 | /* if last group in measure, don't need any more space */ | |
197 | if (gs_p->next == 0) | |
198 | return (0); | |
199 | ||
200 | /* | |
201 | * Find the first group in this measure, and total time preceding the | |
202 | * group we were given. We need this to figure out which voice we are | |
203 | * in, and, if there is another voice, whether it has only spaces | |
204 | * during the time of our group, which affects how the curves should | |
205 | * look. | |
206 | */ | |
207 | starttime = Zero; | |
208 | for (gtemp_p = gs_p->prev, this_p = gs_p; gtemp_p != 0; | |
209 | this_p = gtemp_p, gtemp_p = gtemp_p->prev) | |
210 | starttime = radd(starttime, gtemp_p->fulltime); | |
211 | ||
212 | /* point at other voice, or null pointer if none */ | |
213 | if (staff_p->groups_p[0] == this_p) | |
214 | that_p = staff_p->groups_p[1]; /* might be 0 */ | |
215 | else if (staff_p->groups_p[1] == this_p) | |
216 | that_p = staff_p->groups_p[0]; /* might be 0 */ | |
217 | else | |
218 | that_p = 0; /* we are voice 3, ignore other voices */ | |
219 | ||
220 | if (that_p == 0 || hasspace(that_p, starttime, | |
221 | radd(starttime, gs_p->fulltime))) | |
222 | interfere = NO; | |
223 | else | |
224 | interfere = YES; | |
225 | ||
226 | pad = 0; /* start with no padding */ | |
227 | ||
228 | /* | |
229 | * Loop through every note in this group. If it's tied, check each | |
230 | * note to see if either it or the note it's tied to is ineligible for | |
231 | * phrase-like curves. If so, there will be a horizontally aligned | |
232 | * curve, and we need to pad. The note must be the same in both | |
233 | * groups, so there's no need to consider vertical distances at this | |
234 | * point. Then loop through the 0 or more slurs from this note to | |
235 | * note(s) in the next group. For each one, find the vertical distance | |
236 | * between the two notes. The padding it needs is based on this and on | |
237 | * whether phrase-like curves can be drawn. Keep track of the maximum | |
238 | * padding needed by any pair of notes. | |
239 | * We also need to pad if the stems are UP-DOWN, because that leaves no | |
240 | * room for the curve. | |
241 | */ | |
242 | for (n = 0; n < gs_p->nnotes; n++) { | |
243 | note_p = &gs_p->notelist[n]; | |
244 | if (note_p->tie == YES) { | |
245 | if (gs_p->stemdir == UP && gs_p->next->stemdir == DOWN | |
246 | || phrase_tieslur_note(gs_p, n, STARTITEM, interfere) | |
247 | == NO || phrase_tieslur_note(gs_p->next, tied_to_nidx( | |
248 | gs_p, n), ENDITEM, interfere) == NO) | |
249 | pad = MAX(pad, TIESLURPAD); | |
250 | } | |
251 | for (s = 0; s < note_p->nslurto; s++) { | |
252 | /* | |
253 | * If it's a slur to/from nowhere, don't deal with it | |
254 | * here. It is considered along with the width of the | |
255 | * individual note. | |
256 | */ | |
257 | if (IS_NOWHERE(note_p->slurtolist[s].octave)) | |
258 | continue; /* from nowhere */ | |
259 | ||
260 | stepdiff = abs( | |
261 | ( note_p->octave * 7 + | |
262 | Letshift[ note_p->letter - 'a' ] ) - | |
263 | ( note_p->slurtolist[s].octave * 7 + | |
264 | Letshift[ note_p->slurtolist[s].letter | |
265 | - 'a' ] ) | |
266 | ); | |
267 | if (gs_p->stemdir == UP && gs_p->next->stemdir == DOWN | |
268 | || phrase_tieslur_note(gs_p, n, STARTITEM, interfere) | |
269 | == NO || phrase_tieslur_note(gs_p->next, | |
270 | slurred_to_nidx(gs_p, n, s), ENDITEM, interfere) == NO){ | |
271 | pad = MAX(pad, stepdiff <= 3 ? TIESLURPAD : | |
272 | TIESLURPAD + (stepdiff - 3) * STEPSIZE / 2); | |
273 | } else { | |
274 | pad = MAX(pad, stepdiff <= 3 ? 0 : | |
275 | (stepdiff - 3) * STEPSIZE / 2); | |
276 | } | |
277 | } | |
278 | } | |
279 | ||
280 | return (pad); /* max padding needed by any pair of notes */ | |
281 | } | |
282 | \f | |
283 | /* | |
284 | * Name: phrase_tieslur_note() | |
285 | * | |
286 | * Abstract: Is the given note the end note and eligible for "new" tie/slur? | |
287 | * | |
288 | * Returns: YES or NO | |
289 | * | |
290 | * Description: This function determines whether a tie or slur to/from the | |
291 | * given note is to be drawn like a phrase mark (as opposed to | |
292 | * drawing it vertically aligned with the note). | |
293 | */ | |
294 | ||
295 | static int | |
296 | phrase_tieslur_note(gs_p, nidx, side, interfere) | |
297 | ||
298 | struct GRPSYL *gs_p; /* point at note's group */ | |
299 | int nidx; /* index to this note in notelist */ | |
300 | int side; /* STARTITEM (curve here to right) or ENDITEM */ | |
301 | int interfere; /* does the other voice have notes/rests here?*/ | |
302 | ||
303 | { | |
304 | /* check for each bad condition, returning NO if it exists */ | |
305 | ||
306 | /* inner note of a group */ | |
307 | if (nidx != 0 && nidx != gs_p->nnotes - 1) | |
308 | return (NO); | |
309 | ||
310 | /* bottom note of voice 1 and other voice interferes */ | |
311 | if (gs_p->vno == 1 && nidx == gs_p->nnotes - 1 && interfere) | |
312 | return (NO); | |
313 | ||
314 | /* top note of voice 2 and other voice interferes */ | |
315 | if (gs_p->vno == 2 && nidx == 0 && interfere) | |
316 | return (NO); | |
317 | ||
318 | /* antistem note if "with" list is present */ | |
319 | /* (don't need to check normwith; if it weren't YES we would have */ | |
320 | /* returned above) */ | |
321 | if (gs_p->nwith != 0 && gs_p->stemdir == UP && nidx == gs_p->nnotes - 1) | |
322 | return (NO); | |
323 | ||
324 | /* antistem note if "with" list is present */ | |
325 | /* (don't need to check normwith; if it weren't YES we would have */ | |
326 | /* returned above) */ | |
327 | if (gs_p->nwith != 0 && gs_p->stemdir == DOWN && nidx == 0) | |
328 | return (NO); | |
329 | ||
330 | /* stem in the way of left end of curve */ | |
331 | if (side == STARTITEM && gs_p->basictime >= 2 && gs_p->stemdir == UP && | |
332 | nidx == 0 && gs_p->nnotes > 1) | |
333 | return (NO); | |
334 | ||
335 | /* stem in the way of right end of curve */ | |
336 | if (side == ENDITEM && gs_p->basictime >= 2 && gs_p->stemdir == DOWN && | |
337 | nidx == gs_p->nnotes - 1 && gs_p->nnotes > 1) | |
338 | return (NO); | |
339 | ||
340 | return (YES); | |
341 | } | |
342 | \f | |
343 | /* | |
344 | * Name: tied_to_nidx() | |
345 | * | |
346 | * Abstract: Return the note index of the note the given note is tied to. | |
347 | * | |
348 | * Returns: index into gs_p->next->notelist | |
349 | * | |
350 | * Description: This function is given a valid group (not the last one in the | |
351 | * measure) and an index into its notelist to a note that is tied | |
352 | * to the next group. It returns the index into the next group's | |
353 | * notelist to the note that the first group's note is tied to. | |
354 | */ | |
355 | ||
356 | static int | |
357 | tied_to_nidx(gs_p, nidx) | |
358 | ||
359 | struct GRPSYL *gs_p; /* point at note's group */ | |
360 | int nidx; /* index to this note in notelist */ | |
361 | ||
362 | { | |
363 | struct NOTE *nl_ptr; /* ptr to next group's notelist */ | |
364 | int n; | |
365 | ||
366 | ||
367 | nl_ptr = gs_p->next->notelist; | |
368 | ||
369 | for (n = 0; n < gs_p->next->nnotes; n++) { | |
370 | if (gs_p->notelist[nidx].letter == nl_ptr[n].letter && | |
371 | gs_p->notelist[nidx].octave == nl_ptr[n].octave) | |
372 | return (n); | |
373 | } | |
374 | ||
375 | pfatal("tied_to_nidx: can't find note tied to"); | |
376 | return (0); /* to keep lint happy */ | |
377 | } | |
378 | \f | |
379 | /* | |
380 | * Name: slurred_to_nidx() | |
381 | * | |
382 | * Abstract: Return the note index of the note the given note is slurred to. | |
383 | * | |
384 | * Returns: index into gs_p->next->notelist | |
385 | * | |
386 | * Description: This function is given a valid group (not the last one in the | |
387 | * measure) and an index into its notelist to a note that is tied | |
388 | * to the next group. It returns the index into the next group's | |
389 | * notelist to the note that the first group's note is tied to. | |
390 | */ | |
391 | ||
392 | static int | |
393 | slurred_to_nidx(gs_p, nidx, sidx) | |
394 | ||
395 | struct GRPSYL *gs_p; /* point at note's group */ | |
396 | int nidx; /* index to this note in notelist */ | |
397 | int sidx; /* index to slurred to note in slurto list */ | |
398 | ||
399 | { | |
400 | struct NOTE *nl_ptr; /* ptr to next group's notelist */ | |
401 | int n; | |
402 | ||
403 | ||
404 | nl_ptr = gs_p->next->notelist; | |
405 | ||
406 | for (n = 0; n < gs_p->next->nnotes; n++) { | |
407 | if (gs_p->notelist[nidx].slurtolist[sidx].letter == | |
408 | nl_ptr[n].letter && | |
409 | gs_p->notelist[nidx].slurtolist[sidx].octave == | |
410 | nl_ptr[n].octave) | |
411 | return (n); | |
412 | } | |
413 | ||
414 | pfatal("slurred_to_nidx: can't find note slurred to"); | |
415 | return (0); /* to keep lint happy */ | |
416 | } | |
417 | \f | |
418 | /* | |
419 | * Name: hasspace() | |
420 | * | |
421 | * Abstract: Finds out if the given voice has space during given time. | |
422 | * | |
423 | * Returns: YES or NO | |
424 | * | |
425 | * Description: This function is given a linked list of groups to check | |
426 | * during a given time interval. If the list consists entirely | |
427 | * of space(s) during the time interval, the function returns | |
428 | * YES. Otherwise it returns NO. If vtime2 is greater than the | |
429 | * length of a measure, the extra, nonexistent time is regarded | |
430 | * as all spaces. If the linked list of groups doesn't exist | |
431 | * (gs_p is a null pointer), the function returns YES, since | |
432 | * there is nothing there but "space". | |
433 | */ | |
434 | ||
435 | int | |
436 | hasspace(gs_p, vtime, vtime2) | |
437 | ||
438 | register struct GRPSYL *gs_p; /* starts pointing at the first GRPSYL list */ | |
439 | RATIONAL vtime, vtime2; /* time when to start and stop checking for space */ | |
440 | ||
441 | { | |
442 | RATIONAL t; /* accumulate time */ | |
443 | int oldcont; /* content of previous group */ | |
444 | ||
445 | ||
446 | /* "no linked list exists" counts as all spaces */ | |
447 | if (gs_p == 0) | |
448 | return (YES); | |
449 | ||
450 | oldcont = GC_SPACE; /* prevent useless 'used before set' warning */ | |
451 | ||
452 | /* accumulate time until crossing vtime boundary */ | |
453 | for (t = Zero; LT(t, vtime); gs_p = gs_p->next) { | |
454 | if (gs_p->grpvalue == GV_ZERO) | |
455 | continue; | |
456 | t = radd(t, gs_p->fulltime); | |
457 | oldcont = gs_p->grpcont; | |
458 | } | |
459 | ||
460 | if (GT(t, vtime) && oldcont != GC_SPACE) | |
461 | return (NO); | |
462 | ||
463 | for ( ; gs_p != 0 && LT(t, vtime2); gs_p = gs_p->next) { | |
464 | if (gs_p->grpvalue == GV_ZERO) | |
465 | continue; | |
466 | if (gs_p->grpcont != GC_SPACE) | |
467 | return (NO); | |
468 | t = radd(t, gs_p->fulltime); | |
469 | } | |
470 | ||
471 | return (YES); | |
472 | } | |
473 | \f | |
474 | /* | |
475 | * Name: closestgroup() | |
476 | * | |
477 | * Abstract: Find closest nongrace group in this voice to given time value. | |
478 | * | |
479 | * Returns: pointer to the closest nongrace GRPSYL | |
480 | * | |
481 | * Description: This function finds the GRPSYL in the given linked list that is | |
482 | * closest, timewise, to the given count number, ignoring grace | |
483 | * groups. | |
484 | */ | |
485 | ||
486 | struct GRPSYL * | |
487 | closestgroup(count, firstgs_p, timeden) | |
488 | ||
489 | double count; /* which count of the measure */ | |
490 | struct GRPSYL *firstgs_p; /* first GRPSYL of relevant voice in measure */ | |
491 | int timeden; /* denominator of current time signature */ | |
492 | ||
493 | { | |
494 | RATIONAL reqtime; /* time requested */ | |
495 | RATIONAL tottime; /* total time in measure so far */ | |
496 | RATIONAL otottime; /* old total time in measure so far */ | |
497 | struct GRPSYL *gs_p; /* point along group list */ | |
498 | struct GRPSYL *ogs_p; /* (old) point along group list */ | |
499 | ||
500 | ||
501 | /* skip over any initial grace groups */ | |
502 | if (firstgs_p->grpvalue == GV_ZERO) | |
503 | firstgs_p = nextnongrace(firstgs_p); | |
504 | ||
505 | /* if at or before the first count, it's closest to first group */ | |
506 | if (count <= 1) | |
507 | return (firstgs_p); | |
508 | ||
509 | /* get requested time to nearest tiny part of a count, in lowest terms*/ | |
510 | reqtime.n = 4 * MAXBASICTIME * (count - 1) + 0.5; | |
511 | reqtime.d = 4 * MAXBASICTIME * timeden; | |
512 | rred(&reqtime); | |
513 | ||
514 | /* | |
515 | * Loop through this voice accumulating time values. As soon as we | |
516 | * equal or exceed the requested time value, check whether the | |
517 | * requested time is closer to the new accumulated time, or that before | |
518 | * this last group. Return the closest one. | |
519 | */ | |
520 | tottime = Zero; | |
521 | for (ogs_p = firstgs_p, gs_p = nextnongrace(ogs_p); gs_p != 0; | |
522 | ogs_p = gs_p, gs_p = nextnongrace(gs_p)) { | |
523 | otottime = tottime; | |
524 | tottime = radd(tottime, ogs_p->fulltime); | |
525 | if (GE(tottime, reqtime)) { | |
526 | if (GT( rsub(reqtime,otottime), rsub(tottime,reqtime) )) | |
527 | return (gs_p); | |
528 | else | |
529 | return (ogs_p); | |
530 | } | |
531 | } | |
532 | ||
533 | /* requested time is after last group; return last group */ | |
534 | return (ogs_p); | |
535 | } | |
536 | \f | |
537 | /* | |
538 | * Name: chkallspace() | |
539 | * | |
540 | * Abstract: Check if voice is all spaces for the voice this stuff is on. | |
541 | * | |
542 | * Returns: YES or NO | |
543 | * | |
544 | * Description: This function checks where one voice seems to be all spaces | |
545 | * during the duration of a phrase mark, or other stuff which must | |
546 | * be associated with a definite group. The tricky thing is that | |
547 | * until we've decided which voice the stuff is intended to | |
548 | * apply to, we don't exactly know what the endpoints of the | |
549 | * stuff are going to be. All we know is the "count" values the | |
550 | * user asked for, which may or may not equal the positions of | |
551 | * GRPSYLs in the voices. So we look at voices 1 and 2 and take | |
552 | * the worst (widest) case as the endpoints. (This is called only | |
553 | * when both of these voices exist. We ignore any voice 3.) | |
554 | */ | |
555 | ||
556 | int | |
557 | chkallspace(msbeg_p, stuff_p, vno) | |
558 | ||
559 | struct MAINLL *msbeg_p; /* staff at beginning of the stuff */ | |
560 | struct STUFF *stuff_p; /* the STUFF */ | |
561 | int vno; /* voice being tested for being all spaces */ | |
562 | ||
563 | { | |
564 | static RATIONAL tiny = {1, 4 * MAXBASICTIME}; | |
565 | struct MAINLL *msend_p; /* staff at end of the phrase */ | |
566 | int timeden; /* denom of time sig at end of stuff */ | |
567 | RATIONAL begtime, endtime; /* time into measures of begin & end */ | |
568 | RATIONAL temptime; /* temp var for storing time */ | |
569 | ||
570 | ||
571 | /* | |
572 | * Find what measure this stuff ends in. Along the way, keep | |
573 | * track of the time signature denominator, in case it changes. | |
574 | */ | |
575 | msend_p = getendstuff(msbeg_p, stuff_p, &timeden); | |
576 | ||
577 | /* | |
578 | * If we hit a multirest, bail out, arbitrarily returning NO. This | |
579 | * stuff will be thrown away later anyway. | |
580 | */ | |
581 | if (msend_p == 0) | |
582 | return (NO); | |
583 | ||
584 | /* | |
585 | * If the second voice doesn't exist (because vscheme changed), | |
586 | * it's like all spaces in that voice. So if we're asking about that | |
587 | * voice, return YES. If asking about the first voice, return NO. | |
588 | */ | |
589 | if (msend_p->u.staff_p->groups_p[1] == NULL) { | |
590 | return (vno == 1 ? YES : NO); | |
591 | } | |
592 | ||
593 | /* | |
594 | * Find time values that are sure to contain the stuff. Take the | |
595 | * outermost values of the two voices. | |
596 | */ | |
597 | begtime = lefttime(stuff_p->start.count, | |
598 | msbeg_p->u.staff_p->groups_p[0], Score.timeden); | |
599 | temptime = lefttime(stuff_p->start.count, | |
600 | msbeg_p->u.staff_p->groups_p[1], Score.timeden); | |
601 | if (LT(temptime, begtime)) | |
602 | begtime = temptime; | |
603 | endtime = righttime(stuff_p->end.count, | |
604 | msend_p->u.staff_p->groups_p[0], timeden); | |
605 | temptime = righttime(stuff_p->end.count, | |
606 | msend_p->u.staff_p->groups_p[1], timeden); | |
607 | if (GT(temptime, endtime)) | |
608 | endtime = temptime; | |
609 | ||
610 | /* | |
611 | * If the beginning and end are in the same measure and at the same | |
612 | * time, this phrase would normally be thrown away later, but we need | |
613 | * to deal with it because the case of a phrase from a grace to its | |
614 | * main note. It doesn't make sense to ask what a zero time contains, | |
615 | * so to handle this, add a tiny time value to the end time. | |
616 | */ | |
617 | if (msbeg_p == msend_p && EQ(begtime, endtime)) | |
618 | endtime = radd(endtime, tiny); | |
619 | ||
620 | return (allspace(vno, msbeg_p, begtime, msend_p, endtime)); | |
621 | } | |
622 | \f | |
623 | /* | |
624 | * Name: allspace() | |
625 | * | |
626 | * Abstract: Finds out if the given voice has space for the given time. | |
627 | * | |
628 | * Returns: YES or NO | |
629 | * | |
630 | * Description: This function is a multi-measure version of hasspace(), and in | |
631 | * fact works by calling hasspace() repeatedly. It is given the | |
632 | * linked list of groups for the voice in the first measure in | |
633 | * question. It checks whether the voice consists entirely of | |
634 | * spaces from the duration point given for this first measure, | |
635 | * until the endpoint, which may or may not be in the same measure. | |
636 | */ | |
637 | ||
638 | int | |
639 | allspace(vno, msbeg_p, begtime, msend_p, endtime) | |
640 | ||
641 | int vno; /* voice number, numbering from voice 1 == 0 */ | |
642 | struct MAINLL *msbeg_p; /* point at MLL (staff) where duration begins */ | |
643 | RATIONAL begtime; /* time where duration begins */ | |
644 | struct MAINLL *msend_p; /* point at MLL (staff) where duration ends */ | |
645 | RATIONAL endtime; /* time where duration ends */ | |
646 | ||
647 | { | |
648 | struct MAINLL *mainll_p; /* point along MLL */ | |
649 | int staffno; | |
650 | ||
651 | ||
652 | /* if the time starts and ends in the same measure, let hasspace do it*/ | |
653 | if (msbeg_p == msend_p) { | |
654 | return (hasspace(msbeg_p->u.staff_p->groups_p[vno], | |
655 | begtime, endtime)); | |
656 | } | |
657 | ||
658 | /* | |
659 | * If the first measure contains non-spaces, return NO. Rather than | |
660 | * keeping track of time signatures, we're going to pretend that we | |
661 | * are in the longest possible time. This relies on the fact that | |
662 | * hasspace() in effect assumes that any phony time past the end of | |
663 | * the actual measure is spaces. | |
664 | */ | |
665 | if (hasspace(msbeg_p->u.staff_p->groups_p[vno], begtime, Maxtime) == NO) | |
666 | return (NO); | |
667 | ||
668 | staffno = msbeg_p->u.staff_p->staffno; | |
669 | ||
670 | /* if any intermediate measures contain non-spaces, return NO */ | |
671 | for (mainll_p = msbeg_p->next; mainll_p != 0 && mainll_p != msend_p; | |
672 | mainll_p = mainll_p->next) { | |
673 | ||
674 | /* skip everything but STAFFs for our staff number */ | |
675 | if (mainll_p->str != S_STAFF || | |
676 | mainll_p->u.staff_p->staffno != staffno) | |
677 | continue; | |
678 | ||
679 | if (hasspace(mainll_p->u.staff_p->groups_p[vno], Zero, Maxtime) | |
680 | == NO) | |
681 | return (NO); | |
682 | } | |
683 | ||
684 | if (mainll_p == 0) | |
685 | pfatal("bug found in allspace"); | |
686 | ||
687 | /* the result is now determined by the last measure */ | |
688 | return (hasspace(msend_p->u.staff_p->groups_p[vno], Zero, endtime)); | |
689 | } | |
690 | \f | |
691 | /* | |
692 | * Name: getendstuff() | |
693 | * | |
694 | * Abstract: Find staff and time signature denominator for end of a stuff. | |
695 | * | |
696 | * Returns: pointer to MLL structure for staff containing end of stuff, or 0 | |
697 | * | |
698 | * Description: This function finds the staff for the end of the given stuff. | |
699 | * As a byproduct, it also finds the denominator of the time | |
700 | * signature at that place. If a multirest is encountered, a null | |
701 | * pointer is returned, and timeden is not guaranteed. | |
702 | * If the end of the piece is encountered, it returns the last | |
703 | * staff. | |
704 | */ | |
705 | ||
706 | struct MAINLL * | |
707 | getendstuff(mainll_p, stuff_p, timeden_p) | |
708 | ||
709 | struct MAINLL *mainll_p;/* staff at beginning of stuff, gets changed to end */ | |
710 | struct STUFF *stuff_p; /* the STUFF */ | |
711 | int *timeden_p; /* gets set to denom of time sig at end of stuff */ | |
712 | ||
713 | { | |
714 | int staffno; /* staff number where stuff is */ | |
715 | struct MAINLL *mst_p; /* point at the last staffno staff seen */ | |
716 | int timenum; /* remember the last time sig numerator */ | |
717 | int b; /* count bar lines */ | |
718 | ||
719 | ||
720 | /* bail out if multirest */ | |
721 | if (mainll_p->u.staff_p->groups_p[0]->basictime < -1) | |
722 | return (0); | |
723 | ||
724 | timenum = Score.timenum; /* init to current time sig numerator*/ | |
725 | *timeden_p = Score.timeden; /* init to current time sig denom */ | |
726 | ||
727 | /* if stuff doesn't cross any bar lines, we can return right away */ | |
728 | if (stuff_p->end.bars == 0) | |
729 | return (mainll_p); | |
730 | ||
731 | mst_p = mainll_p; /* remember last staff of this number */ | |
732 | ||
733 | staffno = mainll_p->u.staff_p->staffno; | |
734 | ||
735 | /* | |
736 | * Count past the right number of bar lines, keeping the time sig | |
737 | * denominator up to date. | |
738 | */ | |
739 | for (b = 0; b < stuff_p->end.bars; b++) { | |
740 | for (mainll_p = mainll_p->next; | |
741 | mainll_p != 0 && mainll_p->str != S_BAR; | |
742 | mainll_p = mainll_p->next) { | |
743 | ||
744 | if (mainll_p->str == S_SSV && | |
745 | mainll_p->u.ssv_p->used[TIME] == YES) { | |
746 | timenum = mainll_p->u.ssv_p->timenum; | |
747 | *timeden_p = mainll_p->u.ssv_p->timeden; | |
748 | } | |
749 | ||
750 | /* bail out if multirest encountered */ | |
751 | if (mainll_p->str == S_STAFF && mainll_p->u.staff_p-> | |
752 | groups_p[0]->basictime < -1) | |
753 | return (0); | |
754 | ||
755 | /* remember last staff of this number */ | |
756 | if (mainll_p->str == S_STAFF && mainll_p->u.staff_p-> | |
757 | staffno == staffno) | |
758 | mst_p = mainll_p; | |
759 | } | |
760 | /* if end of song, set to last bar line and return this staff*/ | |
761 | if (mainll_p == 0) { | |
762 | stuff_p->end.count = timenum + 1; | |
763 | return (mst_p); | |
764 | } | |
765 | } | |
766 | ||
767 | /* | |
768 | * mainll_p points at the bar line preceding the place where the stuff | |
769 | * ends. Continue forward to find the correct STAFF. | |
770 | */ | |
771 | for (mainll_p = mainll_p->next ; | |
772 | mainll_p != 0 && mainll_p->str != S_BAR; | |
773 | mainll_p = mainll_p->next) { | |
774 | ||
775 | if (mainll_p->str == S_SSV && | |
776 | mainll_p->u.ssv_p->used[TIME] == YES) | |
777 | *timeden_p = mainll_p->u.ssv_p->timeden; | |
778 | ||
779 | if (mainll_p->str == S_STAFF && | |
780 | mainll_p->u.staff_p->staffno == staffno) | |
781 | break; | |
782 | } | |
783 | ||
784 | /* if end of song, set to last bar line and return this staff */ | |
785 | if (mainll_p == 0) { | |
786 | stuff_p->end.count = timenum + 1; | |
787 | return (mst_p); /* hit end of song, return last meas */ | |
788 | } | |
789 | if (mainll_p->str == S_BAR) | |
790 | pfatal("stuff crosses FEED where number of staffs changes"); | |
791 | if (mainll_p->u.staff_p->groups_p[0]->basictime < -1) | |
792 | return (0); | |
793 | ||
794 | return (mainll_p); | |
795 | } | |
796 | \f | |
797 | /* | |
798 | * Name: lefttime() | |
799 | * | |
800 | * Abstract: Find time value of the nongrace group left of the given count. | |
801 | * | |
802 | * Returns: time value into measure | |
803 | * | |
804 | * Description: This function finds the nongrace GRPSYL in the given linked | |
805 | * list that is at or left of the given count number. If the | |
806 | * count is less 1, we return time zero, even though technically | |
807 | * time zero is to the right of the given count number. | |
808 | */ | |
809 | ||
810 | static RATIONAL | |
811 | lefttime(count, firstgs_p, timeden) | |
812 | ||
813 | double count; /* which count of the measure */ | |
814 | struct GRPSYL *firstgs_p; /* first GRPSYL of relevant voice in measure */ | |
815 | int timeden; /* denominator of current time signature */ | |
816 | ||
817 | { | |
818 | RATIONAL reqtime; /* time requested */ | |
819 | RATIONAL tottime; /* total time in measure so far */ | |
820 | RATIONAL otottime; /* old total time in measure so far */ | |
821 | struct GRPSYL *gs_p; /* point along group list */ | |
822 | struct GRPSYL *ogs_p; /* (old) point along group list */ | |
823 | ||
824 | ||
825 | /* skip over any initial grace groups */ | |
826 | if (firstgs_p->grpvalue == GV_ZERO) | |
827 | firstgs_p = nextnongrace(firstgs_p); | |
828 | ||
829 | /* if at or before the first count, have to use first group */ | |
830 | if (count <= 1) | |
831 | return (Zero); | |
832 | ||
833 | /* | |
834 | * Get requested time to the nearest half of the smallest fraction of a | |
835 | * count that a user can specify, +1, in lowest terms. The +1 is so | |
836 | * that if the user isn't too accurate, we still land on the intended | |
837 | * group. | |
838 | */ | |
839 | reqtime.n = 2 * MAXBASICTIME * (count - 1) + 0.5 + 1.0; | |
840 | reqtime.d = 2 * MAXBASICTIME * timeden; | |
841 | rred(&reqtime); | |
842 | ||
843 | /* | |
844 | * Loop through this voice accumulating time values. As soon as we | |
845 | * equal or exceed the requested time value, return the previous | |
846 | * group's time. | |
847 | */ | |
848 | otottime = tottime = Zero; | |
849 | for (ogs_p = firstgs_p, gs_p = nextnongrace(ogs_p); gs_p != 0; | |
850 | ogs_p = gs_p, gs_p = nextnongrace(gs_p)) { | |
851 | otottime = tottime; | |
852 | tottime = radd(tottime, ogs_p->fulltime); | |
853 | if (GE(tottime, reqtime)) | |
854 | return (otottime); | |
855 | } | |
856 | ||
857 | /* requested time is after last group; return time of last group */ | |
858 | return (otottime); | |
859 | } | |
860 | \f | |
861 | /* | |
862 | * Name: righttime() | |
863 | * | |
864 | * Abstract: Find time value of the nongrace group right of the given count. | |
865 | * | |
866 | * Returns: time value into measure | |
867 | * | |
868 | * Description: This function finds the nongrace GRPSYL in the given linked | |
869 | * list that is at or right of the given count number. If the | |
870 | * count is greater then the rightmost group in the measure, we | |
871 | * return the time up to the rightmost group, even though | |
872 | * technically that time is to the left of the given count number. | |
873 | */ | |
874 | ||
875 | static RATIONAL | |
876 | righttime(count, firstgs_p, timeden) | |
877 | ||
878 | double count; /* which count of the measure */ | |
879 | struct GRPSYL *firstgs_p; /* first GRPSYL of relevant voice in measure */ | |
880 | int timeden; /* denominator of current time signature */ | |
881 | ||
882 | { | |
883 | RATIONAL reqtime; /* time requested */ | |
884 | RATIONAL tottime; /* total time in measure so far */ | |
885 | struct GRPSYL *gs_p; /* point along group list */ | |
886 | struct GRPSYL *ogs_p; /* (old) point along group list */ | |
887 | ||
888 | ||
889 | /* skip over any initial grace groups */ | |
890 | if (firstgs_p->grpvalue == GV_ZERO) | |
891 | firstgs_p = nextnongrace(firstgs_p); | |
892 | ||
893 | /* if at or before the first count, use first group */ | |
894 | if (count <= 1) | |
895 | return (Zero); | |
896 | ||
897 | /* | |
898 | * Get requested time to the nearest half of the smallest fraction of a | |
899 | * count that a user can specify, -1, in lowest terms. The -1 is so | |
900 | * that if the user isn't too accurate, we still land on the intended | |
901 | * group. | |
902 | */ | |
903 | reqtime.n = 2 * MAXBASICTIME * (count - 1) + 0.5 - 1.0; | |
904 | reqtime.d = 2 * MAXBASICTIME * timeden; | |
905 | rred(&reqtime); | |
906 | ||
907 | /* | |
908 | * Loop through this voice accumulating time values. As soon as we | |
909 | * equal or exceed the requested time value, return that new time, | |
910 | * although don't go beyond the last group's time value. | |
911 | */ | |
912 | tottime = Zero; | |
913 | for (ogs_p = firstgs_p, gs_p = nextnongrace(ogs_p); gs_p != 0; | |
914 | ogs_p = gs_p, gs_p = nextnongrace(gs_p)) { | |
915 | tottime = radd(tottime, ogs_p->fulltime); | |
916 | if (GE(tottime, reqtime)) | |
917 | return (tottime); | |
918 | } | |
919 | ||
920 | /* requested time is after last group; but must return last group */ | |
921 | return (tottime); | |
922 | } | |
923 | \f | |
924 | /* | |
925 | * Name: accdimen() | |
926 | * | |
927 | * Abstract: Find the dimensions of a note's accidental. | |
928 | * | |
929 | * Returns: void | |
930 | * | |
931 | * Description: This function finds the ascent, descent, and width of an | |
932 | * accidental, and returns them through pointers. If a pointer | |
933 | * is null, it doesn't try to fill it in. An accidental char of | |
934 | * '\0' gives zero for each dimension. The function takes into | |
935 | * account whether the accidental is normal or small size, and | |
936 | * whether it has parentheses around it. | |
937 | * NOTE: This function ignores staffscale. | |
938 | */ | |
939 | ||
940 | void | |
941 | accdimen(note_p, ascent_p, descent_p, width_p) | |
942 | ||
943 | struct NOTE *note_p; /* the note whose accidental we're working on*/ | |
944 | float *ascent_p; /* ascent, to be filled in */ | |
945 | float *descent_p; /* descent, to be filled in */ | |
946 | float *width_p; /* width, to be filled in */ | |
947 | ||
948 | { | |
949 | char accchar; /* accidental character number */ | |
950 | int size; /* which size of character */ | |
951 | float halfhigh; /* half the height of a parenthesis */ | |
952 | ||
953 | ||
954 | if (note_p->accidental == '\0') { | |
955 | if (ascent_p != 0) { | |
956 | *ascent_p = 0.0; | |
957 | } | |
958 | if (descent_p != 0) { | |
959 | *descent_p = 0.0; | |
960 | } | |
961 | if (width_p != 0) { | |
962 | *width_p = 0.0; | |
963 | } | |
964 | return; | |
965 | } | |
966 | ||
967 | /* find character name and size of this accidental */ | |
968 | accchar = acc2char(note_p->accidental); | |
969 | size = (note_p->notesize == GS_NORMAL ? DFLT_SIZE : SMALLSIZE); | |
970 | ||
971 | /* get the requested dimensions of this accidental */ | |
972 | if (ascent_p != 0) { | |
973 | *ascent_p = ascent(FONT_MUSIC, size, accchar); | |
974 | } | |
975 | if (descent_p != 0) { | |
976 | *descent_p = descent(FONT_MUSIC, size, accchar); | |
977 | } | |
978 | if (width_p != 0) { | |
979 | *width_p = width(FONT_MUSIC, size, accchar); | |
980 | } | |
981 | ||
982 | /* | |
983 | * If it has parentheses around it, account for that. Assume the left | |
984 | * and right parens are symmetrical. They will be centered on the line | |
985 | * or space of the note. | |
986 | */ | |
987 | if (note_p->acc_has_paren) { | |
988 | if (width_p != 0) { | |
989 | *width_p += 2 * width(FONT_TR, size, '('); | |
990 | } | |
991 | halfhigh = height(FONT_TR, size, '(') / 2.0; | |
992 | if (ascent_p != 0 && halfhigh > *ascent_p) { | |
993 | *ascent_p = halfhigh; | |
994 | } | |
995 | if (descent_p != 0 && halfhigh > *descent_p) { | |
996 | *descent_p = halfhigh; | |
997 | } | |
998 | } | |
999 | } | |
1000 | \f | |
1001 | /* | |
1002 | * Name: staffvertspace() | |
1003 | * | |
1004 | * Abstract: Find the minimum amount of vertical space a staff should have. | |
1005 | * | |
1006 | * Returns: the amount of vertical distance in inches | |
1007 | * | |
1008 | * Description: This function finds the minimum amount of vertical space that | |
1009 | * should be allocated for a staff, based on how many lines it has | |
1010 | * and whether it is tablature. This does not take into account | |
1011 | * the extra space required by things sticking out farther; it's | |
1012 | * just for the staff itself, plus the extra white space required | |
1013 | * by staffs that have few lines. The SSVs must be up to date. | |
1014 | * NOTE: This function takes staffscale into account. | |
1015 | */ | |
1016 | ||
1017 | double | |
1018 | staffvertspace(s) | |
1019 | ||
1020 | int s; /* staff number */ | |
1021 | ||
1022 | { | |
1023 | float space; /* the answer */ | |
1024 | ||
1025 | ||
1026 | /* | |
1027 | * Base space on number of steps between top and bottom lines. But for | |
1028 | * tablature, it must be scaled because the lines are farther apart. | |
1029 | */ | |
1030 | space = (svpath(s, STAFFLINES)->stafflines - 1) * 2 * STEPSIZE; | |
1031 | if (is_tab_staff(s)) | |
1032 | space *= TABRATIO; | |
1033 | ||
1034 | /* but don't ever return less than a (scaled) regular 5 line staff */ | |
1035 | return (svpath(s, STAFFSCALE)->staffscale * MAX(space, 8.0 * STEPSIZE)); | |
1036 | } | |
1037 | \f | |
1038 | /* | |
1039 | * Name: halfstaffhi() | |
1040 | * | |
1041 | * Abstract: Find half of the staff height. | |
1042 | * | |
1043 | * Returns: half the staff height in inches | |
1044 | * | |
1045 | * Description: This function finds half of the staff's height, based on how | |
1046 | * many lines it has and whether it is tablature. This does not | |
1047 | * take into account the extra space required by things sticking | |
1048 | * out farther; it's just for the staff itself, except that one | |
1049 | * line staffs are given a minimum instead of the zero you would | |
1050 | * expect. The SSVs must be up to date. | |
1051 | * NOTE: This function takes staffscale into account. | |
1052 | */ | |
1053 | ||
1054 | double | |
1055 | halfstaffhi(s) | |
1056 | ||
1057 | int s; /* staff number */ | |
1058 | ||
1059 | { | |
1060 | float space; /* the answer */ | |
1061 | ||
1062 | ||
1063 | /* | |
1064 | * Base space on the number of steps between the top line and the | |
1065 | * middle of the staff. But for tablature, it must be scaled because | |
1066 | * the lines are farther apart. | |
1067 | */ | |
1068 | space = (svpath(s, STAFFLINES)->stafflines - 1) * STEPSIZE; | |
1069 | if (is_tab_staff(s)) | |
1070 | space *= TABRATIO; | |
1071 | ||
1072 | /* but don't ever return less than one (scaled) stepsize */ | |
1073 | return (MAX(space, STEPSIZE) * svpath(s, STAFFSCALE)->staffscale); | |
1074 | } | |
1075 | \f | |
1076 | /* | |
1077 | * Name: ratbend() | |
1078 | * | |
1079 | * Abstract: Convert a bend distance to rational. | |
1080 | * | |
1081 | * Returns: the rational number answer; 0/1 if null bend or no bend | |
1082 | * | |
1083 | * Description: This function, given a NOTE structure from a tab staff, returns | |
1084 | * the amount of the bend (if any) as a rational number. | |
1085 | */ | |
1086 | ||
1087 | RATIONAL | |
1088 | ratbend(note_p) | |
1089 | ||
1090 | struct NOTE *note_p; | |
1091 | ||
1092 | { | |
1093 | RATIONAL answer; | |
1094 | ||
1095 | ||
1096 | if (note_p->BEND == 0) | |
1097 | return (Zero); | |
1098 | ||
1099 | answer.d = BENDDEN(*note_p); | |
1100 | answer.n = BENDNUM(*note_p) + BENDINT(*note_p) * answer.d; | |
1101 | rred(&answer); | |
1102 | ||
1103 | return (answer); | |
1104 | } | |
1105 | \f | |
1106 | /* | |
1107 | * Name: notehorz() | |
1108 | * | |
1109 | * Abstract: Find horizontal boundary of note and associated things. | |
1110 | * | |
1111 | * Returns: the RE or RW | |
1112 | * | |
1113 | * Description: This function finds the horizontal boundary of a note, | |
1114 | * including accidentals, dots, etc., all the things that can be | |
1115 | * on the note. The note's own RE and RW only tell about the note | |
1116 | * head itself. | |
1117 | * NOTE: This function takes staffscale into account. The SSVs | |
1118 | * need not be up to date, but Staffscale and Stdpad must be set. | |
1119 | */ | |
1120 | ||
1121 | double | |
1122 | notehorz(gs_p, note_p, coord) | |
1123 | ||
1124 | struct GRPSYL *gs_p; /* the group the note is in */ | |
1125 | struct NOTE *note_p; /* point at the note */ | |
1126 | int coord; /* RE or RW */ | |
1127 | ||
1128 | { | |
1129 | int s; /* index into slurtolist */ | |
1130 | double h; /* the answer */ | |
1131 | ||
1132 | ||
1133 | if (coord == RE) { | |
1134 | if (note_p->note_has_paren == YES && | |
1135 | ! is_tab_staff(gs_p->staffno)) { | |
1136 | /* | |
1137 | * If there are parens around the note, start there. | |
1138 | * Note: this field does not apply on tab staff; it | |
1139 | * is only there for carrying over to tabnote staff. | |
1140 | * Tab staff uses FRET_HAS_PAREN, and this distance is | |
1141 | * included in the size of the "note" (fret) itself. | |
1142 | */ | |
1143 | h = note_p->erparen; | |
1144 | } else { | |
1145 | /* | |
1146 | * If non-tablature and there are dots, start from the | |
1147 | * first dot. Otherwise start from the note. | |
1148 | */ | |
1149 | if (is_tab_staff(gs_p->staffno) == NO && | |
1150 | gs_p->dots > 0) { | |
1151 | h = gs_p->xdotr + 6 * Stdpad; | |
1152 | if (gs_p->dots > 1) { | |
1153 | h += (gs_p->dots - 1) * (2 * Stdpad + | |
1154 | width(FONT_MUSIC, DFLT_SIZE, C_DOT)); | |
1155 | } | |
1156 | } else { | |
1157 | h = note_p->c[RE] + Stdpad; | |
1158 | } | |
1159 | } | |
1160 | ||
1161 | /* | |
1162 | * If this note has a slur to nowhere (and there can be at most | |
1163 | * one such), include its length. | |
1164 | */ | |
1165 | for (s = 0; s < note_p->nslurto; s++) { | |
1166 | if (note_p->slurtolist[s].octave == OUT_UPWARD || | |
1167 | note_p->slurtolist[s].octave == OUT_DOWNWARD) { | |
1168 | h += Staffscale * (SLIDEXLEN + STDPAD); | |
1169 | break; | |
1170 | } | |
1171 | } | |
1172 | ||
1173 | } else { /* RW */ | |
1174 | ||
1175 | if (note_p->note_has_paren == YES && | |
1176 | ! is_tab_staff(gs_p->staffno)) { | |
1177 | /* if parens around note, start there */ | |
1178 | h = note_p->wlparen; | |
1179 | } else if (is_tab_staff(gs_p->staffno) == NO && | |
1180 | note_p->accidental != '\0') { | |
1181 | /* if there's an accidental, start there */ | |
1182 | /* (this includes any parens around the accidental) */ | |
1183 | h = note_p->waccr; | |
1184 | } else { | |
1185 | /* the note head itself, with padding */ | |
1186 | h = note_p->c[RW] - Stdpad; | |
1187 | } | |
1188 | ||
1189 | /* | |
1190 | * If this note has a slur from nowhere (and there can be at | |
1191 | * most one such), include its length. | |
1192 | */ | |
1193 | for (s = 0; s < note_p->nslurto; s++) { | |
1194 | if (note_p->slurtolist[s].octave == IN_UPWARD || | |
1195 | note_p->slurtolist[s].octave == IN_DOWNWARD) { | |
1196 | h -= Staffscale * (SLIDEXLEN + STDPAD); | |
1197 | break; | |
1198 | } | |
1199 | } | |
1200 | } | |
1201 | ||
1202 | return (h); | |
1203 | } | |
1204 | \f | |
1205 | /* | |
1206 | * Name: allsmall() | |
1207 | * | |
1208 | * Abstract: Do the given groups (of notes) consist entirely of small notes? | |
1209 | * | |
1210 | * Returns: YES or NO | |
1211 | * | |
1212 | * Description: This function is given pointer to two GRPSYLs in a linked list. | |
1213 | * They may point to the same GRPSYL, or the second may point to a | |
1214 | * later GRPSYL in the list. The function returns YES if all the | |
1215 | * notes in these GRPSYLs and any intervening GRPSYLs are small. | |
1216 | * Any GRPSYLs that are not for notes are ignored. | |
1217 | */ | |
1218 | ||
1219 | int | |
1220 | allsmall(gs1_p, gs2_p) | |
1221 | ||
1222 | struct GRPSYL *gs1_p; /* starting group */ | |
1223 | struct GRPSYL *gs2_p; /* ending group (may be same as starting group) */ | |
1224 | ||
1225 | { | |
1226 | struct GRPSYL *gs_p; /* point along the list */ | |
1227 | int n; /* index into notelist */ | |
1228 | ||
1229 | ||
1230 | /* check every group, and return NO if anything is normal size */ | |
1231 | for (gs_p = gs1_p; gs_p != gs2_p->next; gs_p = gs_p->next) { | |
1232 | if (gs_p->grpcont == GC_NOTES && gs_p->grpsize == GS_NORMAL) { | |
1233 | for (n = 0; n < gs_p->nnotes; n++) { | |
1234 | if (gs_p->notelist[n].notesize == GS_NORMAL) | |
1235 | return (NO); | |
1236 | } | |
1237 | } | |
1238 | } | |
1239 | ||
1240 | return (YES); /* everything must have been small */ | |
1241 | } | |
1242 | \f | |
1243 | /* | |
1244 | * Name: finalstemadjust() | |
1245 | * | |
1246 | * Abstract: Make final adjustments to the stem length. | |
1247 | * | |
1248 | * Returns: void | |
1249 | * | |
1250 | * Description: This function makes final adjustments to the stem length that | |
1251 | * all the cases have in common. Coming in, it is set to the | |
1252 | * stem length not counting the part between notes of a multinote | |
1253 | * group, and it doesn't account for the thickness of a beam. | |
1254 | * The SSVs must be up to date. | |
1255 | * NOTE: This function takes staffscale into account. | |
1256 | */ | |
1257 | ||
1258 | void | |
1259 | finalstemadjust(gs_p) | |
1260 | ||
1261 | struct GRPSYL *gs_p; /* group whose stemlen should be adjusted */ | |
1262 | ||
1263 | { | |
1264 | float stepdiff; /* distance between outer notes in steps */ | |
1265 | ||
1266 | ||
1267 | /* if it is negative (note on wrong side of beam), zap it */ | |
1268 | if (gs_p->stemlen < 0) | |
1269 | gs_p->stemlen = 0; | |
1270 | ||
1271 | /* add distance between outer notes of group */ | |
1272 | stepdiff = gs_p->notelist[0].c[RY] - | |
1273 | gs_p->notelist[ gs_p->nnotes - 1 ].c[RY]; | |
1274 | gs_p->stemlen += stepdiff; | |
1275 | ||
1276 | /* | |
1277 | * Decr the length by half the thickness of the beam, but don't let it | |
1278 | * get less than the distance between the outer notes of the group. | |
1279 | */ | |
1280 | gs_p->stemlen -= (W_WIDE * POINT / 2.0) * | |
1281 | (gs_p->grpsize == GS_NORMAL ? 1.0 : SM_FACTOR) * | |
1282 | svpath(gs_p->staffno, STAFFSCALE)->staffscale; | |
1283 | gs_p->stemlen = MAX(gs_p->stemlen, stepdiff); | |
1284 | } | |
1285 | \f | |
1286 | /* | |
1287 | * Name: getstemshift() | |
1288 | * | |
1289 | * Abstract: Find how far a stem is from the group's X coordinate. | |
1290 | * | |
1291 | * Returns: the distance in inches | |
1292 | * | |
1293 | * Description: This function finds how far a group's stem is shifted | |
1294 | * horizontally from the group's X coordinate. | |
1295 | * NOTE: This function takes staffscale into account. | |
1296 | */ | |
1297 | ||
1298 | double | |
1299 | getstemshift(gs_p) | |
1300 | ||
1301 | struct GRPSYL *gs_p; /* group whose stemlen should be adjusted */ | |
1302 | ||
1303 | { | |
1304 | /* return half the width of the widest note in the group */ | |
1305 | return (svpath(gs_p->staffno, STAFFSCALE)->staffscale * | |
1306 | widest_head(gs_p) / 2.0); | |
1307 | } | |
1308 | \f | |
1309 | /* | |
1310 | * Name: vscheme_voices() | |
1311 | * | |
1312 | * Abstract: Given a vscheme, how many voices are in it? | |
1313 | * | |
1314 | * Returns: number of voices | |
1315 | * | |
1316 | * Description: This function is given one of the valid vschemes, and it | |
1317 | * returns the number of voices that vscheme allows. | |
1318 | */ | |
1319 | ||
1320 | int | |
1321 | vscheme_voices(vscheme) | |
1322 | ||
1323 | int vscheme; /* the given vscheme */ | |
1324 | ||
1325 | { | |
1326 | switch (vscheme) { | |
1327 | case V_1: | |
1328 | return (1); | |
1329 | ||
1330 | case V_2OPSTEM: | |
1331 | case V_2FREESTEM: | |
1332 | return (2); | |
1333 | ||
1334 | case V_3OPSTEM: | |
1335 | case V_3FREESTEM: | |
1336 | return (3); | |
1337 | ||
1338 | default: | |
1339 | pfatal("invalid vscheme in vscheme_voices()"); | |
1340 | } | |
1341 | ||
1342 | return (0); | |
1343 | } | |
1344 | \f | |
1345 | /* | |
1346 | * Name: chmgrp2staffm() | |
1347 | * | |
1348 | * Abstract: Given a chord and group, find the group's staff. | |
1349 | * | |
1350 | * Returns: pointer to staff's MLL item | |
1351 | * | |
1352 | * Description: This function is given the MLL item for a chord, and a group | |
1353 | * connected to the chord. It returns the MLL item for the staff | |
1354 | * that the group is connected to. The group can belong to any of | |
1355 | * the staff's voices. | |
1356 | */ | |
1357 | ||
1358 | struct MAINLL * | |
1359 | chmgrp2staffm(mll_p, gs_p) | |
1360 | ||
1361 | struct MAINLL *mll_p; /* starts as MLL for the chord */ | |
1362 | struct GRPSYL *gs_p; /* starts as GRPSYL the given group */ | |
1363 | ||
1364 | { | |
1365 | /* find the first group in this measure */ | |
1366 | for ( ; gs_p->prev != 0; gs_p = gs_p->prev) | |
1367 | ; | |
1368 | ||
1369 | /* find the staff that it belongs to */ | |
1370 | for ( ; mll_p != 0; mll_p = mll_p->next) { | |
1371 | ||
1372 | if (mll_p->str == S_STAFF && | |
1373 | (mll_p->u.staff_p->groups_p[0] == gs_p || | |
1374 | mll_p->u.staff_p->groups_p[1] == gs_p || | |
1375 | mll_p->u.staff_p->groups_p[2] == gs_p)) | |
1376 | break; | |
1377 | } | |
1378 | if (mll_p == 0) | |
1379 | pfatal("can't find staff in chmgrp2staffm"); | |
1380 | ||
1381 | return (mll_p); | |
1382 | } | |
1383 | \f | |
1384 | /* | |
1385 | * Name: shiftgs() | |
1386 | * | |
1387 | * Abstract: Shift a GRPSYL's relative horizontal coords. | |
1388 | * | |
1389 | * Returns: void | |
1390 | * | |
1391 | * Description: This function is a GRPSYL and a shift amount. It shifts the | |
1392 | * GRPSYL's relative horizontal coords by that amount, also | |
1393 | * including any preceding grace GRPSYLs. | |
1394 | */ | |
1395 | ||
1396 | void | |
1397 | shiftgs(gs_p, shift) | |
1398 | ||
1399 | struct GRPSYL *gs_p; /* the main GRPSYL */ | |
1400 | double shift; | |
1401 | ||
1402 | { | |
1403 | struct GRPSYL *ggs_p; /* point at a grace group */ | |
1404 | ||
1405 | ||
1406 | gs_p->c[RX] += shift; | |
1407 | gs_p->c[RW] += shift; | |
1408 | gs_p->c[RE] += shift; | |
1409 | ||
1410 | /* apply shift to any preceding grace groups */ | |
1411 | for (ggs_p = gs_p->prev; ggs_p != 0 && ggs_p->grpvalue == GV_ZERO; | |
1412 | ggs_p = ggs_p->prev) { | |
1413 | ggs_p->c[RX] += shift; | |
1414 | ggs_p->c[RW] += shift; | |
1415 | ggs_p->c[RE] += shift; | |
1416 | } | |
1417 | } | |
1418 | \f | |
1419 | /* | |
1420 | * Name: nearestline() | |
1421 | * | |
1422 | * Abstract: Round a vertical offset to the nearest staff line. | |
1423 | * | |
1424 | * Returns: the resulting offset | |
1425 | * | |
1426 | * Description: This function rounds its parameter off to a multiple of 2 | |
1427 | * stepsizes. | |
1428 | * NOTE: This function takes staffscale into account. The SSVs | |
1429 | * need not be up to date, but Stepsize must be set. | |
1430 | */ | |
1431 | ||
1432 | double | |
1433 | nearestline(offset) | |
1434 | ||
1435 | double offset; /* offset from center staff line */ | |
1436 | ||
1437 | { | |
1438 | if (offset >= 0) { | |
1439 | offset /= (2 * Stepsize); | |
1440 | offset = (int)(offset + 0.5); | |
1441 | offset *= (2 * Stepsize); | |
1442 | } else { | |
1443 | offset = -offset; | |
1444 | offset /= (2 * Stepsize); | |
1445 | offset = (int)(offset + 0.5); | |
1446 | offset *= (2 * Stepsize); | |
1447 | offset = -offset; | |
1448 | } | |
1449 | ||
1450 | return (offset); | |
1451 | } | |
1452 | \f | |
1453 | /* | |
1454 | * Name: vfyoffset() | |
1455 | * | |
1456 | * Abstract: Verify horizontal offsets are not in conflict. | |
1457 | * | |
1458 | * Returns: void | |
1459 | * | |
1460 | * Description: This function prints a warning if the horizontal offsets of | |
1461 | * voices 1 and 2 are in conflict. In that case it also zaps | |
1462 | * the bad offsets. | |
1463 | */ | |
1464 | ||
1465 | void | |
1466 | vfyoffset(g_p) | |
1467 | ||
1468 | struct GRPSYL *g_p[]; /* array of pointers to two groups */ | |
1469 | ||
1470 | { | |
1471 | /* the only errors are cases where "+" and "-" are used */ | |
1472 | if (g_p[0]->ho_usage != HO_LEFT && g_p[0]->ho_usage != HO_RIGHT) | |
1473 | return; | |
1474 | ||
1475 | /* can't both be "+" or both be "-" */ | |
1476 | if (g_p[0]->ho_usage == g_p[1]->ho_usage) { | |
1477 | ||
1478 | l_warning( | |
1479 | g_p[1]->inputfile, g_p[1]->inputlineno, | |
1480 | "voices 1 and 2 cannot both have horizontal offset '%c'; ignoring them", | |
1481 | g_p[0]->ho_usage == HO_LEFT ? '-' :'+'); | |
1482 | ||
1483 | g_p[0]->ho_usage = HO_NONE; | |
1484 | g_p[1]->ho_usage = HO_NONE; | |
1485 | } | |
1486 | } | |
1487 | \f | |
1488 | /* | |
1489 | * Name: adjslope() | |
1490 | * | |
1491 | * Abstract: Adjust the slope of a beam. | |
1492 | * | |
1493 | * Returns: the new slope | |
1494 | * | |
1495 | * Description: This function is given the slope of a beam as determined by | |
1496 | * linear regression. It adjusts it according to the "beamslope" | |
1497 | * parameter. | |
1498 | */ | |
1499 | ||
1500 | double | |
1501 | adjslope(g_p, oldslope, betweencsb) | |
1502 | ||
1503 | struct GRPSYL *g_p; /* pointer to GRPSYL to get staff and voice from */ | |
1504 | double oldslope; /* the given slope */ | |
1505 | int betweencsb; /* is this beam CSB and between the staffs? */ | |
1506 | ||
1507 | { | |
1508 | struct SSV *ssv_p; /* for getting fact and max */ | |
1509 | float beamfact; /* to multiply by */ | |
1510 | float beammax; /* max angle in degrees */ | |
1511 | float newangle; /* the adjusted angle */ | |
1512 | ||
1513 | ||
1514 | /* find parameter values */ | |
1515 | ssv_p = vvpath(g_p->staffno, g_p->vno, BEAMSLOPE); | |
1516 | beamfact = ssv_p->beamfact; | |
1517 | beammax = ssv_p->beammax; | |
1518 | ||
1519 | /* | |
1520 | * If cross staff beaming and the beam is between the staffs, we allow | |
1521 | * a somewhat bigger angle, because it may be necessary to avoid | |
1522 | * collisions. | |
1523 | */ | |
1524 | if (betweencsb == YES) | |
1525 | beammax *= 1.4; | |
1526 | ||
1527 | /* new angle = old angle * beamfact */ | |
1528 | newangle = (atan(oldslope) * 180.0 / PI) * beamfact; | |
1529 | ||
1530 | /* force it to stay within the limit */ | |
1531 | if (newangle > beammax) | |
1532 | newangle = beammax; | |
1533 | else if (newangle < -beammax) | |
1534 | newangle = -beammax; | |
1535 | ||
1536 | /* return as slope */ | |
1537 | return (tan(newangle * PI / 180.0)); | |
1538 | } | |
1539 | \f | |
1540 | /* | |
1541 | * Name: curve_y_at_x() | |
1542 | * | |
1543 | * Abstract: Given a curve and an X value, return the Y value there. | |
1544 | * | |
1545 | * Returns: the Y value | |
1546 | * | |
1547 | * Description: This function should only be called for curves where the X | |
1548 | * value of each point in the curve list is greater than the | |
1549 | * previous point's X value, although it's okay if the curve | |
1550 | * itself is not strictly increasing in X value all the time as | |
1551 | * you go from the start to the end. | |
1552 | * | |
1553 | * If the X value given is less than the first point's, it returns | |
1554 | * the Y of the first point. If the X value is greater than the | |
1555 | * last point's, it returns the Y of the last point. Otherwise, | |
1556 | * it returns a Y value of the curve at that X value. I say "a" | |
1557 | * Y value, because if the curve isn't strictly increasing, there | |
1558 | * can be multiple answers, and it just returns one of them. | |
1559 | * | |
1560 | * The function assumes that the curve points will be connected by | |
1561 | * cubic curves, according to the algorithm in calccurve() and | |
1562 | * findcontrol(). | |
1563 | */ | |
1564 | ||
1565 | double | |
1566 | curve_y_at_x(first_p, x) | |
1567 | ||
1568 | struct CRVLIST *first_p; /* left endpoint of curve */ | |
1569 | double x; /* X coord at which we need Y */ | |
1570 | ||
1571 | { | |
1572 | float y; /* the answer */ | |
1573 | float a, b, c; /* coefficients for a cubic */ | |
1574 | struct CRVLIST *left_p, *right_p; /* endpoints of a cubic segment */ | |
1575 | float rotangle; /* rotate new system to get old (in radians) */ | |
1576 | float tranx, trany; /* a point translated to another coord system */ | |
1577 | float pointx, pointy; /* trans & rotated in another coord system */ | |
1578 | float lineslope, intercept; /* of a line through the given x */ | |
1579 | float cos_rotangle, sin_rotangle; /* for saving these values */ | |
1580 | float deltax, deltay; /* of endpoints of segment between 2 points */ | |
1581 | float len; /* length of segment between 2 points */ | |
1582 | ||
1583 | ||
1584 | /* | |
1585 | * If the first point of the curve is at or already beyond the given x, | |
1586 | * return the first point's y. | |
1587 | */ | |
1588 | if (first_p->x >= x) { | |
1589 | return (first_p->y); | |
1590 | } | |
1591 | right_p = 0; /* for lint */ | |
1592 | for (left_p = first_p; left_p->next != 0; left_p = left_p->next) { | |
1593 | right_p = left_p->next; | |
1594 | /* if x is right at this point, use this point's y */ | |
1595 | if (right_p->x == x) { | |
1596 | return (right_p->y); | |
1597 | } | |
1598 | /* if x is within this interval, break out */ | |
1599 | if (left_p->x < x && x < right_p->x) { | |
1600 | break; | |
1601 | } | |
1602 | } | |
1603 | /* if this happens, x is beyond the last point, so use last point's y */ | |
1604 | if (left_p->next == 0) { | |
1605 | return (left_p->y); | |
1606 | } | |
1607 | ||
1608 | /* | |
1609 | * The given x is between the x coords of two of the points in the | |
1610 | * curvelist. So we need to find the cubic arc that calccurve() and | |
1611 | * findcontrol() would use, if this curve is going to be used. The | |
1612 | * cubic arc is determined in a translated/rotated coordinate system | |
1613 | * where left_p is (0,0) and right_p is on the positive X axix. | |
1614 | * rotangle is the angle from the segment between left_p and right_p | |
1615 | * to the real X axis. The cubic, in the translated/rotated system, is | |
1616 | * y = a x^3 + b x^2 + c x. It turns out that the constant term is | |
1617 | * always zero. | |
1618 | */ | |
1619 | rotangle = findcubic(left_p, right_p, &a, &b, &c); | |
1620 | ||
1621 | /* | |
1622 | * If left_p->y == right_p->y, rotangle is zero, meaning no rotation was | |
1623 | * necessary, only translation. In that case we can simply plug into | |
1624 | * the cubic we found, adjusting for the translation. A fudge factor is | |
1625 | * needed so that we don't take the tangent of almost 90 degrees below. | |
1626 | */ | |
1627 | if (fabs(rotangle) < 0.001) { | |
1628 | pointx = x - left_p->x; | |
1629 | pointy = ((a * pointx + b) * pointx + c) * pointx; | |
1630 | y = pointy + left_p->y; | |
1631 | return (y); | |
1632 | } | |
1633 | ||
1634 | /* | |
1635 | * Rotation was necessary. In the original coord system, picture a | |
1636 | * vertical line at the given x value. It intersects the cubic, | |
1637 | * possibly in more than one place. We want the y value at the | |
1638 | * intersection. In the translated/rotated system, this line has a | |
1639 | * slope as determine below. | |
1640 | */ | |
1641 | if (rotangle < 0.0) { | |
1642 | lineslope = tan(PI / 2.0 + rotangle); | |
1643 | } else { | |
1644 | lineslope = tan(-PI / 2.0 + rotangle); | |
1645 | } | |
1646 | ||
1647 | /* | |
1648 | * In the real coord system, the vertical line hits (x, 0). Find this | |
1649 | * point in the translated/rotated system. | |
1650 | */ | |
1651 | /* first translate */ | |
1652 | tranx = x - left_p->x; | |
1653 | trany = -left_p->y; | |
1654 | /* then rotate */ | |
1655 | cos_rotangle = cos(rotangle); /* save to avoid recalculation */ | |
1656 | sin_rotangle = sin(rotangle); | |
1657 | pointx = tranx * cos_rotangle - trany * sin_rotangle; | |
1658 | pointy = trany * cos_rotangle + tranx * sin_rotangle; | |
1659 | ||
1660 | /* find Y intercept in the translated/rotated system */ | |
1661 | intercept = pointy - lineslope * pointx; | |
1662 | ||
1663 | /* | |
1664 | * Now, in the tran/rot coord system, we need to find the intersection | |
1665 | * of this line | |
1666 | * y = lineslope * x + intercept | |
1667 | * and the cubic | |
1668 | * y = a * x^3 + b * x^2 + c * x | |
1669 | * Setting the two values of y equal, we get | |
1670 | * lineslope * x + intercept = a * x^3 + b * x^2 + c * x | |
1671 | * or | |
1672 | * a * x^3 + b * x^2 + (c - lineslope) * x - intercept = 0 | |
1673 | */ | |
1674 | /* find intersection point in the tran/rot coord system */ | |
1675 | deltax = right_p->x - left_p->x; | |
1676 | deltay = right_p->y - left_p->y; | |
1677 | len = sqrt(SQUARED(deltax) + SQUARED(deltay)); | |
1678 | pointx = solvecubic(a, b, c-lineslope, -intercept, | |
1679 | 0.0, len, Stdpad / 2.0); | |
1680 | pointy = lineslope * pointx + intercept; | |
1681 | ||
1682 | /* rotate backwards, getting Y value */ | |
1683 | trany = pointy * cos_rotangle - pointx * sin_rotangle; | |
1684 | ||
1685 | /* translate back to the original coord system */ | |
1686 | y = trany + left_p->y; | |
1687 | ||
1688 | return (y); | |
1689 | } | |
1690 | \f | |
1691 | /* | |
1692 | * Name: findcubic() | |
1693 | * | |
1694 | * Abstract: Given neighboring curve points, find cubic and rotation angle. | |
1695 | * | |
1696 | * Returns: angle from new coord system's X axis to old system's (radians) | |
1697 | * | |
1698 | * Description: This function uses a new coordinate system, where the left | |
1699 | * curve point is (0, 0), and the right curve point is on the | |
1700 | * positive X axis. It finds the coefficients for the cubic arc | |
1701 | * that will be put through these points. It returns the angle | |
1702 | * that the old coord system needs to be rotated by to get to | |
1703 | * the new system. | |
1704 | */ | |
1705 | ||
1706 | double | |
1707 | findcubic(left_p, right_p, a_p, b_p, c_p) | |
1708 | ||
1709 | struct CRVLIST *left_p; /* left endpoint of cubic arc */ | |
1710 | struct CRVLIST *right_p; /* right endpoint of cubic arc */ | |
1711 | float *a_p, *b_p, *c_p; /* return the answers here, the coefficients */ | |
1712 | ||
1713 | { | |
1714 | double langle; /* half angle from prev segment to this one */ | |
1715 | double rangle; /* half angle from this segment to next one */ | |
1716 | float deltax, deltay; /* for this segment */ | |
1717 | float len; /* of this segment */ | |
1718 | float lslope, rslope; /* slope of tangent line at left & right point*/ | |
1719 | float thisang, prevang, nextang; /* angle of segment with horizontal */ | |
1720 | ||
1721 | ||
1722 | langle = rangle = 0.0; /* for lint */ | |
1723 | ||
1724 | /* get angle of this segment */ | |
1725 | thisang = atan((right_p->y - left_p->y) / (right_p->x - left_p->x)); | |
1726 | ||
1727 | if (left_p->prev != 0) { | |
1728 | /* there is a previous segment; find its angle */ | |
1729 | prevang = atan((left_p->y - left_p->prev->y) / | |
1730 | (left_p->x - left_p->prev->x)); | |
1731 | /* half the change in angle */ | |
1732 | langle = (thisang - prevang) / 2.0; | |
1733 | } | |
1734 | if (right_p->next != 0) { | |
1735 | /* there is a next segment; find its angle */ | |
1736 | nextang = atan((right_p->next->y - right_p->y) / | |
1737 | (right_p->next->x - right_p->x)); | |
1738 | /* half the change in angle */ | |
1739 | rangle = (nextang - thisang) / 2.0; | |
1740 | } | |
1741 | if (left_p->prev == 0) { | |
1742 | /* no previous segment; use same angle as on the right */ | |
1743 | langle = rangle; | |
1744 | } | |
1745 | if (right_p->next == 0) { | |
1746 | /* no next segment; use same angle as on the left */ | |
1747 | rangle = langle; | |
1748 | } | |
1749 | ||
1750 | /* get lengths of this segment */ | |
1751 | deltax = right_p->x - left_p->x; | |
1752 | deltay = right_p->y - left_p->y; | |
1753 | len = sqrt(SQUARED(deltax) + SQUARED(deltay)); | |
1754 | ||
1755 | /* | |
1756 | * Rotate and translate the axes so that the starting point (left_p) | |
1757 | * is at the origin, and the ending point (right_p) is on the positive | |
1758 | * x axis. Their coords are (0, 0) and (len, 0). We are going to | |
1759 | * find a cubic equation that intersects the endpoints, and has the | |
1760 | * required slope at those points. The equation is | |
1761 | * y = a x^3 + b x^2 + c x + d | |
1762 | * so the slope is | |
1763 | * y' = 3 a x^2 + 2 b x + c | |
1764 | * By plugging the two points into these, you get 4 equations in the 4 | |
1765 | * unknowns a, b, c, d. | |
1766 | */ | |
1767 | /* find the slope of the tangent lines at the first & second points */ | |
1768 | lslope = -tan(langle); | |
1769 | rslope = tan(rangle); | |
1770 | ||
1771 | /* set values of a, b, c (d turns out to be always 0) */ | |
1772 | *a_p = (lslope + rslope) / SQUARED(len); | |
1773 | *b_p = (-2.0 * lslope - rslope) / len; | |
1774 | *c_p = lslope; | |
1775 | ||
1776 | return (-thisang); | |
1777 | } | |
1778 | \f | |
1779 | /* | |
1780 | * Name: solvecubic() | |
1781 | * | |
1782 | * Abstract: Find a solution to a cubic equation within a given interval. | |
1783 | * | |
1784 | * Returns: the solution | |
1785 | * | |
1786 | * Description: This function is given the coefficients of a cubic equation and | |
1787 | * the boundaries of an interval. The function must be positive | |
1788 | * at one end and negative at the other (or zero is okay at | |
1789 | * either). It uses the "modified regula falsi" algorithm to find | |
1790 | * a solution, meaning that it keeps narrowing down the interval. | |
1791 | * It stops when the inverval size <= the threshhold given. But | |
1792 | * in case something goes wrong, it also stops after 20 loops. | |
1793 | */ | |
1794 | ||
1795 | double | |
1796 | solvecubic(a, b, c, d, lo, hi, thresh) | |
1797 | ||
1798 | double a, b, c, d; /* in equation a x^3 + b x^2 + c x + d = 0 */ | |
1799 | double lo, hi; /* low and high boundaries of interval to look in */ | |
1800 | double thresh; /* how close must result be to the true answer */ | |
1801 | ||
1802 | #define FUNC(x) (((a * x + b) * x + c) * x + d) | |
1803 | { | |
1804 | float lovert, hivert; /* y values */ | |
1805 | float mid, midvert; /* a point in the middle and its y value */ | |
1806 | float oldmidvert; /* midvert in previous loop */ | |
1807 | int n; | |
1808 | ||
1809 | ||
1810 | lovert = FUNC(lo); | |
1811 | hivert = FUNC(hi); | |
1812 | ||
1813 | /* | |
1814 | * If the function is positive at both endpoints or negative at both | |
1815 | * endpoints, it was called incorrectly. But we're going to allow for | |
1816 | * a small violation of this due to presumed roundoff error. If one | |
1817 | * endpoint if "very near" zero, we'll pretend it was zero and return | |
1818 | * it as the solution. | |
1819 | */ | |
1820 | if (lovert * hivert > 0.0) { | |
1821 | if (fabs(lovert) < thresh) | |
1822 | return (lo); | |
1823 | if (fabs(hivert) < thresh) | |
1824 | return (hi); | |
1825 | pfatal("solvecubic was called with an invalid interval"); | |
1826 | } | |
1827 | ||
1828 | mid = lo; | |
1829 | midvert = lovert; | |
1830 | ||
1831 | for (n = 0; n < 20 && hi - lo > thresh; n++) { | |
1832 | oldmidvert = midvert; | |
1833 | ||
1834 | /* | |
1835 | * Find a point somewhere in the interval by passing a segment | |
1836 | * through (lo, lovert) and (hi, hivert) and seeing where it | |
1837 | * hits the X axis. | |
1838 | */ | |
1839 | mid = (lovert * hi - hivert * lo) / (lovert - hivert); | |
1840 | midvert = FUNC(mid); | |
1841 | ||
1842 | /* | |
1843 | * Set either the hi or the lo equal to the mid. If the value | |
1844 | * at mid is the same sign as the previous one, divide the | |
1845 | * vert value by 2, so we can eventually get the segment to | |
1846 | * hit on the other side. | |
1847 | */ | |
1848 | if ((lovert > 0.0) != (midvert > 0.0)) { | |
1849 | hi = mid; | |
1850 | hivert = midvert; | |
1851 | if ((midvert > 0.0) == (oldmidvert > 0.0)) { | |
1852 | lovert /= 2.0; | |
1853 | } | |
1854 | } else { | |
1855 | lo = mid; | |
1856 | lovert = midvert; | |
1857 | if ((midvert > 0.0) == (oldmidvert > 0.0)) { | |
1858 | hivert /= 2.0; | |
1859 | } | |
1860 | } | |
1861 | } | |
1862 | ||
1863 | return (mid); | |
1864 | } | |
1865 | \f | |
1866 | /* | |
1867 | * Name: css_affects_stemtip() | |
1868 | * | |
1869 | * Abstract: Do CSS notes (if any) affect the position of the stem's tip? | |
1870 | * | |
1871 | * Returns: YES or NO | |
1872 | * | |
1873 | * Description: This function is given a pointer to a GRPSYL. It must be a | |
1874 | * note group, but can be grace or nongrace. It may be a member | |
1875 | * of a beamed set, or not a member of a beamed set. It decides | |
1876 | * whether the position of the tip of the stem (or where the stem | |
1877 | * would be for a non-stemmed note) is affected by CSS notes. | |
1878 | */ | |
1879 | ||
1880 | int | |
1881 | css_affects_stemtip(gs_p) | |
1882 | ||
1883 | struct GRPSYL *gs_p; /* starts at the given group */ | |
1884 | ||
1885 | { | |
1886 | /* | |
1887 | * For the single (unbeamed) group case, the position of the tip of the | |
1888 | * stem is affected if either the CSS notes are on the stem side, or if | |
1889 | * all the notes are CSS. | |
1890 | */ | |
1891 | if (gs_p->beamloc == NOITEM) { | |
1892 | return (STEMSIDE_CSS(gs_p) || NNN(gs_p) == 0 ? YES : NO); | |
1893 | } | |
1894 | ||
1895 | /* CSB is never CSS */ | |
1896 | if (gs_p->beamto != CS_SAME) { | |
1897 | return (NO); | |
1898 | } | |
1899 | ||
1900 | /* | |
1901 | * For the beamed case, either all or none of the groups can have CSS | |
1902 | * notes on the stem side, if there are any other-staff notes at all | |
1903 | * in the group. This is because all the groups' stems go | |
1904 | * the same direction, and we don't allow the beaming together of | |
1905 | * groups where some have stemto == CS_ABOVE and others have CS_BELOW. | |
1906 | * Theoretically a group with all CSS notes could affect the position | |
1907 | * of the beam regardless of whether its CSS notes are stemside or not; | |
1908 | * but we will pretend that it can't. We'll fake things out in | |
1909 | * setbeam(). This way, we can handle beaming and set the beam | |
1910 | * position and good group boundaries on the beamside during the | |
1911 | * CSSpss==NO pass. Then the placement of "stuff" on that side will | |
1912 | * be better. | |
1913 | */ | |
1914 | /* if we're not at the start of the beamed set, go back to there */ | |
1915 | while (gs_p->beamloc != STARTITEM) { | |
1916 | gs_p = prevsimilar(gs_p); | |
1917 | } | |
1918 | /* check each member to see if any have stemside CSS */ | |
1919 | while (gs_p != 0) { | |
1920 | if (STEMSIDE_CSS(gs_p)) { | |
1921 | return (YES); | |
1922 | } | |
1923 | if (gs_p->beamloc == ENDITEM) { | |
1924 | break; | |
1925 | } | |
1926 | gs_p = nextsimilar(gs_p); | |
1927 | } | |
1928 | return (NO); | |
1929 | } | |
1930 | \f | |
1931 | /* | |
1932 | * Name: css_affects_tieslurbend() | |
1933 | * | |
1934 | * Abstract: Do CSS notes (if any) affect the position of this tie/slur/bend? | |
1935 | * | |
1936 | * Returns: YES or NO | |
1937 | * | |
1938 | * Description: This function decides whether the given tie, slur, or bend is | |
1939 | * affected by CSS notes in any of the groups it covers. | |
1940 | */ | |
1941 | ||
1942 | int | |
1943 | css_affects_tieslurbend(stuff_p, mll_p) | |
1944 | ||
1945 | struct STUFF *stuff_p; /* the tie, slur, or bend */ | |
1946 | struct MAINLL *mll_p; /* MLL item where this tie/slur/bend starts */ | |
1947 | ||
1948 | { | |
1949 | struct GRPSYL *sg_p; /* starting group of the tie/slur/bend */ | |
1950 | struct GRPSYL *eg_p; /* starting group of the tie/slur/bend */ | |
1951 | struct NOTE *snote_p; /* starting note of the tie/slur/bend */ | |
1952 | struct NOTE *enote_p; /* ending note of the tie/slur/bend */ | |
1953 | int idx; /* index of note in the group */ | |
1954 | ||
1955 | ||
1956 | /* if not cross staff stemming, don't waste time checking */ | |
1957 | if (CSSused == NO) { | |
1958 | return (NO); | |
1959 | } | |
1960 | ||
1961 | /* second half (after crossing scorefeed); was handled by first half */ | |
1962 | if (stuff_p->carryin == YES) { | |
1963 | return (NO); | |
1964 | } | |
1965 | ||
1966 | sg_p = stuff_p->beggrp_p; | |
1967 | snote_p = stuff_p->begnote_p; | |
1968 | ||
1969 | /* find the index of the note in the group */ | |
1970 | for (idx = 0; idx < sg_p->nnotes; idx++) { | |
1971 | if (&sg_p->notelist[idx] == snote_p) { | |
1972 | break; | |
1973 | } | |
1974 | } | |
1975 | if (idx == sg_p->nnotes) { | |
1976 | pfatal("can't find tied/slurred/bent note in group"); | |
1977 | } | |
1978 | ||
1979 | /* if this starting note is CSS, return YES */ | |
1980 | if (IS_CSS_NOTE(sg_p, idx)) { | |
1981 | return (YES); | |
1982 | } | |
1983 | ||
1984 | /* | |
1985 | * Find the end note of the tie/slur/bend. If none, we don't care | |
1986 | * about the end note. | |
1987 | */ | |
1988 | eg_p = nextgrpsyl(sg_p, &mll_p); | |
1989 | if (eg_p == 0) { | |
1990 | return (NO); | |
1991 | } | |
1992 | ||
1993 | /* find the note tied/slurred/bent to */ | |
1994 | if (stuff_p->curveno == -1) { /* this is a tie */ | |
1995 | enote_p = find_matching_note(eg_p, snote_p->letter, | |
1996 | snote_p->octave, (char *)0); | |
1997 | } else { /* this is a slur or bend */ | |
1998 | enote_p = find_matching_note(eg_p, | |
1999 | snote_p->slurtolist[stuff_p->curveno].letter, | |
2000 | snote_p->slurtolist[stuff_p->curveno].octave, | |
2001 | (char *)0); | |
2002 | } | |
2003 | ||
2004 | if (enote_p == 0) { | |
2005 | return (NO); | |
2006 | } | |
2007 | ||
2008 | /* find the index of the note in the group */ | |
2009 | for (idx = 0; idx < eg_p->nnotes; idx++) { | |
2010 | if (&eg_p->notelist[idx] == enote_p) { | |
2011 | break; | |
2012 | } | |
2013 | } | |
2014 | if (idx == eg_p->nnotes) { | |
2015 | pfatal("can't find tied/slurred/bent-to note in group"); | |
2016 | } | |
2017 | ||
2018 | /* if this ending note is CSS, return YES */ | |
2019 | if (IS_CSS_NOTE(eg_p, idx)) { | |
2020 | return (YES); | |
2021 | } | |
2022 | ||
2023 | return (NO); | |
2024 | } | |
2025 | /* | |
2026 | * Name: css_affects_phrase() | |
2027 | * | |
2028 | * Abstract: Do CSS notes (if any) affect the position of this phrase mark? | |
2029 | * | |
2030 | * Returns: YES or NO | |
2031 | * | |
2032 | * Description: This function decides whether the given phrase mark is | |
2033 | * affected by CSS notes in any of the groups it covers. | |
2034 | */ | |
2035 | ||
2036 | int | |
2037 | css_affects_phrase(stuff_p, mll_p) | |
2038 | ||
2039 | struct STUFF *stuff_p; /* the phrase */ | |
2040 | struct MAINLL *mll_p; /* for the group at start of phrase */ | |
2041 | ||
2042 | { | |
2043 | struct GRPSYL *gs_p; /* point at end group covered by phrase */ | |
2044 | int place; /* PL_ABOVE or PL_BELOW */ | |
2045 | int staffno; /* staff number */ | |
2046 | int vidx; /* voice index */ | |
2047 | ||
2048 | ||
2049 | place = stuff_p->place; | |
2050 | gs_p = stuff_p->beggrp_p; | |
2051 | staffno = gs_p->staffno; | |
2052 | vidx = gs_p->vno - 1; | |
2053 | ||
2054 | /* loop once for each group covered by the phrase */ | |
2055 | while (gs_p != 0) { | |
2056 | /* return YES right away if we find an affected group */ | |
2057 | switch (gs_p->stemto) { | |
2058 | case CS_SAME: | |
2059 | break; | |
2060 | case CS_ABOVE: | |
2061 | if (place == PL_ABOVE || NNN(gs_p) == 0) { | |
2062 | return (YES); | |
2063 | } | |
2064 | break; | |
2065 | case CS_BELOW: | |
2066 | if (place == PL_BELOW || NNN(gs_p) == 0) { | |
2067 | return (YES); | |
2068 | } | |
2069 | break; | |
2070 | } | |
2071 | ||
2072 | /* if we've seen the last group, we are done */ | |
2073 | if (gs_p == stuff_p->endgrp_p) { | |
2074 | break; | |
2075 | } | |
2076 | ||
2077 | /* find the next note group in this voice */ | |
2078 | do { | |
2079 | gs_p = gs_p->next; | |
2080 | } while (gs_p != 0 && gs_p->grpcont != GC_NOTES); | |
2081 | ||
2082 | /* if we hit the end of the measure, look for next measure */ | |
2083 | if (gs_p == 0) { | |
2084 | /* find the bar line */ | |
2085 | while (mll_p != 0 && mll_p->str != S_BAR) { | |
2086 | mll_p = mll_p->next; | |
2087 | } | |
2088 | /* find the matching staff in the next measure */ | |
2089 | while (mll_p != 0 && ! (mll_p->str == S_STAFF && | |
2090 | mll_p->u.staff_p->staffno == staffno)) { | |
2091 | mll_p = mll_p->next; | |
2092 | } | |
2093 | /* defensive check, should not happen */ | |
2094 | if (mll_p == 0) { | |
2095 | break; | |
2096 | } | |
2097 | /* point at the first group in new measure */ | |
2098 | gs_p = mll_p->u.staff_p->groups_p[vidx]; | |
2099 | } | |
2100 | } | |
2101 | ||
2102 | return (NO); | |
2103 | } | |
2104 | \f | |
2105 | /* | |
2106 | * Name: nextsimilar() | |
2107 | * | |
2108 | * Abstract: Return next group in a GRPSYL list that is "like" the current. | |
2109 | * | |
2110 | * Returns: pointer to GRPSYL of next desired group, 0 if none | |
2111 | * | |
2112 | * Description: This function loop down the GRPSYL linked list from the given | |
2113 | * starting point. It returns the next GRPSYL in the list that has | |
2114 | * the same grpcont and grpvalue as the given one, or 0 if none. | |
2115 | */ | |
2116 | ||
2117 | struct GRPSYL * | |
2118 | nextsimilar(gs_p) | |
2119 | ||
2120 | struct GRPSYL *gs_p; /* current group */ | |
2121 | ||
2122 | { | |
2123 | int curvalue; | |
2124 | int curcont; | |
2125 | ||
2126 | curvalue = gs_p->grpvalue; | |
2127 | curcont = gs_p->grpcont; | |
2128 | gs_p = gs_p->next; | |
2129 | while (gs_p != 0 && | |
2130 | (gs_p->grpvalue != curvalue || gs_p->grpcont != curcont)) { | |
2131 | gs_p = gs_p->next; | |
2132 | } | |
2133 | return (gs_p); | |
2134 | } | |
2135 | \f | |
2136 | /* | |
2137 | * Name: prevsimilar() | |
2138 | * | |
2139 | * Abstract: Return prev group in a GRPSYL list that is "like" the current. | |
2140 | * | |
2141 | * Returns: pointer to GRPSYL of previous desired group, 0 if none | |
2142 | * | |
2143 | * Description: This function loops down the GRPSYL linked list from the given | |
2144 | * starting point. It returns the prev GRPSYL in the list that has | |
2145 | * the same grpcont and grpvalue as the given one, or 0 if none. | |
2146 | */ | |
2147 | ||
2148 | struct GRPSYL * | |
2149 | prevsimilar(gs_p) | |
2150 | ||
2151 | struct GRPSYL *gs_p; /* current group */ | |
2152 | ||
2153 | { | |
2154 | int curvalue; | |
2155 | int curcont; | |
2156 | ||
2157 | curvalue = gs_p->grpvalue; | |
2158 | curcont = gs_p->grpcont; | |
2159 | gs_p = gs_p->prev; | |
2160 | while (gs_p != 0 && | |
2161 | (gs_p->grpvalue != curvalue || gs_p->grpcont != curcont)) { | |
2162 | gs_p = gs_p->prev; | |
2163 | } | |
2164 | return (gs_p); | |
2165 | } | |
2166 | \f | |
2167 | /* | |
2168 | * Name: gs2ch() | |
2169 | * | |
2170 | * Abstract: Given a GRPSYL and its staff's MLL, find chord for that time. | |
2171 | * | |
2172 | * Returns: pointer to CHORD structure | |
2173 | * | |
2174 | * Description: This function is given a GRPSYL and the MLL structure for the | |
2175 | * GRPSYL's staff. It finds the CHORD structure that heads the | |
2176 | * list of GRPSYLs occurring at that time in that measure. Note | |
2177 | * that if the given GRPSYL is grace, it won't actually occur in | |
2178 | * that linked list of GRPSYLs; but in that case the following | |
2179 | * non-grace GRPSYL will. | |
2180 | */ | |
2181 | ||
2182 | struct CHORD * | |
2183 | gs2ch(mll_p, gs_p) | |
2184 | ||
2185 | struct MAINLL *mll_p; /* the MLL for the given GRPSYL */ | |
2186 | struct GRPSYL *gs_p; /* the given GRPSYL */ | |
2187 | ||
2188 | { | |
2189 | struct CHORD *ch_p; /* loop through chords */ | |
2190 | struct GRPSYL *gs2_p; /* point along a GRPSYL list */ | |
2191 | RATIONAL time; /* time offset where our group is */ | |
2192 | ||
2193 | ||
2194 | /* find chord headcell for this measure */ | |
2195 | while (mll_p != 0 && mll_p->str == S_STAFF) { | |
2196 | mll_p = mll_p->prev; | |
2197 | } | |
2198 | if (mll_p == 0 || mll_p->str != S_CHHEAD) { | |
2199 | pfatal("missing chord head cell in gs2ch"); | |
2200 | } | |
2201 | ||
2202 | /* find time offset of our group by summing all previous groups */ | |
2203 | time = Zero; | |
2204 | for (gs2_p = gs_p->prev; gs2_p != 0; gs2_p = gs2_p->prev) { | |
2205 | time = radd(time, gs2_p->fulltime); | |
2206 | } | |
2207 | ||
2208 | /* | |
2209 | * Find the chord that contains our group (or, if we are a grace group, | |
2210 | * the following normal group). | |
2211 | */ | |
2212 | for (ch_p = mll_p->u.chhead_p->ch_p; | |
2213 | ch_p != 0 && NE(ch_p->starttime, time); | |
2214 | ch_p = ch_p->ch_p) { | |
2215 | ; | |
2216 | } | |
2217 | if (ch_p == 0) { | |
2218 | pfatal("can't find chord in gs2ch"); | |
2219 | } | |
2220 | ||
2221 | return (ch_p); | |
2222 | } | |
2223 | \f | |
2224 | /* | |
2225 | * Name: stemroom() | |
2226 | * | |
2227 | * Abstract: Try to find how much room a "wrong way" stem needs. | |
2228 | * | |
2229 | * Returns: The room needed, measured in stepsizes. | |
2230 | * | |
2231 | * Description: This function is given a nongrace note group whose stem has | |
2232 | * been forced the wrong way (down for the top group or up for the | |
2233 | * bottom group) despite the other voice being nonspace. It tries | |
2234 | * to find how long the stem will be so that we can decide whether | |
2235 | * the groups need to be horizontally offset. It works well for | |
2236 | * nonbeamed groups, but for beamed groups it can only guess. It | |
2237 | * is to be used in places where we need to know the stem length | |
2238 | * (to the extent possible) even though beamstem.c hasn't run yet. | |
2239 | * | |
2240 | * WARNING: This code is similar to the nongrace section of | |
2241 | * proclist() in beamstem.c. If you change one, you probably | |
2242 | * will need to change the other. | |
2243 | */ | |
2244 | ||
2245 | double | |
2246 | stemroom(gs_p) | |
2247 | ||
2248 | struct GRPSYL *gs_p; /* the group in question */ | |
2249 | ||
2250 | { | |
2251 | float room; /* the answer, in stepsizes */ | |
2252 | int bf; /* number of beams/flags */ | |
2253 | ||
2254 | ||
2255 | /* | |
2256 | * If user specified stem length, use that. | |
2257 | */ | |
2258 | if (IS_STEMLEN_KNOWN(gs_p->stemlen)) { | |
2259 | return (gs_p->stemlen / STEPSIZE); | |
2260 | } | |
2261 | ||
2262 | /* | |
2263 | * If this group is part of a beamed set, there is no way to know how | |
2264 | * long the stem will be, since the beaming hasn't been done yet, and | |
2265 | * can't be done until all horizontal placement has been done. So | |
2266 | * return the default stem length and hope for the best. | |
2267 | */ | |
2268 | if (gs_p->beamloc != NOITEM) { | |
2269 | return (DEFSTEMLEN); | |
2270 | } | |
2271 | ||
2272 | /* | |
2273 | * Only half notes and shorter have stems, but whole and double | |
2274 | * whole notes still need to have a pseudo stem length set if | |
2275 | * alternation beams are to be drawn between two neighboring | |
2276 | * groups, or the group has slashes. | |
2277 | */ | |
2278 | if (gs_p->basictime <= 1 && gs_p->slash_alt == 0) { | |
2279 | /* no (pseudo)stem */ | |
2280 | return (0.0); | |
2281 | } | |
2282 | ||
2283 | /* find default stemlen for this voice */ | |
2284 | room = vvpath(gs_p->staffno, gs_p->vno, STEMLEN)->stemlen; | |
2285 | if (room == 0.0) { | |
2286 | return (0.0); | |
2287 | } | |
2288 | ||
2289 | /* if small notes, reduce this default */ | |
2290 | room *= (allsmall(gs_p, gs_p) == YES ? SM_STEMFACTOR : 1.0); | |
2291 | ||
2292 | /* add more, if needed, for flags/beams/slashes/alternations */ | |
2293 | if (gs_p->basictime >= 8) { | |
2294 | bf = drmo(gs_p->basictime) - 2; /* no. of beams/flags*/ | |
2295 | } else { | |
2296 | bf = 0; /* none on quarter or longer */ | |
2297 | } | |
2298 | bf += abs(gs_p->slash_alt); /* slashes or alternations */ | |
2299 | if (gs_p->slash_alt > 0 && gs_p->basictime >= 16) { | |
2300 | bf++; /* slashes need an extra one if 16, 32, ... */ | |
2301 | } | |
2302 | if (bf > 2) { | |
2303 | room += (bf - 2) * FLAGSEP / STEPSIZE; | |
2304 | } | |
2305 | ||
2306 | /* | |
2307 | * If the note may have flag(s), is stem up, and has dot(s), we must | |
2308 | * prevent the flag(s) from hitting the dot(s), by lengthening the stem. | |
2309 | */ | |
2310 | if (gs_p->basictime >= 8 && gs_p->stemdir == UP && gs_p->dots != 0) { | |
2311 | if (gs_p->notelist[0].stepsup % 2 == 0) { | |
2312 | /* note is on a line */ | |
2313 | if (gs_p->basictime == 8) { | |
2314 | room += 1.0; | |
2315 | } else { | |
2316 | room += 2.0; | |
2317 | } | |
2318 | } else { | |
2319 | /* note is on a space */ | |
2320 | if (gs_p->basictime > 8) { | |
2321 | room += 1.0; | |
2322 | } | |
2323 | } | |
2324 | } | |
2325 | ||
2326 | return (room); | |
2327 | } |